From smakeig at ucsd.edu  Mon Jan  4 12:52:36 2010
From: smakeig at ucsd.edu (Scott Makeig)
Date: Mon, 4 Jan 2010 12:52:36 -0800
Subject: [Eeglablist] EEGLAB website up again ...
Message-ID: <9e09b8f01001041252v67a738ecw6e9346fcca30865c@mail.gmail.com>

All -

We apologize that, for an unfortunate combination of reasons beyond our
control, the EEGLAB and SCCN websites were down for nearly two weeks during
the recent UCSD holiday shutdown. It is now up again and we do not
anticipate further problems.

Our best wishes for a Happy New Year,

Scott, Arnaud, and Dev - the EEGLAB team @ SCCN

-- 
Scott Makeig, Research Scientist and Director, Swartz Center for
Computational Neuroscience, Institute for Neural Computation, University of
California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100104/ee874f49/attachment.html>

From dgroppe at cogsci.ucsd.edu  Tue Jan  5 08:12:51 2010
From: dgroppe at cogsci.ucsd.edu (David Groppe)
Date: Tue, 5 Jan 2010 08:12:51 -0800
Subject: [Eeglablist] A quick question about baseline correction after
	ICA
In-Reply-To: <9e09b8f00912231002g5db910e0uaaa280ab185aa8ae@mail.gmail.com>
References: <43e75ab40912220159h1c110073ndbbb57136cb8b96f@mail.gmail.com>
	<9e09b8f00912231002g5db910e0uaaa280ab185aa8ae@mail.gmail.com>
Message-ID: <3d4c78cd1001050812w667acb03ne515964fa8cbec14@mail.gmail.com>

Hi Zhu,
   The results Scott mentions are in the following paper:
Groppe, D.M., Makeig, S., &  Kutas, M. (2009) Identifying reliable
independent components via split-half comparisons. NeuroImage, 45
pp.1199-1211.
http://www.cogsci.ucsd.edu/~dgroppe/PUBLICATIONS/Groppe2009.pdf

cheers,
  -David


On Wed, Dec 23, 2009 at 10:02 AM, Scott Makeig <smakeig at gmail.com> wrote:
> Zhu -
>
> David Groppe pointed out that running ICA after baseline-correcting
> individual data epochs essentially adds a new (baseline mean) map to each
> trial, one that depends on the length of the epoch as well as on the
> underlying data sources. He found that he got better ICA decompositions when
> he (1) applied high-pass filtering (as however appropriate) to the
> continuous data, (2) extracted event-related epochs of interest, (3)
> performed ICA decomposition on the concatenated epochs, and then (4) applied
> baseline correction to the epochs. This seems theoretically sound -- I
> suggest trying this approach.
>
> Scott Makeig
>
> ps. Below you don't say exactly what you mean by A and B ...
>
>
> On Tue, Dec 22, 2009 at 1:59 AM, zhu zhu <zhuzude at gmail.com> wrote:
>>
>> Dear All,
>> I have a puzzle when extract epochs?from ICA-back projected data. Let'
>> say, I selected one component after ICA, when I extract epochs for a
>> specified marker (condition), should I use baseline correction (default =
>> -200 0)? The reason I ask this question is that I the data was baseline
>> corrected before ICA, and sometimes the baseline correction during
>> extracting epochs will change the relative amplidute( i.e. A > B without
>> baseline correction, but A < B after using baseline correction). Any comment
>> is highly appreciated!
>> Thank you!
>> Best regards,
>> Zude
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>
>
>
> --
> Scott Makeig, Research Scientist and Director, Swartz Center for
> Computational Neuroscience, Institute for Neural Computation, University of
> California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



From d.arisi at ospedale.cremona.it  Tue Jan  5 09:55:14 2010
From: d.arisi at ospedale.cremona.it (d.arisi at ospedale.cremona.it)
Date: Tue, 5 Jan 2010 18:55:14 +0100 (CET)
Subject: [Eeglablist] head models for children
Message-ID: <2866.87.8.100.186.1262714114.squirrel@webmail.ospedale.cremona.it>

We are planning to use low resolution electromagnetic tomography (Loreta)
in different studies in children (age 6-12); a few works in literature
exist about Loreta use in children, but we cannot address the question
relative to different head size between adults and children; the software
uses a database of adult MRIs to map cortical electrical activity; is it
possible to use the same database for children or a new specific database
is necessary?

thank you in advance for every suggestion

Daniele Arisi
Child Neurology and Psychiatry Unit
Cremona Hospital (Italy)



From a0302186 at unet.univie.ac.at  Mon Jan  4 16:20:25 2010
From: a0302186 at unet.univie.ac.at (Bernhard Meyer)
Date: Tue, 05 Jan 2010 01:20:25 +0100
Subject: [Eeglablist] Interpolation and ICA
Message-ID: <4B4285C9.9030907@unet.univie.ac.at>

As I need to use the algorithm eeg_interp(EEG, Channel, 'spherical') for
channel-interpolation in combination with ICA I found a possible
contradiction regarding the question what to apply first, ICA or
interpolation.

Arno Delorme wrote in 2006 (Archive): 'Note that it is not recommended
to perform interpolation before running ICA. The attached function will
take care of you ICA information when you perform interpolation.'

Andreas Widmann wrote in 2009 (Archive): 'However, be aware that a
possibly existing ICA weight matrix will be invalid as only part of the
dataset is interpolated!'

In my opinion I would run ICA after interpolation. On the one hand I
am afraid to have invalid ICA-data in the end, especially if I
interpolate just a part of the epochs of the channel. On the other hand
I want to have all channels as clean as possible, otherwise I could lose
degrees of freedom for useless data.
I hope the EEGLAB-community can clarify that step.

Best,
Bernhard Meyer



From arno at ucsd.edu  Tue Jan  5 13:12:22 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Tue, 5 Jan 2010 13:12:22 -0800
Subject: [Eeglablist] Interpolation and ICA
In-Reply-To: <4B4285C9.9030907@unet.univie.ac.at>
References: <4B4285C9.9030907@unet.univie.ac.at>
Message-ID: <4175DA81-96DA-44CD-B2D2-07A8746364DF@ucsd.edu>

Dear Bernhard,


> As I need to use the algorithm eeg_interp(EEG, Channel, 'spherical')  
> for
> channel-interpolation in combination with ICA I found a possible
> contradiction regarding the question what to apply first, ICA or
> interpolation.


Apply ICA first. Once you have run ICA you may (1) remove artifactual  
ICA components if this is what you want to do (2) then interpolate  
channels.

If you interpolate channels first, the rank of the data might not be  
equal to the number of channel. Even if it is (because spherical  
interpolation is not linear interpolation), you have introduced some  
non-linearity in the data (in the sense that an underlying source  
might not project linearly to scalp channels any more) and this is  
going to make the ICA solution more noisy.

Best,

Arno



From arno at ucsd.edu  Tue Jan  5 13:47:48 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Tue, 5 Jan 2010 13:47:48 -0800
Subject: [Eeglablist] Interpolation and ICA
In-Reply-To: <1372.159.71.184.228.1262727333.squirrel@sccn.ucsd.edu>
References: <4B4285C9.9030907@unet.univie.ac.at>
	<4175DA81-96DA-44CD-B2D2-07A8746364DF@ucsd.edu>
	<1372.159.71.184.228.1262727333.squirrel@sccn.ucsd.edu>
Message-ID: <3E208515-0614-4657-BBD3-0E841FD24412@ucsd.edu>

The ICA matrix is going to be unchanged by interpolation.
However, after interpolation, you cannot remove ICA artifact  
components from your data (otherwise only the artifactual components  
are only going to be removed only from the channels that were used to  
compute ICA). By contrast, if you remove components from your data  
prior to interpolation, this is fine.
In other words, ICA components are not interpolated,

Arno

On Jan 5, 2010, at 1:35 PM, Julie Onton wrote:

> does EEGLAB automatically fill in the EEG.weights/sphere with values  
> for the
> newly-interpolated channel?  I think that was one of the concerns that
> Bernhard raised.
>
> J
>
> -- 
> Julie Onton, PhD
> http://sccn.ucsd.edu/~julie
>
>> Dear Bernhard,
>>
>>
>>> As I need to use the algorithm eeg_interp(EEG, Channel, 'spherical')
>>> for
>>> channel-interpolation in combination with ICA I found a possible
>>> contradiction regarding the question what to apply first, ICA or
>>> interpolation.
>>
>>
>> Apply ICA first. Once you have run ICA you may (1) remove artifactual
>> ICA components if this is what you want to do (2) then interpolate
>> channels.
>>
>> If you interpolate channels first, the rank of the data might not be
>> equal to the number of channel. Even if it is (because spherical
>> interpolation is not linear interpolation), you have introduced some
>> non-linearity in the data (in the sense that an underlying source
>> might not project linearly to scalp channels any more) and this is
>> going to make the ICA solution more noisy.
>>
>> Best,
>>
>> Arno
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>>
>



From demiral.007 at googlemail.com  Wed Jan  6 03:17:39 2010
From: demiral.007 at googlemail.com (Baris Demiral)
Date: Wed, 6 Jan 2010 11:17:39 +0000
Subject: [Eeglablist] supplies related
Message-ID: <f423beba1001060317j5e7529c7n6cd9b771f8472b07@mail.gmail.com>

Dear eeglab members,

We are using BioSemi system and need to buy some gel and double sided
adhesives (they are very convenient I should admit).
Do you have any recommendations, cheaper, better solutions?

Also, friends from the 2009 summer school, could you please e-mail me? Also,
some people in the workshop suggested planning some small scale
activities/workshops in UK on eeglab applications. I know it is a bit late I
am sending this message, but I experienced some problems with my account
which caused this delay.

Cheers,
Baris

-- 
SB Demiral, PhD.
Department of Psychology
7 George Square
The University of Edinburgh
Edinburgh, EH8 9JZ
UK
Phone: +44 (0131) 6503063
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100106/7c4300bd/attachment.html>

From widmann at uni-leipzig.de  Wed Jan  6 04:38:59 2010
From: widmann at uni-leipzig.de (Andreas Widmann)
Date: Wed, 6 Jan 2010 13:38:59 +0100
Subject: [Eeglablist] Interpolation and ICA
In-Reply-To: <4B4285C9.9030907@unet.univie.ac.at>
References: <4B4285C9.9030907@unet.univie.ac.at>
Message-ID: <9EFF5A1B-6328-4454-9DAB-EFA69A0F4F4E@uni-leipzig.de>

Dear Bernhard and Arno,

imho, it is also relevant (a) why you want to interpolate/the type of artifact and (b) whether you want to interpolate the channel for the whole dataset or only parts/trials/epochs.

(a) E.g. for unplugged or broken active electrodes containing noise and no signal I would prefer removing the channel and inserting an interpolated channel after ICA. Arno, would it be safe to feed (partially) noise only channels into ICA?

(b) If you interpolate ONLY PART of the dataset/epochs/trials AFTER ICA computation, imho ICA solution is no longer strictly valid for the dataset. Did I misunderstand something here?

Two comments:
Using the sphspline plugin instead of eeg_interp you should be able to interpolate the channel for ICA components too.

Spherical spline interpolation does not necessarily introduce non-linearity (the name is misleading). E.g., the back-projected dataset with a channel interpolated for all components is identical to the dataset with the channel interpolated.

Best,
Andreas

Am 05.01.2010 um 01:20 schrieb Bernhard Meyer:

> As I need to use the algorithm eeg_interp(EEG, Channel, 'spherical') for
> channel-interpolation in combination with ICA I found a possible
> contradiction regarding the question what to apply first, ICA or
> interpolation.
> 
> Arno Delorme wrote in 2006 (Archive): 'Note that it is not recommended
> to perform interpolation before running ICA. The attached function will
> take care of you ICA information when you perform interpolation.'
> 
> Andreas Widmann wrote in 2009 (Archive): 'However, be aware that a
> possibly existing ICA weight matrix will be invalid as only part of the
> dataset is interpolated!'
> 
> In my opinion I would run ICA after interpolation. On the one hand I
> am afraid to have invalid ICA-data in the end, especially if I
> interpolate just a part of the epochs of the channel. On the other hand
> I want to have all channels as clean as possible, otherwise I could lose
> degrees of freedom for useless data.
> I hope the EEGLAB-community can clarify that step.
> 
> Best,
> Bernhard Meyer
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
> 




From zhuzude at gmail.com  Tue Jan  5 23:48:38 2010
From: zhuzude at gmail.com (zhu zhu)
Date: Wed, 6 Jan 2010 08:48:38 +0100
Subject: [Eeglablist] A quick question about baseline correction after
	ICA
In-Reply-To: <3d4c78cd1001050812w667acb03ne515964fa8cbec14@mail.gmail.com>
References: <43e75ab40912220159h1c110073ndbbb57136cb8b96f@mail.gmail.com>
	<9e09b8f00912231002g5db910e0uaaa280ab185aa8ae@mail.gmail.com>
	<3d4c78cd1001050812w667acb03ne515964fa8cbec14@mail.gmail.com>
Message-ID: <43e75ab41001052348u40fdfde7r6b276b84dfb03428@mail.gmail.com>

Dear Scott and David,

Thank you so much!

Best regards,
Zude

2010/1/5 David Groppe <dgroppe at cogsci.ucsd.edu>

> Hi Zhu,
>   The results Scott mentions are in the following paper:
> Groppe, D.M., Makeig, S., &  Kutas, M. (2009) Identifying reliable
> independent components via split-half comparisons. NeuroImage, 45
> pp.1199-1211.
> http://www.cogsci.ucsd.edu/~dgroppe/PUBLICATIONS/Groppe2009.pdf
>
> cheers,
>  -David
>
>
> On Wed, Dec 23, 2009 at 10:02 AM, Scott Makeig <smakeig at gmail.com> wrote:
> > Zhu -
> >
> > David Groppe pointed out that running ICA after baseline-correcting
> > individual data epochs essentially adds a new (baseline mean) map to each
> > trial, one that depends on the length of the epoch as well as on the
> > underlying data sources. He found that he got better ICA decompositions
> when
> > he (1) applied high-pass filtering (as however appropriate) to the
> > continuous data, (2) extracted event-related epochs of interest, (3)
> > performed ICA decomposition on the concatenated epochs, and then (4)
> applied
> > baseline correction to the epochs. This seems theoretically sound -- I
> > suggest trying this approach.
> >
> > Scott Makeig
> >
> > ps. Below you don't say exactly what you mean by A and B ...
> >
> >
> > On Tue, Dec 22, 2009 at 1:59 AM, zhu zhu <zhuzude at gmail.com> wrote:
> >>
> >> Dear All,
> >> I have a puzzle when extract epochs from ICA-back projected data. Let'
> >> say, I selected one component after ICA, when I extract epochs for a
> >> specified marker (condition), should I use baseline correction (default
> =
> >> -200 0)? The reason I ask this question is that I the data was baseline
> >> corrected before ICA, and sometimes the baseline correction during
> >> extracting epochs will change the relative amplidute( i.e. A > B without
> >> baseline correction, but A < B after using baseline correction). Any
> comment
> >> is highly appreciated!
> >> Thank you!
> >> Best regards,
> >> Zude
> >>
> >> _______________________________________________
> >> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> >> To unsubscribe, send an empty email to
> >> eeglablist-unsubscribe at sccn.ucsd.edu
> >> For digest mode, send an email with the subject "set digest mime" to
> >> eeglablist-request at sccn.ucsd.edu
> >
> >
> >
> > --
> > Scott Makeig, Research Scientist and Director, Swartz Center for
> > Computational Neuroscience, Institute for Neural Computation, University
> of
> > California San Diego, La Jolla CA 92093-0961,
> http://sccn.ucsd.edu/~scott
> >
> > _______________________________________________
> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> > To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> > For digest mode, send an email with the subject "set digest mime" to
> > eeglablist-request at sccn.ucsd.edu
> >
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100106/b6c0f0c4/attachment.html>

From Elmar.Lang at biologie.uni-regensburg.de  Thu Jan  7 07:42:29 2010
From: Elmar.Lang at biologie.uni-regensburg.de (Elmar Lang)
Date: Thu, 07 Jan 2010 16:42:29 +0100
Subject: [Eeglablist] (no subject)
Message-ID: <4B460EF5020000D80000ADF0@gwsmtp1.uni-regensburg.de>

Dear colleagues this is just to bring to your attention a special session entitled Empirical Mode Decomposition which will be held during WCCI 2010 in Barcelona. Please, spread around this information to anyone interested to actively participate.

Best regards
Elmar Lang



Prof. Dr. E. W. Lang
Computationl Intelligence and Machine Learning - CIML
Institute of Biophysics
University of Regensburg
D-93040 Regensburg
Germany

email: elmar.lang at biologie.uni-regensburg.de
http://www-aglang.uni-regensburg.de
phone: +49-941-943-2599
fax: +49-941-943-2479
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100107/4db27c39/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Special_Session_EMD.pdf
Type: application/pdf
Size: 67505 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100107/4db27c39/attachment.pdf>

From g.rousselet at psy.gla.ac.uk  Thu Jan  7 11:02:24 2010
From: g.rousselet at psy.gla.ac.uk (Guillaume Rousselet)
Date: Thu, 7 Jan 2010 19:02:24 +0000
Subject: [Eeglablist] supplies related
In-Reply-To: <f423beba1001060317j5e7529c7n6cd9b771f8472b07@mail.gmail.com>
References: <f423beba1001060317j5e7529c7n6cd9b771f8472b07@mail.gmail.com>
Message-ID: <83D4EF8B-079F-4A67-B881-7DE5D88A800A>

You can get cheaper gel, needles and syringes from MedCat: <http://www.medcat.nl/supplies/En/ 
 >


On 6 Jan 2010, at 11:17, Baris Demiral wrote:

>
> Dear eeglab members,
>
> We are using BioSemi system and need to buy some gel and double  
> sided adhesives (they are very convenient I should admit).
> Do you have any recommendations, cheaper, better solutions?
>
> Also, friends from the 2009 summer school, could you please e-mail  
> me? Also, some people in the workshop suggested planning some small  
> scale activities/workshops in UK on eeglab applications. I know it  
> is a bit late I am sending this message, but I experienced some  
> problems with my account which caused this delay.
>
> Cheers,
> Baris
>
> -- 
> SB Demiral, PhD.
> Department of Psychology
> 7 George Square
> The University of Edinburgh
> Edinburgh, EH8 9JZ
> UK
> Phone: +44 (0131) 6503063
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu





************************************************************************************
Guillaume A. Rousselet, Ph.D.

Lecturer

Centre for Cognitive Neuroimaging (CCNi)
Department of Psychology
Faculty of Information & Mathematical Sciences (FIMS)
University of Glasgow
58 Hillhead Street
Glasgow, UK
G12 8QB

The University of Glasgow, charity number SC004401

http://web.me.com/rousseg/GARs_website/

Email: g.rousselet at psy.gla.ac.uk
Fax. +44 (0)141 330 4606
Tel. +44 (0)141 330 6652
Cell +44 (0)791 779 7833


?640,000 bytes of memory ought to be enough for anybody.?
	
	Bill Gates, 1981
************************************************************************************




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100107/f46817e5/attachment.html>

From cornelia.mccormick at googlemail.com  Thu Jan  7 05:59:06 2010
From: cornelia.mccormick at googlemail.com (Cornelia McCormick)
Date: Thu, 7 Jan 2010 08:59:06 -0500
Subject: [Eeglablist] Time frequency analysis on a studyset
Message-ID: <46a4cf61001070559r704ada75o11b38ccb2a818a1@mail.gmail.com>

Dear all!

We are interested in hippocampal recordings during a memory task and
collected already data from eight temporal lobe epilepsy patients. I am now
trying to run time frequency analyses with eeglab on these patients. For
every individual, this works fine but when I am creating a studyset, I
cannot click on either the tools or edit button (they are just not
highlighted anymore). I am really stuck here and cannot figure out why it
wouldn't work for more than one subject. Does anybody have some ideas? That
would be really great. Do I have to do ICA before I can access the TFA? The
thing is, I know already exactly which channel I want to analsye so I don't
need a ICA, is that right?

Thanks very much in advance for any response.

Cornelia McCormick
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100107/748bb48e/attachment.html>

From a0302186 at unet.univie.ac.at  Fri Jan  8 07:55:01 2010
From: a0302186 at unet.univie.ac.at (Bernhard Meyer)
Date: Fri, 08 Jan 2010 16:55:01 +0100
Subject: [Eeglablist] Interpolation and ICA
Message-ID: <4B475555.9020307@unet.univie.ac.at>

Dear All,

I appreciate the information I got and want to conclude the common 
recommendation to calculate ICA, remove artifactual ICA components and 
then interpolate channels. Furthermore I want to add following thoughts:

1.) The broken or unplugged channel should be removed before 
ICA-calculation, as Andreas Widmann wrote. Since I combined removing the 
channel and interpolation, it was important for me to have that explicit.

2.) Consequently using the interpolation in pop_precomp() is the last 
chance to remove artifactual-clusters. Maybe it would be worth to add 
that information.

3.) If ICA-components are not interpolated, there is no need for 
interpolation in ICA-exclusive research. E.g. if I don't plot ERPs of 
that channel but I want to identify a component.

4.) To interpolate a part of a bad channel (e.g. during one block of 
epochs with no signal from broken electrode) does not seem to be 
possible up to now. What about the idea (Point b Andreas Widmann) to 
fill that part of the channel with noise or to just remove it before 
ICA-calculation and interpolate that segment after?

It would be kind if you evaluate these points to prevent possible 
misconceptions.

Best,
Bernhard


From andreas.galka at googlemail.com  Fri Jan  8 01:42:44 2010
From: andreas.galka at googlemail.com (Andreas Galka)
Date: Fri, 8 Jan 2010 10:42:44 +0100
Subject: [Eeglablist] PhD position in Kiel, Germany
Message-ID: <7b22104d1001080142h396af161p3ea012b6b0cd993b@mail.gmail.com>

The University of Kiel, Germany is seeking applications for

1 Doctoral Position (salary class TV-L 13 / 2)

within the newly established DFG-funded Collaborative Research Centre (CRC) 855
"Magnetoelectric sensors for biomagnetic signal recording".

Areas of work:
- Participation in the project "improvement of brain source
localisation in epileptology by time series analysis"
- Analysis of electroencephalographic (EEG) and
magnetoencephalographic (MEG) time series data with respect
  to source localisation, physiological signal components and artefacts
- Development and implementation of new algorithms for source analysis
(solution of an inverse problem) and for
  other tasks from the field of time series analysis
- Contributions to the presentation of research results at conferences
and publications
- Opportunity to obtain a doctorate degree in physics, mathematics or
engineering

Profile:
- A Master's or comparable degree in physics, mathematics, statistics
or a closely related discipline (completed
  or pending completion)
- Experience with scientific programming, in particular MATLAB or a
similar language
- Experience with time series analysis would be a plus, but is not required
- Potential for original research
- Possessing initiative and being able to work independently if necessary
- Good communication and teamwork skills
- Fluency in spoken and written English; German would be a plus but is
not required

The place of work is the neuropediatric department of the University
of Kiel, Germany.
The position is limited to 3 years, with the possibility of extension,
and will commence on 1 April 2010, or
as soon as possible thereafter.

Applications should include a letter of motivation, CV, detailed
description of study achievements and work
experience, a list of theses and published scholarly papers, copies of
certificates and contact information of
two referees. Applications will be considered until the position is filled.

Applications and requests for further information should be sent to

Prof. Dr. Michael Siniatchkin
Klinik fuer Neuropaediatrie der Christian-Albrechts-Universit?t zu Kiel
Schwanenweg 20
24105 Kiel
Email: m.siniatchkin at pedneuro.uni-kiel.de
Tel. ++49-431-597-1771



From cornelia.mccormick at googlemail.com  Fri Jan  8 05:57:54 2010
From: cornelia.mccormick at googlemail.com (Cornelia McCormick)
Date: Fri, 8 Jan 2010 08:57:54 -0500
Subject: [Eeglablist] Time frequency analysis on a studyset
Message-ID: <46a4cf61001080557o60011b87o21b8e83fcd82c6d1@mail.gmail.com>

Dear all,

We are interested in hippocampal theta frequency signaling during memory
tasks and collected intracranial data from eight temporal lobe epilepsy
patients. I already analyzed each individual dataset with time frequency
analyses, built in eeglab, however, I cannot run a group analysis. I created
a studyset of all datasets but then I cannot click on the "edit" or "plot"
button anymore (they are just not highlighted anymore). I am really stuck
here and hope that anybody has any idea.

Thanks in advance for any comment.
Cornelia McCormick
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100108/26d7ab7a/attachment.html>

From mklados at med.auth.gr  Fri Jan  8 00:43:42 2010
From: mklados at med.auth.gr (Klados Manousos)
Date: Fri, 8 Jan 2010 10:43:42 +0200
Subject: [Eeglablist] MEDICON 2010 XII MEDITERRANEAN CONFERENCE ON MEDICAL
	AND BIOLOGICAL ENGINEERING AND COMPUTING 2nd CALL
Message-ID: <3b0db4861001080043tf65b84k6c3ffa195c523311@mail.gmail.com>

2ND CALL FOR PAPERS

It is our great pleasure to invite you to participate in the 12th
Mediterranean Conference on Medical and Biological Engineering and Computing
- the MEDICON 2010. After 21 years, the MEDICON conference is coming back to
Greece, this time in a grand resort of Greece, the Porto Carras in
Chalkidiki. The MEDICON conferences are international events of high
scientific standards with long lasting tradition held every third year in
one of the Mediterranean countries under the auspices of the International
Federation for Medical and Biological Engineering. MEDICON 2010 is intended
to provide an international forum for discussing the latest results in the
field of medical and biological engineering and computing. The scientific
program of MEDICON 2010 will consist of invited keynote talks given by
leading scientists in the field, and regular and special track sessions and
workshops that cover a broad array of issues which relate technology and
computing to (bio)medicine. Authors are invited to submit their research
contributions or practical (industrial) experience reports. All papers will
be peer reviewed by at least two referees. The conference provides its
attendees with an opportunity to experience state-of-the-art research and
development in a variety of topics directly and indirectly related to their
own work, as well as an opportunity to come up-to-date on important
technological issues involved in the enhancement of education in biomedical
engineering and related fields.


INTERESTING:


   - *MEDICON 2010 Home* <http://www.medicon2010.org/index.html>
   - 2nd Call for Papers Medicon
2010<http://www.medicon2010.org/2nd_cfp_medicon2010-final.pdf>


-- 
Klados A. Manousos
Research Assistant
Group of Applied Neurosciences
Lab of Medical Informatics, Medical School
Aristotle University of Thessaloniki
Thessaloniki, Greece
_________________________________________________
Tel: +30-2310-999332
Website: http://lomiweb.med.auth.gr/gan/mklados
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100108/7fa6062a/attachment.html>

From thermann at techfak.uni-bielefeld.de  Fri Jan  8 09:25:31 2010
From: thermann at techfak.uni-bielefeld.de (Thomas Hermann)
Date: Fri, 8 Jan 2010 18:25:31 +0100
Subject: [Eeglablist] Open Research Position @ Bielefeld University: EEG
	data analysis for predicting bipolar disorder episodes
Message-ID: <0A5A7949-9A30-4366-A017-CAEDAC010BA6@techfak.uni-bielefeld.de>

== Open Research Position ==
EEG data analysis for predicting bipolar disorder episodes


The research groups Neuroinformatics (Prof. Dr. Helge Ritter)
and Ambient Intelligence (Dr. Thomas Hermann) at Bielefeld
University invite applications for one

                **Postdoctoral Position**

for the development of innovative methods for EEG signal analysis
and classification to be carried out in the context of the European
cooperation project MONARCA. This highly interdisciplinary project
aims to develop advanced technology and data analysis methods for
the realization of an innovative multi-parametric, long-term
monitoring system relevant to the disease of bipolar disorder.
Planned duration of the position will be 32 months. The position
is available immediately. Payment will be according to TVL-13.

The hosting groups have extensive expertise in adaptive methods
for EEG signal analysis and datamining. Being part of the Excellence
Cluster "Cognitive Interaction Technology (CITEC)" we can offer
an exciting and highly interdisciplinary research environment.

We are looking for applicants at the post-doctoral level with
a degree from informatics, electrical engineering, applied mathematics
or physics and a strong interest in brain signal analysis method
development within a medical diagnosis application context.
The ideal applicant should have a significant research background
in at least one of the fields of pattern recognition, time series
analysis, machine learning and datamining and thorough experience
in the implementation of algorithms in C, C++ and/or Matlab.
We will also consider exceptionally qualified applicants at the
university masters or diploma levels.

Bielefeld University is committed to equal opportunity. We strongly
encourage applications from qualified women and persons with  
disabilities.

To apply, please send a letter stating your motivation and your research
interests, a complete CV (preferably in pdf format) and the names and
email addresses of three referees to the attention of

Mrs. Susanne Strunk
sstrunk at techfak.uni-bielefeld.de
Faculty of Technology and
Excellence Center for Cognitive Interaction Technology (CITEC)
Bielefeld University

--------------------------------------------------------------
Helge Ritter
Neuroinformatics Group, CITEC
Bielefeld University, Germany
http://www.techfak.uni-bielefeld.de/ags/ni

Thomas Hermann
Ambient Intelligence Group, CITEC
Bielefeld University, Bielefeld, Germany
http://www.techfak.uni-bielefeld.de/ags/ami
--------------------------------------------------------------





From vsalaiselvam at yahoo.com  Sat Jan  9 03:13:54 2010
From: vsalaiselvam at yahoo.com (salai selvam)
Date: Sat, 9 Jan 2010 03:13:54 -0800 (PST)
Subject: [Eeglablist] eeglablist Digest, Vol 63, Issue 5
In-Reply-To: <mailman.0.1262980802.27546.eeglablist@sccn.ucsd.edu>
Message-ID: <957436.76470.qm@web53101.mail.re2.yahoo.com>

Dear Cornelia, This could be the error due to the format in which the data set is presented to the time-frequency analysis function of the EEGLAB. Please check the help attached to this mail or better the m file of the corresponding function to know in which format the data set must presented to the function. Are you using the latest version of EEGLAB? If not please check out the website for the one: http://sccn.ucsd.edu/eeglab/. Please feel free to contact me if you have still problems through this mail ID. All the best.


With regards V Sallai Selvam Asst Prof, Dept. of ECE, Sriram Engg College, Chennai, India.

--- On Fri, 1/8/10, eeglablist-request at sccn.ucsd.edu <eeglablist-request at sccn.ucsd.edu> wrote:

From: eeglablist-request at sccn.ucsd.edu <eeglablist-request at sccn.ucsd.edu>
Subject: eeglablist Digest, Vol 63, Issue 5
To: eeglablist at sccn.ucsd.edu
Date: Friday, January 8, 2010, 12:00 PM

Send eeglablist mailing list submissions to
??? eeglablist at sccn.ucsd.edu

To subscribe or unsubscribe via the World Wide Web, visit
??? http://sccn.ucsd.edu/mailman/listinfo/eeglablist
or, via email, send a message with subject or body 'help' to
??? eeglablist-request at sccn.ucsd.edu

You can reach the person managing the list at
??? eeglablist-owner at sccn.ucsd.edu

When replying, please edit your Subject line so it is more specific
than "Re: Contents of eeglablist digest..."
Today's Topics:

???1. Time frequency analysis on a studyset (Cornelia McCormick)
_______________________________________________
eeglablist mailing list eeglablist at sccn.ucsd.edu
Eeglablist page: http://www.sccn.ucsd.edu/eeglab/eeglabmail.html
To unsubscribe, send an empty email to eeglablist-unsub at sccn.ucsd.edu
To switch to non-digest mode, send an empty email to eeglablist-nodigest at sccn.ucsd.edu


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100109/def08c15/attachment.html>

From chaitanya1686 at gmail.com  Sun Jan 10 12:58:33 2010
From: chaitanya1686 at gmail.com (chaitanya bhavaraju)
Date: Sun, 10 Jan 2010 14:58:33 -0600
Subject: [Eeglablist] Data analysis using EEG lab
Message-ID: <79cd0c501001101258k64fa5a54o40d9de0817c4da69@mail.gmail.com>

Hi,

I am performing a test to compare the change in the dominant alpha
frequency during two different conditions in the test. For this I did
ICA analysis using the EEGLAB and then I exported the file into a
notepad. The data is so huge (my experiment is for 7 min and sampling
frequency is 500 Hz). From that data I selected a sample (10,000
points ) and I performed FFT. I would like to know if there are any
mistakes in this.
Is there any tutorial available in the web??


thank you,
Chaitanya


From qiweijingwx at 126.com  Mon Jan 11 02:11:53 2010
From: qiweijingwx at 126.com (qiweijingwx)
Date: Mon, 11 Jan 2010 18:11:53 +0800 (CST)
Subject: [Eeglablist] removing artifacts with ICA
Message-ID: <7750938.392501263204713631.JavaMail.coremail@bj126app39.126.com>

Dear All,
I want using ICA to remove noise such as eye movement, muscle movement et al. in 
eeglab. I have two questions about this process. 
First, which seems better, running ICA to remove noise before or after epoch the data?
Second, if I want to remove artifacts before epoching data, but the file after merging is too big, could I remove noise separately in each file, then merge them together?
 
Thanks in advance!
 
best nancy
 
2010/01/11



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100111/45c54d9d/attachment.html>

From bpruce at indiana.edu  Tue Jan 12 08:32:39 2010
From: bpruce at indiana.edu (Pruce,  Benjamin)
Date: Tue, 12 Jan 2010 11:32:39 -0500
Subject: [Eeglablist] Epoching correct answers
Message-ID: <20100112113239.428qj7om1w0408ok@webmail.iu.edu>

Hello,

I am new to EEGlab and have a question about epochs.  I have a Rapid 
Serial Visual Presentation sentence comprehension study and I would 
like to only include correct answers for analysis of epochs.  My main 
problem is that the event I wish to epoch occurs 6 to 10 flags before 
the subject actually responds. The extra flags and response information 
I do not want included for the event, only to test as correct or 
incorrect.  How could I go about doing this?  I was going to make one 
long epoch that starts at the first flag and includes up until the 
response, but I am still not sure how I can then reject trials that do 
not pair correctly.  I realize I could do it by hand using trial 
rejection, but I was wondering if there is a faster way to do this?  
Any kind direction will be greatly appreciated as I am still learning 
the major uses of the software.

Best,
Ben Pruce



From dr.ilya at yahoo.com  Tue Jan 12 02:20:53 2010
From: dr.ilya at yahoo.com (Ilya Adamchic)
Date: Tue, 12 Jan 2010 02:20:53 -0800 (PST)
Subject: [Eeglablist] Export from BESA, import in to EEGLAB
Message-ID: <456490.45629.qm@web63808.mail.re1.yahoo.com>

Dear all!
 
Could you share your experiences regarding data transfer from BESA to EEGLAB. 
We use BESA to do a data preprocessing and artifact correction (which we find nicely done in BESA) 
then want to import this data to the EEGLAB. This does not run smooth. 
Did you have experiences with those two systems? 
What would be the best way to export continuous data from BESA and import it in to EEGLAB?

Please reply to iamed2 at mail.ru
 
Thanks very much in advance for any response.


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100112/f1e9f7c1/attachment.html>

From dr.ilya at yahoo.com  Tue Jan 12 02:28:51 2010
From: dr.ilya at yahoo.com (Ilya Adamchic)
Date: Tue, 12 Jan 2010 02:28:51 -0800 (PST)
Subject: [Eeglablist] ICA decomposition of possibly 2 different
	conditions....
Message-ID: <104116.51255.qm@web63801.mail.re1.yahoo.com>

Dear All.

We have an experiment set up, when a patient should sit 2 min with eyes closed
and 2 min with eyes open and so on several times. This is a spontaneous
recording: no actions are performed in both conditions. When we perform ICA decomposition, is it reasonable to leave these 2
conditions mixed as a continuous file and let ICA decompose the continuous file. Or we need
to separate eyes closed from eyes open conditions and only then perform ICA on
the short epochs, which might be to short (2 min) for the 128 channel
decomposition. 

Thank you all for your replies in advance.


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100112/4459bb60/attachment.html>

From tarikbelbahar at gmail.com  Tue Jan 12 05:52:55 2010
From: tarikbelbahar at gmail.com (Tarik S Bel-Bahar)
Date: Tue, 12 Jan 2010 13:52:55 +0000
Subject: [Eeglablist] removing artifacts with ICA
In-Reply-To: <7750938.392501263204713631.JavaMail.coremail@bj126app39.126.com>
References: <7750938.392501263204713631.JavaMail.coremail@bj126app39.126.com>
Message-ID: <38e1ab891001120552k7bf92d5cr39d71a7618e7f334@mail.gmail.com>

Greetings EEGlabers,

Just two quick questions someone may be able to help with:

1. What is the best way to save out the data that results from spectopo ?
For example, frequency values are plotted onscreen as a result,
but exactly where is the data ? The help for the function does suggest that
data can be outputed, but it's not clear exactly how.
Can anyone help with a suggestion ?

2. Does anyone have EGI-based 128 and 256 location files for EEGlab ?

Respectfully,
Tarik

********************************************************************************************************************************
Tarik Bel-Bahar, PhD
Department of Clinical, Educational, and Health Psychology/ University
College London
AFC-UCL Developmental Neuroscience Unit / Anna Freud Centre, 21, Maresfield
Gardens/ London NW3 5SD
Tel: +44 (0) 20 7443 2212 / Receptionist: +44 (0) 20 7794 2313/ Fax: + 44
(0) 20 77946506
********************************************************************************************************************************

_______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100112/671e7d1f/attachment.html>

From mklados at gmail.com  Tue Jan 12 14:42:17 2010
From: mklados at gmail.com (Klados Manousos)
Date: Wed, 13 Jan 2010 00:42:17 +0200
Subject: [Eeglablist] ICA decomposition of possibly 2 different
	conditions....
In-Reply-To: <104116.51255.qm@web63801.mail.re1.yahoo.com>
References: <104116.51255.qm@web63801.mail.re1.yahoo.com>
Message-ID: <3b0db4861001121442r44b088dcrff7438feff3890c5@mail.gmail.com>

Dear IIya

According to my opinion i suppose that you have to separate them because
other pools of neuron (different sources) are activated with eyes opened and
other with eyes closed...And if you see the mathematical background because
you have a two condition experiment the underlying sources will be more than
in one codition....so some sources will be merged in one component...i think
that it is not preferable...it is better to reduce "normally" your data
dimensionality by separating the recording session....and a 2 mins record is
not a short epoch for ICA...There is an heuristic rule which says that ICA
can be succesfully applied in epochs with sample points grater than
channel^2 => Sample Points>= Channels^2.

I hope to helped you...

2010/1/12 Ilya Adamchic <dr.ilya at yahoo.com>

> Dear All.
>
>
> We have an experiment set up, when a patient should sit 2 min with eyes
> closed and 2 min with eyes open and so on several times. This is a
> spontaneous recording: no actions are performed in both conditions. When we
> perform ICA decomposition, is it reasonable to leave these 2 conditions
> mixed as a continuous file and let ICA decompose the continuous file. Or we
> need to separate eyes closed from eyes open conditions and only then perform
> ICA on the short epochs, which might be to short (2 min) for the 128 channel
> decomposition.
>
> Thank you all for your replies in advance.
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Klados A. Manousos
 Research Assistant
Group of Applied Neurosciences
Lab of Medical Informatics, Medical School
Aristotle University of Thessaloniki
Thessaloniki, Greece
_________________________________________________
Tel: +30-2310-999332
Website: http://lomiweb.med.auth.gr/gan/mklados
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100113/a94d75e4/attachment.html>

From mklados at gmail.com  Tue Jan 12 16:46:06 2010
From: mklados at gmail.com (Klados Manousos)
Date: Wed, 13 Jan 2010 02:46:06 +0200
Subject: [Eeglablist] removing artifacts with ICA
In-Reply-To: <7750938.392501263204713631.JavaMail.coremail@bj126app39.126.com>
References: <7750938.392501263204713631.JavaMail.coremail@bj126app39.126.com>
Message-ID: <3b0db4861001121646t4487e115tf332e2b87454b261@mail.gmail.com>

Dear Nancy I think that it is better to remove artifacts before
epoching...This doesnt mean that the application of ICA in epochs (with
proper length) is wrong...But i cant find a reason to run ICA (for artifact
rejection purposes) many times in all epochs when you can clean the signals
in a single run... On the other hand an epoch it is possible to be clean
from eye-blinks so ICA will not give you an observable artifactual
component...that doesnt mean that some components are not artifactual... For
your secong question you dont tell us exactly why you want to do
that...because ICA gots an out of memmory message in long term data??? If
that happens you can choose another BSS algorithm for the separation
procedure...Try to use JADE or ACSOBIRO which are both implemented in EEGLAB
and you can find them in runica function.

I hope to helped you...
2010/1/11 qiweijingwx <qiweijingwx at 126.com>

> Dear All,
> I want using ICA to remove noise such as eye movement, muscle movement et
> al. in
> eeglab. I have two questions about this process.
> First, which seems better, running ICA to remove noise before or after
> epoch the data?
> Second, if I want to remove artifacts before epoching data, but the file
> after merging is too big, could I remove noise separately in each file, then
> merge them together?
>
> Thanks in advance!
>
> best nancy
>
> 2010/01/11
>
>
>
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Klados A. Manousos
Graduate Student, Research Assistant
Group of Applied Neurosciences
Lab of Medical Informatics, Medical School
Aristotle University of Thessaloniki
Thessaloniki, Greece
_________________________________________________
Tel: +30-2310-999332
Website: http://lomiweb.med.auth.gr/gan/mklados
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100113/1a676d81/attachment.html>

From marco.congedo at gmail.com  Tue Jan 12 23:42:22 2010
From: marco.congedo at gmail.com (Marco Congedo)
Date: Wed, 13 Jan 2010 08:42:22 +0100
Subject: [Eeglablist] ICA decomposition of possibly 2 different
	conditions....
In-Reply-To: <3b0db4861001121442r44b088dcrff7438feff3890c5@mail.gmail.com>
References: <104116.51255.qm@web63801.mail.re1.yahoo.com>
	<3b0db4861001121442r44b088dcrff7438feff3890c5@mail.gmail.com>
Message-ID: <1e7e63f81001122342qaee3d7j69a42499213c294f@mail.gmail.com>

Hello Klados,

   you can use the fact that you have two conditions explicitly to recover
sources characteristic of each condition. The appropriate framework is
described in length in

On the blind source separation of human electroencephalogram by approximate
joint diagonalization of second order
statistics.<http://www.ncbi.nlm.nih.gov/pubmed/18993114?itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_RVDocSum&ordinalpos=2>

Congedo M, Gouy-Pailler C, Jutten C.
Clin Neurophysiol. 2008 Dec;119(12):2677-86.

Cheers,
Marco Congedo
Senior Scientist, cnrs
GIPSA-lab Grenoble

On Tue, Jan 12, 2010 at 11:42 PM, Klados Manousos <mklados at gmail.com> wrote:

> Dear IIya
>
> According to my opinion i suppose that you have to separate them because
> other pools of neuron (different sources) are activated with eyes opened and
> other with eyes closed...And if you see the mathematical background because
> you have a two condition experiment the underlying sources will be more than
> in one codition....so some sources will be merged in one component...i think
> that it is not preferable...it is better to reduce "normally" your data
> dimensionality by separating the recording session....and a 2 mins record is
> not a short epoch for ICA...There is an heuristic rule which says that ICA
> can be succesfully applied in epochs with sample points grater than
> channel^2 => Sample Points>= Channels^2.
>
> I hope to helped you...
>
> 2010/1/12 Ilya Adamchic <dr.ilya at yahoo.com>
>
>>  Dear All.
>>
>>
>> We have an experiment set up, when a patient should sit 2 min with eyes
>> closed and 2 min with eyes open and so on several times. This is a
>> spontaneous recording: no actions are performed in both conditions. When we
>> perform ICA decomposition, is it reasonable to leave these 2 conditions
>> mixed as a continuous file and let ICA decompose the continuous file. Or we
>> need to separate eyes closed from eyes open conditions and only then perform
>> ICA on the short epochs, which might be to short (2 min) for the 128 channel
>> decomposition.
>>
>> Thank you all for your replies in advance.
>>
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>>
>
>
>
> --
> Klados A. Manousos
>  Research Assistant
> Group of Applied Neurosciences
> Lab of Medical Informatics, Medical School
> Aristotle University of Thessaloniki
> Thessaloniki, Greece
> _________________________________________________
> Tel: +30-2310-999332
> Website: http://lomiweb.med.auth.gr/gan/mklados
>
>
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Marco Congedo
http://www.lis.inpg.fr/pages_perso/congedo/MC_Home.html
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100113/9739f4ce/attachment.html>

From smakeig at gmail.com  Thu Jan 14 08:09:07 2010
From: smakeig at gmail.com (Scott Makeig)
Date: Thu, 14 Jan 2010 08:09:07 -0800
Subject: [Eeglablist] Postdoctoral position available
Message-ID: <9e09b8f01001140809j74b2a5dek841b9515ae3412c8@mail.gmail.com>

A position in the Swartz Center for Computational Neuroscience, in the
Institute for Neural Computation, UCSD, is available immediately. The topic
is *EEG brain dynamics underlying gestural communication*. The applicant
will work with Dr. Rafael N??ez (Dept. of Cognitive Science) and Dr. Scott
Makeig (Institute for Neural Computation) to construct, run, and analyze
first experiments on gestural communication using the new Mobile Brain/Body
Imaging concept (Makeig et al., *Int J Psychophysiol*, 2009) based on
concepts of embodied cognition (N??ez, 2008, *Handbook of Cognitive Science:
An Embodied Approach*). The ideal candidate would have a degree in some
cognitive science-related discipline--particularly in natural language,
gesture, and non-verbal communication-- plus computational skills that will
allow him or her to interact with the rest of team working with
computational methods and EEG analysis. Results will contribute to basic
cognitive neuroscience, with possible applications to brain-computer
interface design.

Applicants should send their applications (CV, statement of purpose, and
names of two potential references) by January 31th 2010 to:

Dr. Rafael N??ez
(nunez at cogsci.ucsd.edu)
Department of Cognitive Science
University of California, San Diego
La Jolla, CA 92093
U.S.A.

Cc to Scott Makeig (smakeig at ucsd.edu)

-- 
Scott Makeig, Research Scientist and Director, Swartz Center for
Computational Neuroscience, Institute for Neural Computation, University of
California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100114/acd95d39/attachment.html>

From schalk at wadsworth.org  Fri Jan 15 05:24:12 2010
From: schalk at wadsworth.org (Gerwin Schalk)
Date: Fri, 15 Jan 2010 08:24:12 -0500
Subject: [Eeglablist] Annual g.tec BCI Award ...
Message-ID: <25FF6DBF-63D3-44A8-8DE6-CA514386752A@wadsworth.org>

Dear EEGlab users,

This is a reminder that the deadline for the first Brain-Computer Interface Award sponsored by is Feb. 1, 2010. Our team will be refereeing the award this year. Please find further information on http://www.gtec.at/bci_award2010.htm


Sincerely,
Gerwin Schalk
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Gerwin Schalk, Ph.D.
Research Scientist V
Wadsworth Center, NYS Dept. of Health
Dept. of Neurology, Albany Medical College
Dept. of Neurosurgery, Washington Univ. in St. Louis
Dept. of Biomed. Eng., Rensselaer Polytechnic Institute
Dept. of Biomed. Sci., State Univ. of New York at Albany
C650 Empire State Plaza
Albany, New York 12201
phone  (518) 486-2559
fax    (518) 486-4910
e-mail schalk at wadsworth.org
www    http://www.bci2000.org
www    http://www.brainmuri.org
www    http://www.gerv.org
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~



IMPORTANT NOTICE: This e-mail and any attachments may contain
confidential or sensitive information which is, or may be, legally
privileged or otherwise protected by law from further disclosure.  It
is intended only for the addressee.  If you received this in error or
from someone who was not authorized to send it to you, please do not
distribute, copy or use it or any attachments.  Please notify the
sender immediately by reply e-mail and delete this from your
system. Thank you for your cooperation.





From pacodiaz at ub.edu  Fri Jan 15 06:28:07 2010
From: pacodiaz at ub.edu (=?ISO-8859-1?Q?Paco_D=EDaz?=)
Date: Fri, 15 Jan 2010 15:28:07 +0100
Subject: [Eeglablist] OPEN POST-DOC POSITION IN BARCELONA
Message-ID: <4B507B77.6090605@ub.edu>

Hope you don't mind if I announce here the following 3 years post doc 
position in our group:

The Barcelona BrainLab (BBL) [www.ub.edu/brainlab] invites applications 
for a 3-year post-doctoral position commencing during the spring 2010 in 
the field of the Cognitive Neuroscience of Auditory Perception. 
Principal Investigator: Prof. Dr. Carles Escera. Successful applicants 
should have completed a Ph.D. in Neurosciences, Cognitive, Computer or 
Biological Sciences, Engineering or related disciplines.

For a more detailed information please find the attached document.
Contact Person: Mrs. Marta Turr?
e-mail: brainlab at ub.edu
Subject: ERANET-postdoc application

Best regards,
Paco.

-- 
---------------------------------------------------------
 Francisco Javier D?az Santaella
 Institute for Brain,Cognition and Behavior (IR3C)
 University of Barcelona
 and
 Cognitive Neuroscience Research Group
 Department of Psychiatry and Clinical Psychobiology
 University of Barcelona
 P. Vall d'Hebron 171 * 08035 Barcelona * Spain
 Telf.: +34 93 3125035 * Cell Phone: +34 678 89 47 57
 email: pacodiaz at ub.edu
 http://www.ub.edu/brainlab
---------------------------------------------------------

-------------- next part --------------
A non-text attachment was scrubbed...
Name: call_postdoc_EN_pdf.pdf
Type: application/pdf
Size: 107988 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100115/f187c37c/attachment.pdf>

From jordicostafa at gmail.com  Fri Jan 15 05:50:07 2010
From: jordicostafa at gmail.com (Jordi Costa)
Date: Fri, 15 Jan 2010 14:50:07 +0100
Subject: [Eeglablist] Post-doctoral position in Barcelona
Message-ID: <9073064e1001150550i2793cdf7jbb2853d25f0952f7@mail.gmail.com>

Please find attached an announcement of an open post-doctoral position in
Barcelona in Cognitive Neuroscience,

thanks in advance,

J.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100115/f2e21788/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: post-doctoral position in cog neurosci.pdf
Type: application/pdf
Size: 107988 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100115/f2e21788/attachment.pdf>

From arno at ucsd.edu  Sat Jan 16 19:28:09 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Sat, 16 Jan 2010 19:28:09 -0800
Subject: [Eeglablist] ICA decomposition of possibly 2 different
	conditions....
In-Reply-To: <1e7e63f81001122342qaee3d7j69a42499213c294f@mail.gmail.com>
References: <104116.51255.qm@web63801.mail.re1.yahoo.com>
	<3b0db4861001121442r44b088dcrff7438feff3890c5@mail.gmail.com>
	<1e7e63f81001122342qaee3d7j69a42499213c294f@mail.gmail.com>
Message-ID: <7EAB7F8D-9904-40B7-BF7E-E90081242583@ucsd.edu>

Dear Klados,

yes, you should keep the two conditions together. Otherwise, it is  
going to be hard to compare ICA component activities in the two  
conditions - you would have to find matching components and you would  
never know if the difference you observe between the two conditions  
for a given component pair is due to the difference in component  
carateristics (scalp projection, etc...) or if it due to the  
difference of brin activity in the two conditions. If you have the  
same components for both conditions, the first problem does not arise.  
So keep both conditions together and process the continuous file.

Hope this helps,

Arno

On Jan 12, 2010, at 11:42 PM, Marco Congedo wrote:

> Hello Klados,
>
>    you can use the fact that you have two conditions explicitly to  
> recover sources characteristic of each condition. The appropriate  
> framework is described in length in
> On the blind source separation of human electroencephalogram by  
> approximate joint diagonalization of second order statistics.
>
> Congedo M, Gouy-Pailler C, Jutten C.
>
> Clin Neurophysiol. 2008 Dec;119(12):2677-86.
>
> Cheers,
> Marco Congedo
> Senior Scientist, cnrs
> GIPSA-lab Grenoble
>
> On Tue, Jan 12, 2010 at 11:42 PM, Klados Manousos  
> <mklados at gmail.com> wrote:
> Dear IIya
>
> According to my opinion i suppose that you have to separate them  
> because other pools of neuron (different sources) are activated with  
> eyes opened and other with eyes closed...And if you see the  
> mathematical background because you have a two condition experiment  
> the underlying sources will be more than in one codition....so some  
> sources will be merged in one component...i think that it is not  
> preferable...it is better to reduce "normally" your data  
> dimensionality by separating the recording session....and a 2 mins  
> record is not a short epoch for ICA...There is an heuristic rule  
> which says that ICA can be succesfully applied in epochs with sample  
> points grater than channel^2 => Sample Points>= Channels^2.
>
> I hope to helped you...
>
> 2010/1/12 Ilya Adamchic <dr.ilya at yahoo.com>
> Dear All.
>
>
> We have an experiment set up, when a patient should sit 2 min with  
> eyes closed and 2 min with eyes open and so on several times. This  
> is a spontaneous recording: no actions are performed in both  
> conditions. When we perform ICA decomposition, is it reasonable to  
> leave these 2 conditions mixed as a continuous file and let ICA  
> decompose the continuous file. Or we need to separate eyes closed  
> from eyes open conditions and only then perform ICA on the short  
> epochs, which might be to short (2 min) for the 128 channel  
> decomposition.
>
> Thank you all for your replies in advance.
>
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
>
>
>
> -- 
> Klados A. Manousos
>  Research Assistant
> Group of Applied Neurosciences
> Lab of Medical Informatics, Medical School
> Aristotle University of Thessaloniki
> Thessaloniki, Greece
> _________________________________________________
> Tel: +30-2310-999332
> Website: http://lomiweb.med.auth.gr/gan/mklados
>
>
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
>
>
>
> -- 
> Marco Congedo
> http://www.lis.inpg.fr/pages_perso/congedo/MC_Home.html
> <ATT00001..txt>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100116/89da7c26/attachment.html>

From demiral.007 at googlemail.com  Sun Jan 17 05:55:37 2010
From: demiral.007 at googlemail.com (Baris Demiral)
Date: Sun, 17 Jan 2010 13:55:37 +0000
Subject: [Eeglablist] Epoching correct answers
In-Reply-To: <20100112113239.428qj7om1w0408ok@webmail.iu.edu>
References: <20100112113239.428qj7om1w0408ok@webmail.iu.edu>
Message-ID: <f423beba1001170555w3a44d9faqfb894ab562628065@mail.gmail.com>

After you epoch the whole data with the longest interval, you go through the
epochs and use EEG.epoch(1,j).eventtype(1,k) from 6<k<10 and check whether
any correct response (hopefully you have a specific trigger for these)
occurred (you can use for/isequal), if not you put the epoch to the
incorrect category. Best way to understand what I am saying is to open an
epoched file and investigate the EEG.epoch structure. Then you can re-epoch
again and make shorter epochs for correct trials.
Best
Baris

On Tue, Jan 12, 2010 at 4:32 PM, Pruce, Benjamin <bpruce at indiana.edu> wrote:

> Hello,
>
> I am new to EEGlab and have a question about epochs.  I have a Rapid
> Serial Visual Presentation sentence comprehension study and I would
> like to only include correct answers for analysis of epochs.  My main
> problem is that the event I wish to epoch occurs 6 to 10 flags before
> the subject actually responds. The extra flags and response information
> I do not want included for the event, only to test as correct or
> incorrect.  How could I go about doing this?  I was going to make one
> long epoch that starts at the first flag and includes up until the
> response, but I am still not sure how I can then reject trials that do
> not pair correctly.  I realize I could do it by hand using trial
> rejection, but I was wondering if there is a faster way to do this?
> Any kind direction will be greatly appreciated as I am still learning
> the major uses of the software.
>
> Best,
> Ben Pruce
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
SB Demiral, PhD.
Department of Psychology
7 George Square
The University of Edinburgh
Edinburgh, EH8 9JZ
UK
Phone: +44 (0131) 6503063
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100117/32c1dacd/attachment.html>

From dr.ilya at yahoo.com  Mon Jan 18 06:54:27 2010
From: dr.ilya at yahoo.com (Ilya Adamchic)
Date: Mon, 18 Jan 2010 06:54:27 -0800 (PST)
Subject: [Eeglablist] ICA decomposition of possibly 2 different
	conditions....
In-Reply-To: <7EAB7F8D-9904-40B7-BF7E-E90081242583@ucsd.edu>
References: <104116.51255.qm@web63801.mail.re1.yahoo.com>
	<3b0db4861001121442r44b088dcrff7438feff3890c5@mail.gmail.com>
	<1e7e63f81001122342qaee3d7j69a42499213c294f@mail.gmail.com>
	<7EAB7F8D-9904-40B7-BF7E-E90081242583@ucsd.edu>
Message-ID: <501832.80963.qm@web63802.mail.re1.yahoo.com>

Dear all,

as
it follows from J Onton?s paper: ??? jointly decomposing data from awake
and
sleeping conditions might not be optimal if the EEG source locations in these
portions of the data differed??? we do need to separate 2 conditions and than
perform ICA on
each of them separately.  
I
guess it should not be a problem cutting out some of the data (artifacts) from
EEG recording using EEGLAB function Reject an then running ICA on this data, but then the question of
enough data arises. It was suggested in several publications to use k*n2 data
points (where k is a coefficient and n is a number of channels), to get a
stable results of the ICA.  It was suggested to use k of 20 by some (128
channel EEG) (McMenamin 2009) or even bigger. What experiences do you have? How
much data will one need to get stable ICA components in 128 channel recording?
And
thanks a lot for the previous answers, you help a lot. 


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100118/1bad3ad1/attachment.html>

From julie at sccn.ucsd.edu  Mon Jan 18 12:42:18 2010
From: julie at sccn.ucsd.edu (Julie Onton)
Date: Mon, 18 Jan 2010 12:42:18 -0800 (PST)
Subject: [Eeglablist] ICA decomposition of possibly 2 different
 conditions....
In-Reply-To: <501832.80963.qm@web63802.mail.re1.yahoo.com>
References: <104116.51255.qm@web63801.mail.re1.yahoo.com>
	<3b0db4861001121442r44b088dcrff7438feff3890c5@mail.gmail.com>
	<1e7e63f81001122342qaee3d7j69a42499213c294f@mail.gmail.com>
	<7EAB7F8D-9904-40B7-BF7E-E90081242583@ucsd.edu>
	<501832.80963.qm@web63802.mail.re1.yahoo.com>
Message-ID: <53834.66.75.133.220.1263847338.squirrel@sccn.ucsd.edu>

With regard to the comment about decomposing dissimilar data separately,
perhaps a distinction should be made. Arno is correct that by far the easiest
approach is to look at *activity* differences within single sources. However,
what I point out in the paper is that vastly different behavioral conditions
(ie, sleep and wake) may show fundamentally different active sources. Within
most experimental paradigms, the difference between conditions is MUCH less
than the difference between sleep and wake, therefore warranting a single
decomposition for both conditions. For your own curiosity, try decomposing the
2 conditions separately and see how similar your components are (assuming you
have enough data to get clean decompositions).

Now, the amount of *good* data that you will need is, of course, a slightly
hazy subject. I have previously recommended a points per weight factor of 25
or more, but this is based on 71 channels (dimensions) and 256 sampling rate.
More channels require more data, of course, but I'm not sure if the points per
weight factor remains the same when the channel number scales up. The more the
better usually... I got good decompositions for ~215 channel EEG with about an
hour's worth of data. A half hour would likely not have been enough. For 128
channels, you can get away with less data... perhaps a half hour even, but not
less, I would guess. Sorry for the highly anecdotal answer, but I have found
that there is a lot of variability between subjects even with comparable
amounts of data... therefore pointing to individual differences in the
'quality' of data that ICA is good at decomposing. But that's just a theory.

Good luck, Julie

-- 
Julie Onton, PhD
http://sccn.ucsd.edu/~julie

> Dear all,
>
> as
> it follows from J Onton?s paper: ??? jointly decomposing data from awake
> and
> sleeping conditions might not be optimal if the EEG source locations in these
> portions of the data differed??? we do need to separate 2 conditions and than
> perform ICA on
> each of them separately.
> I
> guess it should not be a problem cutting out some of the data (artifacts) from
> EEG recording using EEGLAB function Reject an then running ICA on this data,
> but then the question of
> enough data arises. It was suggested in several publications to use k*n2 data
> points (where k is a coefficient and n is a number of channels), to get a
> stable results of the ICA.  It was suggested to use k of 20 by some (128
> channel EEG) (McMenamin 2009) or even bigger. What experiences do you have?
> How
> much data will one need to get stable ICA components in 128 channel recording?
> And
> thanks a lot for the previous answers, you help a lot.
>
>
>       _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu



From dr.ilya at yahoo.com  Mon Jan 18 14:18:30 2010
From: dr.ilya at yahoo.com (Ilya Adamchic)
Date: Mon, 18 Jan 2010 14:18:30 -0800 (PST)
Subject: [Eeglablist] ICA decomposition of possibly 2 different
	conditions....
In-Reply-To: <53834.66.75.133.220.1263847338.squirrel@sccn.ucsd.edu>
References: <104116.51255.qm@web63801.mail.re1.yahoo.com>
	<3b0db4861001121442r44b088dcrff7438feff3890c5@mail.gmail.com>
	<1e7e63f81001122342qaee3d7j69a42499213c294f@mail.gmail.com>
	<7EAB7F8D-9904-40B7-BF7E-E90081242583@ucsd.edu>
	<501832.80963.qm@web63802.mail.re1.yahoo.com>
	<53834.66.75.133.220.1263847338.squirrel@sccn.ucsd.edu>
Message-ID: <855891.6240.qm@web63802.mail.re1.yahoo.com>

Dear Julie,
thank you a lot for your comment. From my experience, you are right about the
amount of data needed for a nice decomposition. I have tried so far with less
amount of data (about 6 min with 1kHz sampling rate, which gave me a weight
factor of about 21) and did not get a perfect decomposition (used it, to get
rid of EMG from my EEG recording, and EMG components still contained quite a
lot of "neuro" activity). Now will try more data, the only constrain,
is the memory of the computer needed for MATLAB (working now on this problem).
Thanks
again for your time.
Best
regards,
Ilya


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100118/ed83a02d/attachment.html>

From german.gomezherrero at tut.fi  Tue Jan 19 03:58:30 2010
From: german.gomezherrero at tut.fi (German Gomez Herrero)
Date: Tue, 19 Jan 2010 13:58:30 +0200
Subject: [Eeglablist] ICA decomposition of possibly 2 different
	conditions....
In-Reply-To: <53834.66.75.133.220.1263847338.squirrel@sccn.ucsd.edu>
References: <104116.51255.qm@web63801.mail.re1.yahoo.com>	<3b0db4861001121442r44b088dcrff7438feff3890c5@mail.gmail.com>	<1e7e63f81001122342qaee3d7j69a42499213c294f@mail.gmail.com>	<7EAB7F8D-9904-40B7-BF7E-E90081242583@ucsd.edu>	<501832.80963.qm@web63802.mail.re1.yahoo.com>
	<53834.66.75.133.220.1263847338.squirrel@sccn.ucsd.edu>
Message-ID: <000201ca98fe$b7b1abf0$271503d0$@gomezherrero@tut.fi>

Hello,
I just wanted to share my opinion on the important issue of data points to use with ICA. My personal approach is to be as cautious as possible and I would usually use even more data samples than the recommended by Julie (e.g. I would use a k=100 points per weight assuming a sampling frequency of 250 Hz or so). Of course this is a rather arbitrary value and there is no way to know where is the best trade-off between giving ICA enough time to "learn" the components (i.e. providing enough data samples) and not violating the assumption that the components are stationary and their number smaller than the number of EEG sensors (i.e. not trying to analyze a too long EEG segment). Recently, my colleagues and I made some (very simple) toy experiments to investigate how many data samples you would need to obtain an accurate source estimate when the sources are EEG time-series:

http://www.cs.tut.fi/~gomezher/enrica/ieeespl2009.pdf 

Basically what we did was to mix sources that were just EEG time-series taken randomly taken from non-overlapping epochs of a long EEG recording. In this way our sources were almost independent (assuming that autocorrelations at very long lags in the EEG are negligible) while retaining very similar spectral properties to those one would expect from the true underlying sources. Because volume conduction is linear and instantaneous, these effects have a relatively minor impact on the spectral properties of the scalp EEG, which are mainly determined by the spectral properties of the underlying brain sources. What we found is that you may need up to k=300 points per weight to get "accurate enough" estimates. I have to admit that we were relatively exigent with the definition of "accurate enough" but, still, I think that the results point to the direction that k=25 might be too little in some cases. Of course, our simulations are very simplistic and, for instance, do not consider the possibility that more sources may become active as the analysis window length increases. 

Using a k=100 or more might be completely unaffordable when you have many data channels. In that case I would first use PCA to reduce the number of components. Specially in slow-wave sleep EEG I would expect to be able to explain most of the EEG variance with relatively few principal components. But of course, in many other cases you might lose too much data by rejecting components with PCA. My own experience tells that the most common mistake is too use too few data samples and only rarely one can be accused of using too many. In the extreme of using far too few data samples, overlearning may happen which, in some cases, can lead to completely wrong but visually appealing results:

http://www.cs.tut.fi/~gomezher/projects/eeg/cimed05.pdf 



Best wishes,
Germ?n



> With regard to the comment about decomposing dissimilar data
> separately,
> perhaps a distinction should be made. Arno is correct that by far the
> easiest
> approach is to look at *activity* differences within single sources.
> However,
> what I point out in the paper is that vastly different behavioral
> conditions
> (ie, sleep and wake) may show fundamentally different active sources.
> Within
> most experimental paradigms, the difference between conditions is MUCH
> less
> than the difference between sleep and wake, therefore warranting a
> single
> decomposition for both conditions. For your own curiosity, try
> decomposing the
> 2 conditions separately and see how similar your components are
> (assuming you
> have enough data to get clean decompositions).
> 
> Now, the amount of *good* data that you will need is, of course, a
> slightly
> hazy subject. I have previously recommended a points per weight factor
> of 25
> or more, but this is based on 71 channels (dimensions) and 256 sampling
> rate.
> More channels require more data, of course, but I'm not sure if the
> points per
> weight factor remains the same when the channel number scales up. The
> more the
> better usually... I got good decompositions for ~215 channel EEG with
> about an
> hour's worth of data. A half hour would likely not have been enough.
> For 128
> channels, you can get away with less data... perhaps a half hour even,
> but not
> less, I would guess. Sorry for the highly anecdotal answer, but I have
> found
> that there is a lot of variability between subjects even with
> comparable
> amounts of data... therefore pointing to individual differences in the
> 'quality' of data that ICA is good at decomposing. But that's just a
> theory.
> 
> Good luck, Julie
> 
> --
> Julie Onton, PhD
> http://sccn.ucsd.edu/~julie
> 
> > Dear all,
> >
> > as
> > it follows from J Onton?s paper: ??? jointly decomposing data from
> awake
> > and
> > sleeping conditions might not be optimal if the EEG source locations
> in these
> > portions of the data differed??? we do need to separate 2 conditions
> and than
> > perform ICA on
> > each of them separately.
> > I
> > guess it should not be a problem cutting out some of the data
> (artifacts) from
> > EEG recording using EEGLAB function Reject an then running ICA on
> this data,
> > but then the question of
> > enough data arises. It was suggested in several publications to use
> k*n2 data
> > points (where k is a coefficient and n is a number of channels), to
> get a
> > stable results of the ICA.  It was suggested to use k of 20 by some
> (128
> > channel EEG) (McMenamin 2009) or even bigger. What experiences do you
> have?
> > How
> > much data will one need to get stable ICA components in 128 channel
> recording?
> > And
> > thanks a lot for the previous answers, you help a lot.
> >
> >
> >       _______________________________________________
> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> > To unsubscribe, send an empty email to eeglablist-
> unsubscribe at sccn.ucsd.edu
> > For digest mode, send an email with the subject "set digest mime" to
> > eeglablist-request at sccn.ucsd.edu
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-
> unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu




From mail at nikobusch.net  Tue Jan 19 05:24:08 2010
From: mail at nikobusch.net (Niko Busch)
Date: Tue, 19 Jan 2010 14:24:08 +0100
Subject: [Eeglablist] Postdoc position (cognitive neuroscience/visual
 cognition) - Humboldt University Berlin, Germany
Message-ID: <4B55B278.1070304@nikobusch.net>

A postdoctoral position is available immediately in the Cognitive 
Neuroscience Group (Professor Niko Busch) at the Berlin School of Mind 
and Brain.

Our research focuses on the neural and cognitive mechanisms of visual 
perception and cognition. We combine psychophysical techniques with 
state-of-the-art EEG analyses. Topics include visual object and scene 
recognition, change blindness, masking, time perception and the temporal 
aspects of perception and cognition. In addition, we investigate the 
role of EEG oscillations in perception and cognition. The research 
facilities comprise a 128-channel EEG and psychophysics lab (access to 
other facilities like MRI, TMS, etc. may be possible).

More information about the School and its research environment can be 
found on the website: http://www.mind-and-brain.de

The successful candidate will be part of a young and dynamic research 
group. He or she will be strongly encouraged to develop his/her own 
research lines within the framework of the group?s interests.

Requirements:
- a Ph.D. in Cognitive Neuroscience or any other relevant field,
- solid experience with at least one experimental technique relevant to 
the above-mentioned research topics (e.g., psychophysics, EEG, TMS, 
fMRI, etc.),
- publication(s) in international journals,
- good programming and statistical skills,
- excellent skills in written and spoken English,
- strong motivation

The full-time position is remunerated according to public service 
research salary BAT-O IIa AnwTV HU and available until 31 October 2011 
(extension possible). The position requires teaching at Master?s and 
Ph.D. level (in German or English; topics according to 
expertise/background).

Applicants should send a letter quoting the code number DR/002/10 and 
describing their research experience, a CV, a brief statement of 
motivation, and the names and contact information of two referees as PDF 
to: niko.busch at hu-berlin.de.

The full job ad may be found here:
http://tinyurl.com/yjgqo5j



From sjwebb at u.washington.edu  Wed Jan 20 14:02:53 2010
From: sjwebb at u.washington.edu (Sara Jane Webb)
Date: Wed, 20 Jan 2010 14:02:53 -0800
Subject: [Eeglablist] Autism EEG/MEG special interest group
Message-ID: <B8CF5801-6CA5-4751-A2F1-A4FAEEE57B3E@u.washington.edu>

Hello all,

The International Society for Autism Research (INSAR) will be forming  
6 special interest groups. I would like to put together an EEG/MEG  
special interest group focused on autism spectrum disorder research.  
If this special interest group is selected, INSAR will host the group  
during one of the lunch hours at the 2010 International Meeting For  
Autism Research.

Please RSVP about your interest in forming a special interest autism  
group focused on the EEG/MEG methodologies.

RSVP to sjwebb at u.washington.edu

Name:
Title:
Department/University:

1) Interest in Special Interest group
Yes
No

2) Will be attending the 2010 IMFAR meeting in Philadelphia
Yes
No.

Please forward this email to ANY colleagues who may be interested.  
Thanks for your time.

Sincerely,

Sara Jane Webb, PhD
Research Assistant Professor of Psychiatry and Behavioral Sciences
and UW Autism Center, Research Program
http://depts.washington.edu/pbslab/
Box 357920; CHDD 314C; University of Washington
Seattle WA 98195
206.221.6461
sjwebb at u.washington.edu

Confidentiality Notice:  Because email is not secure, please be aware  
that we cannot guarantee the confidentiality of information sent by  
email.  If you are not the intended recipient, please notify the  
sender by reply email, and then destroy all copies of the message and  
any attachments.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100120/0aa40cae/attachment.html>

From arno at ucsd.edu  Thu Jan 21 08:56:25 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Thu, 21 Jan 2010 08:56:25 -0800
Subject: [Eeglablist] Time frequency analysis on a studyset
In-Reply-To: <46a4cf61001080557o60011b87o21b8e83fcd82c6d1@mail.gmail.com>
References: <46a4cf61001080557o60011b87o21b8e83fcd82c6d1@mail.gmail.com>
Message-ID: <A8FCBFD7-D626-4CEC-8C94-68F23A63143F@ucsd.edu>

Dear Cornelia,

STUDY sets works differently than datasets and they have their own  
dedicated menu to compute and visualize time-frequency  
transformations. After creating a STUDY set, you may use menu item  
"STUDY > Precompute channel measures" to precompute the time-frequency  
decomposition for all channels and all subjects. This may take some  
time but once you have done it, you will not have to do it again. You  
may then use menu item "STUDY > Plot channel measures" to plot time- 
frequency decomposition for each individual subjects or for all  
subjects pooled together. You will also be able to perform statistical  
group analysis.

The STUDY menu also contains the same type of menu for processing and  
visualizing independent components.

This part of the EEGLAB wiki deals with working with a STUDY

http://sccn.ucsd.edu/wiki/Chapter_04:_Independent_Component_Clustering

I hope this helps.
Best regards,

Arno

On Jan 8, 2010, at 5:57 AM, Cornelia McCormick wrote:

> Dear all,
>
> We are interested in hippocampal theta frequency signaling during  
> memory tasks and collected intracranial data from eight temporal  
> lobe epilepsy patients. I already analyzed each individual dataset  
> with time frequency analyses, built in eeglab, however, I cannot run  
> a group analysis. I created a studyset of all datasets but then I  
> cannot click on the "edit" or "plot" button anymore (they are just  
> not highlighted anymore). I am really stuck here and hope that  
> anybody has any idea.
>
> Thanks in advance for any comment.
> Cornelia McCormick



From dr.ilya at yahoo.com  Thu Jan 21 13:12:25 2010
From: dr.ilya at yahoo.com (Ilya Adamchic)
Date: Thu, 21 Jan 2010 13:12:25 -0800 (PST)
Subject: [Eeglablist] ICA decomposition of possibly 2 different
	conditions....
In-Reply-To: <000201ca98fe$b7b1abf0$271503d0$@gomezherrero@tut.fi>
References: <104116.51255.qm@web63801.mail.re1.yahoo.com>
	<3b0db4861001121442r44b088dcrff7438feff3890c5@mail.gmail.com>
	<1e7e63f81001122342qaee3d7j69a42499213c294f@mail.gmail.com>
	<7EAB7F8D-9904-40B7-BF7E-E90081242583@ucsd.edu>
	<501832.80963.qm@web63802.mail.re1.yahoo.com>
	<53834.66.75.133.220.1263847338.squirrel@sccn.ucsd.edu>
	<000201ca98fe$b7b1abf0$271503d0$@gomezherrero@tut.fi>
Message-ID: <469140.36491.qm@web63802.mail.re1.yahoo.com>

Hallo, 
there is probably one more issue in the number of points to use, if we talk
about EEG: sampling rate. 
If we assume "useful" range of frequencies in EEG between ~ 0 and ~
150, then having more points in the shorter period just in sense of higher sampling
rate, may not help. One might need to take in to account number of frames as
well as the length of the recording. 
Does someone know about some study, done to investigate this?
All the best,
Ilya





________________________________
From: German Gomez Herrero <german.gomezherrero at tut.fi>
To: eeglablist at sccn.ucsd.edu
Sent: Tue, January 19, 2010 12:58:30 PM
Subject: Re: [Eeglablist] ICA decomposition of possibly 2 different conditions....

Hello,
I just wanted to share my opinion on the important issue of data points to use with ICA. My personal approach is to be as cautious as possible and I would usually use even more data samples than the recommended by Julie (e.g. I would use a k=100 points per weight assuming a sampling frequency of 250 Hz or so). Of course this is a rather arbitrary value and there is no way to know where is the best trade-off between giving ICA enough time to "learn" the components (i.e. providing enough data samples) and not violating the assumption that the components are stationary and their number smaller than the number of EEG sensors (i.e. not trying to analyze a too long EEG segment). Recently, my colleagues and I made some (very simple) toy experiments to investigate how many data samples you would need to obtain an accurate source estimate when the sources are EEG time-series:

http://www.cs.tut.fi/~gomezher/enrica/ieeespl2009.pdf 

Basically what we did was to mix sources that were just EEG time-series taken randomly taken from non-overlapping epochs of a long EEG recording. In this way our sources were almost independent (assuming that autocorrelations at very long lags in the EEG are negligible) while retaining very similar spectral properties to those one would expect from the true underlying sources. Because volume conduction is linear and instantaneous, these effects have a relatively minor impact on the spectral properties of the scalp EEG, which are mainly determined by the spectral properties of the underlying brain sources. What we found is that you may need up to k=300 points per weight to get "accurate enough" estimates. I have to admit that we were relatively exigent with the definition of "accurate enough" but, still, I think that the results point to the direction that k=25 might be too little in some cases. Of course, our simulations are very simplistic and, for
 instance, do not consider the possibility that more sources may become active as the analysis window length increases. 

Using a k=100 or more might be completely unaffordable when you have many data channels. In that case I would first use PCA to reduce the number of components. Specially in slow-wave sleep EEG I would expect to be able to explain most of the EEG variance with relatively few principal components. But of course, in many other cases you might lose too much data by rejecting components with PCA. My own experience tells that the most common mistake is too use too few data samples and only rarely one can be accused of using too many. In the extreme of using far too few data samples, overlearning may happen which, in some cases, can lead to completely wrong but visually appealing results:

http://www.cs.tut.fi/~gomezher/projects/eeg/cimed05.pdf 



Best wishes,
Germ?n



> With regard to the comment about decomposing dissimilar data
> separately,
> perhaps a distinction should be made. Arno is correct that by far the
> easiest
> approach is to look at *activity* differences within single sources.
> However,
> what I point out in the paper is that vastly different behavioral
> conditions
> (ie, sleep and wake) may show fundamentally different active sources.
> Within
> most experimental paradigms, the difference between conditions is MUCH
> less
> than the difference between sleep and wake, therefore warranting a
> single
> decomposition for both conditions. For your own curiosity, try
> decomposing the
> 2 conditions separately and see how similar your components are
> (assuming you
> have enough data to get clean decompositions).
> 
> Now, the amount of *good* data that you will need is, of course, a
> slightly
> hazy subject. I have previously recommended a points per weight factor
> of 25
> or more, but this is based on 71 channels (dimensions) and 256 sampling
> rate.
> More channels require more data, of course, but I'm not sure if the
> points per
> weight factor remains the same when the channel number scales up. The
> more the
> better usually... I got good decompositions for ~215 channel EEG with
> about an
> hour's worth of data. A half hour would likely not have been enough.
> For 128
> channels, you can get away with less data... perhaps a half hour even,
> but not
> less, I would guess. Sorry for the highly anecdotal answer, but I have
> found
> that there is a lot of variability between subjects even with
> comparable
> amounts of data... therefore pointing to individual differences in the
> 'quality' of data that ICA is good at decomposing. But that's just a
> theory.
> 
> Good luck, Julie
> 
> --
> Julie Onton, PhD
> http://sccn.ucsd.edu/~julie
> 
> > Dear all,
> >
> > as
> > it follows from J Onton?s paper: ??? jointly decomposing data from
> awake
> > and
> > sleeping conditions might not be optimal if the EEG source locations
> in these
> > portions of the data differed??? we do need to separate 2 conditions
> and than
> > perform ICA on
> > each of them separately.
> > I
> > guess it should not be a problem cutting out some of the data
> (artifacts) from
> > EEG recording using EEGLAB function Reject an then running ICA on
> this data,
> > but then the question of
> > enough data arises. It was suggested in several publications to use
> k*n2 data
> > points (where k is a coefficient and n is a number of channels), to
> get a
> > stable results of the ICA.  It was suggested to use k of 20 by some
> (128
> > channel EEG) (McMenamin 2009) or even bigger. What experiences do you
> have?
> > How
> > much data will one need to get stable ICA components in 128 channel
> recording?
> > And
> > thanks a lot for the previous answers, you help a lot.
> >
> >
> >       _______________________________________________
> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> > To unsubscribe, send an empty email to eeglablist-
> unsubscribe at sccn.ucsd.edu
> > For digest mode, send an email with the subject "set digest mime" to
> > eeglablist-request at sccn.ucsd.edu
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-
> unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu


_______________________________________________
Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100121/5860f691/attachment.html>

From ole.jensen at donders.ru.nl  Fri Jan 22 00:27:36 2010
From: ole.jensen at donders.ru.nl (Ole Jensen)
Date: Fri, 22 Jan 2010 09:27:36 +0100
Subject: [Eeglablist] data analysis competition/Biomag
Message-ID: <4B596178.4070904@donders.ru.nl>

Dear all, 

We would like to announce a data analysis competition for the Biomag2010 meeting in Dubrovnik (March 28-April 1):

http://megcommunity.org/index.php?option=com_content&view=article&id=2&Itemid=24

It focuses on connectivity analysis and multivariate classification approaches.

Please consider attending or encourage interest researchers to participate.

Best regards,

Ole Jensen and Jan-Mathijs Schoffelen

-- 
 Ole Jensen
 Principal Investigator
 Neuronal Oscillations Group
 Donders Institute for Brain, Cognition and Behaviour 
 Centre for Cognitive Neuroimaging 
 P.O. Box 9101
 NL-6500 HB Nijmegen
 The Netherlands

 Office  : +31 24 36 10884
 MEG lab : +31 24 36 10988

 Fax     : +31 24 36 10989

 e-mail : ole.jensen at donders.ru.nl
 URL    : http://ojensen.ruhosting.nl/




From ole.jensen at donders.ru.nl  Fri Jan 22 00:29:37 2010
From: ole.jensen at donders.ru.nl (Ole Jensen)
Date: Fri, 22 Jan 2010 09:29:37 +0100
Subject: [Eeglablist] Biomag2010 satellite meeting: Analysis toolboxes
Message-ID: <4B5961F1.6030201@donders.ru.nl>

Dear all,

We would like to announce the satellite meeting March 28 in connection with Biomag 2010:

Analysis toolboxes for MEG data

http://megcommunity.org/index.php?option=com_content&view=article&id=7&Itemid=23

Please register soon if you would like to attend.

All the best,

Ole Jensen

-- 
 Ole Jensen
 Principal Investigator
 Neuronal Oscillations Group
 Donders Institute for Brain, Cognition and Behaviour 
 Centre for Cognitive Neuroimaging 
 P.O. Box 9101
 NL-6500 HB Nijmegen
 The Netherlands

 Office  : +31 24 36 10884
 MEG lab : +31 24 36 10988

 Fax     : +31 24 36 10989

 e-mail : ole.jensen at donders.ru.nl
 URL    : http://ojensen.ruhosting.nl/




From crangle at stanfordalumni.org  Thu Jan 21 18:32:28 2010
From: crangle at stanfordalumni.org (Colleen E Crangle)
Date: Thu, 21 Jan 2010 18:32:28 -0800
Subject: [Eeglablist] Using MATLAB's clustergram on eeg data
Message-ID: <79EA78CF3E3C400CADEEF3B0378A16FF@ColleenPC>

Has anyone made use of MATLAB's clustergram function for eeg data analysis? Usually applied to gene expression data, with each row corresponding to a gene and each column corresponding to a sample, clustergram seems as if it might usefully be applied to a matrix where each row contains single-trial data for a single subject. Across a large number of epochs, could the analysis provide useful information? Are there potential problems interpreting the results?

CGobj = clustergram(Data)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100121/80d3599d/attachment.html>

From chaitanya1686 at gmail.com  Thu Jan 21 18:43:05 2010
From: chaitanya1686 at gmail.com (chaitanya bhavaraju)
Date: Thu, 21 Jan 2010 20:43:05 -0600
Subject: [Eeglablist] Frequency Analysis
Message-ID: <79cd0c501001211843l7218fbc9w109f635c1e0d6936@mail.gmail.com>

Dear All,

I am new to EEG analysis. Firstly, I would like to know that is it
sufficient to use ICA analysis for minimizing the Artifacts. I used runica
buit in function of the EEGLAB for ICA analysis, do i have to use another
one. After ICA analysis I exported the data into a text file and then tried
FFT on that data. Is it the correct procedure to follow. Prior to perform
the FFT, i created a hanning window for the required length and then
multiplied it with the specific channel epoch of same length and then I did
FFT.

These questions may be too basic for you but I am not from electrical
background and I really need help from you people in analyzing the data.

thank you,

Chaitanya
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100121/c2874f4c/attachment.html>

From cerenakdeniz at yahoo.com  Fri Jan 22 08:08:31 2010
From: cerenakdeniz at yahoo.com (ceren akdeniz)
Date: Fri, 22 Jan 2010 08:08:31 -0800 (PST)
Subject: [Eeglablist] question about pop_newtimef
Message-ID: <261536.62689.qm@web65608.mail.ac4.yahoo.com>

Dear EEGlab members,

I'm a new user of EEGlab, and I have two questions. Please fell free to contribute.

I'm using "pop_newtimef" to plot a specific time limit  [10000 -150000ms] of my data. However because of default number of output times ( 'timesout',200)
 i can never be sure whether the image that I have plotted is representing the time points that I am interested. 
When I change it, it says "Subscripted assignment dimension mismatch."
the whole code is like the following,


for i=1:EEG.nbchan

[ersp itc powbase times1 freqs]=pop_newtimef( EEG, 1, i, [10000  150000], [0] ,'type', 'phasecoher', 'topovec',i, 'elocs', EEG.chanlocs, 'chaninfo', EEG.chaninfo, 'title','Channel power and inter-trial phase coherence','padratio', 1, 'plotphase','off','maxfreq',45,'baseline',NaN,'timesout',200);
spect1(i,:,:)=ersp;

end

figure
imagesc(times1,freqs,squeeze(mean(spect1)));axis xy
colorbar

at the end spect1 is;

  Name         Size                    Bytes  Class     Attributes

  spect1      30x922x200            44256000     double    

Another thing; is there an easy way to change/control the range of color map of power spectrum  in 'pop_newtimef'?

Thank you!


Best regards,
Ceren Akdeniz


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100122/5d7f4f68/attachment.html>

From wambua.kazi at gmail.com  Sat Jan 23 11:25:05 2010
From: wambua.kazi at gmail.com (Wambua Kazi)
Date: Sat, 23 Jan 2010 11:25:05 -0800
Subject: [Eeglablist] Question about false discovery rate
Message-ID: <292ab2591001231125n54d6e550w141eee0844ddeb0b@mail.gmail.com>

Dear EEGLABers,
    I am interested in using false discovery rate (FDR) to control for
multiple comparisons and have a couple of questions:
1) What is the version of FDR control used by EEGLAB's fdr.m (written by
Delorme & Nichols).  Could you give me a citation?

2) FDR correction assumes that the measurements in each comparison are
independent or "positively dependent."   What does "positively dependent"
mean?  I am applying FDR correction to EEG data and the results of the
different comparisons are surely not independent.  Are they likely to be
positively dependent?

    thank you for your time,
           Wambua
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100123/ec1e0f45/attachment.html>

From tarikbelbahar at gmail.com  Sun Jan 24 03:42:39 2010
From: tarikbelbahar at gmail.com (Tarik S Bel-Bahar)
Date: Sun, 24 Jan 2010 11:42:39 +0000
Subject: [Eeglablist] Frequency Analysis
In-Reply-To: <79cd0c501001211843l7218fbc9w109f635c1e0d6936@mail.gmail.com>
References: <79cd0c501001211843l7218fbc9w109f635c1e0d6936@mail.gmail.com>
Message-ID: <38e1ab891001240342s768ac459t79c7427a0601924c@mail.gmail.com>

Hello Chaitanya,

The eeglab spectopo function (see help spectopo) can help you compute
frequency data and graphics,
and as per previous communications on eeglablist, computes hanning windows
using matlab's pwelch function.
The eeglab time-frequency function may also be of help.
Regarding ICA, I assume you are removing unwanted components from data that
is analyzed.
Let me know if you have success, and if you are able to complete your
mission within eeglab.

All the best,
Tarik



********************************************************************************************************************************
Tarik Bel-Bahar, PhD
Department of Clinical, Educational, and Health Psychology/ University
College London
AFC-UCL Developmental Neuroscience Unit / Anna Freud Centre, 21, Maresfield
Gardens/ London NW3 5SD
Tel: +44 (0) 20 7443 2212 / Receptionist: +44 (0) 20 7794 2313/ Fax: + 44
(0) 20 77946506
********************************************************************************************************************************


On Fri, Jan 22, 2010 at 2:43 AM, chaitanya bhavaraju <
chaitanya1686 at gmail.com> wrote:

> Dear All,
>
> I am new to EEG analysis. Firstly, I would like to know that is it
> sufficient to use ICA analysis for minimizing the Artifacts. I used runica
> buit in function of the EEGLAB for ICA analysis, do i have to use another
> one. After ICA analysis I exported the data into a text file and then tried
> FFT on that data. Is it the correct procedure to follow. Prior to perform
> the FFT, i created a hanning window for the required length and then
> multiplied it with the specific channel epoch of same length and then I did
> FFT.
>
> These questions may be too basic for you but I am not from electrical
> background and I really need help from you people in analyzing the data.
>
> thank you,
>
> Chaitanya
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100124/ecf2acf9/attachment.html>

From ralf.baales at netcologne.de  Tue Jan 26 01:18:37 2010
From: ralf.baales at netcologne.de (Ralf Baales)
Date: Tue, 26 Jan 2010 10:18:37 +0100
Subject: [Eeglablist] fieldtrip massages  in matlab 7.5.0 (R2007b)
Message-ID: <!&!AAAAAAAAAAAYAAAAAAAAACRPUt+fOP9KhGq0ssR+LVfCgAAAEAAAAI+lm6iroKVCju1cSxuY5JIBAAAAAA==@netcologne.de>

When running Eeglab within  Matlab 7.5.0 (R2007b) a lot of warning massages
occur for functions like:

 

Warning: Function C:\Program Files

(x86)\eeglab\external\fieldtrip-20090727\external\biosig\private\isequal.m
has the same name as a

MATLAB builtin. We suggest you rename the function to avoid a potential name
conflict.

 

Is this a hazard conflict? What should I do to avoid it?

 

-Ralf-

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100126/f9430f2e/attachment.html>

From hzhang.lib at gmail.com  Mon Jan 25 18:46:18 2010
From: hzhang.lib at gmail.com (hui zhang)
Date: Mon, 25 Jan 2010 18:46:18 -0800
Subject: [Eeglablist] the activity power spectrum of the ICA component looks
	not smooth
Message-ID: <b248a64c1001251846n6ff808b3l67134fffab37d457@mail.gmail.com>

Hi there,

After  I run the ICA. The activity power spectrum looks not smooth enough.
Please see the att.. Does anybody have any guess why it looks like that?
What should I do then? Many thanks in advance!

Best,
-Hui
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100125/87bcd987/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Picture 2.png
Type: image/png
Size: 81119 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100125/87bcd987/attachment.png>

From chaitanya1686 at gmail.com  Sat Jan 23 11:23:53 2010
From: chaitanya1686 at gmail.com (chaitanya bhavaraju)
Date: Sat, 23 Jan 2010 13:23:53 -0600
Subject: [Eeglablist] Frequency Analysis
In-Reply-To: <79cd0c501001211843l7218fbc9w109f635c1e0d6936@mail.gmail.com>
References: <79cd0c501001211843l7218fbc9w109f635c1e0d6936@mail.gmail.com>
Message-ID: <79cd0c501001231123r14cd8e98rb314027b4529b743@mail.gmail.com>

Dear All,

I am new to EEG analysis. Firstly, I would like to know that is it
sufficient to use ICA analysis for minimizing the Artifacts. I used runica
buit in function of the EEGLAB for ICA analysis, do i have to use another
one. After ICA analysis I exported the data into a text file and then tried
FFT on that data. Is it the correct procedure to follow. Prior to perform
the FFT, i created a hanning window for the required length and then
multiplied it with the specific channel epoch of same length and then I did
FFT.

These questions may be too basic for you but I am not from electrical
background and I really need help from you people in analyzing the data.

thank you,

Chaitanya
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100123/54d149c6/attachment.html>

From arno at ucsd.edu  Tue Jan 26 12:38:09 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Tue, 26 Jan 2010 12:38:09 -0800
Subject: [Eeglablist] the activity power spectrum of the ICA component
	looks not smooth
In-Reply-To: <b248a64c1001251846n6ff808b3l67134fffab37d457@mail.gmail.com>
References: <b248a64c1001251846n6ff808b3l67134fffab37d457@mail.gmail.com>
Message-ID: <3368DDC7-EF14-4BFA-9A43-1F8306D0EEF5@ucsd.edu>

Dear Hui,

there was a bug in EEGLAB version 6.03b which you are probably using  
where the spectrum was only computed using 1 trial.
This was fixed about a year ago. You should download the most up to  
date version of EEGLAB.
Best,

Arno

On Jan 25, 2010, at 6:46 PM, hui zhang wrote:

> Hi there,
>
> After  I run the ICA. The activity power spectrum looks not smooth  
> enough. Please see the att.. Does anybody have any guess why it  
> looks like that? What should I do then? Many thanks in advance!
>
> Best,
> -Hui
> <Picture 2.png><ATT00001..txt>



From demiral.007 at googlemail.com  Wed Jan 27 07:34:34 2010
From: demiral.007 at googlemail.com (Baris Demiral)
Date: Wed, 27 Jan 2010 15:34:34 +0000
Subject: [Eeglablist] ERSP parameters in STUDY
Message-ID: <f423beba1001270734t3dac0a2cmb3b58bbc4721b844@mail.gmail.com>

Dear all,

When I run ERSP pre-computation for a study with default parameters I end up
with a frequency-time plot starting from 12Hz. No value below is shown.
I use EEGLABv7.2. What do you think is the problem?

Baris

-- 
SB Demiral, PhD.
Department of Psychology
7 George Square
The University of Edinburgh
Edinburgh, EH8 9JZ
UK
Phone: +44 (0131) 6503063
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100127/65ca71b7/attachment.html>

From guillaumechaumet at gmail.com  Wed Jan 27 01:12:22 2010
From: guillaumechaumet at gmail.com (guillaume chaumet)
Date: Wed, 27 Jan 2010 10:12:22 +0100
Subject: [Eeglablist] Overflow detection when edf import
Message-ID: <37af89ad1001270112x60316a08v1ecc8b572976348f@mail.gmail.com>

Dear All,
I have got an overflowdetection message during edf file importation plus
importation take very long time for short file = 20Mo.
I follow hardware and software recommendations with an core2duo 2.6Ghz, 4Go
of Ram, Ubuntu 64bits and Matlab 64bits.
Perhaps the problem is that I don't have signal processor toolbox ?

Regards

Guillaume Chaumet
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100127/944fc1aa/attachment.html>

From mataothefifth at yahoo.co.jp  Tue Jan 26 17:56:47 2010
From: mataothefifth at yahoo.co.jp (Makoto Miyakoshi)
Date: Wed, 27 Jan 2010 10:56:47 +0900 (JST)
Subject: [Eeglablist] the activity power spectrum of the ICA component
	looks not smooth
In-Reply-To: <b248a64c1001251846n6ff808b3l67134fffab37d457@mail.gmail.com>
Message-ID: <20100127015647.98728.qmail@web3711.mail.tnz.yahoo.co.jp>

Dear Hui,

Oh yes this is familiar to me. This is most probably because spectrum
computation mistakenly calculate only the first trial of the data. I
thought this was fixed in the updated version of EEGLAB. If you have a
reason not to use the other versions, please check EEGLAB BUGZILLA 698.
Below are the copy from there.

spectopo.m line 650

where, in my case, epoch_subset = [1 1 1 1 1 1...]. This analyzes only the
first trial. This is unacceptably misleading, so let me report this as a
blocker. I remember the same thing happened to me before and I fixed it for
myself, but this time I totally forgot it and wondered what was wrong with
my data. As a temporally solution, I added 

epoch_subset = find(epoch_subset == 1);

in the line643 after 'else'.

This does not happen when I use the ver. 6.01. I'm not sure though if this
always happens.

Makoto 




--- hui zhang <hzhang.lib at gmail.com> wrote:

> Hi there,
> 
> After  I run the ICA. The activity power spectrum looks not smooth
> enough.
> Please see the att.. Does anybody have any guess why it looks like
> that?
> What should I do then? Many thanks in advance!
> 
> Best,
> -Hui
> > _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
eeglablist-request at sccn.ucsd.edu



From bradley.voytek at gmail.com  Wed Jan 27 10:55:03 2010
From: bradley.voytek at gmail.com (Bradley Voytek)
Date: Wed, 27 Jan 2010 10:55:03 -0800
Subject: [Eeglablist] ERSP parameters in STUDY
In-Reply-To: <f423beba1001270734t3dac0a2cmb3b58bbc4721b844@mail.gmail.com>
References: <f423beba1001270734t3dac0a2cmb3b58bbc4721b844@mail.gmail.com>
Message-ID: <6d898bac1001271055l2e89fd86y6c2848e62d54e233@mail.gmail.com>

Probably the length of time you're using for your event window and/or
baseline. What are these settings?

::brad

On Wed, Jan 27, 2010 at 07:34, Baris Demiral <demiral.007 at googlemail.com> wrote:
> Dear all,
> When I run ERSP pre-computation for a study with default parameters I end up
> with a frequency-time plot starting from 12Hz. No value below is shown.
> I use EEGLABv7.2. What do you think is the problem?
> Baris
> --
> SB Demiral, PhD.
> Department of Psychology
> 7 George Square
> The University of Edinburgh
> Edinburgh, EH8 9JZ
> UK
> Phone: +44 (0131) 6503063
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>


From bradley.voytek at gmail.com  Wed Jan 27 11:21:47 2010
From: bradley.voytek at gmail.com (Bradley Voytek)
Date: Wed, 27 Jan 2010 11:21:47 -0800
Subject: [Eeglablist] Overflow detection when edf import
In-Reply-To: <37af89ad1001270112x60316a08v1ecc8b572976348f@mail.gmail.com>
References: <37af89ad1001270112x60316a08v1ecc8b572976348f@mail.gmail.com>
Message-ID: <6d898bac1001271121x16c473cak7bfe4b0824978e6a@mail.gmail.com>

Guillaume:

It might also have to do with your import settings... I've seen this
happen before if the import process accidentally believes that one of
your data channels is an event channel, in which case it's trying to
import thousands upon thousands of events...

Try importing only a few seconds worth of data and take a look at the
events structure.

::brad

On Wed, Jan 27, 2010 at 01:12, guillaume chaumet
<guillaumechaumet at gmail.com> wrote:
> Dear All,
> I have got an overflowdetection message during edf file?importation plus
> importation take very long time for short file = 20Mo.
> I follow hardware and software recommendations with an core2duo 2.6Ghz, 4Go
> of Ram, Ubuntu 64bits and Matlab 64bits.
> Perhaps the problem is that I don't have signal processor toolbox ?
>
> Regards
>
> Guillaume Chaumet
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



From arno at ucsd.edu  Thu Jan 28 09:49:18 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Thu, 28 Jan 2010 09:49:18 -0800
Subject: [Eeglablist] fieldtrip massages  in matlab 7.5.0 (R2007b)
In-Reply-To: <!&!AAAAAAAAAAAYAAAAAAAAACRPUt+fOP9KhGq0ssR+LVfCgAAAEAAAAI+lm6iroKVCju1cSxuY5JIBAAAAAA==@netcologne.de>
References: <!&!AAAAAAAAAAAYAAAAAAAAACRPUt+fOP9KhGq0ssR+LVfCgAAAEAAAAI+lm6iroKVCju1cSxuY5JIBAAAAAA==@netcologne.de>
Message-ID: <46CDBDE1-0AD1-436C-8451-90AAB59626CA@ucsd.edu>

Dear Ralf,

I will take a Field Trip Massage myself :-).
Just kidding. Yes, we are aware of these messages which occurs on  
Windows 7 only I think, when using Matlab 7.5.
I suggest you delete the folder \eeglab\external 
\fieldtrip-20090727\external\biosig\. This is going to have minor  
consequences on importing data using the File-IO module.
I have not done it in the repository because I want to preserve the  
integrity of the Fieldtrip folder (and not distribute a - yet another  
- modified version of Fieldtrip).
I have told the Fieldtrip developers about it, but it is actually a  
problem with Biosig, yet another module external to Fieldtrip... This  
is getting complicated.

Just delete the folder and you will be fine.
Best,

Arno

On Jan 26, 2010, at 1:18 AM, Ralf Baales wrote:

> When running Eeglab within  Matlab 7.5.0 (R2007b) a lot of warning  
> massages occur for functions like:
>
> Warning: Function C:\Program Files
> (x86)\eeglab\external\fieldtrip-20090727\external\biosig\private 
> \isequal.m has the same name as a
> MATLAB builtin. We suggest you rename the function to avoid a  
> potential name conflict.
>
> Is this a hazard conflict? What should I do to avoid it?
>
> -Ralf-
>
> <ATT00001..txt>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100128/74d3833d/attachment.html>

From bradley.voytek at gmail.com  Thu Jan 28 13:28:09 2010
From: bradley.voytek at gmail.com (Bradley Voytek)
Date: Thu, 28 Jan 2010 13:28:09 -0800
Subject: [Eeglablist] ERSP parameters in STUDY
In-Reply-To: <f423beba1001280539r38a0baf3mf5211df04044a3a9@mail.gmail.com>
References: <f423beba1001270734t3dac0a2cmb3b58bbc4721b844@mail.gmail.com>
	<6d898bac1001271055l2e89fd86y6c2848e62d54e233@mail.gmail.com>
	<f423beba1001280539r38a0baf3mf5211df04044a3a9@mail.gmail.com>
Message-ID: <6d898bac1001281328k430e8789u7c53c52fd049f7ba@mail.gmail.com>

Well if you set the basline to be only 100ms long, then you can't
reasonably expect to look at frequencies below about 10Hz. The
rationale being that a 10Hz oscillation has--by definition--10
oscillations per second, or a period of 100ms. If you work with the
definition that to get a reasonable measure of TF baseline activity
you need at least one full oscillation, then you can only look at
frequencies with periods of 100ms or shorter.

In short, use a longer baseline.

::brad

On Thu, Jan 28, 2010 at 05:39, Baris Demiral <demiral.007 at googlemail.com> wrote:
> Data is filtered ?highpass 0.3Hz and lowpass 40Hz, and epochs were set from
> -300ms to 1000ms to event onset. Baseline is -100-0ms. I wonder there are
> some parameters I should set beforehand. ?Any ideas? Or at least, am I the
> only one experiencing this problem, or is this a bug? That is what I am
> trying to understand.
>
> On Wed, Jan 27, 2010 at 6:55 PM, Bradley Voytek <bradley.voytek at gmail.com>
> wrote:
>>
>> Probably the length of time you're using for your event window and/or
>> baseline. What are these settings?
>>
>> ::brad
>>
>> On Wed, Jan 27, 2010 at 07:34, Baris Demiral <demiral.007 at googlemail.com>
>> wrote:
>> > Dear all,
>> > When I run ERSP pre-computation for a study with default parameters I
>> > end up
>> > with a frequency-time plot starting from 12Hz. No value below is shown.
>> > I use EEGLABv7.2. What do you think is the problem?
>> > Baris
>> > --
>> > SB Demiral, PhD.
>> > Department of Psychology
>> > 7 George Square
>> > The University of Edinburgh
>> > Edinburgh, EH8 9JZ
>> > UK
>> > Phone: +44 (0131) 6503063
>> >
>> > _______________________________________________
>> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> > To unsubscribe, send an empty email to
>> > eeglablist-unsubscribe at sccn.ucsd.edu
>> > For digest mode, send an email with the subject "set digest mime" to
>> > eeglablist-request at sccn.ucsd.edu
>> >
>
>
>
> --
> SB Demiral, PhD.
> Department of Psychology
> 7 George Square
> The University of Edinburgh
> Edinburgh, EH8 9JZ
> UK
> Phone: +44 (0131) 6503063
>



From demiral.007 at googlemail.com  Thu Jan 28 05:39:00 2010
From: demiral.007 at googlemail.com (Baris Demiral)
Date: Thu, 28 Jan 2010 13:39:00 +0000
Subject: [Eeglablist] ERSP parameters in STUDY
In-Reply-To: <6d898bac1001271055l2e89fd86y6c2848e62d54e233@mail.gmail.com>
References: <f423beba1001270734t3dac0a2cmb3b58bbc4721b844@mail.gmail.com>
	<6d898bac1001271055l2e89fd86y6c2848e62d54e233@mail.gmail.com>
Message-ID: <f423beba1001280539r38a0baf3mf5211df04044a3a9@mail.gmail.com>

Data is filtered  highpass 0.3Hz and lowpass 40Hz, and epochs were set from
-300ms to 1000ms to event onset. Baseline is -100-0ms. I wonder there are
some parameters I should set beforehand.  Any ideas? Or at least, am I the
only one experiencing this problem, or is this a bug? That is what I am
trying to understand.

On Wed, Jan 27, 2010 at 6:55 PM, Bradley Voytek <bradley.voytek at gmail.com>wrote:

> Probably the length of time you're using for your event window and/or
> baseline. What are these settings?
>
> ::brad
>
> On Wed, Jan 27, 2010 at 07:34, Baris Demiral <demiral.007 at googlemail.com>
> wrote:
> > Dear all,
> > When I run ERSP pre-computation for a study with default parameters I end
> up
> > with a frequency-time plot starting from 12Hz. No value below is shown.
> > I use EEGLABv7.2. What do you think is the problem?
> > Baris
> > --
> > SB Demiral, PhD.
> > Department of Psychology
> > 7 George Square
> > The University of Edinburgh
> > Edinburgh, EH8 9JZ
> > UK
> > Phone: +44 (0131) 6503063
> >
> > _______________________________________________
> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> > To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> > For digest mode, send an email with the subject "set digest mime" to
> > eeglablist-request at sccn.ucsd.edu
> >
>



-- 
SB Demiral, PhD.
Department of Psychology
7 George Square
The University of Edinburgh
Edinburgh, EH8 9JZ
UK
Phone: +44 (0131) 6503063
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100128/8d133723/attachment.html>

From hzhang.lib at gmail.com  Thu Jan 28 15:31:13 2010
From: hzhang.lib at gmail.com (hui zhang)
Date: Thu, 28 Jan 2010 15:31:13 -0800
Subject: [Eeglablist] the activity power spectrum of the ICA component
	looks not smooth
In-Reply-To: <20100127015647.98728.qmail@web3711.mail.tnz.yahoo.co.jp>
References: <b248a64c1001251846n6ff808b3l67134fffab37d457@mail.gmail.com>
	<20100127015647.98728.qmail@web3711.mail.tnz.yahoo.co.jp>
Message-ID: <b248a64c1001281531n75f1d0d7g66b41f224c226b5b@mail.gmail.com>

Cool! It's really helpful. Thank you all!

2010/1/26 Makoto Miyakoshi <mataothefifth at yahoo.co.jp>

> Dear Hui,
>
> Oh yes this is familiar to me. This is most probably because spectrum
> computation mistakenly calculate only the first trial of the data. I
> thought this was fixed in the updated version of EEGLAB. If you have a
> reason not to use the other versions, please check EEGLAB BUGZILLA 698.
> Below are the copy from there.
>
> spectopo.m line 650
>
> where, in my case, epoch_subset = [1 1 1 1 1 1...]. This analyzes only the
> first trial. This is unacceptably misleading, so let me report this as a
> blocker. I remember the same thing happened to me before and I fixed it for
> myself, but this time I totally forgot it and wondered what was wrong with
> my data. As a temporally solution, I added
>
> epoch_subset = find(epoch_subset == 1);
>
> in the line643 after 'else'.
>
> This does not happen when I use the ver. 6.01. I'm not sure though if this
> always happens.
>
> Makoto
>
>
>
>
> --- hui zhang <hzhang.lib at gmail.com> wrote:
>
> > Hi there,
> >
> > After  I run the ICA. The activity power spectrum looks not smooth
> > enough.
> > Please see the att.. Does anybody have any guess why it looks like
> > that?
> > What should I do then? Many thanks in advance!
> >
> > Best,
> > -Hui
> > > _______________________________________________
> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> > To unsubscribe, send an empty email to
> > eeglablist-unsubscribe at sccn.ucsd.edu
> > For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100128/984ab55b/attachment.html>

From isacor at us.es  Thu Jan 28 01:17:30 2010
From: isacor at us.es (isacor at us.es)
Date: Thu, 28 Jan 2010 10:17:30 +0100
Subject: [Eeglablist] ERSP in STUDY
Message-ID: <f5fda4b58a46.4b61643a@us.es>

Hello everyone. 
I created a study with 8 subjects and three conditions.The sampling frequency is 1024. The time window ranging from -1800 ms to 500 ms. To compute the ERSP have used the following parameters:
'cycles', [1.5 0.5], 'nfreqs', 100, 'baseline', [-1500 -1400], 'padratio', 4, 'alpha', 0.01
When I visualize ERSP activity for a certain channel, I hope that no significant activity was plotted in green. However this does not happen, or not take into account the level of significance.
Do not know if I used the appropriate parameters. When I do a merged with my 8 subjects for a condition and I use the same parameters from eeglab through 'channel time-frequency' , non-significant features are plotted in green.
In the study to the time of making the channels precompute've left anything out?
I hope your help.
Thank you.

Isabel Cordones Cano
Neurociencia y Comportamiento
Fisiolog?a Animal Y Zoologia
Facultad de Biolog?a, Universidad de Sevilla
Avda. Reina Mercedes 6, 41012-Sevilla
Espa?a




From m.blefari at sssup.it  Thu Jan 28 06:23:03 2010
From: m.blefari at sssup.it (Blefari Maria Laura)
Date: Thu, 28 Jan 2010 15:23:03 +0100
Subject: [Eeglablist] ICA - message
Message-ID: <web-57278489@sssup.it>


Hi,

    I'm using fastICA algorithm in order to remove 
artifact. I have 16 electrodes. 

Sometimes I got this messagge 

Component 13 did not converge in 1000 iterations.
Too many failures to converge(6). Giving up. Adding the 
mean to the data.
eeg_checkset:recompiuting the ICA activation matrix.
Done.

and then stopped.
What happen, any idea? the signal is too noisy and it is 
not possible to perform the decomposition? 

Thanks in advance,
Maria Laura


From perlaki at gamma.ttk.pte.hu  Thu Jan 28 13:19:37 2010
From: perlaki at gamma.ttk.pte.hu (Perlaki Gabor)
Date: Thu, 28 Jan 2010 22:19:37 +0100
Subject: [Eeglablist] Ecg
Message-ID: <20100128211659.M22636@gamma.ttk.pte.hu>

Hi all,

I've a noisy ecg and FMRIB-plugin could detect the qrs comlexes quite well. Is
it any way to find the R-peaks using these complexes, because I think RR-peak
distance would be better for HRV analysis. R-peaks are much higher than any
other peaks, so they aren't affected by noise so much as Q.

Sincerely,

Gabor

--
P?csi Tudom?nyegyetem Term?szettudom?nyi Kar (http://www.ttk.pte.hu/)


From arno at ucsd.edu  Thu Jan 28 20:35:46 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Thu, 28 Jan 2010 20:35:46 -0800
Subject: [Eeglablist] the activity power spectrum of the ICA component
	looks not smooth
In-Reply-To: <20100127015647.98728.qmail@web3711.mail.tnz.yahoo.co.jp>
References: <20100127015647.98728.qmail@web3711.mail.tnz.yahoo.co.jp>
Message-ID: <FF3B5271-2800-475F-BF72-AC377D250979@ucsd.edu>

Dear Hui and Makoto,

yes, this has been fixed since version 7.
Thanks a lot,

Arno



From Keith.McConnell at cchmc.org  Fri Jan 29 06:17:12 2010
From: Keith.McConnell at cchmc.org (Keith McConnell)
Date: Fri, 29 Jan 2010 09:17:12 -0500
Subject: [Eeglablist] Ecg
In-Reply-To: <20100128211659.M22636@gamma.ttk.pte.hu>
References: <20100128211659.M22636@gamma.ttk.pte.hu>
Message-ID: <4B62A798.0F11.006F.0@cchmc.org>

Hello Gabor.

First, RR interval is a great means of doing HRV analysis.  Without
knowing much about the noise content, if you are easily identifying the
qrs complexes then an easy next step would be to simply find the max
value within each complex and assume that as the R-wave peak.  This may
be oversimplifying, though.

Good luck.

Keith

>>> "Perlaki Gabor" <perlaki at gamma.ttk.pte.hu> 1/28/2010 4:19 PM >>>
Hi all,

I've a noisy ecg and FMRIB-plugin could detect the qrs comlexes quite
well. Is
it any way to find the R-peaks using these complexes, because I think
RR-peak
distance would be better for HRV analysis. R-peaks are much higher than
any
other peaks, so they aren't affected by noise so much as Q.

Sincerely,

Gabor

--
P?csi Tudom?nyegyetem Term?szettudom?nyi Kar (http://www.ttk.pte.hu/)
_______________________________________________
Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html 
To unsubscribe, send an empty email to
eeglablist-unsubscribe at sccn.ucsd.edu 
For digest mode, send an email with the subject "set digest mime" to
eeglablist-request at sccn.ucsd.edu


From mklados at gmail.com  Thu Jan 28 23:51:02 2010
From: mklados at gmail.com (Klados Manousos)
Date: Fri, 29 Jan 2010 09:51:02 +0200
Subject: [Eeglablist] ICA - message
In-Reply-To: <web-57278489@sssup.it>
References: <web-57278489@sssup.it>
Message-ID: <3b0db4861001282351m3b89074eye859c1e96b7056f0@mail.gmail.com>

Dear Maria Laura

The problem relies on the real dimensionality of the data. The real
independent sources are much more than the recorded electrodes, so fast-ICA
tries to exctract many sources in a single independent component. If you
consider the central limit theorem this component tends to have gaussian
distribution, which is a great problem...especially when you use algorithms
based on high order statistics

According to my opinion is better to use extended - infomax ICA (the default
ICA of EEGLAB) or if you want something faster you can use ACSOBIRO...which
is based on second order statistics and can retrieve components with
distributions close to gaussian.

I hope i helped...

Manousos

2010/1/28 Blefari Maria Laura <m.blefari at sssup.it>

>
> Hi,
>
>    I'm using fastICA algorithm in order to remove
> artifact. I have 16 electrodes.
>
> Sometimes I got this messagge
>
> Component 13 did not converge in 1000 iterations.
> Too many failures to converge(6). Giving up. Adding the
> mean to the data.
> eeg_checkset:recompiuting the ICA activation matrix.
> Done.
>
> and then stopped.
> What happen, any idea? the signal is too noisy and it is
> not possible to perform the decomposition?
>
> Thanks in advance,
> Maria Laura
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Manousos A. Klados
PhD Candidate -- Research Assistant
Group of Applied Neurosciences
Lab of Medical Informatics
School of Medicine
Aristotle University of Thessaloniki
P.O. Box 323 54124 Thessaloniki Greece
_________________________________________________
Tel: +30-2310-999332
Fax:+30-2310-999263
Website: http://lomiweb.med.auth.gr/gan/mklados
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100129/f2062d16/attachment.html>

From m.blefari at sssup.it  Thu Jan 28 23:59:04 2010
From: m.blefari at sssup.it (Blefari Maria Laura)
Date: Fri, 29 Jan 2010 08:59:04 +0100
Subject: [Eeglablist] ICA - message
In-Reply-To: <3b0db4861001282351m3b89074eye859c1e96b7056f0@mail.gmail.com>
Message-ID: <web-57297328@sssup.it>


Thanks Manousos for your clear and quick answer. I will 
try it.

Maria Laura


On Fri, 29 Jan 2010 09:51:02 +0200
  Klados Manousos <mklados at gmail.com> wrote:
>Dear Maria Laura
>
>The problem relies on the real dimensionality of the 
>data. The real
>independent sources are much more than the recorded 
>electrodes, so fast-ICA
>tries to exctract many sources in a single independent 
>component. If you
>consider the central limit theorem this component tends 
>to have gaussian
>distribution, which is a great problem...especially when 
>you use algorithms
>based on high order statistics
>
>According to my opinion is better to use extended - 
>infomax ICA (the default
>ICA of EEGLAB) or if you want something faster you can 
>use ACSOBIRO...which
>is based on second order statistics and can retrieve 
>components with
>distributions close to gaussian.
>
>I hope i helped...
>
>Manousos
>
>2010/1/28 Blefari Maria Laura <m.blefari at sssup.it>
>
>>
>> Hi,
>>
>>    I'm using fastICA algorithm in order to remove
>> artifact. I have 16 electrodes.
>>
>> Sometimes I got this messagge
>>
>> Component 13 did not converge in 1000 iterations.
>> Too many failures to converge(6). Giving up. Adding the
>> mean to the data.
>> eeg_checkset:recompiuting the ICA activation matrix.
>> Done.
>>
>> and then stopped.
>> What happen, any idea? the signal is too noisy and it is
>> not possible to perform the decomposition?
>>
>> Thanks in advance,
>> Maria Laura
>> _______________________________________________
>> Eeglablist page: 
>>http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set 
>>digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>>
>
>
>
>-- 
>Manousos A. Klados
>PhD Candidate -- Research Assistant
>Group of Applied Neurosciences
>Lab of Medical Informatics
>School of Medicine
>Aristotle University of Thessaloniki
>P.O. Box 323 54124 Thessaloniki Greece
>_________________________________________________
>Tel: +30-2310-999332
>Fax:+30-2310-999263
>Website: http://lomiweb.med.auth.gr/gan/mklados



From paul_c at gmx.de  Fri Jan 29 02:38:15 2010
From: paul_c at gmx.de (Paul Czienskowski)
Date: Fri, 29 Jan 2010 11:38:15 +0100
Subject: [Eeglablist] bem_matrix.exe failing
Message-ID: <4B62BA97.20408@gmx.de>

Hey all,

For I'm investigating age-related differences in EEG-Source localization 
for my diploma thesis I'm trying to generate the BEM forward-model from 
T1 MRI Data with NFT. Now I'm at the stage to actually create the 
forward model, but bem_matrix.exe crashes with the Exception 
STATUS_ACCESS_VIOLATION (Stackdump see below). I'm running Matlab on a 
Windows 7 Professional x64 Laptop (2GB of RAM, 2,3GHz Core2Duo). I was 
trying to run it on our Citrix-Envireonment, but it crashed there too. 
Maybe anyone encountered the problem too and was able to solve it.

Thanks in advance, Paul


Stacktrace:
Exception: STATUS_ACCESS_VIOLATION at eip=610DD894
eax=00000000 ebx=00040000 ecx=00010000 edx=00000000 esi=611B3F08 
edi=00000000
ebp=0028C0B8 esp=0028C0AC 
program=C:\Users\Paul\Programmieren\Matlab\NFT-1.0\bin\bem_matrix.exe, 
pid 4040, thread main
cs=0023 ds=002B es=002B fs=0053 gs=002B ss=002B
Stack trace:
Frame     Function  Args
0028C0B8  610DD894  (00000000, 611B3F08, 00040000, 5DF00000)
00040100  610029C7  (00000032, 000008BC, 000008D0, 00000030)
8648586 [main] bem_matrix 4040 _cygtls::handle_exceptions: Exception: 
STATUS_ACCESS_VIOLATION
8859954 [main] bem_matrix 4040 _cygtls::handle_exceptions: Error while 
dumping state (probably corrupted stack)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: paul_c.vcf
Type: text/x-vcard
Size: 229 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100129/bdde29e7/attachment.vcf>

From pwang.list at googlemail.com  Fri Jan 29 01:49:54 2010
From: pwang.list at googlemail.com (peng wang)
Date: Fri, 29 Jan 2010 10:49:54 +0100
Subject: [Eeglablist] ICA problem
Message-ID: <f81849ea1001290149u587591abh5f9c4b6ecde09338@mail.gmail.com>

Hi there,

   I am using ICA to remove blinks via EEGLab. My dataset has 122 channels,
and it takes so long to compute 122 components.
   (1) So I tried to use the option "ncomps" (say, 24) to reduce the number
of components. However, an error message appears after computing: "Matrix
dimensions must agree".

   (2) Then I tried fastICA instead as following,

==================
 sz = size(EEG.data);
 nchans = sz(1);
 npts = sz(2);
 ntrials = sz(3);
 clear sz;
 nICs = 24;
 data = reshape(EEG.data,nchans,npts*ntrials);
 [ica,V,W] = fastica(data,'numOfIC',nICs,'approach','symm');
 EEG.icasphere = eye(nchans);
 EEG.icaact = single(reshape(ica,nICs,npts,ntrials));
 EEG.icawinv = V;
 EEG.icaweights = W;
 EEG = eeg_checkset( EEG );
 clear V W ica data;

 EEG = pop_saveset( EEG, 'filename','test_raw_ica');
==================

   Everything seems fine. But when I reject the blink component via GUI of
eeglab and load the data again, Something strange happens. It seems the
amplitude of EEG.data become much smaller, about in  -1~1 range.  Thus I
wonder whether there was some normalization behind, and how can I correct
it? The problem would not repeat if I choose the number of components same
as channels event in fast ICA (e.g. change to "nICs = nchans" in the above
code).

   Thank you for your help.

best
Peng
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100129/c07e248a/attachment.html>

From isacor at us.es  Mon Feb  1 00:18:35 2010
From: isacor at us.es (isacor at us.es)
Date: Mon, 01 Feb 2010 09:18:35 +0100
Subject: [Eeglablist] ERSP in STUDY
Message-ID: <f603d32a16b8a.4b669c6b@us.es>

Hello everyone. 
I created a study with 8 subjects and three conditions.The sampling frequency is 1024. The time window ranging from -1800 ms to 500 ms. To compute the ERSP have used the following parameters:
'cycles', [1.5 0.5], 'nfreqs', 100, 'baseline', [-1500 -1400], 'padratio', 4, 'alpha', 0.01
When I visualize ERSP activity for a certain channel, I hope that no significant activity was plotted in green. However this does not happen, or not take into account the level of significance.
Do not know if I used the appropriate parameters. When I do a merged with my 8 subjects for a condition and I use the same parameters from eeglab through 'channel time-frequency' , non-significant features are plotted in green.
In the study to the time of making the channels precompute've left anything out?
I hope your help.
Thank you.

Isabel Cordones Cano
Neurociencia y Comportamiento
Fisiolog?a Animal Y Zoologia
Facultad de Biolog?a, Universidad de Sevilla
Avda. Reina Mercedes 6, 41012-Sevilla
Espa?a



From jdien07 at mac.com  Sun Jan 31 18:04:18 2010
From: jdien07 at mac.com (Joseph Dien)
Date: Sun, 31 Jan 2010 21:04:18 -0500
Subject: [Eeglablist] fieldtrip massages  in matlab 7.5.0 (R2007b)
In-Reply-To: <46CDBDE1-0AD1-436C-8451-90AAB59626CA@ucsd.edu>
References: <!&!AAAAAAAAAAAYAAAAAAAAACRPUt+fOP9KhGq0ssR+LVfCgAAAEAAAAI+lm6iroKVCju1cSxuY5JIBAAAAAA==@netcologne.de>
	<46CDBDE1-0AD1-436C-8451-90AAB59626CA@ucsd.edu>
Message-ID: <3FD211BD-DFEA-4A92-91EA-4A5B40FE3BAD@mac.com>

Yeah, all these cross-toolbox dependencies are getting complex indeed!  As one of the contributors to the FieldTrip I/O code (due to my own EP Toolbox's dependency on it), I figure I should chime in to point out that the current version of FieldTrip has dropped the Biosig module due to these issues.  My suggestion to Arno is to update the version of FieldTrip that is bundled with EEGlab as it is some six months out of date at this point and is missing a lot of fixes, such as this one.

Cheers!

Joe

On Jan 28, 2010, at 12:49 PM, Arnaud Delorme wrote:

> Dear Ralf,
> 
> I will take a Field Trip Massage myself :-).
> Just kidding. Yes, we are aware of these messages which occurs on Windows 7 only I think, when using Matlab 7.5.
> I suggest you delete the folder \eeglab\external\fieldtrip-20090727\external\biosig\. This is going to have minor consequences on importing data using the File-IO module.
> I have not done it in the repository because I want to preserve the integrity of the Fieldtrip folder (and not distribute a - yet another - modified version of Fieldtrip). 
> I have told the Fieldtrip developers about it, but it is actually a problem with Biosig, yet another module external to Fieldtrip... This is getting complicated.
> 
> Just delete the folder and you will be fine.
> Best,
> 
> Arno
> 
> On Jan 26, 2010, at 1:18 AM, Ralf Baales wrote:
> 
>> When running Eeglab within  Matlab 7.5.0 (R2007b) a lot of warning massages occur for functions like:
>>  
>> Warning: Function C:\Program Files
>> (x86)\eeglab\external\fieldtrip-20090727\external\biosig\private\isequal.m has the same name as a
>> MATLAB builtin. We suggest you rename the function to avoid a potential name conflict.
>>  
>> Is this a hazard conflict? What should I do to avoid it?
>>  
>> -Ralf-
>>  
>> <ATT00001..txt>
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu


--------------------------------------------------------------------------------

Joseph Dien,
Senior Research Scientist
Center for Advanced Study of Language
University of Maryland 
7005 52nd Avenue
College Park, MD 20742-0025

E-mail: jdien07 at mac.com
Phone: 301-226-8848
Fax: 301-226-8811
http://homepage.mac.com/jdien07/








-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100131/d651d5b1/attachment.html>

From jdien07 at mac.com  Mon Feb  1 06:56:50 2010
From: jdien07 at mac.com (Joseph Dien)
Date: Mon, 01 Feb 2010 09:56:50 -0500
Subject: [Eeglablist] ICA problem
In-Reply-To: <f81849ea1001290149u587591abh5f9c4b6ecde09338@mail.gmail.com>
References: <f81849ea1001290149u587591abh5f9c4b6ecde09338@mail.gmail.com>
Message-ID: <91D5B97F-6924-45C9-A09E-9583C99DACA4@mac.com>

When you say "so long", how long do you mean?  While ICA is not by its nature a fast procedure, certain datasets can take much longer than usual.  For example, I find that if two channels are perfectly correlated (1 or -1) then an ICA run will take much longer.  This can happen if the data is mean mastoid referenced and both channels are explicitly included in the data because they will have a perfect -1 correlation (see Dien, 1998 for reference issues).  It can also happen if a channel is shorted out during acquisition and the reference channel is explicitly included because then they will have a perfect correlation.  Also if two channels are shorted together during the data acquisition they will be perfectly correlated with each other.  My EP Toolkit (https://sourceforge.net/projects/erppcatoolkit/) has code for dealing with these situations so you might want to look into it.  It implements an automated artifact correction routine that relies on EEGlab's runICA code, among other things.

Cheers!

Joe



On Jan 29, 2010, at 4:49 AM, peng wang wrote:

> Hi there,
> 
>    I am using ICA to remove blinks via EEGLab. My dataset has 122 channels, and it takes so long to compute 122 components. 
>    (1) So I tried to use the option "ncomps" (say, 24) to reduce the number of components. However, an error message appears after computing: "Matrix dimensions must agree". 
> 
>    (2) Then I tried fastICA instead as following,
> 
> ==================
>  sz = size(EEG.data);
>  nchans = sz(1);
>  npts = sz(2);
>  ntrials = sz(3);
>  clear sz;
>  nICs = 24;
>  data = reshape(EEG.data,nchans,npts*ntrials);
>  [ica,V,W] = fastica(data,'numOfIC',nICs,'approach','symm');
>  EEG.icasphere = eye(nchans);
>  EEG.icaact = single(reshape(ica,nICs,npts,ntrials));
>  EEG.icawinv = V;
>  EEG.icaweights = W;
>  EEG = eeg_checkset( EEG );
>  clear V W ica data;
> 
>  EEG = pop_saveset( EEG, 'filename','test_raw_ica');
> ==================
> 
>    Everything seems fine. But when I reject the blink component via GUI of eeglab and load the data again, Something strange happens. It seems the amplitude of EEG.data become much smaller, about in  -1~1 range.  Thus I wonder whether there was some normalization behind, and how can I correct it? The problem would not repeat if I choose the number of components same as channels event in fast ICA (e.g. change to "nICs = nchans" in the above code).
> 
>    Thank you for your help.
> 
> best
> Peng
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu


--------------------------------------------------------------------------------

Joseph Dien,
Senior Research Scientist
University of Maryland 
7005 52nd Avenue
College Park, MD 20742-0025

E-mail: jdien07 at mac.com
Phone: 301-226-8848
Fax: 301-226-8811
http://homepage.mac.com/jdien07/











From Maddie.Groom at nottingham.ac.uk  Mon Feb  1 01:22:49 2010
From: Maddie.Groom at nottingham.ac.uk (Maddie Groom)
Date: Mon, 1 Feb 2010 09:22:49 -0000
Subject: [Eeglablist] PhD studentship - Division of Psychiatry,
	University of Nottingham
Message-ID: <87E5A373B1168C458E9216F820F3D1CF04616CA1@VUIEXCHC.ad.nottingham.ac.uk>


3 year Doctoral Studentship available in "Electrophysiological correlates of cognitive control in children with tics with and without comorbid ADHD symptoms"

School of Community Health Sciences
Division of Psychiatry
University of Nottingham, UK

Start date: October 2010

Applications are invited for a PhD Studentship in the School of Community Health Sciences, Division of Psychiatry. The studentship is intended to support postgraduate investigators seeking to establish a career in neuropsychiatric research using neuroimaging techniques. Within the first three months of the studentship a research training programme will be jointly agreed by the student and the supervisors. This will comprise courses and modules organised by the Graduate School and those available within the Faculty of Medicine and Health Sciences. The Division of Psychiatry is one of four divisions within the School of Community Health Sciences. The Division has a growing capacity and reputation for research in cognitive and developmental neurosciences and holds major grants from Research Councils (MRC, ESRC), leading UK charities (Wellcome Trust) and governmental sources (NIHR). Please visit the Divisional web site for more information see: http://www.nottingham/chs/divisions/psychiatry. The Division has excellent relationships with international academic partners, in particular the European Network of Hyperkinetic Disorders (EuNetHyDis) and Tourettes Action, providing excellent opportunities for the student to develop relationships with researchers at other institutions and to become established in the international scientific community.

The student will join the Mind, Brain, Behaviour Group (MBBG), a group of researchers within the Division of Psychiatry whose primary interests are in developing and refining neuroimaging techniques to uncover the neuro-biological causes of psychiatric disorders. The student will also benefit from strong links between the Division of Psychiatry and Nottinghamshire Healthcare NHS Trust, reflected in part funding of the studentship by the Institute of Mental Health (IMH). The IMH represents collaboration between the University and Nottinghamshire Healthcare Trust which has created a stimulating and supporting environment to foster research of the highest quality into improving outcomes in mental health and learning disability. It brings together NHS researchers from different professional backgrounds with University researchers from a wide variety of academic disciplines. This multidisciplinary approach produces the high quality information required for the implementation of evidence-based practice. For more information about IMH see: http://www.institutemh.org.uk/-home-.html. 

This full-time, three-year studentship will cover the cost of PhD fees, at the Home/European rate of ?3,460 (subject to confirmation) and a tax-free stipend of ?13,290 (09/10 level subject to uplift) and starts in October 2010.

Due to funding restrictions, this studentship is only open to full-time UK/EU students.  Applications from students willing to self-fund can be considered at any time. International students can find details of scholarships available to fund postgraduate research degrees on the University website at: www.nottingham.ac.uk/prospectuses/postgrad.

Applications should be sent, preferably by Email, to Dr M Groom, Division of Psychiatry, E Floor, South Block, Queen's Medical Centre, Derby Road, Nottingham NG7 2UH. Email: Maddie.Groom at Nottingham.ac.uk. 

Applicants should prepare a brief protocol (max one A4 page) outlining a proposal for a pilot study to be carried out in preparation for the main study (please see further details). The proposal should include: background to the pilot study, proposed methods (including planned statistical analysis) and identification of potential difficulties that might be encountered. The proposal must be submitted along with a CV and the names and contact details of two academic referees. Applicants are strongly advised to make contact with Dr Groom to discuss their application, before submitting.


Please quote ref. MED/640. Closing date: 26 February 2010. Interview date: W/C 22 March 2010.
Web link: http://jobs.nottingham.ac.uk/vacancies.aspx?cat=345#j6794

Further details

Project specification

Electrophysiological correlates of cognitive control in children with tics with and without comorbid ADHD symptoms

Supervisors
Dr Madeleine Groom Division of Psychiatry, section of Developmental Psychiatry
Professor Georgina Jackson Division of Psychiatry, section of Behavioural Sciences

Research within the Division of Psychiatry (Jackson et al., 2007, Exp Br Res; Mueller et al., 2006, Curr Biol) and elsewhere (Baym et al., 2008 Brain) (Thibault et al., 2009, Psychiatry Res) has shown enhanced cognitive control in children with TS compared with typically developing children. Cognitive control is an umbrella term for several cognitive functions that work together to produce smooth, efficient goal-directed behaviour. The findings in TS suggest that at least a proportion of these children are capable of exerting greater cognitive control, possibly as a consequence of the continual need to suppress their tics and that this may represent a compensatory mechanism that increases the likelihood of symptom remission later in development. Considering the evidence that ADHD is associated with impaired cognitive control (Groom et al., in press, Biol Psychiatry; Groom et al., 2009, J Ch Psychol Psychiatry; Groom et al., 2008, Biol Psychiatry), comorbid ADHD symptoms may be one factor underlying poor tic suppression and poor long-term prognosis. 

The project will investigate the effects of comorbid ADHD on electrophysiological correlates of cognitive function and tic suppression. The student will be required to develop appropriate paradigms to measure the neural correlates of cognitive control using electrophysiology and apply these paradigms to children with TS with and without comorbid ADHD. 


This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system:
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100201/826faab9/attachment.html>

From paul_c at gmx.de  Sun Jan 31 22:50:59 2010
From: paul_c at gmx.de (Paul Czienskowski)
Date: Mon, 01 Feb 2010 07:50:59 +0100
Subject: [Eeglablist] bem_matrix.exe failing
In-Reply-To: <8796bba51001312018m35083606k14b488e69d98166d@mail.gmail.com>
References: <4B62BA97.20408@gmx.de>
	<8796bba51001312018m35083606k14b488e69d98166d@mail.gmail.com>
Message-ID: <4B6679D3.6010708@gmx.de>

Hi Zeynep,

thanks for your fast and helpful answer. It's good to know why it fails 
and I will try to get it running on some machine with more resources 
now. Fortunately where I am there's no real lack of resources, so this 
wont be any Problem.

Cheers,
Paul

Zeynep Akalin Acar schrieb:
> Hi Paul,
>
> NFT mesh generation module generates meshes with ~19000 nodes. This
> requires a memory of about 3 GB. Also, if you apply IPA, the required
> memory will be about 4.8GB. The version of bem_matrix.exe included in
> the NFT toolbox was compiled with cygwin, and was limited by the
> available physical memory. I am working on getting bem_matrix.exe
> compile with Visual Studio which should make it use the swap space as
> well. It will probably run very slowly on a 2GB machine if it uses
> swap, but it should still work.
>
> Best,
> Zeynep
>
>
> On Fri, Jan 29, 2010 at 2:38 AM, Paul Czienskowski <paul_c at gmx.de> wrote:
>   
>> Hey all,
>>
>> For I'm investigating age-related differences in EEG-Source localization for
>> my diploma thesis I'm trying to generate the BEM forward-model from T1 MRI
>> Data with NFT. Now I'm at the stage to actually create the forward model,
>> but bem_matrix.exe crashes with the Exception STATUS_ACCESS_VIOLATION
>> (Stackdump see below). I'm running Matlab on a Windows 7 Professional x64
>> Laptop (2GB of RAM, 2,3GHz Core2Duo). I was trying to run it on our
>> Citrix-Envireonment, but it crashed there too. Maybe anyone encountered the
>> problem too and was able to solve it.
>>
>> Thanks in advance, Paul
>>
>>
>> Stacktrace:
>> Exception: STATUS_ACCESS_VIOLATION at eip=610DD894
>> eax=00000000 ebx=00040000 ecx=00010000 edx=00000000 esi=611B3F08
>> edi=00000000
>> ebp=0028C0B8 esp=0028C0AC
>> program=C:\Users\Paul\Programmieren\Matlab\NFT-1.0\bin\bem_matrix.exe, pid
>> 4040, thread main
>> cs=0023 ds=002B es=002B fs=0053 gs=002B ss=002B
>> Stack trace:
>> Frame     Function  Args
>> 0028C0B8  610DD894  (00000000, 611B3F08, 00040000, 5DF00000)
>> 00040100  610029C7  (00000032, 000008BC, 000008D0, 00000030)
>> 8648586 [main] bem_matrix 4040 _cygtls::handle_exceptions: Exception:
>> STATUS_ACCESS_VIOLATION
>> 8859954 [main] bem_matrix 4040 _cygtls::handle_exceptions: Error while
>> dumping state (probably corrupted stack)
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>>
>>     


-- 
Paul Czienskowski
Bj?rnsonstr. 25
12163 Berlin

Tel.: (+49)(0)30/221609359
Handy: (+49)(0)1788378772



From zeynep at sccn.ucsd.edu  Sun Jan 31 20:18:36 2010
From: zeynep at sccn.ucsd.edu (Zeynep Akalin Acar)
Date: Sun, 31 Jan 2010 20:18:36 -0800
Subject: [Eeglablist] bem_matrix.exe failing
In-Reply-To: <4B62BA97.20408@gmx.de>
References: <4B62BA97.20408@gmx.de>
Message-ID: <8796bba51001312018m35083606k14b488e69d98166d@mail.gmail.com>

Hi Paul,

NFT mesh generation module generates meshes with ~19000 nodes. This
requires a memory of about 3 GB. Also, if you apply IPA, the required
memory will be about 4.8GB. The version of bem_matrix.exe included in
the NFT toolbox was compiled with cygwin, and was limited by the
available physical memory. I am working on getting bem_matrix.exe
compile with Visual Studio which should make it use the swap space as
well. It will probably run very slowly on a 2GB machine if it uses
swap, but it should still work.

Best,
Zeynep


On Fri, Jan 29, 2010 at 2:38 AM, Paul Czienskowski <paul_c at gmx.de> wrote:
> Hey all,
>
> For I'm investigating age-related differences in EEG-Source localization for
> my diploma thesis I'm trying to generate the BEM forward-model from T1 MRI
> Data with NFT. Now I'm at the stage to actually create the forward model,
> but bem_matrix.exe crashes with the Exception STATUS_ACCESS_VIOLATION
> (Stackdump see below). I'm running Matlab on a Windows 7 Professional x64
> Laptop (2GB of RAM, 2,3GHz Core2Duo). I was trying to run it on our
> Citrix-Envireonment, but it crashed there too. Maybe anyone encountered the
> problem too and was able to solve it.
>
> Thanks in advance, Paul
>
>
> Stacktrace:
> Exception: STATUS_ACCESS_VIOLATION at eip=610DD894
> eax=00000000 ebx=00040000 ecx=00010000 edx=00000000 esi=611B3F08
> edi=00000000
> ebp=0028C0B8 esp=0028C0AC
> program=C:\Users\Paul\Programmieren\Matlab\NFT-1.0\bin\bem_matrix.exe, pid
> 4040, thread main
> cs=0023 ds=002B es=002B fs=0053 gs=002B ss=002B
> Stack trace:
> Frame ? ? Function ?Args
> 0028C0B8 ?610DD894 ?(00000000, 611B3F08, 00040000, 5DF00000)
> 00040100 ?610029C7 ?(00000032, 000008BC, 000008D0, 00000030)
> 8648586 [main] bem_matrix 4040 _cygtls::handle_exceptions: Exception:
> STATUS_ACCESS_VIOLATION
> 8859954 [main] bem_matrix 4040 _cygtls::handle_exceptions: Error while
> dumping state (probably corrupted stack)
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



From smakeig at gmail.com  Mon Feb  1 13:26:32 2010
From: smakeig at gmail.com (Scott Makeig)
Date: Mon, 1 Feb 2010 13:26:32 -0800
Subject: [Eeglablist] ICA problem
In-Reply-To: <91D5B97F-6924-45C9-A09E-9583C99DACA4@mac.com>
References: <f81849ea1001290149u587591abh5f9c4b6ecde09338@mail.gmail.com>
	<91D5B97F-6924-45C9-A09E-9583C99DACA4@mac.com>
Message-ID: <9e09b8f01002011326m4ba79c0dw2d80529feca21a73@mail.gmail.com>

Joe is correct that ICA will not converge if the rank of the data matrix is
less than the number of channels. The runica/binica algorithms are supposed
to test the rank of the input data. If two channels are identical, or if
some subset of n channels are otherwise interdependent, then the rank will
be less than the number of channels and PCA reduction should be applied to
remove the redundancy and allow the ICA decomposition to converge.

Arno -- There was a problem with the Matlab rank() function on 64-bit
machines, I believe. Has this been solved and Is the auto rank detection ->
PCA option currently implemented in runica/binica?  Perhaps we could add a
'toy' rank() function pre-test (e.g. finding the rank of a small full-rank
matrix to detect if rank() is working...) ? If so, run the rank test; if
not, then warn the user or build a work-around rank function that will work
properly?

Scott

On Mon, Feb 1, 2010 at 6:56 AM, Joseph Dien <jdien07 at mac.com> wrote:

> When you say "so long", how long do you mean?  While ICA is not by its
> nature a fast procedure, certain datasets can take much longer than usual.
>  For example, I find that if two channels are perfectly correlated (1 or -1)
> then an ICA run will take much longer.  This can happen if the data is mean
> mastoid referenced and both channels are explicitly included in the data
> because they will have a perfect -1 correlation (see Dien, 1998 for
> reference issues).  It can also happen if a channel is shorted out during
> acquisition and the reference channel is explicitly included because then
> they will have a perfect correlation.  Also if two channels are shorted
> together during the data acquisition they will be perfectly correlated with
> each other.  My EP Toolkit (
> https://sourceforge.net/projects/erppcatoolkit/) has code for dealing with
> these situations so you might want to look into it.  It implements an
> automated artifact correction routine that relies on EEGlab's runICA code,
> among!
>  other things.
>
> Cheers!
>
> Joe
>
>
>
> On Jan 29, 2010, at 4:49 AM, peng wang wrote:
>
> > Hi there,
> >
> >    I am using ICA to remove blinks via EEGLab. My dataset has 122
> channels, and it takes so long to compute 122 components.
> >    (1) So I tried to use the option "ncomps" (say, 24) to reduce the
> number of components. However, an error message appears after computing:
> "Matrix dimensions must agree".
> >
> >    (2) Then I tried fastICA instead as following,
> >
> > ==================
> >  sz = size(EEG.data);
> >  nchans = sz(1);
> >  npts = sz(2);
> >  ntrials = sz(3);
> >  clear sz;
> >  nICs = 24;
> >  data = reshape(EEG.data,nchans,npts*ntrials);
> >  [ica,V,W] = fastica(data,'numOfIC',nICs,'approach','symm');
> >  EEG.icasphere = eye(nchans);
> >  EEG.icaact = single(reshape(ica,nICs,npts,ntrials));
> >  EEG.icawinv = V;
> >  EEG.icaweights = W;
> >  EEG = eeg_checkset( EEG );
> >  clear V W ica data;
> >
> >  EEG = pop_saveset( EEG, 'filename','test_raw_ica');
> > ==================
> >
> >    Everything seems fine. But when I reject the blink component via GUI
> of eeglab and load the data again, Something strange happens. It seems the
> amplitude of EEG.data become much smaller, about in  -1~1 range.  Thus I
> wonder whether there was some normalization behind, and how can I correct
> it? The problem would not repeat if I choose the number of components same
> as channels event in fast ICA (e.g. change to "nICs = nchans" in the above
> code).
> >
> >    Thank you for your help.
> >
> > best
> > Peng
> > _______________________________________________
> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> > To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> > For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
>
>
> --------------------------------------------------------------------------------
>
> Joseph Dien,
> Senior Research Scientist
> University of Maryland
> 7005 52nd Avenue
> College Park, MD 20742-0025
>
> E-mail: jdien07 at mac.com
> Phone: 301-226-8848
> Fax: 301-226-8811
> http://homepage.mac.com/jdien07/
>
>
>
>
>
>
>
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Scott Makeig, Research Scientist and Director, Swartz Center for
Computational Neuroscience, Institute for Neural Computation, University of
California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100201/2e9c0ec5/attachment.html>

From arno at ucsd.edu  Mon Feb  1 15:56:58 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Mon, 1 Feb 2010 15:56:58 -0800
Subject: [Eeglablist] ICA problem
In-Reply-To: <9e09b8f01002011326m4ba79c0dw2d80529feca21a73@mail.gmail.com>
References: <f81849ea1001290149u587591abh5f9c4b6ecde09338@mail.gmail.com>
	<91D5B97F-6924-45C9-A09E-9583C99DACA4@mac.com>
	<9e09b8f01002011326m4ba79c0dw2d80529feca21a73@mail.gmail.com>
Message-ID: <DD3D8BCD-5036-41A9-81DC-17459A5EC3BC@ucsd.edu>

> Arno -- There was a problem with the Matlab rank() function on 64- 
> bit machines, I believe. Has this been solved and Is the auto rank  
> detection -> PCA option currently implemented in runica/binica?   
> Perhaps we could add a 'toy' rank() function pre-test (e.g. finding  
> the rank of a small full-rank matrix to detect if rank() is  
> working...) ? If so, run the rank test; if not, then warn the user  
> or build a work-around rank function that will work properly?

This is correct. This was a problem with Matlab 64-bit in 2007. One of  
the EEGLAB user, Sven Hoffman, implemented an alternate computation of  
the rank which we used whenever the Matlab rank is deficient. However,  
the new Matlab rank is fine now.

Arno




From isacor at us.es  Tue Feb  2 00:19:38 2010
From: isacor at us.es (isacor at us.es)
Date: Tue, 02 Feb 2010 09:19:38 +0100
Subject: [Eeglablist] ERSP in STUDY
Message-ID: <f5f0c43b12bfe.4b67ee2a@us.es>

Hello everyone. 
I created a study with 8 subjects and three conditions.The sampling frequency is 1024. The time window ranging from -1800 ms to 500 ms. To compute the ERSP have used the following parameters:
'cycles', [1.5 0.5], 'nfreqs', 100, 'baseline', [-1500 -1400], 'padratio', 4, 'alpha', 0.01
When I visualize ERSP activity for a certain channel, I hope that no significant activity was plotted in green. However this does not happen, or not take into account the level of significance.
Do not know if I used the appropriate parameters. When I do a merged with my 8 subjects for a condition and I use the same parameters from eeglab through 'channel time-frequency' , non-significant features are plotted in green.
In the study to the time of making the channels precompute've left anything out?
I hope your help.
Thank you.

Isabel Cordones Cano
Neurociencia y Comportamiento
Fisiolog?a Animal Y Zoologia
Facultad de Biolog?a, Universidad de Sevilla
Avda. Reina Mercedes 6, 41012-Sevilla
Espa?a



From pwang.list at googlemail.com  Tue Feb  2 00:14:51 2010
From: pwang.list at googlemail.com (peng wang)
Date: Tue, 2 Feb 2010 09:14:51 +0100
Subject: [Eeglablist] ICA problem
In-Reply-To: <DD3D8BCD-5036-41A9-81DC-17459A5EC3BC@ucsd.edu>
References: <f81849ea1001290149u587591abh5f9c4b6ecde09338@mail.gmail.com>
	<91D5B97F-6924-45C9-A09E-9583C99DACA4@mac.com>
	<9e09b8f01002011326m4ba79c0dw2d80529feca21a73@mail.gmail.com>
	<DD3D8BCD-5036-41A9-81DC-17459A5EC3BC@ucsd.edu>
Message-ID: <f81849ea1002020014y71013b21yc39c5e7b8c1307cb@mail.gmail.com>

Hi guys,

   Thank you for your discussion. However, I did not see a direct answer to
my question. My problem is, when the number of component is smaller than the
number of channels, something wrong would happen. It seems that this problem
appears after computing ICA. I would appreciate to to know how to solve this
problem.
   The ICA would converge in my situation. For Joe's question on how long,
it's about one night for one dataset (122 channels, 240 trials, 3000 data
points for each), if I take the same number of IC as the channels. I have
288 datasets, so it is too long to be acceptable. Probably, I can downsample
the dataset to reduce the time for computing ICA, but then I still need to
pick the noise components (eg. blinks) out of ~120 ones, this manual work
also takes a lot of time. Thus I wish I can have less ICs.
   Thanks again.
btw:   I am using matlab R2009a on Vista SP1, both 64-bit.

On Tue, Feb 2, 2010 at 12:56 AM, Arnaud Delorme <arno at ucsd.edu> wrote:

> > Arno -- There was a problem with the Matlab rank() function on 64-
> > bit machines, I believe. Has this been solved and Is the auto rank
> > detection -> PCA option currently implemented in runica/binica?
> > Perhaps we could add a 'toy' rank() function pre-test (e.g. finding
> > the rank of a small full-rank matrix to detect if rank() is
> > working...) ? If so, run the rank test; if not, then warn the user
> > or build a work-around rank function that will work properly?
>
> This is correct. This was a problem with Matlab 64-bit in 2007. One of
> the EEGLAB user, Sven Hoffman, implemented an alternate computation of
> the rank which we used whenever the Matlab rank is deficient. However,
> the new Matlab rank is fine now.
>
> Arno
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100202/ba4af4b6/attachment.html>

From demiral.007 at googlemail.com  Tue Feb  2 14:19:32 2010
From: demiral.007 at googlemail.com (Baris Demiral)
Date: Tue, 2 Feb 2010 22:19:32 +0000
Subject: [Eeglablist] ERSP parameters in STUDY
In-Reply-To: <6d898bac1001281328k430e8789u7c53c52fd049f7ba@mail.gmail.com>
References: <f423beba1001270734t3dac0a2cmb3b58bbc4721b844@mail.gmail.com>
	<6d898bac1001271055l2e89fd86y6c2848e62d54e233@mail.gmail.com>
	<f423beba1001280539r38a0baf3mf5211df04044a3a9@mail.gmail.com>
	<6d898bac1001281328k430e8789u7c53c52fd049f7ba@mail.gmail.com>
Message-ID: <f423beba1002021419x1326ec9aiebad35b70a53d726@mail.gmail.com>

Thank you Bradley for your comments,
I did try doing it without baselining, and still the same. My experience now
tells me:
1- Baselining does not have any effect on observable frequencies in ersp
analysis
2- cycles [] option and the epoch length have the major effects, such that
you need to specify smaller wavelets (as in cycle [1 0.2] or DFT option
cycle[0]) if you need to look at the smaller frequencies, but this may not
show the whole epoch (time window is smaller in ersp window) and cuts out
the first and the last portion of the original epochs.

My question to you all is: how can we keep the time range same as long as
the epoch length for the lower frequencies? Is the wavelet overlapping has
anything to do with that? How do we set this?I know the factor in cycles [x
(factor)] determines the increments of the cycles as we go up in the
frequency dimension, but which parameter assigns the wavelet overlap?

Baris

On Thu, Jan 28, 2010 at 9:28 PM, Bradley Voytek <bradley.voytek at gmail.com>wrote:

> Well if you set the basline to be only 100ms long, then you can't
> reasonably expect to look at frequencies below about 10Hz. The
> rationale being that a 10Hz oscillation has--by definition--10
> oscillations per second, or a period of 100ms. If you work with the
> definition that to get a reasonable measure of TF baseline activity
> you need at least one full oscillation, then you can only look at
> frequencies with periods of 100ms or shorter.
>
> In short, use a longer baseline.
>
> ::brad
>
> On Thu, Jan 28, 2010 at 05:39, Baris Demiral <demiral.007 at googlemail.com>
> wrote:
> > Data is filtered  highpass 0.3Hz and lowpass 40Hz, and epochs were set
> from
> > -300ms to 1000ms to event onset. Baseline is -100-0ms. I wonder there are
> > some parameters I should set beforehand.  Any ideas? Or at least, am I
> the
> > only one experiencing this problem, or is this a bug? That is what I am
> > trying to understand.
> >
> > On Wed, Jan 27, 2010 at 6:55 PM, Bradley Voytek <
> bradley.voytek at gmail.com>
> > wrote:
> >>
> >> Probably the length of time you're using for your event window and/or
> >> baseline. What are these settings?
> >>
> >> ::brad
> >>
> >> On Wed, Jan 27, 2010 at 07:34, Baris Demiral <
> demiral.007 at googlemail.com>
> >> wrote:
> >> > Dear all,
> >> > When I run ERSP pre-computation for a study with default parameters I
> >> > end up
> >> > with a frequency-time plot starting from 12Hz. No value below is
> shown.
> >> > I use EEGLABv7.2. What do you think is the problem?
> >> > Baris
> >> > --
> >> > SB Demiral, PhD.
> >> > Department of Psychology
> >> > 7 George Square
> >> > The University of Edinburgh
> >> > Edinburgh, EH8 9JZ
> >> > UK
> >> > Phone: +44 (0131) 6503063
> >> >
> >> > _______________________________________________
> >> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> >> > To unsubscribe, send an empty email to
> >> > eeglablist-unsubscribe at sccn.ucsd.edu
> >> > For digest mode, send an email with the subject "set digest mime" to
> >> > eeglablist-request at sccn.ucsd.edu
> >> >
> >
> >
> >
> > --
> > SB Demiral, PhD.
> > Department of Psychology
> > 7 George Square
> > The University of Edinburgh
> > Edinburgh, EH8 9JZ
> > UK
> > Phone: +44 (0131) 6503063
> >
>



-- 
SB Demiral, PhD.
Department of Psychology
7 George Square
The University of Edinburgh
Edinburgh, EH8 9JZ
UK
Phone: +44 (0131) 6503063
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100202/50fd5e1f/attachment.html>

From demiral.007 at googlemail.com  Tue Feb  2 14:04:37 2010
From: demiral.007 at googlemail.com (Baris Demiral)
Date: Tue, 2 Feb 2010 22:04:37 +0000
Subject: [Eeglablist] ERSP in STUDY
In-Reply-To: <f5f0c43b12bfe.4b67ee2a@us.es>
References: <f5f0c43b12bfe.4b67ee2a@us.es>
Message-ID: <f423beba1002021404r39c1e1e6w68870cd592e433a9@mail.gmail.com>

Hi Isabel,

I do not have a direct answer to your question, but as someone trying to
learn the pros and cons of the ersp analysis from study files, I find your
point very important. You mentioned in your message that you merged the data
of 8 subjects per condition and then did channel time-frequency analysis.
This means that you formed data sets of aggregate ERP amplitude-times for
each condition, and compared the two samples. So, in this case, you did a
t-test (is it Arno?) This may have led to a Type 2 error, showing you the
significant things as non-significant. Actually, my question at this point
is: in ERSP stats (parametric) in eeglab do we do Anova having subjects as
random factors?

Best,
Baris

On Tue, Feb 2, 2010 at 8:19 AM, <isacor at us.es> wrote:

> Hello everyone.
> I created a study with 8 subjects and three conditions.The sampling
> frequency is 1024. The time window ranging from -1800 ms to 500 ms. To
> compute the ERSP have used the following parameters:
> 'cycles', [1.5 0.5], 'nfreqs', 100, 'baseline', [-1500 -1400], 'padratio',
> 4, 'alpha', 0.01
> When I visualize ERSP activity for a certain channel, I hope that no
> significant activity was plotted in green. However this does not happen, or
> not take into account the level of significance.
> Do not know if I used the appropriate parameters. When I do a merged with
> my 8 subjects for a condition and I use the same parameters from eeglab
> through 'channel time-frequency' , non-significant features are plotted in
> green.
> In the study to the time of making the channels precompute've left anything
> out?
> I hope your help.
> Thank you.
>
> Isabel Cordones Cano
> Neurociencia y Comportamiento
> Fisiolog?a Animal Y Zoologia
> Facultad de Biolog?a, Universidad de Sevilla
> Avda. Reina Mercedes 6, 41012-Sevilla
> Espa?a
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
SB Demiral, PhD.
Department of Psychology
7 George Square
The University of Edinburgh
Edinburgh, EH8 9JZ
UK
Phone: +44 (0131) 6503063
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100202/1893cfc2/attachment.html>

From isacor at us.es  Wed Feb  3 03:33:38 2010
From: isacor at us.es (isacor at us.es)
Date: Wed, 03 Feb 2010 12:33:38 +0100
Subject: [Eeglablist] ERSP in STUDY
Message-ID: <f67dd3fc1fbe5.4b696d22@us.es>

Hi Baris,
I think that I have not explained clearly my problem with the statistics on
the study. If I performe a statistical analysis to the ERSP in one dataset
giving a p<0.05 with respect to the baseline, the figure is correct (values
no significant  are displayed in green and increases and decreases of power
as red and blue, respectively). However, when I try to do the same
statistical test in a study (only one condition with respect to the
baseline) the resulting picture is the same as if not probability
calculation were made. How can we test for several datasets (from different
subjects performing the same task) if changes in ERSP and ITC are
significant?


Isabel Cordones Cano
Neurociencia y Comportamiento
Fisiolog?a Animal Y Zoologia
Facultad de Biolog?a, Universidad de Sevilla
Avda. Reina Mercedes 6, 41012-Sevilla
Espa?a



From Lars.Michels at kispi.uzh.ch  Wed Feb  3 09:31:02 2010
From: Lars.Michels at kispi.uzh.ch (Michels Lars)
Date: Wed, 3 Feb 2010 18:31:02 +0100
Subject: [Eeglablist] extracting epochs
Message-ID: <FD1E6887BC99C341800CBFD7ABC2F9C201EB0D31@exzh1.kispi.int>

Hello everyone,

 

I have a .set file with 120 epochs (segmented data, epoch length: -400
ms to 1500 ms). Now I want to extract only those epochs with type =
Picture and Code = 11 and 12 (Picture and Code have been read from a
Presentation log file and they appear under event values, i.e. Picture
as type and 11 and 12 as Codes)

 

I tried to extract via Edit -> Select epochs or events and selected type
= Picture and Code 11 and 12. However, I always receive the following
error message: Reference to non-existent field 'minDuration'. I also
selected min = 400 ms and max = 1500 ms (same error). Is there maybe
another way to extract the desired epochs?

 

Any kind direction will be greatly appreciated.

 

Best wishes

Lars
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100203/0ad8cab1/attachment.html>

From Lars.Michels at kispi.uzh.ch  Wed Feb  3 10:29:22 2010
From: Lars.Michels at kispi.uzh.ch (Michels Lars)
Date: Wed, 3 Feb 2010 19:29:22 +0100
Subject: [Eeglablist] merging two different data sets
Message-ID: <FD1E6887BC99C341800CBFD7ABC2F9C201EB0D37@exzh1.kispi.int>

Hello everyone,

 

I have two different types of data (data from 1 EKG channel and one MR
data set) with the same epoch length, which are recorded at the same
time. Next, I have imported both data sets to EEGlab (as mat files). Now
I want to calculate an ICA across the two data sets. However, I do not
want to concatenate both data sets (append data set function in EEGlab)
but rather to treat them as channel 1 and 2. 

 

First question: is there a way in EEGlab to merge the data sets so that
they are stored in one data set but appear as two different channels?

 

Second question: Does is make sense to calculate an ICA on two -or more
in general- one only few channels?

 

Third question: For my example, a topographical plot of the single ICs
seems not to be possible, because I do not have EEG data and therefore
also no electrode location file linked. Most probably, I have to define
arbitrary EEG coordinates for my two channels?

 

Any comment is highly appreciated! 

 

Thank you!

 

Best regards,

Lars

 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100203/28bd5175/attachment.html>

From jamixlee at gmail.com  Wed Feb  3 16:56:01 2010
From: jamixlee at gmail.com (Chung-yeon Lee)
Date: Thu, 4 Feb 2010 09:56:01 +0900
Subject: [Eeglablist] Eliminaton of the Individual Differences
Message-ID: <daa37f581002031656r6df55a33lbb9c040b71b33fa2@mail.gmail.com>

Hello all,

I got some eeg dataset from 10 subjects.
Here I've met a problem which is that data are individually different.
For example, subject A, B and C each have 3 dataset and these data need to
make clusters.
However, they don't due to the individual differences.
How can I eliminate the differences and classify all the data?
Please let me know if you have success with this matter.

Any comment will be greatly appreciated.

Best wishes!

-- 
Chung-yeon Lee
Division of Multimedia Engineering, College of Engineering, Sungkyul
University
email : jamixlee at gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100204/7de72087/attachment.html>

From ronitibon at gmail.com  Thu Feb  4 00:34:27 2010
From: ronitibon at gmail.com (Roni Tibon)
Date: Thu, 4 Feb 2010 10:34:27 +0200
Subject: [Eeglablist] exporting eeglab data to fieldtrip
Message-ID: <b2007e651002040034l1828243dub9c4979e3d55d5b1@mail.gmail.com>

Hello,

I'm trying to export data from eeglab to feildtrip using the
eeglab2fieldtrip function, in order to perform frequency analysis in
fieldtrip, but I'm getting a message saying "freqanalysis fieldbox not
implemented yet".

This was a long intro, just to raise the question -  any chance it's going
to be implemented soon?

Thanks!
Roni
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100204/85a72e5b/attachment.html>

From eskappenman at ucdavis.edu  Thu Feb  4 14:31:00 2010
From: eskappenman at ucdavis.edu (Emily Kappenman)
Date: Thu, 4 Feb 2010 14:31:00 -0800
Subject: [Eeglablist] 2010 ERP Boot Camp: Applications Due March 1
Message-ID: <9dd17e4a1002041431o24b3e509x8ec2c90e3b44fcdc@mail.gmail.com>

The UC-Davis?ERP?Boot?Camp, an NIH-funded summer workshop on the?ERP
technique, will be held July 12-21?2010. (For additional information,
see?www.ERPinfo.org/the-erp-bootcamp).

The?ERP?Boot?Camp?is a 10-day introduction to the?ERP?technique held
each summer at UC Davis. It is intended for beginning and intermediate
ERP?researchers, and for both basic scientists and
clinical/translational researchers.

?The topics will include:

1) Where do ERPs come from? What do they mean?

2)?ERP?components

3) The design and interpretation of?ERP?experiments

4) EEG data acquisition

5) Filtering, artifact rejection, and artifact correction

6) Measuring and analyzing?ERP?components

7)?ERP?localization

8) Setting up and running an?ERP?lab

The?Boot?Camp?consists of lectures on these topics, structured
discussions, individual consultations, and a substantial laboratory
component.?It is led by Steve Luck, and the faculty includes many
distinguished?ERP?researchers from UC Davis and other universities.

Participants at previous?Boot?Camps have come from around the world
and have ranged from beginning graduate students to full professors.
They have included psychologists, neuroscientists, psychiatrists,
neurologists, and speech pathologists.?Typically, we expect that
students and postdocs should have had at least 6 months of significant
ERP?(or related) experience before attending the?Boot?Camp.

We strongly encourage the participation of individuals from
underrepresented groups.

Funding is available from NIMH to defray some or all of the costs of
attending the?Boot?Camp, and scholarships will be provided to all
participants who are U.S. citizens or permanent residents.
Scholarships may also be provided to some international participants,
but this is not guaranteed.?We typically accept 25-28 U.S. citizens
and permanent residents, along with 2-5 international participants.

The application consists of a CV, a 1-2 page statement of background
and interests, and (for students and postdocs) a letter of
recommendation.

We will begin accepting applications for the?2010?session (July 12-21)
in early January. Applications are due on March 1,?2010?and are
submitted electronically via?www.ERPinfo.org/the-erp-bootcamp.

--
--------------------------------------------------------------------
Emily S. Kappenman
UC Davis Center for Mind and Brain
267 Cousteau Place
Davis, CA 95618
eskappenman at ucdavis.edu
--------------------------------------------------------------------



From jamixlee at gmail.com  Thu Feb  4 22:52:42 2010
From: jamixlee at gmail.com (Chung-yeon Lee)
Date: Fri, 5 Feb 2010 15:52:42 +0900
Subject: [Eeglablist] Eliminaton of the Individual Differences
Message-ID: <daa37f581002042252n535a9a6fs9384b72bfa621e38@mail.gmail.com>

Hello all,

I got some eeg dataset from 10 subjects.
Here I've met a problem which is that data are individually different.
For example, subject A, B and C each have 3 dataset
that is related with their emotional status: happy, angry, neutral.
I expect that these data would make 3 clusters according to the emotions.
However, they don't make any clusters and I think it's because of the
individual variances.
How can I eliminate the differences
and classify all the data as they represent each emotional status?
Please let me know if you have succeed with this matter.
Further more, if you send me EEG data which related with this work, it would
be greatly appreciated.

Best wishes!

-- 
Chung-yeon Lee
Division of Multimedia Engineering, College of Engineering, Sungkyul
University
phone: 82-10-7510-7150
email : jamixlee at gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100205/e8af3bb9/attachment.html>

From sjluck at ucdavis.edu  Fri Feb  5 11:47:41 2010
From: sjluck at ucdavis.edu (Steve Luck)
Date: Fri, 5 Feb 2010 11:47:41 -0800
Subject: [Eeglablist] Job opening for research assistant
Message-ID: <261B5CAC-588E-4591-99B5-F81684213CB0@ucdavis.edu>

Job Opening: Research Assistant, Center for Mind & Brain, University of California-Davis

We are seeking a recent college graduate for a research assistant / lab manager position with a duration of 1-2 years.  Our laboratory focuses on visual attention and visual working memory, including collaborative research on cognitive dysfunction in schizophrenia, and we use a combination of psychophysical measures, eye tracking, and ERPs.  The position involves a mix of research and lab management.  The ideal candidate would have excellent interpersonal and organizational skills, substantial research experience using at least one of our laboratory?s main methods, and experience or coursework in computer programming (especially programming of stimuli in Matlab, E-Prime, or similar package). This is an excellent position for someone who would like to obtain additional research experience before going to graduate school.  To apply, send a cover letter, resume, and a letter of recommendation to Steve Luck (sjluck at ucdavis.edu).  The position will remain open until filled, and we anticipate a start date in June, 2010.  Starting salary is $33,672/year. The University of California is an affirmative action/equal opportunity employer.







-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100205/d9c809f0/attachment.html>

From arno at ucsd.edu  Fri Feb  5 16:27:20 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Fri, 5 Feb 2010 16:27:20 -0800
Subject: [Eeglablist] ERSP parameters in STUDY
In-Reply-To: <f423beba1001270734t3dac0a2cmb3b58bbc4721b844@mail.gmail.com>
References: <f423beba1001270734t3dac0a2cmb3b58bbc4721b844@mail.gmail.com>
Message-ID: <E0C8C48F-F92C-460E-AE40-00971A9A3127@ucsd.edu>

Dear Baris,

You can set the default lowest frequency to be lower (3 Hz for  
instance).
The default lowest frequency is calculated based on your epoch length  
and in some case, the default is too high.
I suggest using frequency limits defined manually "'freqs', [3 50]" in  
the ERSP/ITC parameter edit box.

Hope this helps,

Arno


> When I run ERSP pre-computation for a study with default parameters  
> I end up with a frequency-time plot starting from 12Hz. No value  
> below is shown.
> I use EEGLABv7.2. What do you think is the problem?
>
> Baris



From Kelly.Jantzen at wwu.edu  Mon Feb  8 13:55:24 2010
From: Kelly.Jantzen at wwu.edu (Kelly Jantzen)
Date: Mon, 8 Feb 2010 13:55:24 -0800
Subject: [Eeglablist] Status channel in Biosemi
Message-ID: <366AA6C6662BAB4796FF54F8D8BFC742045EB4FA05@ExchMailbox1.univ.dir.wwu.edu>

In previous versions of EEGLab I was able to load biosemi files using the Biosig toolbox extensions in EELab.  Because I use the additional information on the second 8 bits of the status channel I was leaving the status channel intact (i.e. not discarding trigger channel) and decoding it myself.  With a recent download of EEGLab the status channel is no longer available as a seperate channel - despite setting the rmeventchan  option to off.

Does anyone know if this is due to a recent change in EEGlab or in Biosig?  Is there a possible work around?  All suggestions are welcome.

Thanks

KJ

K.J. Jantzen, Ph.D.
Assistant Professor, Psychology
Western Washington University
Academic Instructional Center 584
516 High Street, Bellingham
Washington, 98225
email: Kelly.Jantzen at wwu.edu
phone: (360) 650 4046




From dr.ilya at yahoo.com  Mon Feb  8 16:05:06 2010
From: dr.ilya at yahoo.com (Ilya Adamchic)
Date: Mon, 8 Feb 2010 16:05:06 -0800 (PST)
Subject: [Eeglablist] Channel import
In-Reply-To: <8796bba51001312018m35083606k14b488e69d98166d@mail.gmail.com>
References: <4B62BA97.20408@gmx.de>
	<8796bba51001312018m35083606k14b488e69d98166d@mail.gmail.com>
Message-ID: <920105.18048.qm@web63806.mail.re1.yahoo.com>

Hello dear Colleagues!


I'm having an issue, with what seemed to be easy task. I try to import my
scanned cannel location in to EEGLAB from EGI system (*.spf file in Cartesian system). I have 128 EEG channels and 3 fiducial points, all of this are recognized
o.k. but as soon as I try to accept these channel locations by pressing OK
button in Edit EEG channel window I get the following message: ?The number of
data channels (131) not including fiducials does not correspond to the initial
numbed of channels (128), so for consistency purposes...and so on..will not accepted. ". 

My data
set contains 128 channels + 3 fisucials=131....so what am I missing...?

Thanks a lot for all yours help.
Best regards!

Ilya 



      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100208/39e3421d/attachment.html>

From mklados at med.auth.gr  Tue Feb  9 16:20:30 2010
From: mklados at med.auth.gr (Klados Manousos)
Date: Wed, 10 Feb 2010 02:20:30 +0200
Subject: [Eeglablist] =?utf-8?q?THE_12TH_MEDITERRANEAN_CONFERENCE_ON_MEDIC?=
	=?utf-8?q?AL_AND_BIOLOGICAL_ENGINEERING_AND_COMPUTING_=E2=80=93_ME?=
	=?utf-8?q?DICON_2010?=
Message-ID: <3b0db4861002091620t1e9164cdk11ed98b7e1641cb8@mail.gmail.com>

Dear Colleague,

please, accept our advanced apologies for any multiple cross-postings...

Please, be aware that the deadline for paper submission for the 12th
Mediterranean Conference on Medical and Biological Engineering and
Computing (MEDICON 2010) is fast approaching (14th of February 2010) !

Please, submit your paper contributions online through the conference
web site (http://www.medicon2010.org/paper-submission.html) no later
than 14/2/2010 !

MEDICON 2010 will be held between May 27-30, 2010, in the magnificent
Porto Carras Resort of Chalkidiki in Greece.

We remind you that there are 6 good reasons to submit papers and
attend MEDICON 2010:
-High standard publication of paper contributions (published within
the peer reviewed Springer Verlag IFBME Proceedings Series).
-Very promising prospects for paper submissions as a number of special
issues are planned in international journals (for extended versions of
papers presented at MEDICON 2010; journals considered include:
Biomedical Signal Processing and Control, Computer Methods and
Programs in Biomedicine, Medical & Biological Engineering & Computing,
Hippokratia).
-Relatively low registration fees (only 350? for regular; 200? for student)
-Very good value for money accommodation arrangements (ranging from as
low as 55 up to 116?)
-An excellent social program (with a post-conference trip to Mount
Athos Monasteries)
-An attractive program with very important keynote speeches in a
magnificent resort for the whole event

The MEDICON conferences are of long lasting tradition, held every
third year in one of the Mediterranean countries under the auspices of
the International Federation for Medical and Biological Engineering
(IFMBE). MEDICON 2010 aims to present recent advances in practically
every main topic of BME (Biomaterials, Biomechanics,
Bioinstrumentation, Medical Information Processing and Management,
Technology Assessment, Clinical Engineering, Rehabilitation
Engineering, ehealth,Educational, Ethical and Professional aspects of
BME etc).

We promise to do our best to make MEDICON 2010 a memorable event.

>From the Conference Chairs

Panos Bamidis, Aristotle University of Thessaloniki, Greece
Nicolas Pallikarakis, University of Patras, Greece
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100210/0aa3aa4f/attachment.html>

From dr.ilya at yahoo.com  Thu Feb 11 05:44:34 2010
From: dr.ilya at yahoo.com (Ilya Adamchic)
Date: Thu, 11 Feb 2010 05:44:34 -0800 (PST)
Subject: [Eeglablist] Channel import
Message-ID: <941189.9163.qm@web63804.mail.re1.yahoo.com>

Hi Joe,

thanks for your reply!
I have tried various options, including using 129 channels (128 + reference) and i always get the same warning. The only way EEGLAB accepts channel locations is, when i delete all fiducials, so that number of channel locations exactly corresponds the number of actual channels (columns of data). Since I need to use DIPFIT co-registration function, i need all of them including fiducials, to align my ellectrodes.
Cheers,

Ilya




________________________________
From: Joseph Dien <jdien07 at mac.com>
To: Ilya Adamchic <dr.ilya at yahoo.com>
Cc: eeglablist at sccn.ucsd.edu
Sent: Thu, February 11, 2010 4:43:06 AM
Subject: Re: [Eeglablist] Channel import


What about your 129th channel?  The EGI 128-channel system actually has 129 channels when you include the reference channel (Cz).

Cheers!

Joe

On Feb 8, 2010, at 7:05 PM, Ilya Adamchic wrote:

Hello dear Colleagues!
>
>
>
>I'm having an issue, with what seemed to be easy task. I try to import my scanned cannel location in to EEGLAB from EGI system (*.spf file in Cartesian system). I have 128 EEG channels and 3 fiducial points, all of this are recognized o.k. but as soon as I try to accept these channel locations by pressing OK button in Edit EEG channel window I get the following message: ?The number of data channels (131) not including fiducials does not correspond to the initial numbed of channels (128), so for consistency purposes...and so on..will not accepted. ". 
>
>My data set contains 128 channels + 3 fisucials=131....so what am I missing...?
>
>
>Thanks a lot for all yours help.
>Best regards!
>
>Ilya 
>
>_______________________________________________
>Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
>For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu


--------------------------------------------------------------------------------

Joseph Dien,
Senior Research Scientist
University of Maryland 
7005 52nd Avenue
College Park, MD 20742-0025


E-mail: jdien07 at mac.com
Phone: 301-226-8848
Fax: 301-226-8811
http://homepage.mac.com/jdien07/


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100211/23ca459c/attachment.html>

From jdien07 at mac.com  Wed Feb 10 19:43:06 2010
From: jdien07 at mac.com (Joseph Dien)
Date: Wed, 10 Feb 2010 22:43:06 -0500
Subject: [Eeglablist] Channel import
In-Reply-To: <920105.18048.qm@web63806.mail.re1.yahoo.com>
References: <4B62BA97.20408@gmx.de>
	<8796bba51001312018m35083606k14b488e69d98166d@mail.gmail.com>
	<920105.18048.qm@web63806.mail.re1.yahoo.com>
Message-ID: <BE7E29B7-C17E-44E2-BD66-37F1A97E6134@mac.com>

What about your 129th channel?  The EGI 128-channel system actually has 129 channels when you include the reference channel (Cz).

Cheers!

Joe

On Feb 8, 2010, at 7:05 PM, Ilya Adamchic wrote:

> Hello dear Colleagues!
> 
> I'm having an issue, with what seemed to be easy task. I try to import my scanned cannel location in to EEGLAB from EGI system (*.spf file in Cartesian system). I have 128 EEG channels and 3 fiducial points, all of this are recognized o.k. but as soon as I try to accept these channel locations by pressing OK button in Edit EEG channel window I get the following message: ?The number of data channels (131) not including fiducials does not correspond to the initial numbed of channels (128), so for consistency purposes...and so on..will not accepted. ". 
> My data set contains 128 channels + 3 fisucials=131....so what am I missing...?
> 
> Thanks a lot for all yours help.
> Best regards!
> Ilya 
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu


--------------------------------------------------------------------------------

Joseph Dien,
Senior Research Scientist
University of Maryland 
7005 52nd Avenue
College Park, MD 20742-0025

E-mail: jdien07 at mac.com
Phone: 301-226-8848
Fax: 301-226-8811
http://homepage.mac.com/jdien07/








-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100210/ce754778/attachment.html>

From tarikbelbahar at gmail.com  Thu Feb 11 10:57:24 2010
From: tarikbelbahar at gmail.com (Tarik S Bel-Bahar)
Date: Thu, 11 Feb 2010 18:57:24 +0000
Subject: [Eeglablist] Channel import
In-Reply-To: <920105.18048.qm@web63806.mail.re1.yahoo.com>
References: <4B62BA97.20408@gmx.de>
	<8796bba51001312018m35083606k14b488e69d98166d@mail.gmail.com>
	<920105.18048.qm@web63806.mail.re1.yahoo.com>
Message-ID: <38e1ab891002111057tef7f185i1ec6e838d5a9c250@mail.gmail.com>

Hello Ilya,

I've had a similar issue, and it's not clear from documentation or
eeglablist
what exactly is needed. I think more detailed step by step instructions
are needed to make it so that users don't get stuck at these basic steps.
I've developed some workarounds, but I have not been doing source
localization.

1. First if the data is re-referenced or average referenced before going
into eeglab,
then don't export the reference channel from Netstation.

2 Second you may want to use a different file of electrode locations
that matches your data, i.e., has 131 locations, or that recognizes
fiducials.

Good luck and do let us know what you figure out so others don't suffer!

All the best,
Tarik


********************************************************************************************************************************
Tarik Bel-Bahar, PhD
Department of Clinical, Educational, and Health Psychology/ University
College London
AFC-UCL Developmental Neuroscience Unit / Anna Freud Centre, 21, Maresfield
Gardens/ London NW3 5SD
Tel: +44 (0) 20 7443 2212 / Receptionist: +44 (0) 20 7794 2313/ Fax: + 44
(0) 20 77946506
********************************************************************************************************************************


On Tue, Feb 9, 2010 at 12:05 AM, Ilya Adamchic <dr.ilya at yahoo.com> wrote:

> Hello dear Colleagues!
>
>
> I'm having an issue, with what seemed to be easy task. I try to import my
> scanned cannel location in to EEGLAB from EGI system (*.spf file in *
> Cartesian* system). I have 128 EEG channels and 3 fiducial points, all of
> this are recognized o.k. but as soon as I try to accept these channel
> locations by pressing OK button in Edit EEG channel window I get the
> following message: ?The number of data channels (131) not including
> fiducials does not correspond to the initial numbed of channels (128), so
> for consistency purposes...and so on..will not accepted. ".
>
> My data set contains 128 channels + 3 fisucials=131....so what am I
> missing...?
>
>
> Thanks a lot for all yours help.
>
> Best regards!
>
> Ilya
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100211/cee6eccc/attachment.html>

From mjw19 at psu.edu  Fri Feb 12 09:41:49 2010
From: mjw19 at psu.edu (Michael Wenger)
Date: Fri, 12 Feb 2010 12:41:49 -0500
Subject: [Eeglablist] shielding requirements, weather radar
Message-ID: <47f13611002120941y2ad68510v4d5642f3ede162d8@mail.gmail.com>

All -- can someone point me to a good source for how to estimate
shielding requirements in order to eliminate possible interference
from a high-power weather radar? Feel free to reply off-list, and
thanks in advance for any and all assistance.

-Michael

-- 
M. J. Wenger, Ph.D.
Department of Psychology
Graduate Program in Neuroscience
Huck Institutes of the Life Sciences
Director, Human Electrophysiology Facilitiy
The Pennsylvania State University
University Park PA USA 16803

phone: 814.863.6023
fax: 814.863.7002
e-mail: mjw19 at psu dot edu
web: www.personal.psu.edu/mjw19


From marinafaveri at yahoo.com.br  Thu Feb 11 10:17:38 2010
From: marinafaveri at yahoo.com.br (marina oliveira)
Date: Thu, 11 Feb 2010 10:17:38 -0800 (PST)
Subject: [Eeglablist] Res:  Channel import
In-Reply-To: <920105.18048.qm@web63806.mail.re1.yahoo.com>
References: <4B62BA97.20408@gmx.de>
	<8796bba51001312018m35083606k14b488e69d98166d@mail.gmail.com>
	<920105.18048.qm@web63806.mail.re1.yahoo.com>
Message-ID: <809614.68983.qm@web51501.mail.re2.yahoo.com>

Hi Ilya!
you need to uncheck the box 'channel in data array (set = yes)' for the fiducial channels, since they exist, but don't record any data. their position will remain for loccations purposes, but eeglab will know that they are not in the data matrix.
best wishes,

Marina


Ft. Marina Faveri de Oliveira 
Crefito 88079-F 
Cel: 11.9450.1457




________________________________
De: Ilya Adamchic <dr.ilya at yahoo.com>
Para: eeglablist at sccn.ucsd.edu
Enviadas: Segunda-feira, 8 de Fevereiro de 2010 22:05:06
Assunto: [Eeglablist] Channel import


 
Hello dear Colleagues!


I'm having an issue, with what seemed to be easy task. I try to import my
scanned cannel location in to EEGLAB from EGI system (*.spf file in Cartesian system). I have 128 EEG channels and 3 fiducial points, all of this are recognized
o.k. but as soon as I try to accept these channel locations by pressing OK
button in Edit EEG channel window I get the following message: ?The number of
data channels (131) not including fiducials does not correspond to the initial
numbed of channels (128), so for consistency purposes...and so on..will not accepted. ". 

My data
set contains 128 channels + 3 fisucials=131....so what am I missing...?

Thanks a lot for all yours help.
Best regards!

Ilya 


      ____________________________________________________________________________________
Veja quais s?o os assuntos do momento no Yahoo! +Buscados
http://br.maisbuscados.yahoo.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100211/42df90a3/attachment.html>

From katy.ericson at yahoo.com  Thu Feb 11 10:20:10 2010
From: katy.ericson at yahoo.com (Katy Ericson)
Date: Thu, 11 Feb 2010 10:20:10 -0800 (PST)
Subject: [Eeglablist] Co-registration procedure DIPFIT
Message-ID: <453512.93251.qm@web114108.mail.gq1.yahoo.com>

Dear all,
I'm
new to the EEGLAB, this seems to be a nice software. Anyway i have some
trouble, if one can call it a trouble at all. I need to find the
locations of equivalent dipoles for my ICA components. My recordings
are done on EGI 128 channel system and then the exact locations of the
sensors are modeled with Photogrametry procedure. As i understand, in
order to get more or less correct locations of dipoles, i have to do a
co-registration procedure (my net does not correspond exactly to any of
10-5 or 10-10 systems and sensor locations are "scanned" for every
patient).
I was trying to do the co-registration using a standard
standard-10-5-cap385.elp file as the template file and spherical model.
This analysis takes quite a long time, when one has to pair all 128
channels to template locations, because my system has channels numbered
as E1 till E128 and Cz as a reference. It may be unreasonable use of
time to sit for several hours trying to co-register electrodes to the
standard template. On the other hand there was a study, which
calculated correspondences of E1:E128 channel names to the 10-5 names.
Which may really speed up the process, but the pairing may be done with
a greater distances between electrodes.
It would be a great help from you, if some one could answer my questions:
1. How big can the errors be, if i use no co-registration and file with the scanned locations.
2.
Is there a way to speed up the process or some other solution of
co-registration and dipole fitting in my case (i should mention though,
if it concerns accuracy, we would rather take a longer processing time,
rather than do the work inaccurately).
3. We have a possibility to
export location file from EGI in spherical and in Cartesian
coordinates, will there be a difference in accuracy and time consume
during co-registration procedure between files in these 2 coordinate
systems.
4.
When one does the co-registration, the locations of electrodes are
changed (custom locations are aligned with predefined ones), how does
this affect the accuracy of dipole fitting. 
Thank a lot for your answers and the great mailing lists:)
Cheers!
Katy


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100211/f519808c/attachment.html>

From dr.ilya at yahoo.com  Fri Feb 12 16:00:26 2010
From: dr.ilya at yahoo.com (Ilya Adamchic)
Date: Fri, 12 Feb 2010 16:00:26 -0800 (PST)
Subject: [Eeglablist]  DIPFIT: changing model template .loc file
Message-ID: <512198.32042.qm@web63806.mail.re1.yahoo.com>

Dear David,
I'm st
Did it work with suggestions from Arno concerning using your own template .loc file?
Could you please share your experience in creating a template .loc file and using it with spherical BESA model. 
As i understood from Arno's answer, one has to select a custom model and then add it in to the DIPFIT set up window as a "self made" template .loc file. Did it work in your case?
I think i have similar situation with yours. I have EGIHydroCell custom coordinates for each patient and it takes to long to manually pair all my custom channels with a standard .loc file given in EEGLAB, it simply takes to much time. 
Did you use your own spherical coordinates file or some standard average? 
I think there are some files for EGIHydroCel in BESA, that one could use, but they are in Carthesian coordinates. 

Arno, is there possibly some other suggestion, how to quicker co-register 128 HydroCell channel location file (with electrodes marked E1-E129)?

Thank you guys for your comments!
Best regards,
Ilya


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100212/a35795af/attachment.html>

From dr.ilya at yahoo.com  Thu Feb 11 11:52:19 2010
From: dr.ilya at yahoo.com (Ilya Adamchic)
Date: Thu, 11 Feb 2010 11:52:19 -0800 (PST)
Subject: [Eeglablist] Channel import
In-Reply-To: <38e1ab891002111057tef7f185i1ec6e838d5a9c250@mail.gmail.com>
References: <4B62BA97.20408@gmx.de>
	<8796bba51001312018m35083606k14b488e69d98166d@mail.gmail.com>
	<920105.18048.qm@web63806.mail.re1.yahoo.com>
	<38e1ab891002111057tef7f185i1ec6e838d5a9c250@mail.gmail.com>
Message-ID: <761660.84306.qm@web63808.mail.re1.yahoo.com>

Hi Tarik,
now it works! Many thanks to Marina Oliveira. 
In this case, i had 128 "Real channels", 3 fiducials and 1 reference. 
I unchecked  the box "Channel in data array (set = yes)" for all "channels": fiducials and reference (so i unchecked this box 4 times).
Than it worked. From 3D plot i can see, that the fiducials and reference are placed where they belong.
I agree, this issue is not good enough described in Manual, probably one could add a couple of additional words, describing how can one technically, in the GUI interface, add fiducials and reference to the data set. 
Marina, thanks again!
Cheers!
Ilya




________________________________
From: Tarik S Bel-Bahar <tarikbelbahar at gmail.com>
To: Ilya Adamchic <dr.ilya at yahoo.com>
Cc: eeglablist at sccn.ucsd.edu
Sent: Thu, February 11, 2010 7:57:24 PM
Subject: Re: [Eeglablist] Channel import

Hello Ilya,

I've had a similar issue, and it's not clear from documentation or eeglablist
what exactly is needed. I think more detailed step by step instructions
are needed to make it so that users don't get stuck at these basic steps.
I've developed some workarounds, but I have not been doing source localization.

1. First if the data is re-referenced or average referenced before going into eeglab,
then don't export the reference channel from Netstation.

2 Second you may want to use a different file of electrode locations
that matches your data, i.e., has 131 locations, or that recognizes fiducials.

Good luck and do let us know what you figure out so others don't suffer!

All the best,
Tarik


********************************************************************************************************************************
Tarik Bel-Bahar, PhD
Department of Clinical, Educational, and Health Psychology/ University College London
AFC-UCL Developmental Neuroscience Unit / Anna Freud Centre, 21, Maresfield Gardens/ London NW3 5SD
Tel: +44 (0) 20 7443 2212 / Receptionist: +44 (0) 20 7794 2313/ Fax: + 44 (0) 20 77946506
********************************************************************************************************************************



On Tue, Feb 9, 2010 at 12:05 AM, Ilya Adamchic <dr.ilya at yahoo.com> wrote:

Hello dear Colleagues!
>
>
>
>I'm having an issue, with what seemed to be easy task. I try to import my scanned cannel location in to EEGLAB from EGI system (*.spf file in Cartesian system). I have 128 EEG channels and 3 fiducial points, all of this are recognized o.k. but as soon as I try to accept these channel locations by pressing OK button in Edit EEG channel window I get the following message: ?The number of data channels (131) not including fiducials does not correspond to the initial numbed of channels (128), so for consistency purposes...and so on..will not accepted. ". 
>
>My data set contains 128 channels + 3 fisucials=131....so what am I missing...?
>
>
>Thanks a lot for all yours help.
>Best regards!
>
>Ilya 
>
>
>_______________________________________________
>Eeglablistpage: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
>For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
>



      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100211/d24c7560/attachment.html>

From jdien07 at mac.com  Mon Feb 15 17:39:53 2010
From: jdien07 at mac.com (Joseph Dien)
Date: Mon, 15 Feb 2010 20:39:53 -0500
Subject: [Eeglablist] ICA problem
In-Reply-To: <DD3D8BCD-5036-41A9-81DC-17459A5EC3BC@ucsd.edu>
References: <f81849ea1001290149u587591abh5f9c4b6ecde09338@mail.gmail.com>
	<91D5B97F-6924-45C9-A09E-9583C99DACA4@mac.com>
	<9e09b8f01002011326m4ba79c0dw2d80529feca21a73@mail.gmail.com>
	<DD3D8BCD-5036-41A9-81DC-17459A5EC3BC@ucsd.edu>
Message-ID: <21F33732-F53F-432F-ACFA-AF2DDF907410@mac.com>

Sounds like I may have some helpful feedback then.  While I can't claim to have done a systematic analysis of this issue, I can say that for the datasets where I looked at this in detail, the rank(data) test didn't get triggered for two reasons:

1) all the instances of it are hidden behind an "if lrate> MIN_LRATE" test.  So what happens at least sometimes is that when the rank is deficient, 
the ICA run goes into repeated rounds of the weights blowing up and the learning rate being reduced.  At the end of this extended process, the lrate is too low to trigger the "if" statement and so the rank test is never invoked; instead, the error message is "runica(): QUITTING - weight matrix may not be invertible!"

2) Even when the lrate test is passed, the rank test may still not be triggered.  The reason, in the case that I looked at in close detail, is that the data matrix was rotated into imaginary numbers and somehow this resulted in the rank going (for a 24 channel dataset) from 23 to 24, hence testing as full rank.

These observations were made on the following installation:

Matlab 7.8.0
EEGlab 7_2_9_19b (not the current version but I looked at the current release of runICA and the code appears unchanged in this respect)
Windows XP Professional SP3

Cheers!

Joe

On Feb 1, 2010, at 6:56 PM, Arnaud Delorme wrote:

>> Arno -- There was a problem with the Matlab rank() function on 64- 
>> bit machines, I believe. Has this been solved and Is the auto rank  
>> detection -> PCA option currently implemented in runica/binica?   
>> Perhaps we could add a 'toy' rank() function pre-test (e.g. finding  
>> the rank of a small full-rank matrix to detect if rank() is  
>> working...) ? If so, run the rank test; if not, then warn the user  
>> or build a work-around rank function that will work properly?
> 
> This is correct. This was a problem with Matlab 64-bit in 2007. One of  
> the EEGLAB user, Sven Hoffman, implemented an alternate computation of  
> the rank which we used whenever the Matlab rank is deficient. However,  
> the new Matlab rank is fine now.
> 
> Arno
> 
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu


--------------------------------------------------------------------------------

Joseph Dien,
Senior Research Scientist
Center for Advanced Study of Language
University of Maryland 
7005 52nd Avenue
College Park, MD 20742-0025

E-mail: jdien07 at mac.com
Phone: 301-226-8848
Fax: 301-226-8811
http://homepage.mac.com/jdien07/










From jdien07 at mac.com  Mon Feb 15 17:25:51 2010
From: jdien07 at mac.com (Joseph Dien)
Date: Mon, 15 Feb 2010 20:25:51 -0500
Subject: [Eeglablist] DIPFIT: changing model template .loc file
In-Reply-To: <512198.32042.qm@web63806.mail.re1.yahoo.com>
References: <512198.32042.qm@web63806.mail.re1.yahoo.com>
Message-ID: <12B2EAE9-4B3A-4FD6-9119-C9779524F89F@mac.com>

If you have the BESA .sfp files then you can just do the following:

1) pop_chanedit([]);

2) click on Read Locations.

3) click on Save (as .ced)

I'm attaching a .ced file that I made in this way for the 129-channel Hydrocel.

Cheers!

Joe

-------------- next part --------------
A non-text attachment was scrubbed...
Name: GSN-Hydrocel-129.ced
Type: application/octet-stream
Size: 7101 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100215/208cef92/attachment.obj>
-------------- next part --------------



On Feb 12, 2010, at 7:00 PM, Ilya Adamchic wrote:

> Dear David,
> I'm st
> Did it work with suggestions from Arno concerning using your own template .loc file?
> Could you please share your experience in creating a template .loc file and using it with spherical BESA model. 
> As i understood from Arno's answer, one has to select a custom model and then add it in to the DIPFIT set up window as a "self made" template .loc file. Did it work in your case?
> I think i have similar situation with yours. I have EGI HydroCell custom coordinates for each patient and it takes to long to manually pair all my custom channels with a standard .loc file given in EEGLAB, it simply takes to much time. 
> Did you use your own spherical coordinates file or some standard average? 
> I think there are some files for EGI HydroCel in BESA, that one could use, but they are in Carthesian coordinates. 
> 
> Arno, is there possibly some other suggestion, how to quicker co-register 128 HydroCell channel location file (with electrodes marked E1-E129)?
> 
> Thank you guys for your comments!
> Best regards,
> Ilya
> 
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu


--------------------------------------------------------------------------------

Joseph Dien,
Senior Research Scientist
Center for Advanced Study of Language
University of Maryland 
7005 52nd Avenue
College Park, MD 20742-0025

E-mail: jdien07 at mac.com
Phone: 301-226-8848
Fax: 301-226-8811
http://homepage.mac.com/jdien07/









From dr.ilya at yahoo.com  Tue Feb 16 08:33:33 2010
From: dr.ilya at yahoo.com (Ilya Adamchic)
Date: Tue, 16 Feb 2010 08:33:33 -0800 (PST)
Subject: [Eeglablist] DIPFIT: changing model template .loc file
In-Reply-To: <bf0e3c851002160745lbd60c89wb4ca583fab692db8@mail.gmail.com>
References: <512198.32042.qm@web63806.mail.re1.yahoo.com>
	<bf0e3c851002160745lbd60c89wb4ca583fab692db8@mail.gmail.com>
Message-ID: <824512.65849.qm@web63805.mail.re1.yahoo.com>

Hi Dave,
thank a lot for your detailed answer. This is the sequence
I follow, unfortunately only with file containing ellipsoid head,
working now on changing it it to the sphere head with my custom
channels. I have read in some previous posts and what i see with my
file, that when you insert custom template files, the co-registration
window does some strange things: shows head model as to small, where
one could not see anything because of the size. Did you have such a
problem?
Do you have some  suggestions on literature to read
(probably what you have already read), regarding head models  used in
DIPFIT and what may be important for co-registration procedure. 
Thanks again for your help,
All the best,
Ilya  



________________________________
From: David Towers <dtowers at illinois.edu>
To: Ilya Adamchic <dr.ilya at yahoo.com>
Cc: Arnaud Delorme <arno at ucsd.edu>; eeglablist at sccn.ucsd.edu
Sent: Tue, February 16, 2010 4:45:15 PM
Subject: Re: [Eeglablist] DIPFIT: changing model template .loc file

Ilya,
I did follow the suggestions with some success.  Within dipfit, I selected "Custom Model", and entered all the default files used when selecting "Spherical Four-Shell (BESA)" except for the "Model template channel locations file".  For that, I put together a file based on the electrode locations of our cap.  I've attached the file as an example (note we use a 59-channel cap with electrodes labeled "EEGX", where X is the number, plus two VEOG channels).  Here's an example of the first few lines from our .elp file:

58
     EEG       VEOG1     127.588      65.992      85.000
     EEG       VEOG2    -130.893     -69.140      85.000
     EEG        EEG1    -109.411      45.735      85.000
     EEG        EEG2    -104.947      70.431      85.000
     EEG        EEG3     102.967     -87.992      85.000
     EEG        EEG4     101.340     -67.224      85.000

There are 58 channels listed since the 59th channel is our reference.

-Dave


On Fri, Feb 12, 2010 at 6:00 PM, Ilya Adamchic <dr.ilya at yahoo.com> wrote:

Dear David,
>I'm st
>Did it work with suggestions from Arno concerning using your own template .loc file?
>>Could you please share your experience in creating a template .loc file and using it with spherical BESA model. 
>As i understood from Arno's answer, one has to select a custom model and then add it in to the DIPFIT set up window as a "self made" template .loc file. Did it work in your case?
>>I think i have similar situation with yours. I have EGI HydroCell custom coordinates for each patient and it takes to long to manually pair all my custom channels with a standard .loc file given in EEGLAB, it simply takes to much time. 
>>Did you use your own spherical coordinates file or some standard
> average? 
>I think there are some files for EGI HydroCel in BESA, that one could use, but they are in Carthesian coordinates. 
>
>Arno, is there possibly some other suggestion, how to quicker co-register 128 HydroCell channel location file (with electrodes marked E1-E129)?
>
>Thank you guys for your comments!
>Best regards,
>Ilya
>
>
>


-- 
David N. Towers, Ph.D.
University of Illinois, Urbana-Champaign
Psychology Department
603 East Daniel Street
Champaign, IL 61820
(217) 244-6085
dtowers at illinois.edu



      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100216/be335ee9/attachment.html>

From dtowers at illinois.edu  Tue Feb 16 07:45:15 2010
From: dtowers at illinois.edu (David Towers)
Date: Tue, 16 Feb 2010 09:45:15 -0600
Subject: [Eeglablist] DIPFIT: changing model template .loc file
In-Reply-To: <512198.32042.qm@web63806.mail.re1.yahoo.com>
References: <512198.32042.qm@web63806.mail.re1.yahoo.com>
Message-ID: <bf0e3c851002160745lbd60c89wb4ca583fab692db8@mail.gmail.com>

Ilya,
I did follow the suggestions with some success.  Within dipfit, I selected
"Custom Model", and entered all the default files used when selecting
"Spherical Four-Shell (BESA)" except for the "Model template channel
locations file".  For that, I put together a file based on the electrode
locations of our cap.  I've attached the file as an example (note we use a
59-channel cap with electrodes labeled "EEGX", where X is the number, plus
two VEOG channels).  Here's an example of the first few lines from our .elp
file:

58
     EEG       VEOG1     127.588      65.992      85.000
     EEG       VEOG2    -130.893     -69.140      85.000
     EEG        EEG1    -109.411      45.735      85.000
     EEG        EEG2    -104.947      70.431      85.000
     EEG        EEG3     102.967     -87.992      85.000
     EEG        EEG4     101.340     -67.224      85.000

There are 58 channels listed since the 59th channel is our reference.

-Dave

On Fri, Feb 12, 2010 at 6:00 PM, Ilya Adamchic <dr.ilya at yahoo.com> wrote:

> Dear David,
> I'm st
> Did it work with suggestions from Arno concerning using your own template
> .loc file?
> Could you please share your experience in creating a template .loc file and
> using it with spherical BESA model.
> As i understood from Arno's answer, one has to select a custom model and
> then add it in to the DIPFIT set up window as a "self made" template .loc
> file. Did it work in your case?
> I think i have similar situation with yours. I have EGI HydroCell custom
> coordinates for each patient and it takes to long to manually pair all my
> custom channels with a standard .loc file given in EEGLAB, it simply takes
> to much time.
> Did you use your own spherical coordinates file or some standard average?
> I think there are some files for EGI HydroCel in BESA, that one could use,
> but they are in Carthesian coordinates.
>
> Arno, is there possibly some other suggestion, how to quicker co-register
> 128 HydroCell channel location file (with electrodes marked E1-E129)?
>
> Thank you guys for your comments!
> Best regards,
> Ilya
>
>
>


-- 
David N. Towers, Ph.D.
University of Illinois, Urbana-Champaign
Psychology Department
603 East Daniel Street
Champaign, IL 61820
(217) 244-6085
dtowers at illinois.edu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100216/16fe7098/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: escw_avg.elp
Type: application/octet-stream
Size: 2613 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100216/16fe7098/attachment.obj>

From katy.ericson at yahoo.com  Tue Feb 16 01:45:43 2010
From: katy.ericson at yahoo.com (Katy Ericson)
Date: Tue, 16 Feb 2010 01:45:43 -0800 (PST)
Subject: [Eeglablist] Co-registration procedure DIPFIT
In-Reply-To: <0DBB603CE6952142B61639251E677C3FA45A81@staffexchange6.ul.campus>
References: <453512.93251.qm@web114108.mail.gq1.yahoo.com>
	<0DBB603CE6952142B61639251E677C3FA45A81@staffexchange6.ul.campus>
Message-ID: <20853.91880.qm@web114101.mail.gq1.yahoo.com>

Hi Kelly,
thanks a lot for your reply. I think i understand what you mean (sorry if i don't ;) ). As i understood ,this is a standard way, when you have a 10-10 or 10-5 marking of your electrodes, which corresponds to the standard 10-5 marking of the file, given in DIPFIT. The problem in my case was, that i have a custom system with E1-E129 electrode marking, which does not correspond to the DIPFIT file. In this case wrap function does not recognize those locations and, from my point of view, there are two options:
1. Do a co-regestarationmanually, which takes tons of time
2. Find a .loc file with custom electrodes from EGI in spherical format, that will represent my 129 electrodes with the names E1-E129, and set it as a template file. In this case the wrap function will be able to recognize location names and will do wrapping automatically. I'm not sure that this will work in EEGLAB, right now looking for an appropriate file to do try this out. 
Best regards,
Katy





________________________________
From: Kelly.Kaneswaran <kelly.kaneswaran at ul.ie>
To: Katy Ericson <katy.ericson at yahoo.com>
Sent: Mon, February 15, 2010 3:34:02 PM
Subject: RE: [Eeglablist] Co-registration procedure DIPFIT

  
Katy,
 
I had similar problem before, follow the turorial it explains a way of doing it that speeds it up. First you should have a channel location file with all the locations in an acceptable format .sph .loc etc. from the locate dipole using dipfit menu , choose head and model settings , leave the default settings and choose manual co registration. then a spherical model pops up , choose warp montage , the list shows the models fudicials and you can select corresponding electrode locations that your model has exact. then press ok , the montage is warped to fit the model and press ok it will generate the tailrach co-ordinates.
 
i could be way off so apologies if I mis-understood what you were trying to do 
 
Kelly
 
From:eeglablist-bounces at sccn.ucsd.edu [mailto:eeglablist-bounces at sccn.ucsd.edu] On Behalf Of Katy Ericson
Sent: 11 February 2010 18:20
To: EEGLAB
Subject: [Eeglablist] Co-registration procedure DIPFIT
 
Dear all,
I'm new to the EEGLAB, this seems to be a nice software. Anyway i have some trouble, if one can call it a trouble at all. I need to find the locations of equivalent dipoles for my ICA components. My recordings are done on EGI 128 channel system and then the exact locations of the sensors are modeled with Photogrametry procedure. As i understand, in order to get more or less correct locations of dipoles, i have to do a co-registration procedure (my net does not correspond exactly to any of 10-5 or 10-10 systems and sensor locations are "scanned" for every patient).
I was trying to do the co-registration using a standard standard-10-5-cap385.elp file as the template file and spherical model. This analysis takes quite a long time, when one has to pair all 128 channels to template locations, because my system has channels numbered as E1 till E128 and Cz as a reference. It may be unreasonable use of time to sit for several hours trying to co-register electrodes to the standard template. On the other hand there was a study, which calculated correspondences of E1:E128 channel names to the 10-5 names. Which may really speed up the process, but the pairing may be done with a greater distances between electrodes.
It would be a great help from you, if some one could answer my questions:
1. How big can the errors be, if i use no co-registration and file with the scanned locations.
2. Is there a way to speed up the process or some other solution of co-registration and dipole fitting in my case (i should mention though, if it concerns accuracy, we would rather take a longer processing time, rather than do the work inaccurately).
3. We have a possibility to export location file from EGI in spherical and in Cartesian coordinates, will there be a difference in accuracy and time consume during co-registration procedure between files in these 2 coordinate systems.
4. When one does the co-registration, the locations of electrodes are changed (custom locations are aligned with predefined ones), how does this affect the accuracy of dipole fitting. 
Thank a lot for your answers and the great mailing lists:)
Cheers!
Katy


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100216/6798ceea/attachment.html>

From schalk at wadsworth.org  Wed Feb 17 07:56:37 2010
From: schalk at wadsworth.org (Gerwin Schalk)
Date: Wed, 17 Feb 2010 10:56:37 -0500
Subject: [Eeglablist] 7th BCI2000 Workshop in Monterey, CA, 5/30-5/31/2010
Message-ID: <C72704D3-1383-4982-BA94-332DC865DBC7@wadsworth.org>

Dear colleagues,

BCI2000 is a general-purpose system for brain-computer interface (BCI) and related research. BCI2000 has been in development since 2000 and has been adopted by close to 500 laboratories around the world. It has supported research that is reported in more than 120 peer-reviewed papers. 

We will be holding the 7th BCI2000 Workshop prior to the 4th International BCI Meeting, i.e., on May 30-31, 2010.  Both events will take place at The Asilomar Conference Center on the Monterey Peninsula in Pacific Grove, California. The first day of the BCI2000 workshop (May 30) consists of discussions that describe relevant technical aspects of the BCI2000 system.  The second part of the BCI2000 workshop (May 31) consists of hands?on practical tutorials that implement the two most common BCI approaches currently used in humans.  In these tutorials, participants can use BCI systems to control a cursor on a computer screen and to spell words just by thinking.  Seven BCI systems will be available throughout the day, and participants will operate them under supervision of tutors.  Participants will also have the opportunity to discuss implementation of their experiments with BCI2000 experts.

Please find more information about the workshop program and registration on 

http://www.bci2000.org/BCI2000/Workshop.html


Hope to see you at what will most likely be the biggest BCI2000 Workshop in the ten-year history of the project.

Gerwin Schalk
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Gerwin Schalk, Ph.D.
Research Scientist V
Wadsworth Center, NYS Dept. of Health
Dept. of Neurology, Albany Medical College
Dept. of Neurosurgery, Washington Univ. in St. Louis
Dept. of Biomed. Eng., Rensselaer Polytechnic Institute
Dept. of Biomed. Sci., State Univ. of New York at Albany
C650 Empire State Plaza
Albany, New York 12201
phone  (518) 486-2559
fax    (518) 486-4910
e-mail schalk at wadsworth.org
www    http://www.bci2000.org
www    http://www.brainmuri.org
www    http://www.gerv.org
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~





IMPORTANT NOTICE: This e-mail and any attachments may contain
confidential or sensitive information which is, or may be, legally
privileged or otherwise protected by law from further disclosure.  It
is intended only for the addressee.  If you received this in error or
from someone who was not authorized to send it to you, please do not
distribute, copy or use it or any attachments.  Please notify the
sender immediately by reply e-mail and delete this from your
system. Thank you for your cooperation.





From arno at ucsd.edu  Thu Feb 18 08:46:14 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Thu, 18 Feb 2010 17:46:14 +0100
Subject: [Eeglablist] Co-registration procedure DIPFIT
In-Reply-To: <20853.91880.qm@web114101.mail.gq1.yahoo.com>
References: <453512.93251.qm@web114108.mail.gq1.yahoo.com>
	<0DBB603CE6952142B61639251E677C3FA45A81@staffexchange6.ul.campus>
	<20853.91880.qm@web114101.mail.gq1.yahoo.com>
Message-ID: <0E332878-D2D3-40AB-AA58-932A7F0707A1@ucsd.edu>

Dear Katy and Kelly,

the latest version of EEGLAB does recognize automatically EGI location  
flles (in fact, it will automatically import channel locations for you  
when you import a binary file).
For co-registration in Dipfit, you still have to rotate the channel  
location manually or you can also try to automatically align the  
fiducials to the template in the graphic interface.

Best regards,

Arno

On Feb 16, 2010, at 3:45 AM, Katy Ericson wrote:

> Hi Kelly,
> thanks a lot for your reply. I think i understand what you mean  
> (sorry if i don't ;) ). As i understood ,this is a standard way,  
> when you have a 10-10 or 10-5 marking of your electrodes, which  
> corresponds to the standard 10-5 marking of the file, given in  
> DIPFIT. The problem in my case was, that i have a custom system with  
> E1-E129 electrode marking, which does not correspond to the DIPFIT  
> file. In this case wrap function does not recognize those locations  
> and, from my point of view, there are two options:
> 1. Do a co-regestaration manually, which takes tons of time
> 2. Find a .loc file with custom electrodes from EGI in spherical  
> format, that will represent my 129 electrodes with the names E1- 
> E129, and set it as a template file. In this case the wrap function  
> will be able to recognize location names and will do wrapping  
> automatically. I'm not sure that this will work in EEGLAB, right now  
> looking for an appropriate file to do try this out.
> Best regards,
> Katy
>
>
> From: Kelly.Kaneswaran <kelly.kaneswaran at ul.ie>
> To: Katy Ericson <katy.ericson at yahoo.com>
> Sent: Mon, February 15, 2010 3:34:02 PM
> Subject: RE: [Eeglablist] Co-registration procedure DIPFIT
>
> Katy,
>
> I had similar problem before, follow the turorial it explains a way  
> of doing it that speeds it up. First you should have a channel  
> location file with all the locations in an acceptable  
> format .sph .loc etc. from the locate dipole using dipfit menu ,  
> choose head and model settings , leave the default settings and  
> choose manual co registration. then a spherical model pops up ,  
> choose warp montage , the list shows the models fudicials and you  
> can select corresponding electrode locations that your model has  
> exact. then press ok , the montage is warped to fit the model and  
> press ok it will generate the tailrach co-ordinates.
>
> i could be way off so apologies if I mis-understood what you were  
> trying to do
>
> Kelly
>
> From: eeglablist-bounces at sccn.ucsd.edu [mailto:eeglablist-bounces at sccn.ucsd.edu 
> ] On Behalf Of Katy Ericson
> Sent: 11 February 2010 18:20
> To: EEGLAB
> Subject: [Eeglablist] Co-registration procedure DIPFIT
>
> Dear all,
> I'm new to the EEGLAB, this seems to be a nice software. Anyway i  
> have some trouble, if one can call it a trouble at all. I need to  
> find the locations of equivalent dipoles for my ICA components. My  
> recordings are done on EGI 128 channel system and then the exact  
> locations of the sensors are modeled with Photogrametry procedure.  
> As i understand, in order to get more or less correct locations of  
> dipoles, i have to do a co-registration procedure (my net does not  
> correspond exactly to any of 10-5 or 10-10 systems and sensor  
> locations are "scanned" for every patient).
> I was trying to do the co-registration using a standard  
> standard-10-5-cap385.elp file as the template file and spherical  
> model. This analysis takes quite a long time, when one has to pair  
> all 128 channels to template locations, because my system has  
> channels numbered as E1 till E128 and Cz as a reference. It may be  
> unreasonable use of time to sit for several hours trying to co- 
> register electrodes to the standard template. On the other hand  
> there was a study, which calculated correspondences of E1:E128  
> channel names to the 10-5 names. Which may really speed up the  
> process, but the pairing may be done with a greater distances  
> between electrodes.
> It would be a great help from you, if some one could answer my  
> questions:
> 1. How big can the errors be, if i use no co-registration and file  
> with the scanned locations.
> 2. Is there a way to speed up the process or some other solution of  
> co-registration and dipole fitting in my case (i should mention  
> though, if it concerns accuracy, we would rather take a longer  
> processing time, rather than do the work inaccurately).
> 3. We have a possibility to export location file from EGI in  
> spherical and in Cartesian coordinates, will there be a difference  
> in accuracy and time consume during co-registration procedure  
> between files in these 2 coordinate systems.
> 4. When one does the co-registration, the locations of electrodes  
> are changed (custom locations are aligned with predefined ones), how  
> does this affect the accuracy of dipole fitting.
> Thank a lot for your answers and the great mailing lists:)
> Cheers!
> Katy
>
>
>
> <ATT00001..txt>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100218/7027ab26/attachment.html>

From arno at ucsd.edu  Thu Feb 18 08:46:42 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Thu, 18 Feb 2010 17:46:42 +0100
Subject: [Eeglablist] Co-registration procedure DIPFIT
In-Reply-To: <20853.91880.qm@web114101.mail.gq1.yahoo.com>
References: <453512.93251.qm@web114108.mail.gq1.yahoo.com>
	<0DBB603CE6952142B61639251E677C3FA45A81@staffexchange6.ul.campus>
	<20853.91880.qm@web114101.mail.gq1.yahoo.com>
Message-ID: <E9F6BE0F-799B-4D6B-9F6B-00FDF6746E3B@ucsd.edu>

Dear Katy and Kelly,

the latest version of EEGLAB does recognize automatically EGI location  
flles (in fact, it will automatically import channel locations for you  
when you import a binary file).
For co-registration in Dipfit, you still have to rotate the channel  
location manually or you can also try to automatically align the  
fiducials to the template in the graphic interface.

Best regards,

Arno

On Feb 16, 2010, at 3:45 AM, Katy Ericson wrote:

> Hi Kelly,
> thanks a lot for your reply. I think i understand what you mean  
> (sorry if i don't ;) ). As i understood ,this is a standard way,  
> when you have a 10-10 or 10-5 marking of your electrodes, which  
> corresponds to the standard 10-5 marking of the file, given in  
> DIPFIT. The problem in my case was, that i have a custom system with  
> E1-E129 electrode marking, which does not correspond to the DIPFIT  
> file. In this case wrap function does not recognize those locations  
> and, from my point of view, there are two options:
> 1. Do a co-regestaration manually, which takes tons of time
> 2. Find a .loc file with custom electrodes from EGI in spherical  
> format, that will represent my 129 electrodes with the names E1- 
> E129, and set it as a template file. In this case the wrap function  
> will be able to recognize location names and will do wrapping  
> automatically. I'm not sure that this will work in EEGLAB, right now  
> looking for an appropriate file to do try this out.
> Best regards,
> Katy
>
>
> From: Kelly.Kaneswaran <kelly.kaneswaran at ul.ie>
> To: Katy Ericson <katy.ericson at yahoo.com>
> Sent: Mon, February 15, 2010 3:34:02 PM
> Subject: RE: [Eeglablist] Co-registration procedure DIPFIT
>
> Katy,
>
> I had similar problem before, follow the turorial it explains a way  
> of doing it that speeds it up. First you should have a channel  
> location file with all the locations in an acceptable  
> format .sph .loc etc. from the locate dipole using dipfit menu ,  
> choose head and model settings , leave the default settings and  
> choose manual co registration. then a spherical model pops up ,  
> choose warp montage , the list shows the models fudicials and you  
> can select corresponding electrode locations that your model has  
> exact. then press ok , the montage is warped to fit the model and  
> press ok it will generate the tailrach co-ordinates.
>
> i could be way off so apologies if I mis-understood what you were  
> trying to do
>
> Kelly
>
> From: eeglablist-bounces at sccn.ucsd.edu [mailto:eeglablist-bounces at sccn.ucsd.edu 
> ] On Behalf Of Katy Ericson
> Sent: 11 February 2010 18:20
> To: EEGLAB
> Subject: [Eeglablist] Co-registration procedure DIPFIT
>
> Dear all,
> I'm new to the EEGLAB, this seems to be a nice software. Anyway i  
> have some trouble, if one can call it a trouble at all. I need to  
> find the locations of equivalent dipoles for my ICA components. My  
> recordings are done on EGI 128 channel system and then the exact  
> locations of the sensors are modeled with Photogrametry procedure.  
> As i understand, in order to get more or less correct locations of  
> dipoles, i have to do a co-registration procedure (my net does not  
> correspond exactly to any of 10-5 or 10-10 systems and sensor  
> locations are "scanned" for every patient).
> I was trying to do the co-registration using a standard  
> standard-10-5-cap385.elp file as the template file and spherical  
> model. This analysis takes quite a long time, when one has to pair  
> all 128 channels to template locations, because my system has  
> channels numbered as E1 till E128 and Cz as a reference. It may be  
> unreasonable use of time to sit for several hours trying to co- 
> register electrodes to the standard template. On the other hand  
> there was a study, which calculated correspondences of E1:E128  
> channel names to the 10-5 names. Which may really speed up the  
> process, but the pairing may be done with a greater distances  
> between electrodes.
> It would be a great help from you, if some one could answer my  
> questions:
> 1. How big can the errors be, if i use no co-registration and file  
> with the scanned locations.
> 2. Is there a way to speed up the process or some other solution of  
> co-registration and dipole fitting in my case (i should mention  
> though, if it concerns accuracy, we would rather take a longer  
> processing time, rather than do the work inaccurately).
> 3. We have a possibility to export location file from EGI in  
> spherical and in Cartesian coordinates, will there be a difference  
> in accuracy and time consume during co-registration procedure  
> between files in these 2 coordinate systems.
> 4. When one does the co-registration, the locations of electrodes  
> are changed (custom locations are aligned with predefined ones), how  
> does this affect the accuracy of dipole fitting.
> Thank a lot for your answers and the great mailing lists:)
> Cheers!
> Katy
>
>
>
> <ATT00001..txt>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100218/e08a308e/attachment.html>

From perlaki at gamma.ttk.pte.hu  Fri Feb 19 03:32:34 2010
From: perlaki at gamma.ttk.pte.hu (Perlaki Gabor)
Date: Fri, 19 Feb 2010 12:32:34 +0100
Subject: [Eeglablist] gradient artefact
Message-ID: <20100219112338.M49403@gamma.ttk.pte.hu>

Dear all,

Does anybody know any good methods to filter gradient artefacts from a single
channel ecg? I've found some articles about suppression of MR gradient
artefacts on electrophysiological signal, but those methods need to know
gradient reference signals as well, and I didn't acquire them. 

Regards,
Gabor


--
P?csi Tudom?nyegyetem Term?szettudom?nyi Kar (http://www.ttk.pte.hu/)



From dlozano at ub.edu  Thu Feb 18 04:13:57 2010
From: dlozano at ub.edu (Diego Lozano Soldevilla)
Date: Thu, 18 Feb 2010 13:13:57 +0100
Subject: [Eeglablist] time-frequency of cluster of ICs
Message-ID: <afd4b89d1002180413o3db7ca63g518abdeebc96013f@mail.gmail.com>

Dear all,

I'm doing an independent component clustering of a 15 participants in 2
conditions. I've obtained 5 clusters and I wish to do a time-frequency
analysis of each cluster. I've selected the W matrix of each component of
each cluster and I've applied to longer epochs to do the time-frequency. My
doubt consists in what electrode/s I should select to do the time-frequency
that represent better the source of each cluster: the most activate
electrode in the mean of the topography cluster for example? Do a PCA
of xdimensions of the electrodes?

Thank you all for your replies in advance.

Diego Lozano-Soldevilla
-- 
Neurodynamics Laboratory
Dept. Psychiatry and Clinical Psychobiology
Faculty of Psychology
University of Barcelona
Passeig Vall d'Hebron, 171,
08035 Barcelona - Spain.
Tf: +34 934024493
Fax: +34 934021584
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100218/b4cff468/attachment.html>

From nikolai.novitski at gmail.com  Sun Feb 21 14:20:53 2010
From: nikolai.novitski at gmail.com (Nikolay Novitskiy)
Date: Sun, 21 Feb 2010 23:20:53 +0100
Subject: [Eeglablist] gradient cleaning
Message-ID: <4B81B1C5.6040404@gmail.com>

Dear Gabor,

It is not really a probelm to clean the gradients without volume 
triggers. E.g. Bergen plugin
(http://fmri.uib.no/tools/manual_bergen_plugin.pdf) has an option to 
detect the gradient onset form your raw EEG (ECG, EMG, whatever). It 
doesn't matter how many channels you have, average arteifact subtraction 
works channel by channel.  The only thing  you  may take care of is that 
your data contains only real gradient onsets and doesn't include short  
RF onsets in the beginning of recording. And it is also good to know 
your TR at least approximately.
Please, ask if you have more questions.

Nikolay
 


From smakeig at gmail.com  Sun Feb 21 15:17:18 2010
From: smakeig at gmail.com (Scott Makeig)
Date: Sun, 21 Feb 2010 15:17:18 -0800
Subject: [Eeglablist] time-frequency of cluster of ICs
In-Reply-To: <afd4b89d1002180413o3db7ca63g518abdeebc96013f@mail.gmail.com>
References: <afd4b89d1002180413o3db7ca63g518abdeebc96013f@mail.gmail.com>
Message-ID: <9e09b8f01002211517q3eef7c2dye8f80daab772e0ec@mail.gmail.com>

Diego -

Time/frequency analysis as implemented in EEGLAB is relative event-related
dB change in spectral power from baseline (ERSP) and (absolute) inter-trial
coherence (ITC). It thus can operate on the IC time courses themselves
(e.g., the unitless activations). For IC clusters, the grand mean ERSP and
ITC values are used as representative.

An IC activation times a particular IC channel weight gives the
back-projection of the IC to that channel. One could compute e.g. an
absolute power spectrum (uV^2/Hz) there -- At every scalp channel the
projected IC spectrum is the same except for a scalp map weight constant.

Scott Makeig

On Thu, Feb 18, 2010 at 4:13 AM, Diego Lozano Soldevilla <dlozano at ub.edu>wrote:

> Dear all,
>
> I'm doing an independent component clustering of a 15 participants in 2
> conditions. I've obtained 5 clusters and I wish to do a time-frequency
> analysis of each cluster. I've selected the W matrix of each component of
> each cluster and I've applied to longer epochs to do the time-frequency. My
> doubt consists in what electrode/s I should select to do the time-frequency
> that represent better the source of each cluster: the most activate
> electrode in the mean of the topography cluster for example? Do a PCA of xdimensions of the electrodes?
>
> Thank you all for your replies in advance.
>
> Diego Lozano-Soldevilla
> --
> Neurodynamics Laboratory
> Dept. Psychiatry and Clinical Psychobiology
> Faculty of Psychology
> University of Barcelona
> Passeig Vall d'Hebron, 171,
> 08035 Barcelona - Spain.
> Tf: +34 934024493
> Fax: +34 934021584
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Scott Makeig, Research Scientist and Director, Swartz Center for
Computational Neuroscience, Institute for Neural Computation, University of
California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100221/a848389b/attachment.html>

From demiral.007 at googlemail.com  Sun Feb 21 15:08:22 2010
From: demiral.007 at googlemail.com (Baris Demiral)
Date: Sun, 21 Feb 2010 23:08:22 +0000
Subject: [Eeglablist] Handling Intrapolation & EOG locations
Message-ID: <f423beba1002211508i45c5b53k147d90af8b8d3029@mail.gmail.com>

Dear all,

I read some bug reports about the interpolation function in the new versions
of eeglab. It seems like either EOG electrodes should be taken out (because
they don't have locations) or they should have coordinates in order for the
'interp' function to work. How did you resolve this problem? I would like to
keep EOG electrodes, but not need to define coordinates (or maybe I have
to....) Related to that, I assume that mastoid electrodes are labeled and
defined as M1 and M2 in the file Standard-10-5-Cap385.sfp (?). Are there
special labels in this file for the other EOG electrodes horizontal and
vertical EOGs that I can directly use?

Any comments/experiences are greatly appreciated,
Thanks,
Baris
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100221/ae18a153/attachment.html>

From priyanka.abhang at gmail.com  Fri Feb 26 05:55:59 2010
From: priyanka.abhang at gmail.com (priyanka abhang)
Date: Fri, 26 Feb 2010 05:55:59 -0800
Subject: [Eeglablist] request
Message-ID: <3e75b71d1002260555t3edff2c3g79c68ed1198a1fb3@mail.gmail.com>

Respected sir/madam
 I am priyanka research student studing about EEG and more also
 My request is that, will i get some more information about workshop
and conference about EEG or Brain computer interface


thanks....

-- 
Thanks and Regards,
Priyanka A. Abhang
Department of Computer Science and Information Technology,
Dr.Babasaheb Ambedkar Marathwada University, Aurangabad (M.S.), India.
Mob:9860741611


From dr.ilya at yahoo.com  Mon Mar  1 07:59:10 2010
From: dr.ilya at yahoo.com (Ilya Adamchic)
Date: Mon, 1 Mar 2010 07:59:10 -0800 (PST)
Subject: [Eeglablist] Align fiducials
In-Reply-To: <DD3D8BCD-5036-41A9-81DC-17459A5EC3BC@ucsd.edu>
References: <f81849ea1001290149u587591abh5f9c4b6ecde09338@mail.gmail.com>
	<91D5B97F-6924-45C9-A09E-9583C99DACA4@mac.com>
	<9e09b8f01002011326m4ba79c0dw2d80529feca21a73@mail.gmail.com>
	<DD3D8BCD-5036-41A9-81DC-17459A5EC3BC@ucsd.edu>
Message-ID: <511321.6202.qm@web63807.mail.re1.yahoo.com>

Dear Arno,
I'm trying to align my fiducials to the ones in the DIPFIT predefined spherical electrode locations. Using align fiducials button, I pair 3 fiducials (left and right ears and nasion). After i press OK button the following mistake comes up: Error using electrode align at ==> 547 unknown method. Did i do something wrong?
Thanks for your help,
Ilya



      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100301/80fd3cad/attachment.html>

From brian.murphy at unitn.it  Tue Mar  2 03:44:22 2010
From: brian.murphy at unitn.it (Brian Murphy)
Date: Tue, 2 Mar 2010 12:44:22 +0100
Subject: [Eeglablist] plotting very small-scale signals waveform
Message-ID: <4B8CFA16.3080608@unitn.it>

Hi,

I'm having some trouble viewing very fine-scaled signals. I'm working 
with Elekta MEG recordings (imported via File-IO), where the scale of 
the signals is many orders of magnitude smaller than EEG recordings I 
use. So if I want to view the MEG waveforms (with the plot > channel 
data (scroll) function), I need to set a very very fine scaling factor. 
The interface doesn't allow me to put in anything smaller than 0.005 as 
the scaling factor. Is there a way to around this via the command line? 
I also tried using exponent syntax (5e-3), but the same lower limit 
appears to apply.

I'm not sure if this is a bug, or an intentional feature. Any 
suggestions welcome.

-- 
Brian Murphy
Post-Doctoral Researcher
Language, Interaction and Computation Lab
Centre for Mind/Brain Sciences
University of Trento
http://clic.cimec.unitn.it/brian/



From bradley.voytek at gmail.com  Wed Mar  3 13:21:36 2010
From: bradley.voytek at gmail.com (Bradley Voytek)
Date: Wed, 3 Mar 2010 13:21:36 -0800
Subject: [Eeglablist] plotting very small-scale signals waveform
In-Reply-To: <4B8CFA16.3080608@unitn.it>
References: <4B8CFA16.3080608@unitn.it>
Message-ID: <6d898bac1003031321j3f4a1b3fsf65d575bd39aaaf2@mail.gmail.com>

You could multiply your signal by a scaling factor instead.

::brad

On Tue, Mar 2, 2010 at 03:44, Brian Murphy <brian.murphy at unitn.it> wrote:
> Hi,
>
> I'm having some trouble viewing very fine-scaled signals. I'm working
> with Elekta MEG recordings (imported via File-IO), where the scale of
> the signals is many orders of magnitude smaller than EEG recordings I
> use. So if I want to view the MEG waveforms (with the plot > channel
> data (scroll) function), I need to set a very very fine scaling factor.
> The interface doesn't allow me to put in anything smaller than 0.005 as
> the scaling factor. Is there a way to around this via the command line?
> I also tried using exponent syntax (5e-3), but the same lower limit
> appears to apply.
>
> I'm not sure if this is a bug, or an intentional feature. Any
> suggestions welcome.
>
> --
> Brian Murphy
> Post-Doctoral Researcher
> Language, Interaction and Computation Lab
> Centre for Mind/Brain Sciences
> University of Trento
> http://clic.cimec.unitn.it/brian/
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
>


From Andero.Uusberg at ut.ee  Thu Feb 25 13:03:15 2010
From: Andero.Uusberg at ut.ee (Andero Uusberg)
Date: Thu, 25 Feb 2010 23:03:15 +0200
Subject: [Eeglablist] newtimef ersp output without any or with external
	baseline
Message-ID: <4B86E593.5040705@ut.ee>


Hello,

I have a few (maybe fairly naive) questions regarding running the 
newtimef function without using baseline from the dataset being analyzed.

Firstly, while running newtimef on a channel of a segmented dataset 
without either of the baselines (having both 'baseline' and 'powbase' 
set to NaN) I have trouble understanding the units of the output ersp 
matrix (I call the function from the command line and use freqrange 
2:20, padratio 2, cycles 1 6 and basenorm 'off'). The output ersp matrix 
is full of positive numbers in the range of 40 to 60. Although the 
figure legend states this is 10*log10 transform of power density in 
microvolts, this does not seem likely (the power densities would be 
around 100 000 microvolts). In addition, even with both baselines set to 
NaN the powbase output is created (consisting of numbers in the same 
range). All this probably has something to do with the normalization 
process and is completely meaningful, but could someone explain how 
exactly these values are generated and what is their unit?

Second and related question regards feeding externally created baseline 
into newtimef via the powbase key. I know one way would be to calculate 
the basline with newtimef in the first place and not to care about the 
units of the data :) However, I would also like to be able to use data 
calculated with FFT in Vision Anlayzer as the baseline. Do I understand 
correctly that the data should be in power denisty units (square of 
microvolt per hertz) and should I run the 10*log10 transform on the data 
before feeding in into the newtimef with powbase?

Finally, as I'm not sure my previous e-mail reached the list I repeat a 
related question concerning using std_precomp for calculating ERSPs for 
several datasets using a single external baseline. When referring to a 
channel-by-frequencies (as well as frequencies-by-channles) matrix of 
external power density values with the powbase key in erspparams field 
og std_precom, I receive the following error message: "std_ersp at 372: 
powbase should be of size (ncomps,nfreqs)". Can anyone estimate if this 
could be a bug or should I keep figuring out what I'm doing wrong?

I apologise if these issues have been dealt with before in the list in 
which case I am grateful for a kind reference to older posts.

With kind regards and general gratitude to eeglab :)
Andero Uusberg
University of Tartu
Estonia


From amani at uh.edu  Sat Mar  6 11:34:45 2010
From: amani at uh.edu (Abigail Mani)
Date: Sun, 7 Mar 2010 01:04:45 +0530
Subject: [Eeglablist] Infinity reference - 64 channel EGG - Dezhong Yao
	method
Message-ID: <f82b75b71003061134k562dfbfeu9f88c54e63cceb34@mail.gmail.com>

Hi All,

Has anyone used the REST infinity reference for EEG analysis.
I would like to know if anyone has and how you actually did the
re-referencing, I mean MATLAB procedure.
Also some details like, if it is done on the entire unepoched data, or for
each epoch.
I would be really thankfull, if someone who knows it can explain it to me
more clearly.

Here is the reference paper:
A method to standardize a reference of scalp EEG recordings to a point at
infinityDezhong Yao 2001 *Physiol. Meas.* *22* 693-711
-- 
Abigail





www.gnaniwriter.blogspot.com
"If you haven't found something worth dying for then you are not fit to
live" - Martin Luther King'
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100307/9c3f1bf3/attachment.html>

From eva.kelman at gmail.com  Sat Mar  6 05:24:53 2010
From: eva.kelman at gmail.com (=?UTF-8?B?15DXldeUINen15zXntef?=)
Date: Sat, 6 Mar 2010 15:24:53 +0200
Subject: [Eeglablist] Importing .bdf file
Message-ID: <59ddd6171003060524h2522722bm35b3e76738543c36@mail.gmail.com>

I'm importing a .bdf file using eeglab and it seems that some of the events
are not imported or not recognized as events: I get about 300 events less
than I'm supposed to.

Did anyone encountered something similar?
Does it have some sort of solution or at least explanation?

Thank you,
Eva Kelman
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100306/63d748db/attachment.html>

From smakeig at ucsd.edu  Tue Mar  9 10:36:48 2010
From: smakeig at ucsd.edu (Scott Makeig)
Date: Tue, 9 Mar 2010 10:36:48 -0800
Subject: [Eeglablist] First InterBrain Symposium on ICA/EEG and
	International EEGLAB workshop in Finland
Message-ID: <9e09b8f01003091036w764f50f5pdf6fa96ff1059d52@mail.gmail.com>

*ICA conference (June 12 - 13) and EEGLAB workshop (June 14 - 17) in Jyv?skyl?,
Finland*



http://research.jyu.fi/interbrain/ib2010.htm

http://sccn.ucsd.edu/wiki/Tenth_EEGLAB_Workshop



Welcome -


We would like to welcome you to the ICA conference First InterBrain
Symposium and tenth international EEGLAB workshop. These consecutive events
will take place in Jyv?skyl?, Finland, a university town surrounded by
lakes (http://en.wikipedia.org/wiki/Jyvaskyla) located a short train or
plane ride north of Helsinki.



The two-day International ICA conference will focus on theory of independent
component analysis (ICA) and its applications to brain research, mainly in
EEG/ ERP analysis. The conference program consists of plenary lectures by
invited leading experts of the field, accepted high-quality oral sessions
and posters dealing with ICA and other related analysis techniques. The
abstracts will appear online and in a conference abstract book. The
interactive poster-sessions will offer on opportunity for discussions with
leading experts on various methodological issues and the applicability of
ICA-based analysis to your own data. Follow the links above for more
information.



The conference will be followed by the Tenth International EEGLAB workshop
which will concentrate on ICA-based analysis methods available in the EEGLAB
toolbox developed at the Swartz Center for Computational Neuroscience (SCCN)
at the University of California San Diego (UCSD). The workshop offers both
theoretical lectures and hands-on sessions tutored by experts from SCCN and
associates. The workshop has a limited number of participants. A half-day
excursion to Finnish nature resorts and several social events will be
included to encourage participants to discuss their research.



We cordially invite you to participate in this exciting conference and
ICA/EEGLAB workshop. Workshop participants are strongly encouraged to
participate in the ICA conference, too, as it offers theoretical basis for
much of the material dealt with in the workshop.



The workshop can accept a limited number of participants (50). The
registration is now open for both the conference and the workshop (see under
Registration at http://research.jyu.fi/interbrain/ib2010.htm).



We look forward to meeting you and welcoming you to Jyv?skyl? in June 2010!



Paavo Lepp?nen, Chairman of the Organizing Committee

Department of Psychology, University of Jyv?skyl?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100309/7174a0fc/attachment.html>

From jdauwels at mit.edu  Tue Mar  9 14:35:51 2010
From: jdauwels at mit.edu (Justin Dauwels)
Date: Tue, 09 Mar 2010 17:35:51 -0500
Subject: [Eeglablist] Graduate student positions in the area of medical
	engineering
Message-ID: <4B96CD47.4070504@mit.edu>

Graduate student positions are available in the area of  medical 
engineering.

Students with research interests at the interface of electrical  
engineering and biology are especially encouraged to apply.

Topics include:

o modeling of epileptic seizures (from abstract network models to  
detailed biophysical models);
o neural implants for seizure suppression (from control/game theoretic  
analysis to hardware design);
o diagnosis of brain disorders, from EEG and other brain signals  
(Alzheimer's disease and epilepsy);
o graphical models and nonparametric statistical models for analyzing  
electrophysiological data (e.g., spike sorting and decoding);
o principled statistical models for merging different brain imaging  
modalities (EEG/MEG/fMRI/DTI/single-unit recordings/etc.).

The research is conducted in collaboration with partners at the RIKEN  
Brain Science Institute, MIT, and MGH/Harvard.

Graduate student applicants will pursue PhD.s in Electrical Engineering.

Please send detailed curriculum vitae, statements of research  
interests, three references and relevant publications (if applicable),  
electronically, to:

Prof. Justin Dauwels
Nanyang Technological University
School of Electrical & Electronic Engineering
Singapore

recruitment at dauwels.com
http://www.dauwels.com/Jobs.htm


From aoo34 at cornell.edu  Tue Mar  9 22:19:22 2010
From: aoo34 at cornell.edu (Ayo Ositelu)
Date: Wed, 10 Mar 2010 01:19:22 -0500
Subject: [Eeglablist] DIPFIT2 on a 64-bit Mac computer
Message-ID: <b4e93bb31003092219s67d24f32l5c7ec541a8140a1b@mail.gmail.com>

Dear Colleagues,

I am trying to run DIPFIT2 on a 64-bit Mac computer (intel processor). Upon
running coarse fit, I receive the following error message: "could not locate
MEX file for solid_angle". We have tried following the suggestions from the
Field Trip website with no success. Can anyone please provide assistance.


Thanks in advance,
Ayo
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100310/06cac105/attachment.html>

From balaszone at gmail.com  Wed Mar 10 17:17:45 2010
From: balaszone at gmail.com (balaszone at gmail.com)
Date: Wed, 10 Mar 2010 20:17:45 -0500
Subject: [Eeglablist] General Question
Message-ID: <e847b95c1003101717p77c3ccdem2cc40010f3d98b4@mail.gmail.com>

Dear all,

I have a very non-technical question. I hope somebody might be able to
answer this question.
We run research and clinical ERP studies in our lab.
We would like to know how much does a lab usually charge for ERP research
studies(ofcourse, this being a research experiment as a service to other
research lab/facility)
We are in the process of writing a internal grant and so any information on
this would be really helpful.

Thanks much in advance,
Bala
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100310/d1756721/attachment.html>

From yuanx041 at umn.edu  Fri Mar 12 07:50:22 2010
From: yuanx041 at umn.edu (Han Yuan)
Date: Fri, 12 Mar 2010 09:50:22 -0600
Subject: [Eeglablist] Announcing the release of eConnectome: open-source
	software for brain connectivity imaging
Message-ID: <003901cac1fb$b9ee8890$2dcb99b0$@edu>

Dear Colleagues,

 

We are pleased to announce the release of eConnectome, a free open-source
MATLAB software package for imaging brain functional connectivity from
electrophysiological signals. It provides interactive graphical interfaces
for EEG/ECoG preprocessing, source estimation, connectivity analysis and
visualization. The current release allows connectivity imaging from EEG and
ECoG over sensor and source domains. This package is designed for use by
researchers in neuroscience, psychology, cognitive science, clinical
neurophysiology, neurology and other disciplines. The graphical
interface-based platform requires little programming knowledge or experience
with MATLAB.

 

Highlighted features include: graphical user interface, EEG/ECoG
preprocessing, source estimation (forward modeling, inverse calculation, and
ROI analysis), and connectivity analysis (Directed Transfer Function and
surrogate assessment, ECoG/EEG/source connectivity visualization).

 

eConnectome is developed by the Biomedical Functional Imaging and
Neuroengineering Laboratory at the University of Minnesota, directed by Dr.
Bin He. The visualization module is jointly developed with Drs. Fabio
Babiloni and Laura Astolfi at the University of Rome "La Sapienza". 

 

To download a free copy of eConnectome software, or for more information,
please visit http://econnectome.umn.edu.

 

Sincerely yours,

 

Bin He, PhD

Distinguished McKnight University Professor

Director, Biomedical Functional Imaging and Neuroengineering Laboratory

Director, Center for Neuroengineering

University of Minnesota

http://www.tc.umn.edu/~binhe/

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100312/b69e75fd/attachment.html>

From cerenakdeniz at yahoo.com  Thu Mar 11 00:26:05 2010
From: cerenakdeniz at yahoo.com (ceren akdeniz)
Date: Thu, 11 Mar 2010 00:26:05 -0800 (PST)
Subject: [Eeglablist] newtimef
Message-ID: <493434.93011.qm@web65612.mail.ac4.yahoo.com>

Dear all,

I am using newtimef function for the time-frequency analysis of my eeg data.My data is 640 seconds long and unfortunately I lose first and last 40 seconds of my data during this time-freq analysis.

Do you know any way to lose less may be?

Thank you!

Yours,
Ceren Akdeniz



      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100311/25f42b96/attachment.html>

From Lars.Michels at kispi.uzh.ch  Fri Mar 12 00:56:43 2010
From: Lars.Michels at kispi.uzh.ch (Michels Lars)
Date: Fri, 12 Mar 2010 09:56:43 +0100
Subject: [Eeglablist] paired t-test with corrected p-values
Message-ID: <FD1E6887BC99C341800CBFD7ABC2F9C201EB14E2@exzh1.kispi.int>

Dear all,
 
I calculated in eeglab a paired t-test between two conditions with
corrected p-values (p < 0.05, FDR).
 
I used the study option for this and just tested it for two subjects,
just as a proof of principle
 
The panel on the right of the attached image shows the "significant"
differences between the two conditions.
 
However, I have four questions:
 
1) How do I know the direction of the statistical test, i.e. does the
grey regions indicate the significant differences for L2 > L5 or does
they simply show all significant differences (L2 > L5 and L5 > L2)?
 
2) Is it possible to plot the statistical map on the electrode level,
i.e. to see which electrodes reflecting the strongest effect?
 
3) Is it possible to calculate the test directly for predefined bands
(5-7 Hz etc)?
 
4) Why is the limit of the y-axis 0.1 and not 0.05. Is this because the
effects are always above 0.1?
 
Thanks in advance,
Lars
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100312/075349d0/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: L5-L2_twosub.gif
Type: image/gif
Size: 148707 bytes
Desc: L5-L2_twosub.gif
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100312/075349d0/attachment.gif>

From cmugruza at yahoo.com  Mon Mar 15 08:01:17 2010
From: cmugruza at yahoo.com (Carlos Mugruza)
Date: Mon, 15 Mar 2010 08:01:17 -0700 (PDT)
Subject: [Eeglablist] paired t-test with corrected p-values
Message-ID: <119013.96611.qm@web34203.mail.mud.yahoo.com>

Hello Michael,
I don't very sure if these could be 2 of your four answers:

1) I am not very sure what function you use to make the test. But I use ttest2, which gives you that you like: the fourth parameter is the tail (a flag), that specifies one of three alternative hypotheses:
tail = 0 (default) specifies the alternative,? only looks at the difference;
tail = 1 specifies the alternative, if the first is greater thant the second one;
tail = -1 specifies the alternative, if the first is less than the second one. 

ttest2(x,y,alpha,tail)?


4) I think is only change the axis. Eve you use a pre-exitent code, you can change it.



Regarding you question 3, really I don't understand very well. I imagine that it is take frequency vectors and do the test



Have a nice weekend.



_______________________________________________________________________
Carlos A. Mugruza V.
Demonstrator / PhD Student
School of Psychology
The University of Dundee 
Dundee
DD1 4HN
Scotland, UK
?
Office (+44) 1382 384926
Fax (+44) 1382 229993
Webpage:
http://www.dundee.ac.uk/psychology/people/postgraduates/camugruzavassallo/index.htm

Eeglablist] paired t-test with corrected p-valuesMichels Lars Lars.Michels at kispi.uzh.ch 
Fri Mar 12 00:56:43 PST 2010 


Previous message: [Eeglablist] newtimef 
Next message: [Eeglablist] Announcing the release of eConnectome: open-source software for brain connectivity imaging 
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] 

Dear all,
 
I calculated in eeglab a paired t-test between two conditions with
corrected p-values (p < 0.05, FDR).
 
I used the study option for this and just tested it for two subjects,
just as a proof of principle
 
The panel on the right of the attached image shows the "significant"
differences between the two conditions.
 
However, I have four questions:
 
1) How do I know the direction of the statistical test, i.e. does the
grey regions indicate the significant differences for L2 > L5 or does
they simply show all significant differences (L2 > L5 and L5 > L2)?
 
2) Is it possible to plot the statistical map on the electrode level,
i.e. to see which electrodes reflecting the strongest effect?
 
3) Is it possible to calculate the test directly for predefined bands
(5-7 Hz etc)?
 
4) Why is the limit of the y-axis 0.1 and not 0.05. Is this because the
effects are always above 0.1?
 
Thanks in advance,
Lars


?


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100315/e2c7d20f/attachment.html>

From smakeig at gmail.com  Mon Mar 15 10:56:19 2010
From: smakeig at gmail.com (Scott Makeig)
Date: Mon, 15 Mar 2010 09:56:19 -0800
Subject: [Eeglablist] SCCN reopening and planned workshop
Message-ID: <9e09b8f01003151056w3c2ebd78q2f56e46b627d3a7a@mail.gmail.com>

1. The Swartz Center for Computational Neuroscience has moved into new
quarters on the UCSD campus in La Jolla. Recently, we had an opening event
followed by an open house with demos. The new space is in the new San Diego
Supercomputer Center extension building with adjacent multimedia classroom
and auditorium space available for workshops -- some photos are available
at

http://sccn.ucsd.edu/events/reopening.html

2. We are planning to hold an EEGLAB workshop before or after the Society
for Neuroscience meeting in San Diego in November. Choices include dates and
number of days, parallel sessions for the general (long-track) survey of the
EEGLAB environment, as well as the possibility of parallel tutorials, on the
first day, for new EEGLAB and related toolboxes including forward head
modeling, information flow, and experimental real-time interactive control
and analysis (ERICA) software, the Human electrophysiology, anatomic data,
and integrated tools (HeadIT) resource, (etc). These might be repeated after
the main tutorial if there is sufficient interest.

If you have possible interest in attending, please reply to *eeglab at ucsd.edu
*, checking all the following that apply to you:

__ I would be interested in attending the full (3-4 day) EEGLAB workshop *
before* the 2010 Society for Neuroscience meeting (before Nov. 13).
__ I would be interested in attending the full (3-4 day) EEGLAB workshop *
after* the 2010 Society for Neuroscience meeting (after Nov. 17).
__ I would be interested in attending a day of tutorials on new
EEGLAB-linked new tools and toolboxes *before* the Soc for Neurosci meeting
__ I would be interested in attending a day of tutorials on new
EEGLAB-linked new tools and toolboxes *after* the Soc for Neurosci meeting
__ I would be interested in staying for both the full workshop and the
tutorials on new tools.

This is a survey only -- no final plan or commitment is yet available.

Scott Makeig

-- 
Scott Makeig, Research Scientist and Director, Swartz Center for
Computational Neuroscience, Institute for Neural Computation, University of
California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100315/384b543e/attachment.html>

From smakeig at gmail.com  Mon Mar 15 14:25:12 2010
From: smakeig at gmail.com (Scott Makeig)
Date: Mon, 15 Mar 2010 14:25:12 -0700
Subject: [Eeglablist] SCCN reopening and planned workshop -- Correction!
Message-ID: <9e09b8f01003151425r17b89b5aw726e05928d8cec82@mail.gmail.com>

I was just alerted to the fact that the reply email requested below needs to
go to  eeglab at sccn.ucsd.edu !!  Please forgive the confusion.

Scott Makeig

On Mon, Mar 15, 2010 at 10:56 AM, Scott Makeig <smakeig at gmail.com> wrote:

> 1. The Swartz Center for Computational Neuroscience has moved into new
> quarters on the UCSD campus in La Jolla. Recently, we had an opening event
> followed by an open house with demos. The new space is in the new San Diego
> Supercomputer Center extension building with adjacent multimedia classroom
> and auditorium space available for workshops -- some photos are available
> at
>
> http://sccn.ucsd.edu/events/reopening.html
>
> 2. We are planning to hold an EEGLAB workshop before or after the Society
> for Neuroscience meeting in San Diego in November. Choices include dates and
> number of days, parallel sessions for the general (long-track) survey of the
> EEGLAB environment, as well as the possibility of parallel tutorials, on the
> first day, for new EEGLAB and related toolboxes including forward head
> modeling, information flow, and experimental real-time interactive control
> and analysis (ERICA) software, the Human electrophysiology, anatomic data,
> and integrated tools (HeadIT) resource, (etc). These might be repeated after
> the main tutorial if there is sufficient interest.
>
> If you have possible interest in attending, please reply to *
> eeglab at ucsd.edu*, checking all the following that apply to you:
>
> __ I would be interested in attending the full (3-4 day) EEGLAB workshop *
> before* the 2010 Society for Neuroscience meeting (before Nov. 13).
> __ I would be interested in attending the full (3-4 day) EEGLAB workshop *
> after* the 2010 Society for Neuroscience meeting (after Nov. 17).
> __ I would be interested in attending a day of tutorials on new
> EEGLAB-linked new tools and toolboxes *before* the Soc for Neurosci
> meeting
> __ I would be interested in attending a day of tutorials on new
> EEGLAB-linked new tools and toolboxes *after* the Soc for Neurosci meeting
> __ I would be interested in staying for both the full workshop and the
> tutorials on new tools.
>
> This is a survey only -- no final plan or commitment is yet available.
>
> Scott Makeig
>
> --
> Scott Makeig, Research Scientist and Director, Swartz Center for
> Computational Neuroscience, Institute for Neural Computation, University of
> California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott
>



-- 
Scott Makeig, Research Scientist and Director, Swartz Center for
Computational Neuroscience, Institute for Neural Computation, University of
California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100315/feb11009/attachment.html>

From lakshmanan at kennedykrieger.org  Tue Mar 16 11:52:51 2010
From: lakshmanan at kennedykrieger.org (Balaji Lakshmanan)
Date: Tue, 16 Mar 2010 14:52:51 -0400
Subject: [Eeglablist] Photo diode/sensor
Message-ID: <4B9F9B43020000A00002E633@KKI-GSPGWIA.KKI.ORG>

Hello,
 
We are planning to use a photo diode to test the timing of our stimulus presentation system.
I was just wondering if anybody has any suggestions where we could get/buy such sensors for this purpose?
 
Any suggestions is greatly appreciated.
 
Thanks much,
Balaji


Please consider the environment before printing this E-Mail.



Disclaimer:
The materials in this e-mail are private and may contain Protected Health Information. Please note that e-mail is not necessarily confidential or secure. Your use of e-mail constitutes your acknowledgment of these confidentiality and security limitations. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying, distribution, or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this e-mail in error, please immediately notify the sender via telephone or return e-mail.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100316/14eb02ce/attachment.html>

From pzeman at alumni.uvic.ca  Tue Mar 16 13:02:08 2010
From: pzeman at alumni.uvic.ca (Philip Michael Zeman)
Date: Tue, 16 Mar 2010 13:02:08 -0700
Subject: [Eeglablist] Photo diode/sensor
In-Reply-To: <4B9F9B43020000A00002E633@KKI-GSPGWIA.KKI.ORG>
References: <4B9F9B43020000A00002E633@KKI-GSPGWIA.KKI.ORG>
Message-ID: <E6629B4491354E7098CF086E8365A51D@mine>

Hi Balaji

if you're in the United States or in Canada, try www.digikey.com.

Alternatively, go to your local electronics hardware store and they can get you the part and probably give you a circuit too.


Phil

=-=-=-=-=-=-=-=-=-=
Philip Michael Zeman, B.Eng (B.Sc.), Ph.D.
Applied Brain and Vision Sciences Inc.
ph: +1-250-589-4234
pzeman at abvsciences.com
http://www.abvsciences.com
Recent Research Findings
http://www.spatialbrain.com


  ----- Original Message ----- 
  From: Balaji Lakshmanan 
  To: eeglablist at sccn.ucsd.edu 
  Sent: Tuesday, March 16, 2010 11:52 AM
  Subject: [Eeglablist] Photo diode/sensor


  Hello,

  We are planning to use a photo diode to test the timing of our stimulus presentation system.
  I was just wondering if anybody has any suggestions where we could get/buy such sensors for this purpose?

  Any suggestions is greatly appreciated.

  Thanks much,
  Balaji



  Please consider the environment before printing this E-Mail. 
Disclaimer:The materials in this e-mail are private and may contain Protected Health Information. Please note that e-mail is not necessarily confidential or secure. Your use of e-mail constitutes your acknowledgment of these confidentiality and security limitations. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying, distribution, or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this e-mail in error, please immediately notify the sender via telephone or return e-mail.

------------------------------------------------------------------------------


  _______________________________________________
  Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
  To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
  For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100316/e4e62c22/attachment.html>

From brian.roach at yale.edu  Thu Mar 18 14:38:06 2010
From: brian.roach at yale.edu (Brian Roach)
Date: Thu, 18 Mar 2010 14:38:06 -0700
Subject: [Eeglablist] research jobs in San Francisco, CA
Message-ID: <4BA29D3E.4040305@yale.edu>

Dear EEGlab users,

We are hiring for multiple positions in our San Francisco-based 
schizophrenia brain-imaging EEG research lab.  Please visit the links 
below for additional details:

<http://jobs-ncire.icims.com/jobs/1504/job>http://jobs-ncire.icims.com/jobs/1504/job
http://jobs-ncire.icims.com/jobs/1505/job

Thank you,
Brian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100318/31eadbe4/attachment.html>

From paul at ihr.mrc.ac.uk  Thu Mar 18 07:43:29 2010
From: paul at ihr.mrc.ac.uk (Paul Briley)
Date: Thu, 18 Mar 2010 14:43:29 -0000
Subject: [Eeglablist] Statistics on dipole orientations
Message-ID: <004201cac6a9$60418600$20c49200$@mrc.ac.uk>

Hi, I'm looking to compare dipole orientations between conditions, does
anyone know the best way to do this? I'm currently using BESA to fit
dipoles, which gives me an azimuth and elevation angle for each participant.
However, averaging across participants seems problematic - orientations of 2
degrees and 358 degrees would average to 180 degrees, despite them being
only 4 degrees apart. In addition, dipoles can be flipped in orientation
with no change in residual variance so I would need a method which treats,
for example, 0 degrees and 180 degrees the same.

 

Does anyone know where this problem has been tackled before?

 

Thank-you,

Paul.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100318/452e1d1e/attachment.html>

From sjwebb at u.washington.edu  Thu Mar 18 09:18:17 2010
From: sjwebb at u.washington.edu (Sara Jane Webb)
Date: Thu, 18 Mar 2010 09:18:17 -0700
Subject: [Eeglablist] Special Interest Group EEG/MEG & ASD
Message-ID: <5CCE66C6-D6D2-493A-8879-2218061EF1ED@u.washington.edu>

Hello all,

I am writing to let you know that the special interest group on EEG  
and MEG has been accepted for the 2010 International Meeting For  
Autism Research. We will be meeting on May 20th Thursday during the  
lunch hour.

Please forward to other colleagues who might be interested in  
attending the meeting/lunch.

As well, I am creating a LISTSERV to facilitate communication among  
those interested in EEG, MEG & ASD. To subscribe, please follow this  
link.

I look forward to hearing from you.

Sara

Sara Jane Webb, PhD
Research Assistant Professor of Psychiatry and Behavioral Sciences
and UW Autism Center, Research Program
http://depts.washington.edu/pbslab/
Box 357920; CHDD 314C; University of Washington
Seattle WA 98195
206.221.6461
sjwebb at u.washington.edu

Confidentiality Notice:  Because email is not secure, please be aware  
that we cannot guarantee the confidentiality of information sent by  
email.  If you are not the intended recipient, please notify the  
sender by reply email, and then destroy all copies of the message and  
any attachments.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100318/524b127c/attachment.html>

From bradley.voytek at gmail.com  Fri Mar 19 15:17:48 2010
From: bradley.voytek at gmail.com (Bradley Voytek)
Date: Fri, 19 Mar 2010 15:17:48 -0700
Subject: [Eeglablist] Statistics on dipole orientations
In-Reply-To: <004201cac6a9$60418600$20c49200$@mrc.ac.uk>
References: <004201cac6a9$60418600$20c49200$@mrc.ac.uk>
Message-ID: <6d898bac1003191517q1cff6415m1aba10a0b23fc197@mail.gmail.com>

Paul:

Obviously, for the reason you've stated, you can't take the arithmetic
mean, you need to take the circular mean. If you use the circular
stats toolbox you can get the circular mean, perform one-way or
two-way stats, etc.

http://www.mathworks.com/matlabcentral/fileexchange/10676-circular-statistics-toolbox-directional-statistics

::brad

On Thu, Mar 18, 2010 at 07:43, Paul Briley <paul at ihr.mrc.ac.uk> wrote:
> Hi, I?m looking to compare dipole orientations between conditions, does
> anyone know the best way to do this? I?m currently using BESA to fit
> dipoles, which gives me an azimuth and elevation angle for each participant.
> However, averaging across participants seems problematic ? orientations of 2
> degrees and 358 degrees would average to 180 degrees, despite them being
> only 4 degrees apart. In addition, dipoles can be flipped in orientation
> with no change in residual variance so I would need a method which treats,
> for example, 0 degrees and 180 degrees the same.
>
>
>
> Does anyone know where this problem has been tackled before?
>
>
>
> Thank-you,
>
> Paul.
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



From Tim.Curran at Colorado.EDU  Fri Mar 19 17:17:01 2010
From: Tim.Curran at Colorado.EDU (Tim Curran)
Date: Fri, 19 Mar 2010 18:17:01 -0600
Subject: [Eeglablist] JOB OPENING
Message-ID: <CF9DB7B6-5229-47AD-8FFF-7AA42DEC3984@colorado.edu>

JOB OPENING:  
COGNITIVE/SYSTEMS NEUROSCIENTIST

Science Applications International Corporation (SAIC)

SAIC's Technology and Advanced Systems Business Unit has an opening for a Cognitive/Systems Neuroscientist to join a multidisciplinary team conducting research and building systems based on the translation of neuroscience and neuro-imaging (specifically, EEG) research. The individual will support one or more projects, providing expertise in developing protocols, conducting studies, analyzing data, and writing papers and proposals. Applications of interest include brain-computer interfaces and perception, memory, and attention as they relate to man-machine interactions.

Required Skills:

The candidate must have 6+ years of hands-on relevant research and demonstrated mathematical and computational expertise. Strong written and oral communication skills are a must.

Desired Skills:

Experience with EEG collection and processing is highly desired. Current knowledge of research literature with a history of publication in the field is also desired.

Education Requirements:

Ph.D Degree in Psychology, Neuroscience, or related field is required.

Location:   

Louisville/Boulder, Colorado

Contact:

Laurie Gibson, gibsonld at saic.com

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100319/4b0b3301/attachment.html>

From J.Martinovic at liverpool.ac.uk  Sun Mar 21 15:52:34 2010
From: J.Martinovic at liverpool.ac.uk (Jasna Martinovic)
Date: Sun, 21 Mar 2010 22:52:34 +0000
Subject: [Eeglablist] phd position at university of aberdeen
Message-ID: <4BA6A332.6010300@liv.ac.uk>

Interactions between bottom-up and top-down biases in the processing of 
colour; a fully-funded PhD project at the School of Psychology, 
University of Aberdeen


Project description

The efficiency of attentional selection depends on two factors: the 
saliency of the stimulus (bottom-up processing, driven by exogenous 
cues) and the adopted perceptual set (top-down processing, driven by 
endogenous cues). Relations between these stimulus and task-driven 
factors are still poorly understood. However, it seems that in difficult 
tasks, when stimuli compete for limited perceptual resources, endogenous 
cueing interacts with exogenous cues such as stimulus saliency.  In 
real-life situations, the visual system is habitually presented with the 
difficult task of selecting stimuli from diverse and dynamic scenes. 
Therefore it is reasonable to assume that interactions between bottom-up 
and top-down biases are the norm, rather than an exception.  Preliminary 
work, using psychophysics and steady-state visual evoked potentials in 
the EEG, indicates multiplicative, independent effects of saliency and 
top-down selection in a continuous stimulus domain such as luminance. 
The aim of this project is to look for interactions between colour 
saliency and cueing in colour selection using the same methods. Due to 
the discontinuous nature of colour as a stimulus dimension, there may be 
interactions between bottom-up and top-down factors dependent on both 
low-level and high-level colour representations. This will extend the 
previous work on colour selection derived from visual search paradigms 
into the domain of complex dynamic displays and will complement it with 
insights into underlying neural processes through the use of EEG.

The project is based at the School of Psychology of the University of 
Aberdeen and will be supervised by Dr Jasna Martinovic and Prof Arash 
Sahraie.

School of Psychology has state-of-the-art EEG facilities, including two 
64-electrode Biosemi ActiveTwo EEG systems. Vision Research Laboratories 
within the School are equipped with multiple psychophysical workstations 
and a high-resolution eye tracking system (EyeLink 1000). The succesful 
candidate will join an active group of vision researchers comprising of 
academic and research staff and students.


Information for applicants

Candidates must be eligible for UK/EU fee status and should hold a First 
or Upper Second Class Honours degree, a Masters degree or an equivalent 
qualification.  Candidates should have a background in visual  
perception, neuroscience, cognitive psychology or a related field. 
Programming skills (e.g. Matlab, C) and previous experience with 
electrophysiology, psychophysics and signal processing methods are 
highly desirable.


If you would like to be considered for this position, please send a 
cover letter, an up-to-date CV and names of two referees to Dr Jasna 
Martinovic  (j.martinovic at abdn.ac.uk).


From Tim.Curran at Colorado.EDU  Mon Mar 22 11:42:02 2010
From: Tim.Curran at Colorado.EDU (Tim Curran)
Date: Mon, 22 Mar 2010 12:42:02 -0600
Subject: [Eeglablist] Fwd: [FIELDTRIP] JOB OPENING CLARIFICATION
References: <CF9DB7B6-5229-47AD-8FFF-7AA42DEC3984@colorado.edu>
Message-ID: <A2A1F2CF-9A57-4689-9FFB-BFABEC2CB654@colorado.edu>

Dear All,
I previously posted the following job ad, which I was forwarding for a colleague.  Please note, because this position requires the ability to obtain a US government
security clearance, it is open to US citizens only.    I was not aware of this originally, so sorry for posting it to lists with many non-US members.  Also, I have no official ties with this position, so please do not send application materials to me.  I was just passing it along as a favor to a colleague. 
regards,
Tim



Begin forwarded message:

> From: Tim Curran <Tim.Curran at COLORADO.EDU>
> Date: March 19, 2010 6:17:01 PM MDT
> To: FIELDTRIP at NIC.SURFNET.NL
> Subject: [FIELDTRIP] JOB OPENING
> Reply-To: FieldTrip discussion list <FIELDTRIP at NIC.SURFNET.NL>
> 
> JOB OPENING:  
> COGNITIVE/SYSTEMS NEUROSCIENTIST
> 
> Science Applications International Corporation (SAIC)
> 
> SAIC's Technology and Advanced Systems Business Unit has an opening for a Cognitive/Systems Neuroscientist to join a multidisciplinary team conducting research and building systems based on the translation of neuroscience and neuro-imaging (specifically, EEG) research. The individual will support one or more projects, providing expertise in developing protocols, conducting studies, analyzing data, and writing papers and proposals. Applications of interest include brain-computer interfaces and perception, memory, and attention as they relate to man-machine interactions.
> 
> Required Skills:
> 
> The candidate must have 6+ years of hands-on relevant research and demonstrated mathematical and computational expertise. Strong written and oral communication skills are a must.
> 
> Desired Skills:
> 
> Experience with EEG collection and processing is highly desired. Current knowledge of research literature with a history of publication in the field is also desired.
> 
> Education Requirements:
> 
> Ph.D Degree in Psychology, Neuroscience, or related field is required.
> 
> Location:   
> 
> Louisville/Boulder, Colorado
> 
> Contact:
> 
> Laurie Gibson, gibsonld at saic.com
> 
> ----------------------------------
> 
> The aim of this list is to facilitate the discussion between users of the FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis.
> 
> http://listserv.surfnet.nl/archives/fieldtrip.html
> 
> http://www.ru.nl/fcdonders/fieldtrip/
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100322/9988a289/attachment.html>

From Kris.Baetens at vub.ac.be  Mon Mar 22 10:11:24 2010
From: Kris.Baetens at vub.ac.be (Kris Baetens)
Date: Mon, 22 Mar 2010 18:11:24 +0100
Subject: [Eeglablist] Rejecting components by map
Message-ID: <BAEFHEAAHAEAAGHAAA@webmail.ulb.ac.be>

Hello colleagues,

When choosing to "reject components by map", there are some buttons in the lower region of the resulting window (regarding projection, tresholds etc.). Normally, they are greyed out. 
The help file reads: " if the function POP_REJCOMP is ran prior to this function, some fields of the EEG datasets will be present and the current function will have some more button active to tune up the automatic rejection", which, I assume, refers to these buttons. However, I can't get it to work. 
Do these buttons normally provide useful information to decide which components to reject? If so, how can one get them to work?

Many thanks,

Kris Baetens
Ph.D. fellow of the Research Foundation - Flanders (FWO) 
Dept. Experimental and Applied Psychology
Faculty of Psychology and Educational Sciences
Vrije Universiteit Brussel
Pleinlaan 2, 1050 Elsene 
+32 2 629 23 31


From francesco.vespignani at unitn.it  Fri Mar 26 05:24:09 2010
From: francesco.vespignani at unitn.it (Francesco Vespignani)
Date: Fri, 26 Mar 2010 13:24:09 +0100
Subject: [Eeglablist] non-EEG channel
Message-ID: <4BACA769.5050704@unitn.it>

Dear users and developers,
I'm wondering about the possibility to include in the EEG dataset 
structure non-EEG channels that contains other signals such as EMG, SC, 
voice recordings, eye coordinate ... (same sampling rate or resampled 
according to the EEG data) that could be epoched and averaged but 
systematically excluded from ICA, filtering, mapping and other 
EEG-specific computations and maybe have a descriptor field about the 
type unit of the measure.
Clearly it is possible to workaround by assigning no channel location 
property to these channels and exclude them in every EEG-specific 
operation. But it could be useful to have a way to automatically do all 
this. Are there plans for implementing something like this or does this 
possibility already exists (and I did not found it ....)?
Thanks in advance

Francesco Vespignani


From Adrian.Guggisberg at hcuge.ch  Wed Mar 31 02:52:08 2010
From: Adrian.Guggisberg at hcuge.ch (GUGGISBERG Adrian)
Date: Wed, 31 Mar 2010 11:52:08 +0200
Subject: [Eeglablist] Open research position in clinical neurosciences
Message-ID: <37097D0D760C8E47847FFE86892B22A902746963@EXCH2.huge.ad.hcuge.ch>

A post-doctoral or PhD position is available immediately at the Division of Neurorehabilitation, Department of Clinical Neurosciences, University of Geneva, Switzerland. The research group conducts studies in patients with cognitive and neurological dysfunctions after brain damage as well as studies on cognitive mechanisms in healthy participants. The project will use advanced signal processing techniques to investigate the neural correlates of neurological deficits and the mechanisms of recovery in patient populations with focal lesions or diffuse neuronal loss.

Methods: functional imaging with EEG and fMRI.
Qualification requirements: strong background in EEG/MEG and/or fMRI analyses; French language skills. Experience with the Matlab programming environment is desirable but not essential.
Duration: maximum 3 years.
Salary: according to the scales of the University of Geneva and the Swiss National Science Foundation.

To apply for this position, please submit a letter of interest and a CV by April 15, 2010. All materials should be sent electronically in pdf format.

Contact and further information:
Adrian Guggisberg
Service de Neuror??ducation
D?partment des Neurosciences Cliniques
H?pitaux Universitaires de Gen?ve
Avenue de Beau-S?jour 26
CH-1211 Gen?ve 14
Switzerland
Adrian.Guggisberg at hcuge.ch<mailto:Adrian.Guggisberg at hcuge.ch>
Phone: +41 22 382 3521
Fax: +41 22 382 3644

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100331/a986a9f0/attachment.html>

From Kris.Baetens at vub.ac.be  Wed Mar 31 08:04:23 2010
From: Kris.Baetens at vub.ac.be (Kris Baetens)
Date: Wed, 31 Mar 2010 17:04:23 +0200
Subject: [Eeglablist] Random selection of trials
Message-ID: <CHHDBEAADGEHHAAAAJ@webmail.ulb.ac.be>

Dear colleagues,

We employ a paradigm which inherently leads to a different number of trials in both our conditions (oddball-like). We have two conditions, one with an average of about 150 trials, the other with about 1500 (artefact-free).

- Does anybody have research to support my concern that comparing both conditions with the total number of trials may lead to artificial effects due to the different number of trials (and associated variance and "cleanliness" of the gavg's)? (I have seen such things published before.)
- Does anybody know of an easy way to make a random selection of a predetermined number of trials out of the total number in EEGLAB or MATLAB? (Which would allow for selecting an equal number of trials in both conditions.) Obviously, we don't simply want to take the first or last 150 regular trials, since this would possibly lead to erronous conclusions. 

Thank you very much in advance,

Kris Baetens
Ph.D. fellow of the Research Foundation - Flanders (FWO) 
Dept. Experimental and Applied Psychology
Faculty of Psychology and Educational Sciences
Vrije Universiteit Brussel



From bradley.voytek at gmail.com  Thu Apr  1 11:10:36 2010
From: bradley.voytek at gmail.com (Bradley Voytek)
Date: Thu, 1 Apr 2010 11:10:36 -0700
Subject: [Eeglablist] Random selection of trials
In-Reply-To: <CHHDBEAADGEHHAAAAJ@webmail.ulb.ac.be>
References: <CHHDBEAADGEHHAAAAJ@webmail.ulb.ac.be>
Message-ID: <q2k6d898bac1004011110o72016aapab9a8befff3da6dc@mail.gmail.com>

Kris:

One of the best ways of doing this--from a statistical
standpoint--would be through resampling statistics. I've attached a
quick example script that should be correct.

Also, here's a quick description.

http://en.wikipedia.org/wiki/Resampling_%28statistics%29

First, calculate the real mean difference between your two datasets,
and set this value aside.
Next, put all your data into a big pile (for a total of, in your case,
1650 data points).
Next, randomly grab 150 points and calculate that mean. Grab the
remaining 1500 points and calculate that mean. Then calculate the
difference of these two means.
Repeat a lot (e.g., 10000 times) to get a distribution of possible
mean differences.

This gives you a distribution of possible difference values given the
actual data. Because you're taking means of means, this distribution
of surrogate values approaches normality (see the central limit
theorem), and thus we can calculate a z-score and p-value.

Conceptually what you're doing is asking whether the real difference
you observe between conditions is due to your experimental
manipulation or whether it's an artifact of the possible states your
data can obtain.

::bradley voytek


On Wed, Mar 31, 2010 at 08:04, Kris Baetens <Kris.Baetens at vub.ac.be> wrote:
> Dear colleagues,
>
> We employ a paradigm which inherently leads to a different number of trials in both our conditions (oddball-like). We have two conditions, one with an average of about 150 trials, the other with about 1500 (artefact-free).
>
> - Does anybody have research to support my concern that comparing both conditions with the total number of trials may lead to artificial effects due to the different number of trials (and associated variance and "cleanliness" of the gavg's)? (I have seen such things published before.)
> - Does anybody know of an easy way to make a random selection of a predetermined number of trials out of the total number in EEGLAB or MATLAB? (Which would allow for selecting an equal number of trials in both conditions.) Obviously, we don't simply want to take the first or last 150 regular trials, since this would possibly lead to erronous conclusions.
>
> Thank you very much in advance,
>
> Kris Baetens
> Ph.D. fellow of the Research Foundation - Flanders (FWO)
> Dept. Experimental and Applied Psychology
> Faculty of Psychology and Educational Sciences
> Vrije Universiteit Brussel
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: example_surrogate_stats.m
Type: application/octet-stream
Size: 642 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100401/1ecd5203/attachment.obj>

From Tim.Curran at Colorado.EDU  Thu Apr  1 10:41:31 2010
From: Tim.Curran at Colorado.EDU (Tim Curran)
Date: Thu, 1 Apr 2010 11:41:31 -0600
Subject: [Eeglablist] Random selection of trials
In-Reply-To: <CHHDBEAADGEHHAAAAJ@webmail.ulb.ac.be>
References: <CHHDBEAADGEHHAAAAJ@webmail.ulb.ac.be>
Message-ID: <35B669D8-4D36-4B86-A021-1704CB8F8EA4@colorado.edu>

Hi Kris,

For ERP amplitudes, Luck (2005, p 231) explains that different numbers of trials per condition biases comparisons based on peak amplitude but not mean amplitude.  

Luck, S. J. (2005). An introduction to the event-related potential technique. Cambridge, MA: MIT Press.

I have randomly selected like this before, by starting with the number of trials in the lower condition and randomly selecting the same number of trials from the higher condition, on a per subject basis.  Should be easy to do in matlab, but not sure about within eeglab.

I am also not sure if similar concerns apply to EEG time/frequency analyses.

Tim



On Mar 31, 2010, at 9:04 AM, Kris Baetens wrote:

> Dear colleagues,
> 
> We employ a paradigm which inherently leads to a different number of trials in both our conditions (oddball-like). We have two conditions, one with an average of about 150 trials, the other with about 1500 (artefact-free).
> 
> - Does anybody have research to support my concern that comparing both conditions with the total number of trials may lead to artificial effects due to the different number of trials (and associated variance and "cleanliness" of the gavg's)? (I have seen such things published before.)
> - Does anybody know of an easy way to make a random selection of a predetermined number of trials out of the total number in EEGLAB or MATLAB? (Which would allow for selecting an equal number of trials in both conditions.) Obviously, we don't simply want to take the first or last 150 regular trials, since this would possibly lead to erronous conclusions. 
> 
> Thank you very much in advance,
> 
> Kris Baetens
> Ph.D. fellow of the Research Foundation - Flanders (FWO) 
> Dept. Experimental and Applied Psychology
> Faculty of Psychology and Educational Sciences
> Vrije Universiteit Brussel
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu




From siva.digavalli at bms.com  Thu Apr  1 11:45:54 2010
From: siva.digavalli at bms.com (Digavalli, Siva)
Date: Thu, 1 Apr 2010 14:45:54 -0400
Subject: [Eeglablist] Random selection of trials
In-Reply-To: <CHHDBEAADGEHHAAAAJ@webmail.ulb.ac.be>
References: <CHHDBEAADGEHHAAAAJ@webmail.ulb.ac.be>
Message-ID: <A628C0C82A545D45AEEFC6D45F2D1EB903965AB202@ushpwbmsmmp009.one.ads.bms.com>

Hi Kris
Unequal averaging was a concern for us as well in a rodent protocol. What we did was to identify the standard trial preceding the "odd ball" and use that to make an average. That way, you will have the same number of oddball and standard trials, contiguous in time.  I do not use EEGLAB for this and do not know how to accomplish this. I am sure someone else on this forum might be able to. Good Luck!
Siva

>-----Original Message-----
>From: eeglablist-bounces at sccn.ucsd.edu [mailto:eeglablist-
>bounces at sccn.ucsd.edu] On Behalf Of Kris Baetens
>Sent: Wednesday, March 31, 2010 11:04 AM
>To: eeglablist at sccn.ucsd.edu
>Subject: [Eeglablist] Random selection of trials
>
>Dear colleagues,
>
>We employ a paradigm which inherently leads to a different number of
>trials in both our conditions (oddball-like). We have two conditions,
>one with an average of about 150 trials, the other with about 1500
>(artefact-free).
>
>- Does anybody have research to support my concern that comparing both
>conditions with the total number of trials may lead to artificial
>effects due to the different number of trials (and associated variance
>and "cleanliness" of the gavg's)? (I have seen such things published
>before.)
>- Does anybody know of an easy way to make a random selection of a
>predetermined number of trials out of the total number in EEGLAB or
>MATLAB? (Which would allow for selecting an equal number of trials in
>both conditions.) Obviously, we don't simply want to take the first or
>last 150 regular trials, since this would possibly lead to erronous
>conclusions.
>
>Thank you very much in advance,
>
>Kris Baetens
>Ph.D. fellow of the Research Foundation - Flanders (FWO)
>Dept. Experimental and Applied Psychology
>Faculty of Psychology and Educational Sciences
>Vrije Universiteit Brussel
>
>_______________________________________________
>Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>To unsubscribe, send an empty email to eeglablist-
>unsubscribe at sccn.ucsd.edu
>For digest mode, send an email with the subject "set digest mime" to
>eeglablist-request at sccn.ucsd.edu

This message (including any attachments) may contain confidential, proprietary, privileged and/or private information.  The information is intended to be for the use of the individual or entity designated above.  If you are not the intended recipient of this message, please notify the sender immediately, and delete the message and any attachments.  Any disclosure, reproduction, distribution or other use of this message or any attachments by an individual or entity other than the intended recipient is prohibited.



From pzeman at alumni.uvic.ca  Thu Apr  1 11:36:04 2010
From: pzeman at alumni.uvic.ca (Philip Michael Zeman)
Date: Thu, 1 Apr 2010 11:36:04 -0700
Subject: [Eeglablist] Random selection of trials
In-Reply-To: <CHHDBEAADGEHHAAAAJ@webmail.ulb.ac.be>
References: <CHHDBEAADGEHHAAAAJ@webmail.ulb.ac.be>
Message-ID: <7FFE0A1B2AAB4E36B1DB51CBD906EDC3@mine>

Hello Kris

This is an issue that I've encountered myself and I've had to come up with 
some logical reasoning.  I am also very interested in hearing what other 
people have to say on this topic so that I know what options exist.

My reasoning is as follows:

If there is any chance of learning or adaptation of the neural systems 
involved in the paradigm to the stimuli or the context of the stimuli, then 
I would chose to use only the first 150 trials of each of the conditions you 
are examining.  There is plenty of ERP research out there that shows that 
the second-half vs. the first have of an experiment has characteristically 
different waveforms. (See, face recognition/unknown face ERP literature, 
e.g., Tanaka - I think that participants begin to learn what could be 
considered the 'odd-ball' in this experiment).

Your concern about averaging a different number of trials for each 
condition:
*Yes, this is a concern.  My philosophy is to always compare apples to 
apples --- always try to do the same processing in condition 1 and in 
condition 2 if you are going to ask the question, "are these 2 conditions 
different?" using a statistical test.  If in condition 1 there are 150 
trials and in condition 2 there are 1000 trials (and you use them all) then 
when you average the data within each participant and then look for a 
difference in the distribution across participants you already know there is 
a possible difference simply due to what you did to the data.  Depending on 
the signal-to-noise characteristics (and what we consider to be noise, e.g., 
variability in behaviour, uncorrelated sensor noise, learning and 
adaptation), you might have a distribution of values in condition 2 that are 
tightly distributed around the mean for the condition compared to in 
condition 1 that has a distribution of scores that are spread by a large 
amount around the mean of the condition.

If you are not concerned about learning and adaptation in your experiment, 
(and I think this goes against standard research hypothesis testing rules) 
what I would suggest is doing your hypothesis test in multiple comparisons 
so that you can learn something about your data.  I would answer my 
hypothesis using (1) and then go further by investigating (2) and (3) to see 
what I get.
(1) compare condition 1, trials 1 to 150 vs. condition 2, trials 1 to 150
(2) compare condition 1, trials 1 to 150 vs. condition 2, trials N-150 to N 
(where N is the number of trials in the large set)
(3) compare condition 1, trials 1 to 150 vs. condition 2, random blocks of 
150 trails (and do this multiple times on different random blocks)

Keep me updated as to what you decided to do.


Regards,

Phil

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Philip Michael Zeman B.Eng, Ph.D.
Applied Brain and Vision Sciences Inc.
Brain Function Analysis for Novel Paradigms and Serious Games
Analysis of Pharmaceutical Effects on Brain Function
http://www.abvsciences.com
Latest Brain Research Result:
http://www.spatialbrain.com
Email:   pzeman at alumni.uvic.ca
Phone: +1-250-589-4234
LinkedIn Profile: http://ca.linkedin.com/in/philipmichaelzeman
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

----- Original Message ----- 
From: "Kris Baetens" <Kris.Baetens at vub.ac.be>
To: <eeglablist at sccn.ucsd.edu>
Sent: Wednesday, March 31, 2010 8:04 AM
Subject: [Eeglablist] Random selection of trials


> Dear colleagues,
>
> We employ a paradigm which inherently leads to a different number of 
> trials in both our conditions (oddball-like). We have two conditions, one 
> with an average of about 150 trials, the other with about 1500 
> (artefact-free).
>
> - Does anybody have research to support my concern that comparing both 
> conditions with the total number of trials may lead to artificial effects 
> due to the different number of trials (and associated variance and 
> "cleanliness" of the gavg's)? (I have seen such things published before.)
> - Does anybody know of an easy way to make a random selection of a 
> predetermined number of trials out of the total number in EEGLAB or 
> MATLAB? (Which would allow for selecting an equal number of trials in both 
> conditions.) Obviously, we don't simply want to take the first or last 150 
> regular trials, since this would possibly lead to erronous conclusions.
>
> Thank you very much in advance,
>
> Kris Baetens
> Ph.D. fellow of the Research Foundation - Flanders (FWO)
> Dept. Experimental and Applied Psychology
> Faculty of Psychology and Educational Sciences
> Vrije Universiteit Brussel
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to 
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to 
> eeglablist-request at sccn.ucsd.edu 



From matthew.mollison at colorado.edu  Thu Apr  1 11:36:26 2010
From: matthew.mollison at colorado.edu (Matt Mollison)
Date: Thu, 1 Apr 2010 12:36:26 -0600
Subject: [Eeglablist] Random selection of trials
In-Reply-To: <CHHDBEAADGEHHAAAAJ@webmail.ulb.ac.be>
References: <CHHDBEAADGEHHAAAAJ@webmail.ulb.ac.be>
Message-ID: <x2mbf40af761004011136m2a79d121t3adf7b766941439@mail.gmail.com>

As you imply, there are issues to be aware of when comparing conditions with
vastly different numbers of trials, but I couldn't do the topic justice.

To your second point, you could use the Matlab function randperm to make a
random index. The following code may not be optimal, but it works, so just
as an example:

ind = randperm(max([size(cond1,1),size(cond2,1)]));
ind = ind(1:min([size(cond1,1),size(cond2,1)]));

Don't forget to set the state of the rand function before using randperm;
e.g., rand('state',sum(100*clock)).

-Matt Mollison

--
Univ. of Colorado at Boulder
Dept. of Psychology and Neuroscience
matthew.mollison at colorado.edu
http://psych.colorado.edu/~mollison/


On Wed, Mar 31, 2010 at 9:04 AM, Kris Baetens <Kris.Baetens at vub.ac.be>wrote:

> Dear colleagues,
>
> We employ a paradigm which inherently leads to a different number of trials
> in both our conditions (oddball-like). We have two conditions, one with an
> average of about 150 trials, the other with about 1500 (artefact-free).
>
> - Does anybody have research to support my concern that comparing both
> conditions with the total number of trials may lead to artificial effects
> due to the different number of trials (and associated variance and
> "cleanliness" of the gavg's)? (I have seen such things published before.)
> - Does anybody know of an easy way to make a random selection of a
> predetermined number of trials out of the total number in EEGLAB or MATLAB?
> (Which would allow for selecting an equal number of trials in both
> conditions.) Obviously, we don't simply want to take the first or last 150
> regular trials, since this would possibly lead to erronous conclusions.
>
> Thank you very much in advance,
>
> Kris Baetens
> Ph.D. fellow of the Research Foundation - Flanders (FWO)
> Dept. Experimental and Applied Psychology
> Faculty of Psychology and Educational Sciences
> Vrije Universiteit Brussel
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100401/c266f5bb/attachment.html>

From jackm_ustc at 163.com  Wed Mar 31 16:21:26 2010
From: jackm_ustc at 163.com (=?gbk?B?0+C2vba9?=)
Date: Thu, 1 Apr 2010 07:21:26 +0800 (CST)
Subject: [Eeglablist] 'timesout' in pop_newtimef
Message-ID: <1edb354.5b1.127b6878199.Coremail.jackm_ustc@163.com>

Hi experts,

I am new to the time-frequency analysis of EEGLAB. I have a dataset of hundreds of epoch. The sampling rate is 500. Each epoch starts at -100ms and ends at 298ms. I'd like to see the ERSP of 10hz-50hz at the entire time range, i.e. -100-298ms. So I ran this command

figure; [ersp,itc,powbase,times,freqs,erspboot,itcboot]=pop_newtimef( EEG, 1, 47, [-100 298], [3 0.5] ,'type', 'phasecoher', 'title','Channel P2 power and inter-trial phase coherence', 'alpha',.05,'padratio', 16, 'plotphase','off','freqs',[10 50],'timesout',[-100:2:298]);

However, the output ERSP plot is always from 68ms to 130ms. And I saw a warning message in Matlab command window:

Warning: 168 out of 200 time values were removed (now 68.00 to 130.00 ms) so the lowest
         frequency could be computed with the requested accuracy

And the output 'times' is indeed between 68 and 130
times =
  Columns 1 through 7
   68.8485   70.8586   72.8687   74.8788   76.8889   78.8990   80.9091
  Columns 8 through 14
   82.9192   84.9293   86.9394   88.9495   90.9596   92.9697   94.9798
  Columns 15 through 21
   96.9899   99.0000   99.0000  101.0101  103.0202  105.0303  107.0404
  Columns 22 through 28
  109.0505  111.0606  113.0707  115.0808  117.0909  119.1010  121.1111
  Columns 29 through 32
  123.1212  125.1313  127.1414  129.1515

What does this mean? And how can I get enough data at broader time range? 
I appreciate very much for any response.

Jack
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100401/95ccf87e/attachment.html>

From arno at ucsd.edu  Thu Apr  1 12:49:34 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Thu, 1 Apr 2010 12:49:34 -0700
Subject: [Eeglablist] Random selection of trials
In-Reply-To: <q2k6d898bac1004011110o72016aapab9a8befff3da6dc@mail.gmail.com>
References: <CHHDBEAADGEHHAAAAJ@webmail.ulb.ac.be>
	<q2k6d898bac1004011110o72016aapab9a8befff3da6dc@mail.gmail.com>
Message-ID: <ECA12AFA-6264-4A94-9DF4-DF6307B22C17@ucsd.edu>

Dear Kris,

in EEGLAB try the statcond function. It will perform boostrap,  
permutation (bootstrap and permutation are two resampling techniques),  
or compute parametric statistics and return p-values. If you have 2  
EEGLAB datasets 1 and 2 (on which you have extracted data epochs of  
the same length), on the command line, try

[t df p] = statcond({ ALLEEG(1).data ALLEEG(2).data }, 'mode',  
'bootstrap', 'naccu', 1000);
% or [t df p] = statcond({ ALLEEG(1).data ALLEEG(2).data}, 'mode',  
'perm');
% or [t df p] = statcond({ ALLEEG(1).data ALLEEG(2).data}, 'mode',  
'param');

then use FDR to correct for multiple comparisons

p = fdr(p);

then plot the result with a threshold at 0.05 for a given channel

chan = 10;
erp_1 = mean(ALLEEG(1).data(chan,:,:),3);
erp_2 = mean(ALLEEG(2).data(chan,:,:),3);
plotcurve(ALLEEG(1).times, [erp_1; erp_2], 'maskarray', p(chan,:) <  
0.05);

Hope this helps,

Arno

ps: with many channels, time points, and trials, the bootstrap  
computation may take several minutes

On Apr 1, 2010, at 11:10 AM, Bradley Voytek wrote:

> Kris:
>
> One of the best ways of doing this--from a statistical
> standpoint--would be through resampling statistics. I've attached a
> quick example script that should be correct.
>
> Also, here's a quick description.
>
> http://en.wikipedia.org/wiki/Resampling_%28statistics%29
>
> First, calculate the real mean difference between your two datasets,
> and set this value aside.
> Next, put all your data into a big pile (for a total of, in your case,
> 1650 data points).
> Next, randomly grab 150 points and calculate that mean. Grab the
> remaining 1500 points and calculate that mean. Then calculate the
> difference of these two means.
> Repeat a lot (e.g., 10000 times) to get a distribution of possible
> mean differences.
>
> This gives you a distribution of possible difference values given the
> actual data. Because you're taking means of means, this distribution
> of surrogate values approaches normality (see the central limit
> theorem), and thus we can calculate a z-score and p-value.
>
> Conceptually what you're doing is asking whether the real difference
> you observe between conditions is due to your experimental
> manipulation or whether it's an artifact of the possible states your
> data can obtain.
>
> ::bradley voytek
>
>
> On Wed, Mar 31, 2010 at 08:04, Kris Baetens <Kris.Baetens at vub.ac.be>  
> wrote:
>> Dear colleagues,
>>
>> We employ a paradigm which inherently leads to a different number  
>> of trials in both our conditions (oddball-like). We have two  
>> conditions, one with an average of about 150 trials, the other with  
>> about 1500 (artefact-free).
>>
>> - Does anybody have research to support my concern that comparing  
>> both conditions with the total number of trials may lead to  
>> artificial effects due to the different number of trials (and  
>> associated variance and "cleanliness" of the gavg's)? (I have seen  
>> such things published before.)
>> - Does anybody know of an easy way to make a random selection of a  
>> predetermined number of trials out of the total number in EEGLAB or  
>> MATLAB? (Which would allow for selecting an equal number of trials  
>> in both conditions.) Obviously, we don't simply want to take the  
>> first or last 150 regular trials, since this would possibly lead to  
>> erronous conclusions.
>>
>> Thank you very much in advance,
>>
>> Kris Baetens
>> Ph.D. fellow of the Research Foundation - Flanders (FWO)
>> Dept. Experimental and Applied Psychology
>> Faculty of Psychology and Educational Sciences
>> Vrije Universiteit Brussel
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime"  
>> to eeglablist-request at sccn.ucsd.edu
>>
> <example_surrogate_stats.m><ATT00001..txt>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100401/a58ed0f6/attachment.html>

From bradley.voytek at gmail.com  Thu Apr  1 16:04:55 2010
From: bradley.voytek at gmail.com (Bradley Voytek)
Date: Thu, 1 Apr 2010 16:04:55 -0700
Subject: [Eeglablist] 'timesout' in pop_newtimef
In-Reply-To: <1edb354.5b1.127b6878199.Coremail.jackm_ustc@163.com>
References: <1edb354.5b1.127b6878199.Coremail.jackm_ustc@163.com>
Message-ID: <x2w6d898bac1004011604v10b8e510o452a3b20bc7abcdb@mail.gmail.com>

Jack:

You need a larger window. Because newtimef decomposes a signal using
different wavelet sizes depending on the frequencies of interest, the
lowest frequency, and padratio, determine the size of the edges of
your data.

Try a window from -200 - 400.

::brad



2010/3/31 ??? <jackm_ustc at 163.com>:
> Hi experts,
>
> I am new to the time-frequency analysis of EEGLAB. I have a dataset of
> hundreds of epoch. The sampling rate is 500. Each epoch starts at -100ms and
> ends at 298ms. I'd like to see the ERSP of 10hz-50hz at the entire time
> range, i.e. -100-298ms. So I ran this command
>
> figure; [ersp,itc,powbase,times,freqs,erspboot,itcboot]=pop_newtimef( EEG,
> 1, 47, [-100 298], [3 0.5] ,'type', 'phasecoher', 'title','Channel P2 power
> and inter-trial phase coherence', 'alpha',.05,'padratio', 16,
> 'plotphase','off','freqs',[10 50],'timesout',[-100:2:298]);
>
> However, the output ERSP plot is always from 68ms to 130ms. And I saw a
> warning message in Matlab command window:
>
> Warning: 168 out of 200 time values were removed (now 68.00 to 130.00 ms) so
> the lowest
> ???????? frequency could be computed with the requested accuracy
>
> And the output 'times' is indeed between 68 and 130
> times =
> ? Columns 1 through 7?? 68.8485?? 70.8586?? 72.8687?? 74.8788?? 76.8889
> 78.8990?? 80.9091
> ? Columns 8 through 14
> ?? 82.9192?? 84.9293?? 86.9394?? 88.9495?? 90.9596?? 92.9697?? 94.9798
> ? Columns 15 through 21
> ?? 96.9899?? 99.0000?? 99.0000? 101.0101? 103.0202? 105.0303? 107.0404
> ? Columns 22 through 28
> ? 109.0505? 111.0606? 113.0707? 115.0808? 117.0909? 119.1010? 121.1111
> ? Columns 29 through 32
> ? 123.1212? 125.1313? 127.1414? 129.1515
>
> What does this mean? And how can I get enough data at broader time range?
> I appreciate very much for any response.
>
> Jack
>
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



From Kris.Baetens at vub.ac.be  Fri Apr  2 02:39:51 2010
From: Kris.Baetens at vub.ac.be (Kris Baetens)
Date: Fri, 02 Apr 2010 11:39:51 +0200
Subject: [Eeglablist] Random selection of trials
Message-ID: <BECAAEAAFAAGHBAEEH@webmail.ulb.ac.be>

Hi all, 

I would like to start by thanking everybody for their kind input and time. (For completeness' sake, I 
should note that in the numbers of regulars and deviants, only those were included that followed a 
regular in the first place, as some people voiced concerns about this.)

Some comments:
@ Tim curran: I have read Luck's book and I clearly see why using different trials in case of 
peak measures (which we do not use) might be problematic. However, I can't quite understand why 
average amplitudes would be 'immune' to this sort of problem; I think sensitivity to this 
problem decreases as the number of time points in your average window increases? (In extremis:
an average amplitude of two time points would, in the same logic as he discribes, still be extremely 
vulnerable, no?) 

@ Bradley Voytek: Thank you very much for the idea and the example script. We use a similar method
for difference testing between conditions as outlined by Maris en Oostenveld (2007) and are rather 
pleased about the results. 

@Arnaud Delorme & Matt Mollison: Many thanks for the input. I am relatively new to both eeglab and matlab, 
and even though I have found it is possible to achieve a rather steep learning curve with some effort, it
would probably have taken me hours to figure this out by myself. 

@Philip Michael Zeman: All in all, you formulated my concern rather accurately ("...you might have a 
distribution of values in condition 2 that are tightly distributed around the mean for the condition 
compared to in condition 1 that has a distribution of scores that are spread by a large 
amount around the mean of the condition.") However, selecting the first 150 of each kind of trial
doesn't seem an option; in the case of the "regulars", these would all be from the first minutes of 
the recording, whereas in the other condition they would be spread out over the entire recording. 
Whereas there are clearly issues with learning mechanisms, I cannot assume that learning in
the one condition is independent of the other (regulars follow an implicit sequence, whereas regulars
deviate from the same sequence. Both require "knowledge" of the same sequence.)

@Siva Digivalli: The solution you (and others) suggested, seems the most elegant/easy to implement. 

Thanks,
Kris


Ph.D. fellow of the Research Foundation - Flanders (FWO) 
Dept. Experimental and Applied Psychology
Faculty of Psychology and Educational Sciences
Vrije Universiteit Brussel
Pleinlaan 2, 1050 Elsene 
+32 2 629 23 31


From mjstarr at gmail.com  Fri Apr  2 15:36:40 2010
From: mjstarr at gmail.com (Mark Starr)
Date: Fri, 2 Apr 2010 16:36:40 -0600
Subject: [Eeglablist] importing acq files into eeglab
Message-ID: <n2n4898df141004021536n156bf301j7fbe0b142a32b62d@mail.gmail.com>

hi everyone,
i've run into a pair of errors when importing biopac .acq files into eeglab
using the pop_biosig function. here is the error displayed in the command
window: Warning SOPEN: Automated OVERFLOWDETECTION not supported - check
yourself for saturation artifacts.

and here is the error displayed in a dialog box: Exceeded value of bitmax.

any help to resolve the errors would be much appreciated!

kind regards,
mark
-- 
Mark J. Starr
Graduate Student
Department of Psychology
University of Wisconsin - Madison
1202 West Johnson St.
Madison, WI 53706
http://dionysus.psych.wisc.edu/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100402/6435d8f2/attachment.html>

From philippe.boulinguez at isc.cnrs.fr  Thu Apr  1 23:58:06 2010
From: philippe.boulinguez at isc.cnrs.fr (Philippe Boulinguez)
Date: Fri, 02 Apr 2010 08:58:06 +0200
Subject: [Eeglablist] Open Engineer Position
Message-ID: <4BB5957E.2030601@isc.cnrs.fr>

A 2 year full-time position is available at the:
Centre de Neuroscience Cognitive, UMR CNRS 5229, Lyon, France.
http://cnc.isc.cnrs.fr/index.php

Job description: The successful candidate will be part of a 
multidisciplinary team using neuroimaging, electrophysiological and 
behavioural methods to investigate brain functions in humans and 
animals. The project focuses on the role of inhibitory control of 
action. It will be conducted both in normal subjects and in large 
clinical populations.

The engineer will set-up and run EEG experiments using the Biosemi 
Active Two system and will analyze data using ICA and source 
reconstruction methods. He (she) will be responsible for (1) designing, 
conducting and analyzing experiments in conjunction with the other 
members of the team, and (2) participating to the day-to-day running and 
scientific activity of the research group and laboratory. Secondarily, 
he (she) may also have the opportunity to be involved in collecting and 
processing fMRI data in similar experiments or simultaneous recordings.

Requirements: Candidates should have at least a Master?s degree in 
Biomedical engineering or similar qualification in any related field 
(with possibility to transform the engineer position into a PhD position 
in Neuroscience).
Familiarity with EEG techniques (in particular ICA and source 
reconstruction methods) and a strong knowledge of EEGLab are requested. 
Applicants who have experience in applied programming are encouraged to 
apply (high level skills with Matlab are expected). A supplementary 
working knowledge of fMRI or combined EEG/fMRI recording techniques 
would be welcome but is not a prerequisite.

Expected start date: as soon as possible
Salary will be set according to the scales of the /Centre National de la 
Recherche Scientifique/ (depends on experience).

To apply for this position, please submit a statement of interests and a 
curriculum vitae to:
philippe.boulinguez at isc.cnrs.fr

-- 

Philippe Boulinguez
Centre de Neuroscience Cognitive, UMR CNRS 5229
ISC, 67 Boulevard Pinel
69675 BRON cedex
France
tel: +33 (0)4 37 91 12 22
http://www.cnc.isc.cnrs.fr




From Tim.Curran at Colorado.EDU  Fri Apr  2 16:09:03 2010
From: Tim.Curran at Colorado.EDU (Tim Curran)
Date: Fri, 2 Apr 2010 17:09:03 -0600
Subject: [Eeglablist] Random selection of trials
In-Reply-To: <BECAAEAAFAAGHBAEEH@webmail.ulb.ac.be>
References: <BECAAEAAFAAGHBAEEH@webmail.ulb.ac.be>
Message-ID: <529C8CDA-2545-4F23-B247-E7C71ABC95AC@colorado.edu>


On Apr 2, 2010, at 3:39 AM, Kris Baetens wrote:
> @ Tim curran: I have read Luck's book and I clearly see why using different trials in case of 
> peak measures (which we do not use) might be problematic. However, I can't quite understand why 
> average amplitudes would be 'immune' to this sort of problem; I think sensitivity to this 
> problem decreases as the number of time points in your average window increases? (In extremis:
> an average amplitude of two time points would, in the same logic as he discribes, still be extremely 
> vulnerable, no?) 

I had the same question/doubt when I first read this.  I had long email discussions with Luck about it, and actually ended up writing a simulation in order to convince myself that he was right.  The bottom line is that I cannot explain it myself, but I became convinced that he is correct.
Tim




From dgroppe at cogsci.ucsd.edu  Sat Apr  3 10:36:16 2010
From: dgroppe at cogsci.ucsd.edu (David Groppe)
Date: Sat, 3 Apr 2010 10:36:16 -0700
Subject: [Eeglablist] Random selection of trials
In-Reply-To: <529C8CDA-2545-4F23-B247-E7C71ABC95AC@colorado.edu>
References: <BECAAEAAFAAGHBAEEH@webmail.ulb.ac.be>
	<529C8CDA-2545-4F23-B247-E7C71ABC95AC@colorado.edu>
Message-ID: <q2m3d4c78cd1004031036g72af900ela250b47f6bde11e3@mail.gmail.com>

On Fri, Apr 2, 2010 at 4:09 PM, Tim Curran <Tim.Curran at colorado.edu> wrote:

>
> On Apr 2, 2010, at 3:39 AM, Kris Baetens wrote:
> > @ Tim curran: I have read Luck's book and I clearly see why using
> different trials in case of
> > peak measures (which we do not use) might be problematic. However, I
> can't quite understand why
> > average amplitudes would be 'immune' to this sort of problem; I think
> sensitivity to this
> > problem decreases as the number of time points in your average window
> increases? (In extremis:
> > an average amplitude of two time points would, in the same logic as he
> discribes, still be extremely
> > vulnerable, no?)
>
> I had the same question/doubt when I first read this.  I had long email
> discussions with Luck about it, and actually ended up writing a simulation
> in order to convince myself that he was right.  The bottom line is that I
> cannot explain it myself, but I became convinced that he is correct.
> Tim
>
>
Here's the rationale:
1) When you have fewer trials in one condition than another, the variance of
the estimated ERP is greater.  Thus you're more likely to get extreme values
(both above and below the true mean) in the condition with fewer trials.

2) If you measure the mean amplitude in a particular time window, the
extreme above-the-mean and below-the-mean values tend to cancel out.  So, on
average, your estimates of mean amplitude across the two conditions won't
differ (if they really have the same mean amplitude).

3) If you measure the peak amplitude in a particular time window, you're
measuring the most extreme value of a particular polarity.  Now the extreme
values in one direction (e.g., above the mean) will NOT be canceled out by
extreme values in the other direction (e.g., below the mean) and you'll be
more likely to get bigger peak amplitudes in the condition with fewer trials
(even if they really have the same peak amplitude).

    hope that helps,
          -David




>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100403/09320704/attachment.html>

From belmonte-bulk at mit.edu  Thu Mar 25 05:58:33 2010
From: belmonte-bulk at mit.edu (Matthew Belmonte)
Date: Thu, 25 Mar 2010 08:58:33 -0400 (EDT)
Subject: [Eeglablist] new or recent graduates: EEG research assistant job at
	Cornell University
Message-ID: <201004031252.o33CquGY009306@multics.mit.edu>

[This seems not to have got through to the list when I first sent it this past month - so I'm sending again.  Apologies for any duplication. -MB]

COGNITIVE ELECTROENCEPHALOGRAPHY RESEARCH AIDE, CORNELL UNIVERSITY, ITHACA, NY
The Department of Human Development at Cornell University, Ithaca, New York solicits applications for the post of research aide in its cognitive electroencephalography laboratory, to commence summer 2010.  The successful candidate will support researchers using a state-of-the art 128-channel BioSemi ActiveTwo electroencephalographic recording system, an EyeLink 1000 infrared gaze tracker, and associated computing systems for stimulus delivery and for behavioural and physiological data recording.  The research aide will become the local "guru" for these systems, working with researchers who have theoretical backgrounds in psychology and who need assistance to translate these theoretical ideas into experimental practice in the context of a cognitive electrophysiology laboratory.

Essential skills are computer programming (software is implemented in MATLAB) and superb written and oral technical communication.  Prior knowledge of neuroscience, though helpful, is not essential (and will be taught).  Knowledge of basic signal processing (e.g. interpretation of spectral perturbations and other time-frequency analyses), statistics, experience with amplifier systems, and knowledge of real-time computing or embedded systems also would be great advantages, as would any experience setting up and maintaining computing resources and/or laboratory devices.

This is an excellent opportunity for a recent graduate who aims to acquire methodological experience before applying to graduate school in neuroscience, biological engineering, or scientific computing.  Applications from students of engineering, computer science or the physical sciences would be particularly welcome.  Although applications from all qualified candidates are encouraged and will be considered, those able to make a two-year commitment are preferred.

Applicants should send a c.v., a cover letter, details of three referees, a brief (one-page) technical writing sample, and a brief (one source file) program code sample to Matthew Belmonte <mkb4 at cornell.edu>.  New graduates should also send an unofficial copy of their university transcript.  Although this post will remain open until filled, to ensure full consideration materials should be submitted by Monday 5 April.

This is a half-time post but can be supplemented with other technical employment on campus, depending on the candidate's skills and background, to create a full-time post.  (When applying, please state your interest in full-time or half-time.)  Salary will be commensurate with experience and is expected to be in Cornell University Band C (currently $15.40 to $20.70 per hour).



From belmonte-bulk at mit.edu  Fri Apr  2 08:49:32 2010
From: belmonte-bulk at mit.edu (Matthew Belmonte)
Date: Fri, 2 Apr 2010 11:49:32 -0400 (EDT)
Subject: [Eeglablist] how is trial phase-sorting determined for a RANGE of
	frequencies?
Message-ID: <201004031253.o33Cr4DJ009318@multics.mit.edu>

I have what is perhaps a very basic question about phase-sorting in ERP plots.
When I select Plot->Channel ERP image, and if in the "Sort trials by phase"
section of the pop-up window I were to fill in the range "5 30" rather than a
single, narrow-band frequency, how is the phase information across frequencies
combined to yield a single, linearly ordered sort key on which the trials are
arranged?  I can see how this sort key is unambiguously and straightforwardly
defined for a single frequency (or in general for a narrowband range that's
less than the frequency granularity permitted by the number of time points
under consideration), but I'm having trouble visualising how it's determined
for a broadband range.

--
Matthew Belmonte  <belmonte at mit.edu>  http://www.mit.edu/~belmonte/
Assistant Professor, Department of Human Development, Cornell University
G62A Martha Van Rensselaer Hall, Ithaca, NY 14853-4401


From smakeig at gmail.com  Mon Apr  5 07:47:20 2010
From: smakeig at gmail.com (Scott Makeig)
Date: Mon, 5 Apr 2010 07:47:20 -0700
Subject: [Eeglablist] how is trial phase-sorting determined for a RANGE
	of frequencies?
In-Reply-To: <201004031253.o33Cr4DJ009318@multics.mit.edu>
References: <201004031253.o33Cr4DJ009318@multics.mit.edu>
Message-ID: <m2o9e09b8f01004050747r511d3d77j542e97cded4eff72@mail.gmail.com>

Matthew -

In the erpimage() function, specifying a range of frequencies causes the
function to select a frequency in the given range with highest power in the
data (i.e., the peak frequency). This frequency is given when the power and
ITC time courses are also plotted below the ERP-image panel.

Scott Makeig

On Fri, Apr 2, 2010 at 8:49 AM, Matthew Belmonte <belmonte-bulk at mit.edu>wrote:

> I have what is perhaps a very basic question about phase-sorting in ERP
> plots.
> When I select Plot->Channel ERP image, and if in the "Sort trials by phase"
> section of the pop-up window I were to fill in the range "5 30" rather than
> a
> single, narrow-band frequency, how is the phase information across
> frequencies
> combined to yield a single, linearly ordered sort key on which the trials
> are
> arranged?  I can see how this sort key is unambiguously and
> straightforwardly
> defined for a single frequency (or in general for a narrowband range that's
> less than the frequency granularity permitted by the number of time points
> under consideration), but I'm having trouble visualising how it's
> determined
> for a broadband range.
>
> --
> Matthew Belmonte  <belmonte at mit.edu>  http://www.mit.edu/~belmonte/
> Assistant Professor, Department of Human Development, Cornell University
> G62A Martha Van Rensselaer Hall, Ithaca, NY 14853-4401
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Scott Makeig, Research Scientist and Director, Swartz Center for
Computational Neuroscience, Institute for Neural Computation, University of
California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100405/eacb9541/attachment.html>

From sjluck at ucdavis.edu  Sun Apr  4 20:54:57 2010
From: sjluck at ucdavis.edu (Steve Luck)
Date: Sun, 4 Apr 2010 20:54:57 -0700
Subject: [Eeglablist] Random selection of trials
In-Reply-To: <mailman.0.1270321202.1567.eeglablist@sccn.ucsd.edu>
References: <mailman.0.1270321202.1567.eeglablist@sccn.ucsd.edu>
Message-ID: <8136405A-77FE-4158-AFC5-E9628295F027@ucdavis.edu>

As Tim Curran notes, if you measure mean amplitude rather than peak amplitude, your measure won't be biased by differences in the number of trials.  That is, the measures will be noisier in the conditions with fewer trials, leading to lower p-values, but this won't lead to artificial differences between conditions.

I'd like to thank Tim for referring people to the discussion of this issue in my book.  However, the discussion in the book is brief, and I have had many people ask me for a more extensive explanation.  I've therefore written an essay on this, which I plan to include in the next edition of my book.  To see this essay, simply point your browser to http://erpinfo.org/Members/sjluck/Mean_Peak_Noise.pdf.  I hope people find this helpful!

Steve Luck

> From: Kris Baetens <Kris.Baetens at vub.ac.be>
> Date: April 2, 2010 2:39:51 AM PDT
> To: eeglablist at sccn.ucsd.edu
> Subject: Re: [Eeglablist] Random selection of trials
> 
> @ Tim curran: I have read Luck's book and I clearly see why using different trials in case of 
> peak measures (which we do not use) might be problematic. However, I can't quite understand why 
> average amplitudes would be 'immune' to this sort of problem; I think sensitivity to this 
> problem decreases as the number of time points in your average window increases? (In extremis:
> an average amplitude of two time points would, in the same logic as he discribes, still be extremely 
> vulnerable, no?) 
> 

--------------------------------------------------------------------
Steven J. Luck, Ph.D.
Interim Director, Center for Mind & Brain
Professor, Department of Psychology
University of California, Davis
Room 127
267 Cousteau Place
Davis, CA 95618
(530) 297-4424
sjluck at ucdavis.edu
Web: http://mindbrain.ucdavis.edu/people/sjluck
Calendar: http://www.google.com/calendar/embed?src=stevenjluck%40gmail.com&ctz=America/Los_Angeles
--------------------------------------------------------------------






-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100404/e9d59f55/attachment.html>

From schalk at wadsworth.org  Mon Apr  5 14:17:11 2010
From: schalk at wadsworth.org (Gerwin Schalk)
Date: Mon, 5 Apr 2010 17:17:11 -0400
Subject: [Eeglablist] 7th BCI2000 Workshop ...
Message-ID: <7F7D7BB5-9455-4129-AAD4-C19BDD8D5347@wadsworth.org>

Dear colleagues,

This is a reminder about the 7th BCI2000 Workshop, which will be held on May 30-31 just prior to the 4th International BCI Meeting at The Asilomar Conference Center on the Monterey Peninsula in Pacific Grove, California.

This workshop will contain technical lectures about using BCI2000 for brain-computer interfacing and related neurophysiological experiments, and hand-on tutorials using seven complete BCI systems. We will also discuss novel features of the upcoming version v3.0, and, with some luck, also showcase the BCI2000 book that is about to be published by Springer.

Please find more information about the workshop program and registration on 

http://www.bci2000.org/BCI2000/Workshop.html


Hope to see you at what will most likely be the biggest BCI2000 Workshop in the ten-year history of the project.

Gerwin Schalk
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Gerwin Schalk, Ph.D.
Research Scientist V
Wadsworth Center, NYS Dept. of Health
Dept. of Neurology, Albany Medical College
Dept. of Neurosurgery, Washington Univ. in St. Louis
Dept. of Biomed. Eng., Rensselaer Polytechnic Institute
Dept. of Biomed. Sci., State Univ. of New York at Albany
C650 Empire State Plaza
Albany, New York 12201
phone  (518) 486-2559
fax    (518) 486-4910
e-mail schalk at wadsworth.org
www    http://www.bci2000.org
www    http://www.brainmuri.org
www    http://www.gerv.org
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~





IMPORTANT NOTICE: This e-mail and any attachments may contain
confidential or sensitive information which is, or may be, legally
privileged or otherwise protected by law from further disclosure.  It
is intended only for the addressee.  If you received this in error or
from someone who was not authorized to send it to you, please do not
distribute, copy or use it or any attachments.  Please notify the
sender immediately by reply e-mail and delete this from your
system. Thank you for your cooperation.





From kaztanji at gmail.com  Mon Apr  5 16:47:15 2010
From: kaztanji at gmail.com (Kazuyo Tanji)
Date: Tue, 6 Apr 2010 08:47:15 +0900
Subject: [Eeglablist] complex numbers for tfdata
Message-ID: <p2nfc2f2c2d1004051647ub9f8222fo2cd6dd6938cf9a3b@mail.gmail.com>

Hello all,

I am wondering if someone can tell me how I should interpret 'tfdata', an
output of the function newtimef() which is in the form of complex numbers. I
can tell the matrix is structured as frequency steps x time points x trials
but why complex numbers??
Any help is appreciated.

Thanks,
Kazu

-- 
????
????????????
?990-9585 ??????2-2-2
?? 023-628-5428 ??? 023-628-5425
e-mail: kaztanji at gmail.com
????? 6732


Kazuyo Tanji
Yamagata University
Department of Clinical Neuroscience
2-2-2 Iida-nishi Yamagata,Japan
phone 81-23-628-5428
fax      81-23-628-5425
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100406/3c7dfa47/attachment.html>

From krysta.chauncey at tufts.edu  Sun Apr  4 16:15:05 2010
From: krysta.chauncey at tufts.edu (Krysta Chauncey)
Date: Sun, 4 Apr 2010 19:15:05 -0400
Subject: [Eeglablist] Save multiple datasets in script
Message-ID: <l2nb6a6b35f1004041615h62f9023i888f8a63073c23ef@mail.gmail.com>

I'm trying to extract & save multiple datasets from one session; when I run
the script, I get the following error:

??? Error using ==> pop_saveset at 453
For reasons of consistency, this function  does not save multiple datasets
any more

but the pop_saveset info file says that it can save one or more data sets.
I'm guessing this is one place where the documentation isn't up to date--so
does anybody know what the current way of scripting multiple datasets to
save is?

cheers,
Krysta

----------------------------
Krysta Chauncey, Ph.D.
Brain-Computer Interface Project
Human-Computer Interaction Lab
Computer Science Dept, Tufts University
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100404/4debba56/attachment.html>

From saim_rasheed at hotmail.com  Tue Apr  6 10:49:23 2010
From: saim_rasheed at hotmail.com (Saim Rasheed)
Date: Tue, 6 Apr 2010 22:49:23 +0500
Subject: [Eeglablist] Concatinating data epochs for same conditions in one
 file, from different subjects
Message-ID: <SNT119-W38CDD7F7854AFD172A7C2DEA180@phx.gbl>



Hi,

 

I have recorded data from 7 subjects across three different conditions. For each condition 60 trials. Seperating data epochs for all the three different from each subject is possible. 

Is it possible to concatenate the data epochs of similar conditions from every subject into a single file '*.set' ? I was following a tutorial on http://sccn.ucsd.edu/wiki/Chapter_02:_Writing_EEGLAB_Scripts and tried to run a script under the heading "Example script for processing multiple datasets" on the data I recorded. I am confused with 'subj[1:10]data[1:3].set'  as I created a seperated directory containg three different data epoch files( for both type of extensions'*.set' and '*.fdt') for each of the three conditions for 7 subjects. On running the script, I got the same ERSP and ITC for all the subjects as it utilized only data for one subject ans shows the results for 7 subjects. Probably, I am missing something simple. Any help please. Thanking you in advance.

 

Regards

Saim

Milan.


 
  		 	   		  
_________________________________________________________________
Hotmail: Trusted email with Microsoft?s powerful SPAM protection.
https://signup.live.com/signup.aspx?id=60969
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100406/cda8785d/attachment.html>

From simonor at rennes.ucc.ie  Tue Apr  6 09:26:41 2010
From: simonor at rennes.ucc.ie (Simon O' Regan)
Date: Tue, 6 Apr 2010 17:26:41 +0100
Subject: [Eeglablist] Movement artefact database
Message-ID: <000001cad5a5$f0da02e0$40a1ef8f@SimonOR>

Hi,


My name is Simon O' Regan; I'm a PhD researcher with the EEDSP group at
University College Cork, Ireland, focusing on artefact detection and removal
in EEG. I'm particularly interested in artefacts caused by movement. To this
end, I have collected a movement artefact database of approximately 300
minutes, comprising artefacts resulting from some typical movements which
would be encountered in an ambulatory setting. This database is currently
being annotated, and will be used to train and develop artefact detection
and removal algorithms.

However, for most accurate testing of the algorithms, natural EEG (i.e. EEG
where the patient has not been instructed to perform specific movements, nor
indeed to abstain from certain movements) would be most appropriate. I'm
wondering if anyone on the mailing list is aware of such a labelled artefact
database for movement artefacts.

I am currently aware of the (epilepsy-based) EEG databases at Bonn,
Freiberg, Tampere and Queensland.

I would appreciate any suggestions/help you can give,

Cheers,

Simon

 

__________________________________________________

 

Simon O' Regan

PhD. Student,

Biomedical Signal Processing,

Dept. of Electrical and Electronic Engineering, 

University College Cork, Ireland

 

Tel: (021) 4903156

Email: simonor at rennes.ucc.ie

Web: http://rennes.ucc.ie/~simonor/

__________________________________________________

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100406/5a881309/attachment.html>

From arno at ucsd.edu  Tue Apr  6 12:33:23 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Tue, 6 Apr 2010 12:33:23 -0700
Subject: [Eeglablist] Save multiple datasets in script
In-Reply-To: <l2nb6a6b35f1004041615h62f9023i888f8a63073c23ef@mail.gmail.com>
References: <l2nb6a6b35f1004041615h62f9023i888f8a63073c23ef@mail.gmail.com>
Message-ID: <80FD0284-D6E9-40C4-823F-2D1FDD12AA90@ucsd.edu>

Dear Krista,

we forwarded by mistake your message to the list. I am copying the  
answer I gave you for other people on the list.

Fist I have updated the documentation as you pointed out. The current  
function cannot save multiple datasets in a single file since about  
2004 (it will read these old 2004 multiple dataset files though). You  
can save each dataset in a separate file using the command:

for index = 1:length(EEG)
	pop_saveset(EEG(index), 'filename', sprintf('file%d.set', index));
end;

Hope this helps.
Best,

Arno



From mataothefifth at yahoo.co.jp  Tue Apr  6 19:19:55 2010
From: mataothefifth at yahoo.co.jp (Makoto Miyakoshi)
Date: Wed, 7 Apr 2010 11:19:55 +0900 (JST)
Subject: [Eeglablist] complex numbers for tfdata
In-Reply-To: <p2nfc2f2c2d1004051647ub9f8222fo2cd6dd6938cf9a3b@mail.gmail.com>
Message-ID: <20100407021955.46095.qmail@web3705.mail.tnz.yahoo.co.jp>

Dear Kazu,

It represents phase information.

Makoto



--- Kazuyo Tanji <kaztanji at gmail.com> wrote:

> Hello all,
> 
> I am wondering if someone can tell me how I should interpret
> 'tfdata', an
> output of the function newtimef() which is in the form of complex
> numbers. I
> can tell the matrix is structured as frequency steps x time points x
> trials
> but why complex numbers??
> Any help is appreciated.
> 
> Thanks,
> Kazu
> 
> -- 
> ????
> ????????????
> ?990-9585 ??????2-2-2
> ?? 023-628-5428 ??? 023-628-5425
> e-mail: kaztanji at gmail.com
> ????? 6732
> 
> 
> Kazuyo Tanji
> Yamagata University
> Department of Clinical Neuroscience
> 2-2-2 Iida-nishi Yamagata,Japan
> phone 81-23-628-5428
> fax      81-23-628-5425
> > _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
eeglablist-request at sccn.ucsd.edu



From sherrykhan78 at gmail.com  Mon Apr 12 08:55:35 2010
From: sherrykhan78 at gmail.com (Sheraz Khan)
Date: Mon, 12 Apr 2010 11:55:35 -0400
Subject: [Eeglablist] Postdoctoral Fellowship in Multimodal Imaging at MGH /
	Harvard Medical School
Message-ID: <u2lfb09b4651004120855t9e9eee49nbd92dbf79e8a8a9d@mail.gmail.com>

  Postdoctoral Fellowship: Multimodal Imaging at MGH / Harvard Medical
School

*
*

*Description:*

A two year postdoctoral position pursuing multimodal imaging is available
with the TRANSCEND Research Program (*www.transcendresearch.org*) at the
Martinos Center for Biomedical Imaging in Charlestown, MA (*www.martinos.org
*) which is affiliated with the Massachusetts General Hospital, Harvard and
MIT. There is a second location at a clinical site, the MGH-affiliated Lurie
Family Center for Autism, where we have a 128 lead EGI EEG in a shielded
room as well as a photogrammetry machine.

 The program?s emphasis is on pathophysiologically oriented brain research
and application of advanced imaging acquisition and analysis techniques to
neurological and sensory aspects of autism spectrum disorders. Emphasis will
be on MRI (DTI, spectroscopy, morphometry, ASL) and on co-registering EEG
and MEG with MRI. This position will involve analysis of existing multimodal
imaging data and collection of new data. It will involve working closely
with a multidisciplinary team and with children, and will also involve some
research oriented analysis of data collected for clinical purposes. The age
range of our subjects is primarily from early infancy to mid-adolescence,
with some additional studies including young adults. An ample opportunity
will also be provided to the candidate to self-explore and lead research.


 *Requirements:*

Candidates must have PhD in neuroscience, physics, electrical engineering,
computer science or other related fields. Prior experience in MRI analysis
and EEG signal processing is required. Salary will be consistent with
Massachusetts General Hospital, Harvard Medical School policies for
Postdoctoral trainees and will range between $45,000 to $55,000 depending
upon qualifications and experience. Compensation also includes full staff
benefits, including health insurance, and vacation time.


 *Contact:*

Interested applicants may send a CV and statement of interest addressing
background and specific pertinence of the candidate?s interest to Dr. Martha
R. Herbert at *mherbert1 at partners.org* and cc *transcend at partners.org*.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100412/1634e996/attachment.html>

From dr.ilya at yahoo.com  Mon Apr 12 09:23:55 2010
From: dr.ilya at yahoo.com (Ilya Adamchic)
Date: Mon, 12 Apr 2010 09:23:55 -0700 (PDT)
Subject: [Eeglablist] Component clustering
In-Reply-To: <BAEFHEAAHAEAAGHAAA@webmail.ulb.ac.be>
References: <BAEFHEAAHAEAAGHAAA@webmail.ulb.ac.be>
Message-ID: <429993.56798.qm@web63805.mail.re1.yahoo.com>

Dear all,
I have been looking for a study, which possibly used  ICA component clustering from continuous (not epoched) recordings (several minutes). So far did not found any. Are you guys aware, if someone has done clustering on such (continuous) recordings and if such a clustering is doable in EEGLAB.
Cheers,
Ilya


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100412/d119ad4f/attachment.html>

From brian.murphy at unitn.it  Tue Apr 13 01:35:57 2010
From: brian.murphy at unitn.it (Brian Murphy)
Date: Tue, 13 Apr 2010 10:35:57 +0200
Subject: [Eeglablist] Concatinating data epochs for same conditions in
 one file, from different subjects
In-Reply-To: <mailman.0.1270580402.12848.eeglablist@sccn.ucsd.edu>
References: <mailman.0.1270580402.12848.eeglablist@sccn.ucsd.edu>
Message-ID: <4BC42CED.3030808@unitn.it>

Dear Saim,

the usual way to do group analyses is to join participant sessions into 
a single "study":
    http://sccn.ucsd.edu/wiki/Chapter_12:_Multiple_Datasets

But if you really want to concatenate multiple datasets from different 
people, you can use the pop_mergeset function repeatedly, either from 
the GUI (Edit>Append datasets), or with a script such as:
    files = {'A.set', 'B.set', 'C.set'}
    EEG = pop_loadset(char(files(1)))
    for f=2:length(files)
        file = char(files(f))
        TMP = pop_loadset(file)
        EEG = pop_mergeset(EEG,TMP,0)
    end
    EEG = eeg_checkset(EEG)
    EEG.setname = 'combined'
pop_saveset(EEG,'filename','combined.set')
eeglab redraw
   
See also the earlier post 
http://sccn.ucsd.edu/pipermail/eeglablist/2008/002073.html

good luck,

Brian



From michael.herzog at epfl.ch  Tue Apr 13 11:23:19 2010
From: michael.herzog at epfl.ch (Michael Herzog)
Date: Tue, 13 Apr 2010 20:23:19 +0200
Subject: [Eeglablist] Open Position at the EPFL, Switzerland
Message-ID: <4BC4B697.6000100@epfl.ch>

The Laboratory of Psychophysics at the EPFL in Lausanne, Switzerland,
is searching for a post-doctoral student to lead the EEG unit 
(http://lpsy.epfl.ch/). EEG research relates to all kinds of visual 
processes. In addition, we are also running a TMS facility. Combination 
of TMS and EEG is an option.

The post-doc position is a two to four year appointment and salary is 
approximately 4.600 CHF/month (about 3000Euro) after taxes.
To apply, please send curriculum vitae, list of publications, the names 
of three referees, and a short description of research interests by 
e-mail to: michael.herzog at epfl.ch.



From ShinE at missouri.edu  Tue Apr 13 16:28:37 2010
From: ShinE at missouri.edu (Shin, Eunsam)
Date: Tue, 13 Apr 2010 18:28:37 -0500
Subject: [Eeglablist] Society for Psychophysiological Research Conference
 Grants Application
Message-ID: <126C991FA41FDF46B2030D3BDE42852A0B1808582F@UM-EMAIL05.um.umsystem.edu>

GRANTS TO ATTEND THE ANNUAL MEETING OF THE SOCIETY FOR PSYCHOPHYSIOLOGICAL RESEARCH (SPR) Portland, Oregon (USA), September 29 - October 3, 2010

Deadline for Grant Application: May 1

AIM OF THE GRANTS:
The aim of the SPR travel grant program is to help psychophysiologists from developing countries in Latin America, Asia, and Eastern Europe become members of the SPR and participate actively in its annual meetings. The awards include a travel grant up to $1,500, free registration fee for the 2010 conference, and free SPR membership for 2010. A larger number of grants are available, but applications are evaluated competitively based on merit, geographical distribution and other factors.

CONDITIONS OF APPLICATION:
(a) Applicants must submit an abstract for a poster, symposium, or panel discussion. Instructions and online submission information are available at the SPR web site: www.sprweb.org
(b) The abstract must report research within the domain of human psychophysiology using physiological measures of central (e.g., EEG, MEG, ERP, fMRI etc.) or peripheral nervous system functions (e.g., somatic reflexes, autonomic activity, endocrine, immune functions, etc.).
(c) Applicants at research centers or universities located in Latin America, Asia (except Singapore, Hong Kong, Korea and Japan) or Eastern Europe are eligible. If sufficient funds are available, scholars from these areas who are temporarily working in economically advantaged countries will be considered.
(d) Applicants must either be a member of the SPR or have applied for membership.
(e) The grantees of last year are eligible to apply again although candidates who have not yet received an award have higher priority.

APPLICATION PROCEDURE:
Candidates should send their grant application before MAY 1, 2010, via e-mail to spr2010 at gi.hum.titech.ac.jp (Asia & Eastern Europe) or vilajaime at yahoo.es (Latin America). The following materials should be attached:
(a) A summary in English of their curriculum vitae which should include: publications and congress/symposia/workshops/courses attended abroad indicating presentation type (oral or poster), year of degree, year and grants received, and current position (student or non-student, and university or institute name including that of studying abroad).
(b) A copy of the abstract that will be submitted for the 50th annual meeting via the SPR web site: www.sprweb.org. Note that the final decision regarding each candidate depends on abstract acceptance by program committee.
(c) A copy of membership application if not currently a member. (Although no payment is required, please indicate that you are applying for the SPR travel grant in the "Method of Payment" section of the SPR membership application form. If a candidate does not win the award and does not pay the dues his/her application may be withdrawn.)
 







 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ANNOUNCEMENT_SPR grants_2010.pdf
Type: application/pdf
Size: 61339 bytes
Desc: ANNOUNCEMENT_SPR grants_2010.pdf
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100413/035eba27/attachment.pdf>

From nickbedo at yahoo.com  Wed Apr 14 20:30:23 2010
From: nickbedo at yahoo.com (Nick Bedo)
Date: Wed, 14 Apr 2010 20:30:23 -0700 (PDT)
Subject: [Eeglablist] variables as arguments in pop_importdata()
Message-ID: <51921.9410.qm@web62006.mail.re1.yahoo.com>

Hi all,

I'm importing data as part of a loop, and I'm trying to automate certain parameters of the pop_importdata() function.  Specifically, I'm trying to add a variable as an input argument, but I can't seem to figure it out.  Here is the line of code that I'm working on:

    EEG = pop_importdata('dataformat','array','data','temp2','srate',200,'pnts',0,'xmin',0,'nbchan',0, 'setname', 'currfile');

'currfile' is the name that I want to change with each iteration of the loop.  Any and all contributions are appreciated.

Thanks,
Nick


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100414/94c1c0fe/attachment.html>

From zdmoran at gmail.com  Wed Apr 14 23:17:49 2010
From: zdmoran at gmail.com (Zachary Moran)
Date: Wed, 14 Apr 2010 23:17:49 -0700
Subject: [Eeglablist] Wrapped Phase Angle from newtimef?
Message-ID: <z2s93a92c841004142317j14c12185o338d3d7a594a3b5c@mail.gmail.com>

Hi Everyone,

I'm currently investigating the relationship between phase angle and other
EEG variables.  In order to get phase angle, I've been taking the angle(ITC)
in Matlab which gives me phase angle estimates from -pi to pi radians.
However, when I plot this data against time I find discontinuities at pi and
-pi (see attached jpeg for an example), which has become problematic in
using it as a predictor variable.

Does this suggest that the phase information is "wrapped", and, if so, does
Matlab's "unwrap(phase_angle_data)" function seem appropriate to anyone else
as a means of making my phase angle data better suited as a predictor?
Alternatively, I had initially wanted to constrain phase to between 0 and
2pi radians - does anyone have any advice for how to make this conversion?
Any thoughts or ideas would be very greatly appreciated.

Many thanks!

Zach
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100414/bb0855f4/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: phase_example.jpg
Type: image/jpeg
Size: 20744 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100414/bb0855f4/attachment.jpg>

From mklados at gmail.com  Fri Apr 16 09:35:42 2010
From: mklados at gmail.com (Klados Manousos)
Date: Fri, 16 Apr 2010 19:35:42 +0300
Subject: [Eeglablist] variables as arguments in pop_importdata()
In-Reply-To: <51921.9410.qm@web62006.mail.re1.yahoo.com>
References: <51921.9410.qm@web62006.mail.re1.yahoo.com>
Message-ID: <w2z3b0db4861004160935p4a86f15fuc07eb9617deafbcb@mail.gmail.com>

Dear Nick

According to my opinion it is better to use a numeric system to name your
datasets if you want to automatically insert them into EEGLAB. Suppose that
we have 10 datasets name 1.set,...10.set then with the next code you can
insert them in a loop

for i=1:10
    tic
    a=int2str(i)
    eeglab
    filename=[a '.set'];
    EEG = pop_loadset('filename', filename,'filepath','C:\...)
    eeglab redraw

    (continue code...and do whatever you want)
end;

On the other hand if you dont want, or if you cant change the names of your
variables you can make a structure or a character array...but you have to
search for this further in Matlab's help

I hope to helped you

Best Regards
Manousos

2010/4/15 Nick Bedo <nickbedo at yahoo.com>

> Hi all,
>
> I'm importing data as part of a loop, and I'm trying to automate certain
> parameters of the pop_importdata() function.  Specifically, I'm trying to
> add a variable as an input argument, but I can't seem to figure it out.
>  Here is the line of code that I'm working on:
>
>     EEG =
> pop_importdata('dataformat','array','data','temp2','srate',200,'pnts',0,'xmin',0,'nbchan',0,
> 'setname', 'currfile');
>
> 'currfile' is the name that I want to change with each iteration of the
> loop.  Any and all contributions are appreciated.
>
> Thanks,
> Nick
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Manousos A. Klados
PhD Candidate -- Research Assistant
Group of Applied Neurosciences
Lab of Medical Informatics
School of Medicine
Aristotle University of Thessaloniki
P.O. Box 323 54124 Thessaloniki Greece
_________________________________________________
Tel: +30-2310-999332
Fax:+30-2310-999263
Website: http://lomiweb.med.auth.gr/gan/mklados

________________________________________________________________
??? ????? ???????: ??? ?????? ???? ?? mail ????? ?????????? ?? ??????????.
Acting by Reacting: By not printing this e-mail I help protect the
environment.
________________________________________________________________
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100416/b05128d6/attachment.html>

From pzeman at alumni.uvic.ca  Fri Apr 16 15:23:05 2010
From: pzeman at alumni.uvic.ca (Philip Michael Zeman)
Date: Fri, 16 Apr 2010 15:23:05 -0700
Subject: [Eeglablist] Wrapped Phase Angle from newtimef?
In-Reply-To: <z2s93a92c841004142317j14c12185o338d3d7a594a3b5c@mail.gmail.com>
References: <z2s93a92c841004142317j14c12185o338d3d7a594a3b5c@mail.gmail.com>
Message-ID: <FC3A1510125841A0A900D0ECF5D297A4@mine>

Hi Zach,

a quick look at your data suggests that you have simply run into a fundamental characteristc of measuring phase angle. Phase angle is a continuous and repeating value. Hence, a static, unchanging, reference point is needed.  There is probably a simple tweak you can make to your algorithm that can 'fix' your issue.  Alternatively, there are some transformations you can make:


See:

http://en.wikipedia.org/wiki/Quaternion

I also came across some information about calculating 'circular means' from circular statistics.  I haven't had a chance to check this out, but you might find it helpful.  I think this was posted on the EEGLab group a while back.
http://www.mathworks.com/matlabcentral/fileexchange/10676-circular-statistics-toolbox-directional-statistics


Regards,

Phil

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Philip Michael Zeman B.Eng, Ph.D.
Applied Brain and Vision Sciences Inc.
Brain Function Analysis for Novel Paradigms and Serious Games
Analysis of Pharmaceutical Effects on Brain Function
http://www.abvsciences.com
Latest Brain Research Result:
http://www.spatialbrain.com
Email:   pzeman at alumni.uvic.ca
Phone: +1-250-589-4234
LinkedIn Profile: http://ca.linkedin.com/in/philipmichaelzeman
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

  ----- Original Message ----- 
  From: Zachary Moran 
  To: eeglablist at sccn.ucsd.edu 
  Sent: Wednesday, April 14, 2010 11:17 PM
  Subject: [Eeglablist] Wrapped Phase Angle from newtimef?


  Hi Everyone,

  I'm currently investigating the relationship between phase angle and other EEG variables.  In order to get phase angle, I've been taking the angle(ITC) in Matlab which gives me phase angle estimates from -pi to pi radians.  However, when I plot this data against time I find discontinuities at pi and -pi (see attached jpeg for an example), which has become problematic in using it as a predictor variable.  

  Does this suggest that the phase information is "wrapped", and, if so, does Matlab's "unwrap(phase_angle_data)" function seem appropriate to anyone else as a means of making my phase angle data better suited as a predictor?  Alternatively, I had initially wanted to constrain phase to between 0 and 2pi radians - does anyone have any advice for how to make this conversion?  Any thoughts or ideas would be very greatly appreciated.

  Many thanks!

  Zach



------------------------------------------------------------------------------


  _______________________________________________
  Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
  To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
  For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100416/660b1ae8/attachment.html>

From marial.stavrinou at gmail.com  Sat Apr 17 13:13:53 2010
From: marial.stavrinou at gmail.com (Maria L. Stavrinou)
Date: Sat, 17 Apr 2010 23:13:53 +0300
Subject: [Eeglablist] Wrapped Phase Angle from newtimef?
In-Reply-To: <z2s93a92c841004142317j14c12185o338d3d7a594a3b5c@mail.gmail.com>
References: <z2s93a92c841004142317j14c12185o338d3d7a594a3b5c@mail.gmail.com>
Message-ID: <x2yae2180201004171313x42828e95rffeec6dccde2c072@mail.gmail.com>

Dear Zach,
You can use unwrap to un-wrap the phase that you calculate. To check if all
goes well, try this by having a simple sinusoid as input and check the
result, as shown in the image I am attaching. I would recommend, to do the
phase difference of the unwraped phase,  in radians of as they are now,
between -pi and pi. Using the function mod you can project them into [0,
2pi] interval.

Maria

2010/4/15 Zachary Moran <zdmoran at gmail.com>

> Hi Everyone,
>
> I'm currently investigating the relationship between phase angle and other
> EEG variables.  In order to get phase angle, I've been taking the angle(ITC)
> in Matlab which gives me phase angle estimates from -pi to pi radians.
> However, when I plot this data against time I find discontinuities at pi and
> -pi (see attached jpeg for an example), which has become problematic in
> using it as a predictor variable.
>
> Does this suggest that the phase information is "wrapped", and, if so, does
> Matlab's "unwrap(phase_angle_data)" function seem appropriate to anyone else
> as a means of making my phase angle data better suited as a predictor?
> Alternatively, I had initially wanted to constrain phase to between 0 and
> 2pi radians - does anyone have any advice for how to make this conversion?
> Any thoughts or ideas would be very greatly appreciated.
>
> Many thanks!
>
> Zach
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
-- 
Maria L. Stavrinou
MSc. PhD.
Postdoc Researcher
Unit of Neurophysiology
Department of Physiology
School of Medicine,
University of Patras
Greece
tel. lab +30 - 2610-969153
cellular +30 -6938287930
http://physiology.med.upatras.gr/NU/

https://sites.google.com/site/marialstavrinou/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100417/c2c67e50/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: phase_sine.tif
Type: image/tiff
Size: 31476 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100417/c2c67e50/attachment.tif>

From marial.stavrinou at gmail.com  Sat Apr 17 15:04:55 2010
From: marial.stavrinou at gmail.com (Maria L. Stavrinou)
Date: Sun, 18 Apr 2010 01:04:55 +0300
Subject: [Eeglablist] Wrapped Phase Angle from newtimef?
In-Reply-To: <x2yae2180201004171313x42828e95rffeec6dccde2c072@mail.gmail.com>
References: <z2s93a92c841004142317j14c12185o338d3d7a594a3b5c@mail.gmail.com>
	<x2yae2180201004171313x42828e95rffeec6dccde2c072@mail.gmail.com>
Message-ID: <t2pae2180201004171504q8d94ab68ha0f2ff36fccb3c9f@mail.gmail.com>

2010/4/17 Maria L. Stavrinou <marial.stavrinou at gmail.com>

> Dear Zach,
> You can use unwrap to un-wrap the phase that you calculate. To check if all
> goes well, try this by having a simple sinusoid as input and check the
> result, as shown in the image I am attaching. I would recommend, to do the
> phase difference of the unwraped phase,  in radians of as they are now,
> between -pi and pi. Using the function mod you can project them into [0,
> 2pi] interval.
>
> Maria
>
> 2010/4/15 Zachary Moran <zdmoran at gmail.com>
>
>> Hi Everyone,
>>
>> I'm currently investigating the relationship between phase angle and other
>> EEG variables.  In order to get phase angle, I've been taking the angle(ITC)
>> in Matlab which gives me phase angle estimates from -pi to pi radians.
>> However, when I plot this data against time I find discontinuities at pi and
>> -pi (see attached jpeg for an example), which has become problematic in
>> using it as a predictor variable.
>>
>> Does this suggest that the phase information is "wrapped", and, if so,
>> does Matlab's "unwrap(phase_angle_data)" function seem appropriate to anyone
>> else as a means of making my phase angle data better suited as a predictor?
>> Alternatively, I had initially wanted to constrain phase to between 0 and
>> 2pi radians - does anyone have any advice for how to make this conversion?
>> Any thoughts or ideas would be very greatly appreciated.
>>
>> Many thanks!
>>
>> Zach
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>>
>
>
>
> --
> --
> Maria L. Stavrinou
> MSc. PhD.
> Postdoc Researcher
> Unit of Neurophysiology
> Department of Physiology
> School of Medicine,
> University of Patras
> Greece
> tel. lab +30 - 2610-969153
> cellular +30 -6938287930
> http://physiology.med.upatras.gr/NU/
>
> https://sites.google.com/site/marialstavrinou/
>
>


-- 
-- 
Maria L. Stavrinou
MSc. PhD.
Postdoc Researcher
Unit of Neurophysiology
Department of Physiology
School of Medicine,
University of Patras
Greece
tel. lab +30 - 2610-969153
cellular +30 -6938287930
http://physiology.med.upatras.gr/NU/

https://sites.google.com/site/marialstavrinou/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100418/7cadaee1/attachment.html>

From bradley.voytek at gmail.com  Sat Apr 17 14:57:26 2010
From: bradley.voytek at gmail.com (Bradley Voytek)
Date: Sat, 17 Apr 2010 14:57:26 -0700
Subject: [Eeglablist] Wrapped Phase Angle from newtimef?
In-Reply-To: <z2s93a92c841004142317j14c12185o338d3d7a594a3b5c@mail.gmail.com>
References: <z2s93a92c841004142317j14c12185o338d3d7a594a3b5c@mail.gmail.com>
Message-ID: <v2i6d898bac1004171457r8177d429qab066657d0ed4cdf@mail.gmail.com>

Zachary:

Unwrapping the phase won't really give you what you're looking for
(though the derivative of the unwrapped phase can give you useful
information, such as "instantaneous" frequency).

First of all, I strongly support Dr. Zeman's recommendation of the
circular statistics toolbox. It's very easy to use and quite clear.

If I understand you correctly, you'd like to relate phase to EEG
variables, which I assume means amplitude, power, or latency of the of
the ERP or time-frequency variables.

I'd suggest that you take your phase time-series, such as the one you
plotted, and then extract the instantaneous phase at every time point
of interest (e.g., peak P300 amplitude). Then you can plot the
clustering of those phases in a circular plot. From that you can
extract the mean phase angle and vector length (from 0 to 1) which
would represent the preferred phase and coherence, respectively, at
your P300 peaks. This would be done using the circ_mean and circ_r
commands within the circular statistics toolbox. You can also computer
the significance of the clustering against the assumption of a random
distribution using circ_rtest, for example.

Then you can also extract the P300 peaks themselves, separately.

This would give you two vectors: P300 peak amplitude and phases at
those amplitudes. Now let's say you want to see if the two are
correlated. Obviously because you're phase data are circular, a linear
correlation isn't the correct way to do this. But you *can* correlate
one circular and one linear variable using the circ_corrcl command.

Anyway, good luck! I hope this is helpful.

::brad

On Wed, Apr 14, 2010 at 23:17, Zachary Moran <zdmoran at gmail.com> wrote:
> Hi Everyone,
>
> I'm currently investigating the relationship between phase angle and other
> EEG variables.? In order to get phase angle, I've been taking the angle(ITC)
> in Matlab which gives me phase angle estimates from -pi to pi radians.
> However, when I plot this data against time I find discontinuities at pi and
> -pi (see attached jpeg for an example), which has become problematic in
> using it as a predictor variable.
>
> Does this suggest that the phase information is "wrapped", and, if so, does
> Matlab's "unwrap(phase_angle_data)" function seem appropriate to anyone else
> as a means of making my phase angle data better suited as a predictor?
> Alternatively, I had initially wanted to constrain phase to between 0 and
> 2pi radians - does anyone have any advice for how to make this conversion?
> Any thoughts or ideas would be very greatly appreciated.
>
> Many thanks!
>
> Zach
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



From mataothefifth at yahoo.co.jp  Sun Apr 18 20:19:52 2010
From: mataothefifth at yahoo.co.jp (Makoto Miyakoshi)
Date: Mon, 19 Apr 2010 12:19:52 +0900 (JST)
Subject: [Eeglablist] variables as arguments in pop_importdata()
In-Reply-To: <51921.9410.qm@web62006.mail.re1.yahoo.com>
Message-ID: <20100419031952.37046.qmail@web3710.mail.tnz.yahoo.co.jp>

Dear Nick,

I wonder if you want something like this.

allfiles=dir('*.set');
for n=1:length(allfiles)
    ....
end

During the loop, allfiles(n).name picks up each setfile name in the current
folder. Be sure that you have only necessary setfiles in the current
folder.

Makoto




--- Nick Bedo <nickbedo at yahoo.com> wrote:

> Hi all,
> 
> I'm importing data as part of a loop, and I'm trying to automate
> certain parameters of the pop_importdata() function.  Specifically,
> I'm trying to add a variable as an input argument, but I can't seem
> to figure it out.  Here is the line of code that I'm working on:
> 
>     EEG =
>
pop_importdata('dataformat','array','data','temp2','srate',200,'pnts',0,'xmin',0,'nbchan',0,
> 'setname', 'currfile');
> 
> 'currfile' is the name that I want to change with each iteration of
> the loop.  Any and all contributions are appreciated.
> 
> Thanks,
> Nick
> 
> 
>       > _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
eeglablist-request at sccn.ucsd.edu



From h.brinson at gold.ac.uk  Mon Apr 19 04:33:51 2010
From: h.brinson at gold.ac.uk (Helen Brinson)
Date: Mon, 19 Apr 2010 12:33:51 +0100 (BST)
Subject: [Eeglablist] EEG neurofeedback in stroke
Message-ID: <53399.78.105.24.54.1271676831.squirrel@secure2.gold.ac.uk>


**********************************************************
Research Coordinator - EEG Neurofeedback (NHS National Institute of Health
Research grant)
A novel neurofeedback based intervention to reduce neglect and improve
function in stroke patients.
Reference Number:	NHS Jobs 344-1942MZC
<http://www.jobs.nhs.uk/cgi-bin/vacdetails.cgi?selection=912457435>
Based at William Harvey Hospital and King's College Hospital

The Post holder will assist in research to assess the applicability of
EEG-neurofeedback training to the rehabilitation of acute stroke patients
with spatial neglect. The post will involve recruiting patients from stroke
units, conducting the neurofeedback training sessions on a daily basis
(excluding weekends), and conducting standardised clinical assessments on
patients before and after training, involving receiving and logging complex
and sensitive information. Planning and co-ordinating the research
activities within time constraints is important. There may also be the
opportunity to assist with an fMRI arm to the project.

The successful candidate will ideally have a background in a relevant life
science, (e.g. medicine, psychology or cognitive neuroscience) or in
therapy, and an understanding of clinical research in NHS settings and
experience working with clinical populations. Previous experience with and
understanding of EEG and/or the principles of EEG-neurofeedback is highly
desirable. In addition, a Criminal Records Bureau check and a driving
license is required for this position.

This project is a multi-site study taking place in hospitals in South
London and East Kent. It is a collaboration between Dr David Smithard at
William Harvey Hospital, Prof Lalit Kalra at King's College Hospital, and
Prof John Gruzelier and Dr Karina Linnell at Goldsmiths, University of
London. Informal enquiries can be addressed to Karina Linnell
(k.j.linnell at gold.ac.uk).

The interviews will be held on the 17th May 2010.
Please note that this position will close on or before this date, depending
on the response that we receive. Therefore, please submit your application
promptly - preferably by 26th April - to ensure that you are considered for
this post.


-- 
Helen Brinson MSc
Department of Psychology (Phd cand.)
Goldsmiths College, University of London
h.brinson at gold.ac.uk



From nickbedo at yahoo.com  Mon Apr 19 00:55:13 2010
From: nickbedo at yahoo.com (Nick Bedo)
Date: Mon, 19 Apr 2010 00:55:13 -0700 (PDT)
Subject: [Eeglablist] low-freq resolution issues in newcrossf
Message-ID: <636920.54561.qm@web62002.mail.re1.yahoo.com>

Hi all,

I'm having an issue with some vertical streaks in my cross-coherence spectrogram at the lower frequencies (<10hz).  My problem is similar to an archived problem (http://sccn.ucsd.edu/pipermail/eeglablist/2007/002006.html), but instead of having issues at high frequencies, I'm getting perturbations at the lower frequencies.  

Here is some of the output to give you an idea of my parameters:
Adjust min freq. to 0.06 Hz to match FFT output frequencies
Adjust max freq. to 49.99 Hz to match FFT output frequencies
Using hanning FFT tapering
Generating 1000 time points (128.1 to 1871.9 ms)
Finding closest points for time variable
Time values for time/freq decomposition is not perfectly uniformly distributed
The window size used is 256 samples (256 ms) wide.
Estimating 819 linear-spaced frequencies from 0.1 Hz to 50.0 Hz.

Additionally, in an effort to increase overall resolution, I've been tweaking the default settings on newcrossf():
DEFAULT_NWIN= 1000;
DEFAULT_VARWIN= 0;
DEFAULT_OVERSMP= 64;

I have tried many minor tweaks to fix the issue, but nothing seems to be working.  Any tips are appreciated!

Thanks,
Nick


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100419/c143197a/attachment.html>

From mataothefifth at yahoo.co.jp  Wed Apr 21 00:30:03 2010
From: mataothefifth at yahoo.co.jp (Makoto Miyakoshi)
Date: Wed, 21 Apr 2010 16:30:03 +0900 (JST)
Subject: [Eeglablist] low-freq resolution issues in newcrossf
In-Reply-To: <636920.54561.qm@web62002.mail.re1.yahoo.com>
Message-ID: <20100421073004.18904.qmail@web3712.mail.tnz.yahoo.co.jp>

Dear Nick,

You mean you are using the wavelet analysis, right?
I give you an example of the calculation.
If I'm wrong, someone please correct me (I'm not a specialist).

Suppose you are interested in 4Hz activity.
One cycle of 4Hz is 250ms (1000ms/4Hz).
If you use a 3 cycle wavelet, the length is 250ms*3cycles=750ms.
It means that your 750ms EEG data produces 1 pixel in the TF plot.
Thus, if your data length is 750ms, you obtain 1 time point.
If your data length is 755ms (suppose your EEG sampling is >=200Hz and the
time resolution ie. one time point of the TF plot is 5ms), you obtain 2
time points.
If your data length is 3000ms, you obtain (3000ms-750ms)/5ms=450 time
points.
If you want to obtain 200 time points and you have 3000ms data length, your
time resolution is (3000ms-750ms)/200=11.25ms. If you want 5ms time
resolution and 200 time points, then 5ms*200+750ms=1750ms data length is
required.

I hope now you understand calculating 0.06Hz is hard.
The lowest frequency of interest determines the window size, it should be
carefully determined considering the time-constant of your target EEG
component and (if event-related design) trial duration.

Makoto

--- Nick Bedo <nickbedo at yahoo.com> wrote:

> Hi all,
> 
> I'm having an issue with some vertical streaks in my cross-coherence
> spectrogram at the lower frequencies (<10hz).  My problem is similar
> to an archived problem
> (http://sccn.ucsd.edu/pipermail/eeglablist/2007/002006.html), but
> instead of having issues at high frequencies, I'm getting
> perturbations at the lower frequencies.  
> 
> Here is some of the output to give you an idea of my parameters:
> Adjust min freq. to 0.06 Hz to match FFT output frequencies
> Adjust max freq. to 49.99 Hz to match FFT output frequencies
> Using hanning FFT tapering
> Generating 1000 time points (128.1 to 1871.9 ms)
> Finding closest points for time variable
> Time values for time/freq decomposition is not perfectly uniformly
> distributed
> The window size used is 256 samples (256 ms) wide.
> Estimating 819 linear-spaced frequencies from 0.1 Hz to 50.0 Hz.
> 
> Additionally, in an effort to increase overall resolution, I've been
> tweaking the default settings on newcrossf():
> DEFAULT_NWIN= 1000;
> DEFAULT_VARWIN= 0;
> DEFAULT_OVERSMP= 64;
> 
> I have tried many minor tweaks to fix the issue, but nothing seems to
> be working.  Any tips are appreciated!
> 
> Thanks,
> Nick
> 
> 
>       > _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
eeglablist-request at sccn.ucsd.edu



From guillaumechaumet at gmail.com  Mon Apr 19 02:10:23 2010
From: guillaumechaumet at gmail.com (guillaume chaumet)
Date: Mon, 19 Apr 2010 11:10:23 +0200
Subject: [Eeglablist] [eeglablist] simple question
Message-ID: <v2n37af89ad1004190210gd94bccddx973a6e800ab18f32@mail.gmail.com>

Dear all,
My question is simple.
I have long duration eeg recording and I just want to compute log power
spectrum mean on part of these data (ex: the first hour for alpha range,
theta range, etc.), first on the whole channels and second on each channel.



Guillaume Chaumet
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100419/284d3bea/attachment.html>

From petsol at gmail.com  Mon Apr 19 14:58:25 2010
From: petsol at gmail.com (=?ISO-8859-1?Q?P=E9ter_Solt=E9sz?=)
Date: Mon, 19 Apr 2010 23:58:25 +0200
Subject: [Eeglablist] BinICA for xp
Message-ID: <l2g807610c21004191458tdca04e24q27f1e83e75806d2b@mail.gmail.com>

Hi all,

I would like to know if someone suceeded in compiling newer binica under xp,
and suceeded in calling it directly from EEGlab. As I understand only older
binica version exists under xp and is not directly meant to be called from
EEGLab.

Thanks,

       P?ter Solt?sz
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100419/48ab89fa/attachment.html>

From siu07pa at reading.ac.uk  Wed Apr 21 02:12:42 2010
From: siu07pa at reading.ac.uk (Peter Acs)
Date: 21 Apr 2010 10:12:42 +0100
Subject: [Eeglablist] Headphone/Earphone interference with EEG recording
Message-ID: <Prayer.1.3.1.1004211012420.19655@vimb4.rdg.ac.uk>

Hello all,

I am an undergrad student at the University of Reading looking to conduct 
some EEG experiments.

The experiment involves audio stimuli and I am not sure about the best way 
to present the stimuli to the participants of the experiment.

Would the use of headphones or earphones interfere with the recording of 
the data in any way? If so, I would just use speakers instead, but ideally 
I would want to obstruct any outside noises.

Thank you

Peter


-- 
Peter Acs
Artificial Intelligence and Cybernetics
Student Number: 15016703



From arno at ucsd.edu  Wed Apr 21 11:00:10 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Wed, 21 Apr 2010 11:00:10 -0700
Subject: [Eeglablist] Post-doc Position at University of Pittsburgh
Message-ID: <62BF22A8-99D0-40C7-9958-8FDFA2267B3E@ucsd.edu>

Message posted on Behalf of Raymond Y. Cho, M.D.

The Clinical Cognitive Neuroscience lab (http:// 
kraepelin.wpic.pitt.edu) at the University of Pittsburgh, is currently  
looking for applicants at the post-doctoral level in psychology,  
neuroscience, or related fields with a strong interest in  
investigating mechanisms of cognitive control and their impairments in  
schizophrenia. The post-doctoral fellow will be involved in a large  
EEG study of the normal development of oscillatory brain activity, as  
well as with other projects involving EEG and fMRI investigations of  
cognitive control deficits in first-episode and chronic schizophrenia  
populations.  Candidates should have background in EEG and/or fMRI  
methods, and strong quantitative and programming skills are an asset.  
Experience in Matlab, C/C++, spectral analyses, and functional  
connectivity analyses desirable. Due to the NIH support mechanism,  
applicants should have citizenship or permanent residency in the US.   
Interested applicants should submit a statement of interest an!
  d CV by May 15 to Dr. Raymond Cho at chory at upmc.edu.

Raymond Y. Cho, M.D., M.Sc.
Assistant Professor, Department of Psychiatry
University of Pittsburgh School of Medicine
Western Psychiatric Institute & Clinic
3811 O'Hara St, Oxford 450
Pittsburgh, PA 15213
(412) 586-9250 (o)
(412) 647-7861 (fax)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100421/1067fcd5/attachment.html>

From andygoldfine at gmail.com  Thu Apr 22 07:00:33 2010
From: andygoldfine at gmail.com (Andrew Goldfine)
Date: Thu, 22 Apr 2010 10:00:33 -0400
Subject: [Eeglablist] Online ICA training sets?
Message-ID: <m2zfe9fc08d1004220700jc49b9ee0y2b021e38c130a694@mail.gmail.com>

I'm a new user of EEGLAB and am having difficulty at times determining if a
component should be considered artifact. Are there any training sets out
there of many component pictures and what experts consider them to be
(neurogenic, myogenic, eye, other)?

Thanks,
Andy
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100422/ebf7d652/attachment.html>

From tom.ferree at gmail.com  Wed Apr 21 14:31:10 2010
From: tom.ferree at gmail.com (Thomas Ferree)
Date: Wed, 21 Apr 2010 16:31:10 -0500
Subject: [Eeglablist] Headphone/Earphone interference with EEG recording
In-Reply-To: <Prayer.1.3.1.1004211012420.19655@vimb4.rdg.ac.uk>
References: <Prayer.1.3.1.1004211012420.19655@vimb4.rdg.ac.uk>
Message-ID: <y2tf22c246c1004211431zf7bc5b2di414ed2291ce6bf2f@mail.gmail.com>

Dear Peter,
I believe MRI compatible headphones use air only.
That would prevent the possibility of interference.
There are plenty on the market - I don't have a specific recommendation.
Best, Tom.

On Wed, Apr 21, 2010 at 4:12 AM, Peter Acs <siu07pa at reading.ac.uk> wrote:

> Hello all,
>
> I am an undergrad student at the University of Reading looking to conduct
> some EEG experiments.
>
> The experiment involves audio stimuli and I am not sure about the best way
> to present the stimuli to the participants of the experiment.
>
> Would the use of headphones or earphones interfere with the recording of
> the data in any way? If so, I would just use speakers instead, but ideally
> I would want to obstruct any outside noises.
>
> Thank you
>
> Peter
>
>
> --
> Peter Acs
> Artificial Intelligence and Cybernetics
> Student Number: 15016703
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100421/1fec4cf0/attachment.html>

From mhartman at egi.com  Thu Apr 22 10:00:28 2010
From: mhartman at egi.com (Mike Hartman)
Date: Thu, 22 Apr 2010 10:00:28 -0700
Subject: [Eeglablist] Headphone/Earphone interference with EEG recording
In-Reply-To: y2tf22c246c1004211431zf7bc5b2di414ed2291ce6bf2f@mail.gmail.com
Message-ID: <20100422170028.44f721c4@mail.egi.com>

Hi Peter,

google Etymotic ER-3A

----- Original Message -----
From: Thomas Ferree
[mailto:tom.ferree at gmail.com]
To: Peter Acs
[mailto:siu07pa at reading.ac.uk]
Cc: eeglablist at sccn.ucsd.edu
Sent: Wed, 21
Apr 2010 14:31:10 -0700
Subject: Re: [Eeglablist] Headphone/Earphone
interference with EEG recording


> Dear Peter,
> I believe MRI compatible headphones use air only.
> That would prevent the possibility of interference.
> There are plenty on the market - I don't have a specific recommendation.
> Best, Tom.
> 
> On Wed, Apr 21, 2010 at 4:12 AM, Peter Acs <siu07pa at reading.ac.uk> wrote:
> 
> > Hello all,
> >
> > I am an undergrad student at the University of Reading looking to conduct
> > some EEG experiments.
> >
> > The experiment involves audio stimuli and I am not sure about the best way
> > to present the stimuli to the participants of the experiment.
> >
> > Would the use of headphones or earphones interfere with the recording of
> > the data in any way? If so, I would just use speakers instead, but ideally
> > I would want to obstruct any outside noises.
> >
> > Thank you
> >
> > Peter
> >
> >
> > --
> > Peter Acs
> > Artificial Intelligence and Cybernetics
> > Student Number: 15016703
> >
> > _______________________________________________
> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> > To unsubscribe, send an empty email to
> > eeglablist-unsubscribe at sccn.ucsd.edu
> > For digest mode, send an email with the subject "set digest mime" to
> > eeglablist-request at sccn.ucsd.edu
> >
> 



From nickbedo at yahoo.com  Thu Apr 22 13:13:06 2010
From: nickbedo at yahoo.com (Nick Bedo)
Date: Thu, 22 Apr 2010 13:13:06 -0700 (PDT)
Subject: [Eeglablist] drawing lines between channels on topoplot()
Message-ID: <215737.94565.qm@web62003.mail.re1.yahoo.com>

I'm trying to show causal influences between channels by drawing lines on the topoplot() figure output.  For example, if my calculation shows that Fz causally influences Pz, I want a line coming from Fz and pointing to Pz.  I have a feeling that at least one person has done this before using topoplot() in eeglab.

And specifically, I'm asking how to incorporate this command in a script, not through the GUI.

Any tips are appreciated.
 
Thanks,
Nick


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100422/d9329250/attachment.html>

From Michiel.Spape at nottingham.ac.uk  Tue Apr 27 02:49:55 2010
From: Michiel.Spape at nottingham.ac.uk (Michiel Spape)
Date: Tue, 27 Apr 2010 10:49:55 +0100
Subject: [Eeglablist] Newcrossf and Study
Message-ID: <0CA8E1B4EC20D743912B980E486C5CAF032A80EF@VUIEXCHC.ad.nottingham.ac.uk>

Hi all,

I've been doing cross channel coherence analysis in EEGLAB for a while now, but I keep thinking there might be a better way of doing it. A week ago, I found (I guess, as one stumbles upon many treasures in such a wealth of functions as EEGLAB offers) the 'study' structure and set of GUI operations to be just about what I was looking for - except not cross-channel-wise. My question then is whether it would be sort of possible to... vandalise the Study tools such that rather than running ITC/ERSP analysis on channel 1:28 (given my 28 channels), for two conditions, it would run newcrossf on a single channel (say 1) vs all other channels (i.e. 2:28), such that it will do, for the whole study set, the bootstrapped tests between the two conditions (on the 27 cross-coherences).

 

Any suggestions on where to begin with would be more than welcome.

Thanks,

Michiel

 

Michiel Spap?

Research Fellow

Perception & Action group

University of Nottingham

School of Psychology

 

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system:
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100427/0f4205e0/attachment.html>

From saim_rasheed at hotmail.com  Tue Apr 27 05:14:58 2010
From: saim_rasheed at hotmail.com (Saim Rasheed)
Date: Tue, 27 Apr 2010 18:14:58 +0600
Subject: [Eeglablist] Implementing Classification Algorithms within EEGLAB
Message-ID: <SNT119-W144F1C597C334D3A44DDA7EA030@phx.gbl>


 
Hello EEGLAB community,


I am using EEGLAB for data analysis and now interested in implementing some classification algorithm, for example, Support Vector Machine.
I have recorded EEG data for three different conditions and now want to extract relevant features to perform training and classify them according to the conditions with in EEGLAB.
Perhaps it is not possible with existing EEGLAB code and probably I need to write a customized code for this purpose. I was looking, where and how should I take start with in EEGLAB as I am beginning with classification algorithms and with writing EEGLAB script.


Any answers/suggestions would be appreciated in this regard.
Thanking you in advance.


Regards
Saim, 
Milan



 		 	   		  
_________________________________________________________________
Hotmail: Powerful Free email with security by Microsoft.
https://signup.live.com/signup.aspx?id=60969
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100427/6784e9dd/attachment.html>

From saim.rasheed at unimi.it  Tue Apr 27 10:59:13 2010
From: saim.rasheed at unimi.it (Saim Rasheed)
Date: Tue, 27 Apr 2010 23:59:13 +0600
Subject: [Eeglablist] =?windows-1256?q?Implementing_Classification_Algorit?=
 =?windows-1256?q?hms_within_EEGLAB=FE?=
Message-ID: <SNT119-W299DAB5C49FA7186302F6EA030@phx.gbl>


Hello,


I am using EEGLAB for data analysis and now interested in implementing some classification algorithm, for example, Support Vector Machine.
I have recorded EEG data for three different conditions and now want to extract relevant features to perform training and classify them according to the conditions with in EEGLAB.
Perhaps it is not possible with existing EEGLAB code and probably I need to write a customized code for this purpose. I was looking, where and how should I take start with in EEGLAB as I am beginning with classification algorithms and with writing EEGLAB script.


Any answers/suggestions would be appreciated in this regard.
Thanking you in advance.


Regards
Saim, 
Milan

 		 	   		  
_________________________________________________________________
Hotmail: Powerful Free email with security by Microsoft.
https://signup.live.com/signup.aspx?id=60969
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100427/15082f14/attachment.html>

From smakeig at gmail.com  Tue Apr 27 12:56:45 2010
From: smakeig at gmail.com (Scott Makeig)
Date: Tue, 27 Apr 2010 12:56:45 -0700
Subject: [Eeglablist]
	=?utf-8?q?Implementing_Classification_Algorithms_wit?=
	=?utf-8?q?hin_EEGLAB=E2=80=8F?=
In-Reply-To: <SNT119-W299DAB5C49FA7186302F6EA030@phx.gbl>
References: <SNT119-W299DAB5C49FA7186302F6EA030@phx.gbl>
Message-ID: <k2u9e09b8f01004271256n82053635i7aa20ba838e5448c@mail.gmail.com>

Saim & all  - In a few weeks Christian Kothe and Thorsten Zander plan to
release a comprehensive and easy-to-use Matlab package, BCILAB, that will
function as a plug-in to EEGLAB and will very likely make your task much
easier. You can read a preview here:
http://sccn.ucsd.edu/~scott/pdf/Delorme_BCITools10.pdf -- a forthcoming
chapter on Matlab tools for BCI research.

Scott Makeig

2010/4/27 Saim Rasheed <saim.rasheed at unimi.it>

>  Hello,
>
>
> I am using EEGLAB for data analysis and now interested in implementing some
> classification algorithm, for example, Support Vector Machine.
> I have recorded EEG data for three different conditions and now want to
> extract relevant features to perform training and classify them according to
> the conditions with in EEGLAB.
> Perhaps it is not possible with existing EEGLAB code and probably I need to
> write a customized code for this purpose. I was looking, where and
> how should I take start with in EEGLAB as I am beginning with classification
> algorithms and with writing EEGLAB script.
>
>
> Any answers/suggestions would be appreciated in this regard.
> Thanking you in advance.
>
>
> Regards
> Saim,
> Milan
>
>
> ------------------------------
> Hotmail: Powerful Free email with security by Microsoft. Get it now.<https://signup.live.com/signup.aspx?id=60969>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Scott Makeig, Research Scientist and Director, Swartz Center for
Computational Neuroscience, Institute for Neural Computation, University of
California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100427/fdd64308/attachment.html>

From ralf.baales at netcologne.de  Wed Apr 28 01:01:29 2010
From: ralf.baales at netcologne.de (Ralf Baales)
Date: Wed, 28 Apr 2010 10:01:29 +0200
Subject: [Eeglablist] Filtering continuous Data with FIR high-pass Filter
Message-ID: <!&!AAAAAAAAAAAYAAAAAAAAACRPUt+fOP9KhGq0ssR+LVfCgAAAEAAAAF8+fwvJYQlGp9baH4vgEm8BAAAAAA==@netcologne.de>

EEGLAB version is 8.0.3.4.b

Win 7 Professional 64 bit

 

I am trying to filter and Processing continuous data with the Tools > Filter
the data > Basic FIR filter by entering "1" as the lower edge. 

eegfilt() - performing 1500-point highpass filtering.

eegfilt() - highpass transition band width is 0.2 Hz.

...................20..........

After reaching the last step (1643861:1865960) the filtering process breaks
up with the following error massage:

 

buffer invalid error

invalid input argument #1

 

The items of the continuous Data file are:

 

Channels per frame         30

Frames per epoch            1865960

Epochs                               1

Events                                3621

Sampling rate (Hz)           500

Epoch start (sec)              0.000

Epoch end (sec)                3731.918

Channel locations            Yes

Reference                          nose tip

ICA weights                       no

Data size (Mb)                  225.2

 

What's going wrong?

 

Thanks

Ralf

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100428/6c63846e/attachment.html>

From wu_vivi at yahoo.com  Thu Apr 29 12:01:13 2010
From: wu_vivi at yahoo.com (vivi wu)
Date: Thu, 29 Apr 2010 12:01:13 -0700 (PDT)
Subject: [Eeglablist] K-complex detection
Message-ID: <322288.8242.qm@web45414.mail.sp1.yahoo.com>

Hello,?
I have been using EEGlab for multichannel analysis on EEG signal. Now I am looking for a simple and automatic way to detect K-complex in stage II sleep. I've checked some papers (Bremer et al., 1970; Saccomandi et al., 2008) and wonder if EEGlab has any related functions to detect K-complex. Could someone give me some suggestions? I appreciate that.
regards,Weiwei Wu


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100429/ee880524/attachment.html>

From nickbedo at yahoo.com  Fri Apr 30 08:49:23 2010
From: nickbedo at yahoo.com (Nick Bedo)
Date: Fri, 30 Apr 2010 08:49:23 -0700 (PDT)
Subject: [Eeglablist] Getting coherence between 0-1
Message-ID: <558664.46350.qm@web62001.mail.re1.yahoo.com>

Hi all,

I'm wondering how I might tweak my coherence scripts to compute and output coherence measures between 0-1.  Currently I an getting max outputs ranging ~100-2000 (via pop_newcrossf).  As I've been reading through the eeglab archives, it seems like a pretty minor fix, but I couldn't tell how the message's author went about it.

Any tips are appreciated.

Thanks,
Nick


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100430/e9e62ae9/attachment.html>

From m_mushwani at hotmail.com  Fri Apr 30 15:24:44 2010
From: m_mushwani at hotmail.com (Moeen Riaz)
Date: Sat, 1 May 2010 04:24:44 +0600
Subject: [Eeglablist] eeglablist Digest, Vol 66, Issue 22
In-Reply-To: <mailman.0.1272654005.31619.eeglablist@sccn.ucsd.edu>
References: <mailman.0.1272654005.31619.eeglablist@sccn.ucsd.edu>
Message-ID: <BAY118-W13AA66CAE6DA4D8F34888FEF000@phx.gbl>



Hello, 
I am want to build a simple EEG sensor. i have seen the schematics of two channel EEG in Open EEG but i want to contruct a simple one channel EEG. Please help me and send me a simple circuit diagram or schematics and some guidelines as well if possible. All and any help will highly be appreciated.

Regards,Moeen Riaz
 		 	   		  
_________________________________________________________________
Hotmail: Free, trusted and rich email service.
https://signup.live.com/signup.aspx?id=60969
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100501/07f993f0/attachment.html>

From arno at ucsd.edu  Sat May  1 09:47:52 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Sat, 1 May 2010 09:47:52 -0700
Subject: [Eeglablist] Getting coherence between 0-1
In-Reply-To: <558664.46350.qm@web62001.mail.re1.yahoo.com>
References: <558664.46350.qm@web62001.mail.re1.yahoo.com>
Message-ID: <4476B62B-2477-4FF9-837F-962A555482CF@ucsd.edu>

Dear Nick,

that seems odd. The output should be between 0 and 1. Try this code.

a = rand(1,100,10);
b = rand(1,100,10);
coh = newcrossf(a,b,100,[0 1],100,[3 0.8]);
max(max(coh))

The maximum is 1 or less

Or using FFT

a = rand(1,100,10);
b = rand(1,100,10);
coh = newcrossf(a,b,100,[0 1],100,0);
max(max(coh))

Hope this helps,

Arno

On Apr 30, 2010, at 8:49 AM, Nick Bedo wrote:

> Hi all,
>
> I'm wondering how I might tweak my coherence scripts to compute and  
> output coherence measures between 0-1.  Currently I an getting max  
> outputs ranging ~100-2000 (via pop_newcrossf).  As I've been reading  
> through the eeglab archives, it seems like a pretty minor fix, but I  
> couldn't tell how the message's author went about it.
>
> Any tips are appreciated.
>
> Thanks,
> Nick
>
>
>
> <ATT00001..txt>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100501/2562ae71/attachment.html>

From s_thomasgeorge at sify.com  Mon May  3 01:12:18 2010
From: s_thomasgeorge at sify.com (Thomas George)
Date: Mon, 3 May 2010 13:42:18 +0530
Subject: [Eeglablist] Help on component maps
Message-ID: <i2j5a083cad1005030112nc701e1eci48ffdd0615ea28e0@mail.gmail.com>

Dear All,
I have basic doubts in component maps.
Is it the component 2-D maps which is obtained in * plot>component maps>2-D
maps* is representing the* amplitude spectrum* of various channels which
are  is fed in?.or it represents the *frequency spectrum*.
and
say,
Suppose I import a 16 channel data with channels namely Fp2 , F4, C4,
.............and so on, then component maps obtained are 1, 2,3.............
*Can I consider that the 2-D map  1 is corresponding to** Fp2 and **map 2 to
**F4 etc...*?
please help.
Regards
Thomas George
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100503/531c84d0/attachment.html>

From schwiedrzik at mpih-frankfurt.mpg.de  Mon May  3 05:10:09 2010
From: schwiedrzik at mpih-frankfurt.mpg.de (Caspar M. Schwiedrzik)
Date: Mon, 3 May 2010 14:10:09 +0200
Subject: [Eeglablist] Problems with reading sensor positions for EGI's
	Hydrocell 128 Net
Message-ID: <h2xc56ed4721005030510qb093b68ah7bedf5955f8d96b8@mail.gmail.com>

Dear all,
I am trying to read in the sensor positions for EGI's Hydrocell 128
Net. I have an SFP file that was provided by EGI (Hydrocel GSN 128
10.sfp).
It has the following structure:

E1      6.60688      6.30230     -2.94229
E2      6.04082      7.65872     0.350950
E3      4.41106      8.71481      3.50199

When I read it in in EEGlab, all sensor positions seem to be rotated
by 90 degrees.
Unfortunately, I didn't find any other file on the EGI FTP server.

I guess somebody must have solved this problem. Any hints how to read
in the sensor positions properly or anyone with another file that
works?
Thanks and kind regards,
Caspar M. Schwiedrzik


From hewei12 at hotmail.com  Mon May  3 15:26:46 2010
From: hewei12 at hotmail.com (wei he)
Date: Mon, 3 May 2010 22:26:46 +0000
Subject: [Eeglablist] Problems with reading sensor positions for
	EGI's	Hydrocell 128 Net
In-Reply-To: <h2xc56ed4721005030510qb093b68ah7bedf5955f8d96b8@mail.gmail.com>
References: <h2xc56ed4721005030510qb093b68ah7bedf5955f8d96b8@mail.gmail.com>
Message-ID: <SNT134-w2BAE9A330814C5CC99559BAF20@phx.gbl>


>From the coordinates of first three channels you provided, they look wrong. Use the attached file. Please ignore the first 3 rows as they are for BESA use.
Wei  



> Date: Mon, 3 May 2010 14:10:09 +0200
> From: schwiedrzik at mpih-frankfurt.mpg.de
> To: eeglablist at sccn.ucsd.edu
> Subject: [Eeglablist] Problems with reading sensor positions for EGI's	Hydrocell 128 Net
> 
> Dear all,
> I am trying to read in the sensor positions for EGI's Hydrocell 128
> Net. I have an SFP file that was provided by EGI (Hydrocel GSN 128
> 10.sfp).
> It has the following structure:
> 
> E1      6.60688      6.30230     -2.94229
> E2      6.04082      7.65872     0.350950
> E3      4.41106      8.71481      3.50199
> 
> When I read it in in EEGlab, all sensor positions seem to be rotated
> by 90 degrees.
> Unfortunately, I didn't find any other file on the EGI FTP server.
> 
> I guess somebody must have solved this problem. Any hints how to read
> in the sensor positions properly or anyone with another file that
> works?
> Thanks and kind regards,
> Caspar M. Schwiedrzik
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100503/ad7592cf/attachment.html>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: GSN-HydroCel-129.sfp
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100503/ad7592cf/attachment.ksh>

From dr.ilya at yahoo.com  Mon May  3 14:52:45 2010
From: dr.ilya at yahoo.com (Ilya Adamchic)
Date: Mon, 3 May 2010 14:52:45 -0700 (PDT)
Subject: [Eeglablist] Problems with reading sensor positions for EGI's
	Hydrocell 128 Net
In-Reply-To: <h2xc56ed4721005030510qb093b68ah7bedf5955f8d96b8@mail.gmail.com>
References: <h2xc56ed4721005030510qb093b68ah7bedf5955f8d96b8@mail.gmail.com>
Message-ID: <997523.86328.qm@web63805.mail.re1.yahoo.com>

Hi Schwiedrzik,
Try this file, it is a EGI Hydro_Cell 128 channel file. It should work.
Cheers,
Ilya


________________________________
From: Caspar M. Schwiedrzik <schwiedrzik at mpih-frankfurt.mpg.de>
To: eeglablist at sccn.ucsd.edu
Sent: Mon, May 3, 2010 2:10:09 PM
Subject: [Eeglablist] Problems with reading sensor positions for EGI's Hydrocell 128 Net

Dear all,
I am trying to read in the sensor positions for EGI's Hydrocell 128
Net. I have an SFP file that was provided by EGI (Hydrocel GSN 128
10.sfp).
It has the following structure:

E1      6.60688      6.30230     -2.94229
E2      6.04082      7.65872     0.350950
E3      4.41106      8.71481      3.50199

When I read it in in EEGlab, all sensor positions seem to be rotated
by 90 degrees.
Unfortunately, I didn't find any other file on the EGI FTP server.

I guess somebody must have solved this problem. Any hints how to read
in the sensor positions properly or anyone with another file that
works?
Thanks and kind regards,
Caspar M. Schwiedrzik
_______________________________________________
Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu



      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100503/6900c893/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: GSN-HydroCel-128.sfp
Type: application/octet-stream
Size: 5430 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100503/6900c893/attachment.obj>

From silvia.corbera at yale.edu  Mon May  3 15:29:21 2010
From: silvia.corbera at yale.edu (Silvia Corbera)
Date: Mon, 03 May 2010 17:29:21 -0500
Subject: [Eeglablist] disregard previous please. use this one. Filtering
 with basic FIR filter
Message-ID: <4BDF4E41.7080905@yale.edu>



(please moderator disregard my previous email)

Dear all,
I am trying to filter long latency visual evoked potentials, and using 
Basic FIR filter, and I want to use as a lower edge = 0.1 (i've also 
tried = 1)
and for higher edge = 30.
I know that is recommended to do one first and then the other one. But 
every time I start trying to filter the 0.1 it gives me an error:

" EEG lab error in function firls ( ) at line 152
Out of memory, Type help Memory for your options"

And if I try = 1 (for the lower edge) says:

"invalid argument for input #1

My computer is a Pentium (R) dual core_CPU E5300 @ 2.60GHz 2.59GHx
Installed memory (RAM): 2 GB.

I don't know what to do. DO you think is a memory problem?
thank you so much,
s



Now that I've tried this several times...I think it might be a memory 
problem,

-- 
Silvia Corbera, Ph.D.
Post-Doctorate Associate                                                                                                                    
Yale School of Medicine                                                                                                            
Department of Psychiatry                                                                                                  
Connecticut Mental Health Center                                                                                                  
34 Park Street, New Haven CT 06519                                                                                         
(203) 974 7862                                                                                                     
silvia.corbera at yale.edu




From jdien07 at mac.com  Mon May  3 16:42:57 2010
From: jdien07 at mac.com (Joseph Dien)
Date: Mon, 03 May 2010 19:42:57 -0400
Subject: [Eeglablist] research assistant opening
References: <91DE5CFA11F8DC49ACA9E1DDAD380E54A33AC61E65@EX.casl.umd.edu>
Message-ID: <7F035DED-2BBB-4537-990A-D6D623A6FB53@mac.com>

See attached flyer.

--------------------------------------------------------------------------------

Joseph Dien,
Senior Research Scientist
Center for Advanced Study of Language
University of Maryland 
7005 52nd Avenue
College Park, MD 20742-0025

E-mail: jdien07 at mac.com
Phone: 301-226-8848
Fax: 301-226-8811
http://homepage.mac.com/jdien07/


-------------- next part --------------
A non-text attachment was scrubbed...
Name: FRA-CogNS 2010.pdf
Type: application/pdf
Size: 43064 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100503/a4d96250/attachment.pdf>

From s_thomasgeorge at sify.com  Tue May  4 01:48:10 2010
From: s_thomasgeorge at sify.com (Thomas George)
Date: Tue, 4 May 2010 14:18:10 +0530
Subject: [Eeglablist] Help on component maps
Message-ID: <j2y5a083cad1005040148t552eaeedv1906a408e7d06b80@mail.gmail.com>

Dear All,
I have basic doubts in component maps.
Is it the component 2-D maps which is obtained in * plot>component maps>2-D
maps* is representing the* amplitude spectrum* of various channels which
are  is fed in?.or it represents the *frequency spectrum*.
and
say,
Suppose I import a 16 channel data with channels namely Fp2 , F4, C4,
.............and so on, then component maps obtained are 1, 2,3.............
*Can I consider that the 2-D map  1 is corresponding to** Fp2 and **map 2 to
**F4 etc...*?
please help.
Regards
Thomas George
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100504/420f8e30/attachment.html>

From mschuber at mail.upb.de  Wed May  5 07:36:54 2010
From: mschuber at mail.upb.de (Michael Schubert)
Date: Wed, 05 May 2010 16:36:54 +0200
Subject: [Eeglablist] exporting datasets in .cnt format
Message-ID: <4BE18286.7030004@mail.upb.de>

Hello EEGLAB users,

I was wondering if there's any chance to export datasets back in 
Neuroscan's '.cnt' format. There might be a way using an ASCII or '.edf' 
export which can then be opened with SCAN.

Thanks!
Michael


From jainsanket1 at gmail.com  Wed May  5 10:38:08 2010
From: jainsanket1 at gmail.com (Sanket Jain)
Date: Wed, 5 May 2010 12:38:08 -0500
Subject: [Eeglablist] Help on component maps
In-Reply-To: <j2y5a083cad1005040148t552eaeedv1906a408e7d06b80@mail.gmail.com>
References: <j2y5a083cad1005040148t552eaeedv1906a408e7d06b80@mail.gmail.com>
Message-ID: <z2h54f5410c1005051038t89911b72mf7e0e732657c761e@mail.gmail.com>

It represents neither amplitude nor frequency spectrum. These maps
correspond to the independent components of your data. Refer to Independent
component analysis (Basic reference for use with EEGLAB available on EEGLAB
website)

Sanket

On Tue, May 4, 2010 at 3:48 AM, Thomas George <s_thomasgeorge at sify.com>wrote:

> Dear All,
> I have basic doubts in component maps.
> Is it the component 2-D maps which is obtained in * plot>component
> maps>2-D maps* is representing the* amplitude spectrum* of various
> channels which are  is fed in?.or it represents the *frequency spectrum*.
> and
> say,
> Suppose I import a 16 channel data with channels namely Fp2 , F4, C4,
> .............and so on, then component maps obtained are 1, 2,3.............
> *Can I consider that the 2-D map  1 is corresponding to** Fp2 and **map 2
> to **F4 etc...*?
> please help.
> Regards
> Thomas George
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100505/fe245fb3/attachment.html>

From sjwebb at u.washington.edu  Fri May  7 13:27:51 2010
From: sjwebb at u.washington.edu (Sara Jane Webb)
Date: Fri, 7 May 2010 13:27:51 -0700
Subject: [Eeglablist] Postdoc: University of Washington
Message-ID: <A81D07A2-E4EE-4291-BA84-E62102018923@u.washington.edu>

Physiological Endophenotypes in Autism

2 Postdoctoral Positions available at the Autism Research Programs at  
the University of Washington and Seattle Children?s Research Institute  
(Seattle, Washington).

The University of Washington & Seattle Children?s Research Institute  
bring together scientists from several different disciplines who are  
working collaboratively to discover the genetic cause of autism and  
the factors that impact outcome for children with this disorder. As a  
part of this effort, our lab utilizes psychophysiological and  
behavioral measures for research in the domains of early development,  
neural processes, (endo)phenotypes with applications in clinical  
practice. For more information about the lab please see our website http://depts.washington.edu/pbslab/

Although duties vary based on position, in general, fellows are  
expected to contribute to study design, data collection, data  
analysis, training and supervision of research assistants, co- 
authoring or leading publications, and the dissemination of findings.

1. Post-doctoral position (pending funding) (for up to 2 years) to  
examine broader phenotype characteristics of parents of children with  
autism and infants at-risk for autism.  This project uses EEG/ERP and  
behavioral methodology to study social attention in 1st degree  
relatives of individuals with autism as part of the UW Autism Center  
of Excellence. The successful candidate will have a Ph.D. in  
developmental or cognitive psychology with experience in  
electrophysiology and an interest in clinical neuroscience.
2. Post-doctoral position available (pending funding) (for up to 3  
years) to study the role of attention and regulation in social and  
language growth in young children with autism and typical development.  
This project will use cardiophysiology and behavioral measures to  
better understand heterogeneity of development in autism. The  
successful candidate will have a Ph.D. in clinical, developmental or  
educational psychology (or related field) with a background in  
developmental disabilities. Masters level applicants with experience  
in standardized autism clinical assessment will be considered.

Tentative start date for position would be Summer/Fall 2010.

Enquiries should be directed to: Dr. Sara Jane Webb sjwebb at u.washington.edu 
. Please provide cover letter, CV & contact information for 3  
references.

Sara Jane Webb, PhD
Research Assistant Professor of Psychiatry and Behavioral Sciences
and UW Autism Center, Research Program
http://depts.washington.edu/pbslab/
Box 357920; CHDD 314C; University of Washington
Seattle WA 98195
206.221.6461
sjwebb at u.washington.edu

Confidentiality Notice:  Because email is not secure, please be aware  
that we cannot guarantee the confidentiality of information sent by  
email.  If you are not the intended recipient, please notify the  
sender by reply email, and then destroy all copies of the message and  
any attachments.




From tehuberpro at gmail.com  Fri May  7 08:53:25 2010
From: tehuberpro at gmail.com (ondrej lassak)
Date: Fri, 7 May 2010 17:53:25 +0200
Subject: [Eeglablist] artifacts in time freq plots
Message-ID: <m2p436e576d1005070853o909f741chff76234cd3467ec9@mail.gmail.com>

I fed the TF analysis single sinusoid 6 and 12Hz and the TF plot shows
multiple specral lines (more than two).
How can one rely on the TF when it introduces such massive artifacts both in
pure FFT spectrogram and Wavelet scalogram?
Or am I doing something wrong? When only one freq during the whole time span
is present the TF plots look like really bad moira and the presence of the
freq is apparent only from the summation over time (left from the  main
plot).


The matlab report with function calls and resulting pictures is attached
below (no scripts embedded in the html).
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100507/46a81564/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: baseline eeglab TFs experiments.zip
Type: application/zip
Size: 269024 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100507/46a81564/attachment.zip>

From mschuber at mail.upb.de  Fri May  7 00:56:18 2010
From: mschuber at mail.upb.de (Michael Schubert)
Date: Fri, 07 May 2010 09:56:18 +0200
Subject: [Eeglablist] exporting datasets in .cnt format
Message-ID: <4BE3C7A2.7000504@mail.upb.de>

Hello,

just a short update on this issue:
I downloaded the current EEGLAB (8_0_3_5b) version today and installed 
BIOSIG as described here http://sccn.ucsd.edu/wiki/EEGLAB_Plugins
Now EEGLAB gives me the option to export *.edf files which can then be 
read by Scan (4.4).

Thanks,
Michael


From arno at ucsd.edu  Tue May 11 20:28:28 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Tue, 11 May 2010 20:28:28 -0700
Subject: [Eeglablist] artifacts in time freq plots
In-Reply-To: <m2p436e576d1005070853o909f741chff76234cd3467ec9@mail.gmail.com>
References: <m2p436e576d1005070853o909f741chff76234cd3467ec9@mail.gmail.com>
Message-ID: <F5512F8C-C2D2-4B53-9DB2-EF836F75030D@ucsd.edu>

Dear Ondrej,

I have looked in detail in your detailed analysis.

First, thank you for looking in detail into that. Even though these  
messages are scary (we can never be sure that EEGLAB is bug free), we  
greatly appreciate that people like you take the time to test that the  
functions are doing what they are supposed to do.

Regarding your analysis, all of the problems you encounter are due to  
using the baseline option in a specific way on this artificial data  
(see below). First, you should try the "'baseline', NaN" option which  
does not perform any baseline. It is the first thing to do before  
trying other types of baseline. I am attaching here a screen copy of  
the decomposition on your data. We can clearly see the two frequencies  
that you have generated.


using your data, this is the command line call "figure;  
[ersp,itc,powbase,times,freqs,erspboot,itcboot] = newtimef(x, 8*2048,  
[-500 6000], 2048, 0,'baseline',NaN,'basenorm','off', 'maxfreq' , 
20,'nfreqs',50,'padratio', 32, 'scale', 'abs');"

Then, comes the baseline. The graphs you produce are meaningful until  
you start using the 'basenorm' option. The "basenorm" option is used  
compute and show z scores (we prefer dB ourselves but some other  
researchers prefer to use z-score). In your case your baseline is from  
-500 ms to 500 ms. It means that the standard deviation will be 0 at  
most frequency except at 6 Hz. A standard deviation of 0 (when  
normalizing) makes the weights blow up to close to infinity (your  
power is 10^7 standard deviation). It does not totally blow up to  
infinity since the standard deviation of the baseline is not perfectly  
0 but a very small number. The strange plots you are observing are due  
to that. I have looked into detail in the code of the function and I  
plotted all intermediary results from inside the function itself and  
there is no doubt about that.

Note also that the newtimef function was primarily designed to process  
data trials and not continuous data. This is the reason why the inter- 
trial coherence measure for your data returns meaningless results (ITC  
is only relevant for more than 1 trial).

Let me know if you have other questions or comments,

Arno

On May 7, 2010, at 8:53 AM, ondrej lassak wrote:

> I fed the TF analysis single sinusoid 6 and 12Hz and the TF plot  
> shows multiple specral lines (more than two).
> How can one rely on the TF when it introduces such massive artifacts  
> both in pure FFT spectrogram and Wavelet scalogram?
> Or am I doing something wrong? When only one freq during the whole  
> time span is present the TF plots look like really bad moira and the  
> presence of the freq is apparent only from the summation over time  
> (left from the  main plot).
>
>
> The matlab report with function calls and resulting pictures is  
> attached below (no scripts embedded in the html).
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100511/13fa33b3/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: pastedGraphic.tiff
Type: image/tiff
Size: 23636 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100511/13fa33b3/attachment.tiff>

From pawel.augustynowicz at gmail.com  Wed May 12 12:03:32 2010
From: pawel.augustynowicz at gmail.com (=?iso-8859-2?Q?Pawe=B3_Augustynowicz?=)
Date: Wed, 12 May 2010 21:03:32 +0200
Subject: [Eeglablist] Retreive detailed event info for EGI epoched recordings
Message-ID: <4beafb87.a225e30a.41c3.2824@mx.google.com>

Dear All,
 
I have a short question about import data from EGI simple binary file format
with events. By default, EEGLAB imports EGI's simple binary with events, but
in only one dimension. I mean, that I can see only the name of the event.
Additional data (event fields) are not imported. Does anyone have an idea of
how to attach this data do EEGLAB? I assume that this information is written
in additional channels. But EEGLAB deletes all of these channels except one
with the basic info. 
I tried to insert this data manually, by there is another problem. I have a
few epochs in original NetStation recording. These epochs are essential
because of the experiment design (experiment is too long and we have to do
breaks). When I export this data from NetStation as simple binary, this file
is being broken into files representing epochs in original recording. I can
import this bunch of files with pop_readsegegi command. This function does
the job well, except one little bug: NetStation counts time from the
beginning of the experiment. Pauses during the recording are treated as the
beginning of epochs. In EEGLAB all epochs have continuous time, so there is
no possibility to attach events exported from NetStation, because the time
after the first epoch is different. 
 
Does anyone ran on these problems and managed to resolve it?
 
Regards,
Pawel Augustynowicz
The John Paul II Catholic University of Lublin
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100512/68c931b7/attachment.html>

From mataothefifth at yahoo.co.jp  Tue May 11 19:46:38 2010
From: mataothefifth at yahoo.co.jp (Makoto Miyakoshi)
Date: Wed, 12 May 2010 11:46:38 +0900 (JST)
Subject: [Eeglablist] artifacts in time freq plots
In-Reply-To: <m2p436e576d1005070853o909f741chff76234cd3467ec9@mail.gmail.com>
Message-ID: <20100512024638.73434.qmail@web3709.mail.tnz.yahoo.co.jp>

Dear Ondrej,

I replicated your results. This is shocking actually.
I also found that the problem is probably in baseline subtraction because
when I set 'baseline', NaN, the result seems right.

Makoto


--- ondrej lassak <tehuberpro at gmail.com> wrote:

> I fed the TF analysis single sinusoid 6 and 12Hz and the TF plot
> shows
> multiple specral lines (more than two).
> How can one rely on the TF when it introduces such massive artifacts
> both in
> pure FFT spectrogram and Wavelet scalogram?
> Or am I doing something wrong? When only one freq during the whole
> time span
> is present the TF plots look like really bad moira and the presence
> of the
> freq is apparent only from the summation over time (left from the 
> main
> plot).
> 
> 
> The matlab report with function calls and resulting pictures is
> attached
> below (no scripts embedded in the html).
> > _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
eeglablist-request at sccn.ucsd.edu



From mhartman at egi.com  Wed May 12 15:34:40 2010
From: mhartman at egi.com (Mike Hartman)
Date: Wed, 12 May 2010 15:34:40 -0700
Subject: [Eeglablist] Retreive detailed event info for EGI epoched
	recordings
In-Reply-To: <4beafb87.a225e30a.41c3.2824@mx.google.com>
References: <4beafb87.a225e30a.41c3.2824@mx.google.com>
Message-ID: <005d01caf223$521f98a0$f65ec9e0$@com>

Dear Pavel,

 

the simple binary format only stores the 4-character label of each event,
and the index, of course.The additional components of the events are not
stored.

 

On the EGI website you can get the File Formats manual which explains this
in detail.

 

Using the 'Segmentation Markup' feature in the Net Station Software Waveform
Tools, you can create a new event track in the Net Station file, and have
that track contain new events which are derived from the original events
according to rules that you specify. This is covered in the manuals as well,
but please feel free to contact EGI support for more direction on how to
achieve what you want:

 

supportteam at egi.com

 

PS, depending on which version of Net Station you are using, you can export
the simple binary file as 'epoch-marked'. Then you do not get multiple
simple binary files, only one, and it has events that indicate the epoch
boundaries.

 

Thanks!

 

Michael Hartman

Electrical Geodesics, Inc.

 

From: eeglablist-bounces at sccn.ucsd.edu
[mailto:eeglablist-bounces at sccn.ucsd.edu] On Behalf Of Pawel Augustynowicz
Sent: Wednesday, May 12, 2010 12:04 PM
To: eeglablist at sccn.ucsd.edu
Subject: [Eeglablist] Retreive detailed event info for EGI epoched
recordings

 

Dear All,

 

I have a short question about import data from EGI simple binary file format
with events. By default, EEGLAB imports EGI's simple binary with events, but
in only one dimension. I mean, that I can see only the name of the event.
Additional data (event fields) are not imported. Does anyone have an idea of
how to attach this data do EEGLAB? I assume that this information is written
in additional channels. But EEGLAB deletes all of these channels except one
with the basic info. 

I tried to insert this data manually, by there is another problem. I have a
few epochs in original NetStation recording. These epochs are essential
because of the experiment design (experiment is too long and we have to do
breaks). When I export this data from NetStation as simple binary, this file
is being broken into files representing epochs in original recording. I can
import this bunch of files with pop_readsegegi command. This function does
the job well, except one little bug: NetStation counts time from the
beginning of the experiment. Pauses during the recording are treated as the
beginning of epochs. In EEGLAB all epochs have continuous time, so there is
no possibility to attach events exported from NetStation, because the time
after the first epoch is different. 

 

Does anyone ran on these problems and managed to resolve it?

 

Regards,

Pawel Augustynowicz

The John Paul II Catholic University of Lublin

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100512/cb74ccbe/attachment.html>

From ralphj at rpi.edu  Fri May 14 07:21:14 2010
From: ralphj at rpi.edu (Jason Ralph)
Date: Fri, 14 May 2010 10:21:14 -0400
Subject: [Eeglablist] Retreive detailed event info for EGI epoched
	recordings
In-Reply-To: <4beafb87.a225e30a.41c3.2824@mx.google.com>
References: <4beafb87.a225e30a.41c3.2824@mx.google.com>
Message-ID: <AANLkTimdVFlel6vJHI-F4L2B15UvrGYfDkDCDdI2b7Eh@mail.gmail.com>

We have the same problem.

There are 2 solutions:

*Solution #1*

1. Export .raw file.

2. Create a waveform tool to export events. Select relative time.

3. In Excel (or write a script), convert NetStation's funky time format to
seconds.milliseconds.
4. Import .raw file into EEGLAB
5. Import events into EEGLAB (import events from matlab array or text file)
Your formatted event text file must have the fields type and latency, but
you can add any additional fields desired.

If you need help, we have written scripts to semi-automate this process.


*Solution # 2*

1.  Purchase Amp Server Pro SDK from EGI for $8,000.  ASPSDK allows you to
capture EGI's data without using NetStation
2.  Acquire data directly into BCI2000 or Matlab  itself.
3.  Never use NetStation again!!!!


We've been using solution 1 for some time, but are seriously considering
solution 2.


2010/5/12 Pawe? Augustynowicz <pawel.augustynowicz at gmail.com>

>  Dear All,
>
>
>
> I have a short question about import data from EGI simple binary file
> format with events. By default, EEGLAB imports EGI's simple binary with
> events, but in only one dimension. I mean, that I can see only the name of
> the event. Additional data (event fields) are not imported. Does anyone have
> an idea of how to attach this data do EEGLAB? I assume that this information
> is written in additional channels. But EEGLAB deletes all of these channels
> except one with the basic info.
>
> I tried to insert this data manually, by there is another problem. I have a
> few epochs in original NetStation recording. These epochs are essential
> because of the experiment design (experiment is too long and we have to do
> breaks). When I export this data from NetStation as simple binary, this file
> is being broken into files representing epochs in original recording. I can
> import this bunch of files with pop_readsegegi command. This function does
> the job well, except one little bug: NetStation counts time from the
> beginning of the experiment. Pauses during the recording are treated as the
> beginning of epochs. In EEGLAB all epochs have continuous time, so there is
> no possibility to attach events exported from NetStation, because the time
> after the first epoch is different.
>
>
>
> Does anyone ran on these problems and managed to resolve it?
>
>
>
> Regards,
>
> Pawel Augustynowicz
>
> The John Paul II Catholic University of Lublin
>
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Jason Ralph
CogWorks Laboratory
Cognitive Science Department
Rensselaer Polytechnic Institute
110 8th Street, Troy, NY 12180
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100514/f82121cc/attachment.html>

From j.martinovic at liverpool.ac.uk  Mon May 17 08:03:34 2010
From: j.martinovic at liverpool.ac.uk (Jasna Martinovic)
Date: Mon, 17 May 2010 16:03:34 +0100
Subject: [Eeglablist] a few issues with electrode locations
Message-ID: <4BF15AC6.80000@liverpool.ac.uk>

hi,

I'm trying to reject bad channels and interpolate them using EEGlab.

I first reject the bad channels using 'automatic channel rejection', 
then I end up with a lower number of channels which I subsequently 
interpolate by using the locations from the previous dataset. Is this 
the correct way to do this?

My problem is that once this is done, the new interpolated channels are 
added to the top of the channel structure - say, if I interpolated 5 
channels then 1-5 will now be these new channels. I would like to 
maintain the original structure of channels in all files, as having 
different order between participants will create a right mess. Is there 
a way to do channel interpolation while avoiding this problem?

Also, I want to include the locations for the four eye channels. I have 
found in the EEGlab archive that Andreas Widmann has suggested to use 
the Lo1, Lo2, So1 and So2 locations provided in a file from Robert 
Oostenveld's blog. But the link doesn't work any more as the blog's been 
reorganised. Could someone please re-post an updated link or post the 
locations for these 4 channels?

Thanks in advance for your help!

Kind regards,
Jasna

----
Dr Jasna Martinovic
School of Psychology
University of Aberdeen
William Guild Building
Aberdeen
AB24 2UB

tel: 01224 272240
email: j.martinovic @ abdn.ac.uk
web: http://www.abdn.ac.uk/~psy527/dept/






From jdien07 at mac.com  Mon May 17 10:08:54 2010
From: jdien07 at mac.com (Joseph Dien)
Date: Mon, 17 May 2010 13:08:54 -0400
Subject: [Eeglablist] Lab Tech Position at the University of Maryland
Message-ID: <135539290138500324497685953590578557226-Webmail@me.com>


Position: Faculty Research Assistant 

Position Number: 115823

The University of Maryland Center for Advanced Study of Language (CASL) is seeking a Faculty Research Assistant (FRA) to join the CASL research labs team.  CASL research labs utilize several multimodal research technologies in a highly collaborative and interdisciplinary environment. This includes technologies such as: functional magnetic resonance imaging (fMRI); electroencephalography (EEG); eye tracking; bioneurofeedback (BNB); functional Near-infrared spectrography (fNIRS) and Magneto-encephalography (MEG).

Duties: The FRA will participate in designing research lab protocols utilizing lab equipment, as well as collecting, storing, processing, and analyzing multi-modal research data.  Duties will also include working with research teams to develop experimental and IRB protocols, stimulus materials, study design, and testing procedures as well as providing other research technical expertise to researchers utilizing lab technologies.  Additional duties will include researching and disseminating research technology resources, developing policies and procedures for use of lab technologies, and serving as the labs liaison to the IT department for coordination of specialized lab technology needs.  

Qualifications: Bachelors degree required. Master?s degree preferred with a concentration in Cognitive Neuroscience or a related field, such as Psychology, Cognitive Science, Linguistics, Second Language Acquisition, or Neuroscience. Previous research experience, especially with one or more neuroimaging techniques, and proficiency with at least one experimental design software package (i.e. E-Prime, Direct RT, Presentation, etc.) is required.  Experience with neuroimaging analysis streams (SPM, Brain Voyager, Mat lab) is strongly preferred. Programming skills utilizing C++, JavaScript or other programming languages is a plus. Candidates must be willing to be trained to perform all requisite duties of the position and demonstrate an ability to work in a fast paced, interdisciplinary and demanding environment. 

Candidates must hold U.S. citizenship and be willing to obtain the appropriate security clearance. 

CASL, established in 2003, is the nation?s 10th university-affiliated research center. Its mission is to conduct state-of-the-science research that results in improved performance on language tasks relevant to the work of government language professionals. Our research focuses on improving knowledge of less commonly taught languages; enhancing the acquisition and maintenance of foreign language capability by government professionals; advancing the capacity to use foreign language skills in a government professions; and improving the quality of human language technology.

Application: For earliest consideration, apply online and submit a cover letter, resume or CV, and up to three (3) references and writing samples https://jobs.umd.edu/applicants/jsp/shared/position/JobDetails_css.jsp?postingId=138215

The University of Maryland is an affirmative action, equal opportunity employer. Women and minorities are encouraged to apply.




From j.martinovic at liverpool.ac.uk  Wed May 19 06:44:10 2010
From: j.martinovic at liverpool.ac.uk (Jasna Martinovic)
Date: Wed, 19 May 2010 14:44:10 +0100
Subject: [Eeglablist] postdoctoral position - EEG studies of visual object
 representation (Leipzig, Germany)
Message-ID: <4BF3EB2A.4020502@liverpool.ac.uk>

Postdoctoral position at the Institute of Psychology of the University 
of Leipzig, Germany
EEG lab of Prof Dr Matthias M. Mueller


The Institute of Psychology at the University of Leipzig is seeking to 
appoint a full-time Research Associate to work on a DFG-funded project 
investigating the neural basis of visual object representation awarded 
to Prof Matthias M. M?ller (University of Leipzig) in collaboration with 
Dr Jasna Martinovic (University of Aberdeen). The employment will be 
initially awarded for three years with the possibility of extension, 
with the salary in the postdoctoral range that is typical for positions 
of that level in Germany. The actual salary will be commensurate with 
the applicant?s qualifications and previous work experience.

The appointee should be able to conduct independent research into visual 
object representation using EEG and behavioural methods. The University 
of Leipzig has its own 128-channel EEG laboratory and a fine record in 
investigating brain responses related to visual attention and object 
representation. The lab focuses on the study of temporal dynamics of 
integrative brain activity which are explored with a range of EEG 
responses such as the event-related potentials, steady-state evoked 
potentials and event-related high-frequency oscillatory activity.

Candidates should have a background in Psychology, Biology or 
Neuroscience and hold a PhD or be close to completion of a PhD. Previous 
neuroimaging experience is preferred ? in particular EEG or MEG. 
Familiarity with the use of Matlab for data analysis and experimental 
design would be advantageous, as would be previous experience in vision 
science. Furthermore, excellent writing and presentation skills in 
English would be required. If the candidate would show interest in doing 
a habilitation at the Institute, a limited amount of teaching would also 
be made available.

Applications with a letter of interest, CV, publication list and details 
of two referees should be sent to Prof M.M.M?ller at 
m.mueller at rz.uni-leipzig.de. Informal enquiries can be made to either 
Prof M.M.Mueller or to Dr Jasna Martinovic (email: jasnam at liv.ac.uk). 
The deadline for applications is 5th of June 2010.



From uwe.graichen at tu-ilmenau.de  Wed May 19 00:33:07 2010
From: uwe.graichen at tu-ilmenau.de (Uwe Graichen)
Date: Wed, 19 May 2010 09:33:07 +0200
Subject: [Eeglablist] 5th International Summerschool in Biomedical
	Engineering
In-Reply-To: <3b0db4861001080043tf65b84k6c3ffa195c523311@mail.gmail.com>
References: <3b0db4861001080043tf65b84k6c3ffa195c523311@mail.gmail.com>
Message-ID: <4BF39433.9060604@tu-ilmenau.de>

Dear colleagues,

we are pleased to announce the

5th International Summerschool in Biomedical Engineering
on "Multimodal integration of functional brain measurements"

The summerschool will be held from August 18-25, 2010, in 
Wittenberg/Germany.
For more information, please refer to our website:

http://www.tu-ilmenau.de/fakia/Summerschool-2010.summerschool-20090.0.html

We would be happy to receive your application. Please pass this 
anouncement to your colleagues who might be interested.

Sincerely,

Jens Haueisen and Thomas Kn?sche


From Martin.Schecklmann at medbo.de  Wed May 19 07:29:14 2010
From: Martin.Schecklmann at medbo.de (Schecklmann Martin)
Date: Wed, 19 May 2010 16:29:14 +0200
Subject: [Eeglablist] out of memory
Message-ID: <740C7575D2240A46AD365B1DDD27AF724F3EDE9FB3@bkrexmb3>

Dear colleagues,

I'm a new user of EEGLAB and I think I have some problems with the RAM. I'm working with MatLab 7.10.0 (R2010a) and EEGLAB v8.0.3.5b. I'm using Windows XP with 3.21 GB RAM. When i want to load a dataset of a colleague in set file format (set file 5 MB und the fdt file with 400 MB) i get the message "out of memory" in a "fread Out error" window. The "memory" command in MatLab shows the following informations:
_______________________________________________________

Maximum possible array:              83 MB (8.720e+007 bytes) *
Memory available for all arrays:    510 MB (5.343e+008 bytes) **
Memory used by MATLAB:              288 MB (3.019e+008 bytes)
Physical Memory (RAM):             3292 MB (3.451e+009 bytes)

*  Limited by contiguous virtual address space available.
** Limited by virtual address space available.
_______________________________________________________

It looks that MatLab gets not enough memory. I already used older EEGLAB version, but that do not work as well. Did you have any advice for me?

Thanks a lot,

best regards,

Martin



________________________________
Medizinische Einrichtungen des Bezirks Oberpfalz GmbH (www.medbo.de)
Gesch?ftsf?hrer: Kurt H?upl
Aufsichtsratvorsitzender: Bezirkstagspr?sident Franz L?ffler
Sitz: Universit?tsstra?e 84, 93053 Regensburg
Registergericht: Regensburg HR-Nr.: B9977
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100519/b97997f2/attachment.html>

From Sebastian.Korb at unige.ch  Thu May 20 03:33:21 2010
From: Sebastian.Korb at unige.ch (seba)
Date: Thu, 20 May 2010 12:33:21 +0200
Subject: [Eeglablist] DC shifts in biosemi files: first do a baseline
 correction and then a high pass filter
Message-ID: <E7C9AB38-EAB3-4644-95CE-210707B97B59@unige.ch>

Dear all,

as many of you know and have discussed on this list before (see for example post by Brad, september 12th, 2008),
Biosemi .bdf files look very weird when you open them for the first time in Matlab (with or without Eeglab).
In fact the amplitudes are completely shifted and out of range ("because BioSemi uses some of its bit depth to record the DC offsets", as Brad says).
And the solution for this problem was said to be simple: just apply a high-pass filter (e.g. at 0.1 Hz) and everything gets back into the normal range again.

However, as I found out, if you do so for example with a butterworth filter, you get a huge artifact which forces you to throw away parts of your data in the beginning.
Actually, when the amplitude shift is huge, and even more so if I use a filter with a high order, it looks like the ENTIRE EEG (!!?) is polluted by this artificially introduced artifact.

Thus, I would like to remind all people working with Biosemi files to FIRST do a baseline correction, and only then, apply their filters.
Doing so does not (at least on my data) result in this filter artifact.

Just so that you see it yourself, you can try this little script in Matlab, 
in which I just construct and then shift an artificial signal, and where the signal becomes really ugly after filtering:


%%%%%%%%%%%%%%%
clear all
clc

x=-10*pi:0.001:10*pi;
y=sin(x)+1000000; %introduce an amplitude shift

%%filter with a 5th order butter. In this case I the sampling frequency is
%%defined as 1024
[z,p,k] = butter(10, .1/(1024/2), 'high'); 
[sos, g] = zp2sos(z,p,k);
h1 = dfilt.df2sos(sos,g);
hfvt = fvtool(h1, 'FrequencyScale', 'log'); %plot
eeg_filt = filter(h1, y(:,:)); %apply the filter

%plot the 
figure; subplot(2,1,1);
plot(y(1:end));
subplot(2,1,2);
plot(eeg_filt(1:end));



Sebastian Korb, Ph.D. Student
Swiss Center for Affective Sciences
University of Geneva, Switzerland
Tel.: +41223799812
Email: Sebastian.Korb at unige.ch
http://www.affective-sciences.org/user/34








From b.raven at student.tue.nl  Thu May 20 05:00:59 2010
From: b.raven at student.tue.nl (Raven, B.)
Date: Thu, 20 May 2010 14:00:59 +0200
Subject: [Eeglablist] LORETA in EEGlab
Message-ID: <2C38ED72C6FB82459F1A2B7060A537076F8A32C2C5@EXCHANGE13.campus.tue.nl>

Hello EEGlab users,

I'm doing an internship about understanding LORETA better and to know how I should implement it. I understand there is a plugin module for EEGlab that can perform LORETA on measurements. Can anyone explain to me how I can do this? What I have done so far:

1) read data (cnt-file) into EEGlab
2) perform filtering (1-50Hz and baseline removal)
3) perform ICA

Then I want to use the LORETA module, but I can't click anything in this module. Does anyone know how I can do this?

Kind regards
Boudewijn




From Nikhil.Divekar at uct.ac.za  Thu May 20 05:21:50 2010
From: Nikhil.Divekar at uct.ac.za (Nikhil DIVEKAR)
Date: Thu, 20 May 2010 14:21:50 +0200
Subject: [Eeglablist] newcrossf.m function for calculation of
	corticomuscular	 coherence
Message-ID: <4BF5457E0200006A00131B15@gwiasmtp.uct.ac.za>

Hi
 
I'm running a cortico-muscular coherence analysis using the EEGLAB function 'newcrossf' and am unsure about the specific parameter settings (such as 'type' etc). 
 
Is there a reference article that I can refer to or can anyone advise me on suitable settings or preferred functions for cortico-muscular coherence analyses ?


 

###
UNIVERSITY OF CAPE TOWN 

This e-mail is subject to the UCT ICT policies and e-mail disclaimer published on our website at http://www.uct.ac.za/about/policies/emaildisclaimer/ or obtainable from +27 21 650 4500. This e-mail is intended only for the person(s) to whom it is addressed. If the e-mail has reached you in error, please notify the author. If you are not the intended recipient of the e-mail you may not use, disclose, copy, redirect or print the content. If this e-mail is not related to the business of UCT it is sent by the sender in the sender's individual capacity.

###
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100520/6e0786d1/attachment.html>

From widmann at uni-leipzig.de  Thu May 20 07:34:34 2010
From: widmann at uni-leipzig.de (Andreas Widmann)
Date: Thu, 20 May 2010 16:34:34 +0200
Subject: [Eeglablist] a few issues with electrode locations
In-Reply-To: <4BF15AC6.80000@liverpool.ac.uk>
References: <4BF15AC6.80000@liverpool.ac.uk>
Message-ID: <C1F1400A-D5D5-48C6-BA0B-70105AFE4BA6@uni-leipzig.de>

Hi Jasna,

> My problem is that once this is done, the new interpolated channels are 
> added to the top of the channel structure
Is it necessary to remove the channels? Not sure about EEGLAB built-in channel interpolation but in case spherical spline interpolation is an option you could use sphspline plugin, where you can just replace the channels without prior removing them. Marking bad channels and replacing should be possible in two script lines.

> Also, I want to include the locations for the four eye channels. I have 
> found in the EEGlab archive that Andreas Widmann has suggested to use 
> the Lo1, Lo2, So1 and So2 locations provided in a file from Robert 
> Oostenveld's blog. But the link doesn't work any more as the blog's been 
> reorganised. Could someone please re-post an updated link or post the 
> locations for these 4 channels?
           7         LO1         -42     0.65556     0.65616     0.59081    -0.46947          42         -28           1
           8         LO2          42     0.65556     0.65616    -0.59081    -0.46947         -42         -28           1
           9         SO1         -25     0.58333     0.87543     0.40822    -0.25882          25         -15           1
          10         SO2          25     0.58333     0.87543    -0.40822    -0.25882         -25         -15           1
          11         IO1         -27     0.69444     0.72987     0.37189    -0.57358          27         -35           1
          12         IO2          27     0.69444     0.72987    -0.37189    -0.57358         -27         -35           1

Best,
Andreas


From pvraj011 at gmail.com  Thu May 20 20:14:19 2010
From: pvraj011 at gmail.com (Partha Rajendra)
Date: Thu, 20 May 2010 23:14:19 -0400
Subject: [Eeglablist] Question Regarding Setting up multiple ASCII files for
	analysis
Message-ID: <AANLkTimsvughls7Ycc_UZ7wqULEAIbGuPDoaGAsCwQvS@mail.gmail.com>

Dear EEGLAB users,

My name is Partha Rajendra and I am a high school student working on
creating a seizure prediction algorithm based on my analysis of the
University of Freiburg seizure database. I have no programming experience
but I have started working on MATLAB tutorials and will be getting help from
a software engineer with MATLAB experience. My question is very basic, and
concerns getting the data into a form I can use for my analysis.

I am able to make individual plots of the raw EEG .asc files in EEGLAB, but
am not able to create an EEG montage with more than 1 single plot of EEG
data. I have been looking through an EEGLAB tutorial pdf, but my problem
seems to be more of letting MATLAB/EEGLAB know I am not looking at multiple
different sets, but trying to combine them together for my analysis. In
short, I cannot create a study with multiple EEGs displayed, only individual
plots. I believe this is because I have not set the channel locations, but I
have had some trouble setting this up.

If anyone can help me with setting up the data for analysis, it would be
greatly appreciated. Thanks.

Best Regards,
Partha
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100520/44263f18/attachment.html>

From ande2523 at umn.edu  Mon May 24 10:34:50 2010
From: ande2523 at umn.edu (Jake Anderson)
Date: 24 May 2010 12:34:50 -0500
Subject: [Eeglablist] DC shifts in biosemi files: first do a baseline
	correction and then a high pass filter
In-Reply-To: <E7C9AB38-EAB3-4644-95CE-210707B97B59@unige.ch>
References: <E7C9AB38-EAB3-4644-95CE-210707B97B59@unige.ch>
Message-ID: <Gophermail.2.0.1005241234500.350@vs-w.tc.umn.edu>

Or... you could just start recording the data file 10-20 seconds before the 
first important event and stop recording 10-20 seconds after the last 
important event, and you'd almost always avoid having any filter artifacts 
anywhere near critical data points. Freshly collected, raw .bdf data is 
continuous, right? So what exactly are you baseline correcting?

*************************************************
Jacob E. Anderson, MA
Junior Scientist
Zelazo Lab
Institute of Child Development
University of Minnesota




On May 22 2010, seba wrote:

>Dear all,
>
> as many of you know and have discussed on this list before (see for 
> example post by Brad, september 12th, 2008),
> Biosemi .bdf files look very weird when you open them for the first time 
> in Matlab (with or without Eeglab).
> In fact the amplitudes are completely shifted and out of range ("because 
> BioSemi uses some of its bit depth to record the DC offsets", as Brad 
> says).
> And the solution for this problem was said to be simple: just apply a 
> high-pass filter (e.g. at 0.1 Hz) and everything gets back into the 
> normal range again.
>
> However, as I found out, if you do so for example with a butterworth 
> filter, you get a huge artifact which forces you to throw away parts of 
> your data in the beginning.
> Actually, when the amplitude shift is huge, and even more so if I use a 
> filter with a high order, it looks like the ENTIRE EEG (!!?) is polluted 
> by this artificially introduced artifact.
>
> Thus, I would like to remind all people working with Biosemi files to 
> FIRST do a baseline correction, and only then, apply their filters.
>Doing so does not (at least on my data) result in this filter artifact.
>
> Just so that you see it yourself, you can try this little script in 
> Matlab,
> in which I just construct and then shift an artificial signal, and where 
> the signal becomes really ugly after filtering:
>
>
>%%%%%%%%%%%%%%%
>clear all
>clc
>
>x=-10*pi:0.001:10*pi;
>y=sin(x)+1000000; %introduce an amplitude shift
>
>%%filter with a 5th order butter. In this case I the sampling frequency is
>%%defined as 1024
>[z,p,k] = butter(10, .1/(1024/2), 'high'); 
>[sos, g] = zp2sos(z,p,k);
>h1 = dfilt.df2sos(sos,g);
>hfvt = fvtool(h1, 'FrequencyScale', 'log'); %plot
>eeg_filt = filter(h1, y(:,:)); %apply the filter
>
>%plot the 
>figure; subplot(2,1,1);
>plot(y(1:end));
>subplot(2,1,2);
>plot(eeg_filt(1:end));
>
>
>
>Sebastian Korb, Ph.D. Student
>Swiss Center for Affective Sciences
>University of Geneva, Switzerland
>Tel.: +41223799812
>Email: Sebastian.Korb at unige.ch
>http://www.affective-sciences.org/user/34




From bradley.voytek at gmail.com  Sun May 23 14:33:08 2010
From: bradley.voytek at gmail.com (Bradley Voytek)
Date: Sun, 23 May 2010 14:33:08 -0700
Subject: [Eeglablist] DC shifts in biosemi files: first do a baseline
	correction and then a high pass filter
In-Reply-To: <E7C9AB38-EAB3-4644-95CE-210707B97B59@unige.ch>
References: <E7C9AB38-EAB3-4644-95CE-210707B97B59@unige.ch>
Message-ID: <AANLkTimBajeJ3txBvtmFy5swLo3Az0Rg2DvfBzBE68tG@mail.gmail.com>

Seba:

You're correct. The best way is to simply remove the (true) DC offset
before filtering. To do this you must subtract the mean channel
voltage from each channel. Make sure you convert the channel datatype
to "double" first, since BDF data are natively in "single" format.

See this correspondence:
http://sccn.ucsd.edu/pipermail/eeglablist/2008/002229.html

::brad

On Thu, May 20, 2010 at 03:33, seba <Sebastian.Korb at unige.ch> wrote:
> Dear all,
>
> as many of you know and have discussed on this list before (see for example post by Brad, september 12th, 2008),
> Biosemi .bdf files look very weird when you open them for the first time in Matlab (with or without Eeglab).
> In fact the amplitudes are completely shifted and out of range ("because BioSemi uses some of its bit depth to record the DC offsets", as Brad says).
> And the solution for this problem was said to be simple: just apply a high-pass filter (e.g. at 0.1 Hz) and everything gets back into the normal range again.
>
> However, as I found out, if you do so for example with a butterworth filter, you get a huge artifact which forces you to throw away parts of your data in the beginning.
> Actually, when the amplitude shift is huge, and even more so if I use a filter with a high order, it looks like the ENTIRE EEG (!!?) is polluted by this artificially introduced artifact.
>
> Thus, I would like to remind all people working with Biosemi files to FIRST do a baseline correction, and only then, apply their filters.
> Doing so does not (at least on my data) result in this filter artifact.
>
> Just so that you see it yourself, you can try this little script in Matlab,
> in which I just construct and then shift an artificial signal, and where the signal becomes really ugly after filtering:
>
>
> %%%%%%%%%%%%%%%
> clear all
> clc
>
> x=-10*pi:0.001:10*pi;
> y=sin(x)+1000000; %introduce an amplitude shift
>
> %%filter with a 5th order butter. In this case I the sampling frequency is
> %%defined as 1024
> [z,p,k] = butter(10, .1/(1024/2), 'high');
> [sos, g] = zp2sos(z,p,k);
> h1 = dfilt.df2sos(sos,g);
> hfvt = fvtool(h1, 'FrequencyScale', 'log'); %plot
> eeg_filt = filter(h1, y(:,:)); %apply the filter
>
> %plot the
> figure; subplot(2,1,1);
> plot(y(1:end));
> subplot(2,1,2);
> plot(eeg_filt(1:end));
>
>
>
> Sebastian Korb, Ph.D. Student
> Swiss Center for Affective Sciences
> University of Geneva, Switzerland
> Tel.: +41223799812
> Email: Sebastian.Korb at unige.ch
> http://www.affective-sciences.org/user/34
>
>
>
>
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
>


From balazs at cogpsyphy.hu  Tue May 25 15:21:24 2010
From: balazs at cogpsyphy.hu (=?ISO-8859-1?Q?Bal=E1zs_L=E1szl=F3?=)
Date: Wed, 26 May 2010 00:21:24 +0200
Subject: [Eeglablist] IOP conference - extended deadline for submissions
Message-ID: <4BFC4D64.2060908@cogpsyphy.hu>

Dear EEGLABlisters,
Deadline for submissions to the 15th World Congress of Pschophysiology 
(IOP 2010) is extended until the end of May. Please read the information 
in attachment. Looking forward to see you in Budapest.
Kind regards,
Laszlo

-- 
      Laszlo Balazs, Ph.D.          /      dr. Bal?zs L?szl?
 Institute for Psychology HAS       /   MTA Pszichol?giai Kutat?int?zet
                 P O B 398, Budapest, Hungary, H-1394
                Tel:+36(1)354-2410 | Fax:+36(1)354-2416
                      http://www.cogpsyphy.hu/balazs




__________ Information from ESET NOD32 Antivirus, version of virus signature database 5133 (20100520) __________

The message was checked by ESET NOD32 Antivirus.

http://www.eset.com

-------------- next part --------------
A non-text attachment was scrubbed...
Name: IOP 2010 invitation_extended deadline for submission.pdf
Type: application/octet-stream
Size: 55609 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100526/990fc3bb/attachment.obj>

From hecke at nld.ds.mpg.de  Tue May 25 03:46:52 2010
From: hecke at nld.ds.mpg.de (hecke)
Date: Tue, 25 May 2010 12:46:52 +0200
Subject: [Eeglablist] =?iso-8859-1?q?Fall_Course_on_Computational_Neurosci?=
 =?iso-8859-1?q?ence_in_G=F6ttingen=2C_Germany?=
Message-ID: <4BFBAA9C.1010701@nld.ds.mpg.de>

Applications are invited for the eighth fall course on

           COMPUTATIONAL NEUROSCIENCE in G?ttingen, Germany

                     September 20th - 24th, 2010

                                    organized by Hecke Schrobsdorff


The course is intended to provide graduate students and young
researchers from all parts of neuroscience with working knowledge of
theoretical and computational methods in neuroscience and to acquaint
them with recent developments in this field.

The course includes tutorials and lectures of the following researchers:

    * Sophie Deneve, Ecole Normale Superieur, Paris
    * Jeremy Niven, Dean Selwyn College, Cambridge
    * Hansj?rg Scherberger, German Primate Center, G?ttingen
    * Elad Schneidman, Weizmann Institute of Science, Rehovot
    * Susanne Still, University of Hawaii

The course takes place at the Department of Nonlinear Dynamics of the
Max-Planck Institute for Dynamics and Self-Organization, Bunsenstr. 10,
 D-37073 G?ttingen. A course fee of 100 Euro includes participation in
the tutorials, study materials, and part of the social events. The
number of participants is limited to about 30. Course language is English.

To apply please fill out the application form at:
http://www.bccn-goettingen.de/events-1/cns-course
by August 10, 2010.

Best wishes and looking forward to seeing you in G?ttingen

Hecke


From Garg-G at email.ulster.ac.uk  Mon May 24 03:06:41 2010
From: Garg-G at email.ulster.ac.uk (Gaurav Garg)
Date: Mon, 24 May 2010 10:06:41 +0000
Subject: [Eeglablist] out of memory
In-Reply-To: <740C7575D2240A46AD365B1DDD27AF724F3EDE9FB3@bkrexmb3>
References: <740C7575D2240A46AD365B1DDD27AF724F3EDE9FB3@bkrexmb3>
Message-ID: <4A33A2B5BDE2124DA87615302AA5F2E3129B37BB@DB2PRD0103MB043.eurprd01.prod.exchangelabs.com>

Dear Dr. Martin,

I have tried many things but I think the only solution which worked best for me is to move to 64-bit OS with supporting 64-Bit MATLAB as well.

Other solutions are generally, splitting that data in smaller frames/files then using it for processing and then after all the processing with all the frames you need to merge them may help you, also downsampling (e.g. 200Hz) to less number  of data points can reduce the memory requirement as less number of data points will be left there but this step will reduce the resolution as well.

Hope this would be able to solve your problem.

Thanks,

Best Regards,
Gaurav Garg



==================================================
Postgraduate Research Scholar
Room No. MS125
School of Computing and Intelligent Systems
University of Ulster
Magee Campus
Northland Road
Derry
Northern Ireland BT48 7JL



Tel(office): +442871371207
Email: gauravgarg4 at gmail.com<mailto:gauravgarg4 at gmail.com>

________________________________
From: Schecklmann Martin [Martin.Schecklmann at medbo.de]
Sent: 19 May 2010 15:29
To: 'eeglablist at sccn.ucsd.edu'
Subject: [Eeglablist] out of memory

Dear colleagues,

I'm a new user of EEGLAB and I think I have some problems with the RAM. I'm working with MatLab 7.10.0 (R2010a) and EEGLAB v8.0.3.5b. I'm using Windows XP with 3.21 GB RAM. When i want to load a dataset of a colleague in set file format (set file 5 MB und the fdt file with 400 MB) i get the message "out of memory" in a "fread Out error" window. The "memory" command in MatLab shows the following informations:
_______________________________________________________

Maximum possible array:              83 MB (8.720e+007 bytes) *
Memory available for all arrays:    510 MB (5.343e+008 bytes) **
Memory used by MATLAB:              288 MB (3.019e+008 bytes)
Physical Memory (RAM):             3292 MB (3.451e+009 bytes)

*  Limited by contiguous virtual address space available.
** Limited by virtual address space available.
_______________________________________________________

It looks that MatLab gets not enough memory. I already used older EEGLAB version, but that do not work as well. Did you have any advice for me?

Thanks a lot,

best regards,

Martin



________________________________
Medizinische Einrichtungen des Bezirks Oberpfalz GmbH (www.medbo.de)
Gesch?ftsf?hrer: Kurt H?upl
Aufsichtsratvorsitzender: Bezirkstagspr?sident Franz L?ffler
Sitz: Universit?tsstra?e 84, 93053 Regensburg
Registergericht: Regensburg HR-Nr.: B9977
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100524/c4fc042f/attachment.html>

From achim.andre at uqam.ca  Wed May 26 11:44:31 2010
From: achim.andre at uqam.ca (=?iso-8859-1?Q?Achim=2C_Andr=E9?=)
Date: Wed, 26 May 2010 14:44:31 -0400
Subject: [Eeglablist] DC shifts in biosemi files: first do a
	baselinecorrection and then a high pass filter
In-Reply-To: <AANLkTimBajeJ3txBvtmFy5swLo3Az0Rg2DvfBzBE68tG@mail.gmail.com>
References: <E7C9AB38-EAB3-4644-95CE-210707B97B59@unige.ch>
	<AANLkTimBajeJ3txBvtmFy5swLo3Az0Rg2DvfBzBE68tG@mail.gmail.com>
Message-ID: <B3C24EA955FF0C4EA14658997CD3E25E3D2FD386@CAHIER.gst.uqam.ca>

Note that, for this purpose, it is not necessary to calculate the exact mean of each channel across all the recording. Estimating the channel means over 10-20 seconds should do well enough. The high-pass filter will do the rest.

Andr? Achim

-----Message d'origine-----
De?: eeglablist-bounces at sccn.ucsd.edu [mailto:eeglablist-bounces at sccn.ucsd.edu] De la part de Bradley Voytek
Envoy??: 23 mai 2010 17:33
??: seba
Cc?: eeglablist
Objet?: Re: [Eeglablist] DC shifts in biosemi files: first do a baselinecorrection and then a high pass filter

Seba:

You're correct. The best way is to simply remove the (true) DC offset
before filtering. To do this you must subtract the mean channel
voltage from each channel. Make sure you convert the channel datatype
to "double" first, since BDF data are natively in "single" format.

See this correspondence:
http://sccn.ucsd.edu/pipermail/eeglablist/2008/002229.html

::brad

On Thu, May 20, 2010 at 03:33, seba <Sebastian.Korb at unige.ch> wrote:
> Dear all,
>
> as many of you know and have discussed on this list before (see for example post by Brad, september 12th, 2008),
> Biosemi .bdf files look very weird when you open them for the first time in Matlab (with or without Eeglab).
> In fact the amplitudes are completely shifted and out of range ("because BioSemi uses some of its bit depth to record the DC offsets", as Brad says).
> And the solution for this problem was said to be simple: just apply a high-pass filter (e.g. at 0.1 Hz) and everything gets back into the normal range again.
>
> However, as I found out, if you do so for example with a butterworth filter, you get a huge artifact which forces you to throw away parts of your data in the beginning.
> Actually, when the amplitude shift is huge, and even more so if I use a filter with a high order, it looks like the ENTIRE EEG (!!?) is polluted by this artificially introduced artifact.
>
> Thus, I would like to remind all people working with Biosemi files to FIRST do a baseline correction, and only then, apply their filters.
> Doing so does not (at least on my data) result in this filter artifact.
>
> Just so that you see it yourself, you can try this little script in Matlab,
> in which I just construct and then shift an artificial signal, and where the signal becomes really ugly after filtering:
>
>
> %%%%%%%%%%%%%%%
> clear all
> clc
>
> x=-10*pi:0.001:10*pi;
> y=sin(x)+1000000; %introduce an amplitude shift
>
> %%filter with a 5th order butter. In this case I the sampling frequency is
> %%defined as 1024
> [z,p,k] = butter(10, .1/(1024/2), 'high');
> [sos, g] = zp2sos(z,p,k);
> h1 = dfilt.df2sos(sos,g);
> hfvt = fvtool(h1, 'FrequencyScale', 'log'); %plot
> eeg_filt = filter(h1, y(:,:)); %apply the filter
>
> %plot the
> figure; subplot(2,1,1);
> plot(y(1:end));
> subplot(2,1,2);
> plot(eeg_filt(1:end));
>
>
>
> Sebastian Korb, Ph.D. Student
> Swiss Center for Affective Sciences
> University of Geneva, Switzerland
> Tel.: +41223799812
> Email: Sebastian.Korb at unige.ch
> http://www.affective-sciences.org/user/34
>
>
>
>
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
>
_______________________________________________
Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu



From drampsych at gmail.com  Wed May 26 12:31:38 2010
From: drampsych at gmail.com (Alan McAllister)
Date: Wed, 26 May 2010 15:31:38 -0400
Subject: [Eeglablist] Grass Amplifier Model 15A94
Message-ID: <AANLkTilL1dwl2uAzGohE7wLWkEEY1QAxrKInZsltHfst@mail.gmail.com>

Does anyone out there use the Grass Amplifier Model 15A94?  I am
interested in discussing how to set it up.

Alan McAllister


From arno at ucsd.edu  Thu May 27 22:51:09 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Thu, 27 May 2010 22:51:09 -0700
Subject: [Eeglablist] DC shifts in biosemi files: first do a
	baselinecorrection and then a high pass filter
In-Reply-To: <B3C24EA955FF0C4EA14658997CD3E25E3D2FD386@CAHIER.gst.uqam.ca>
References: <E7C9AB38-EAB3-4644-95CE-210707B97B59@unige.ch>
	<AANLkTimBajeJ3txBvtmFy5swLo3Az0Rg2DvfBzBE68tG@mail.gmail.com>
	<B3C24EA955FF0C4EA14658997CD3E25E3D2FD386@CAHIER.gst.uqam.ca>
Message-ID: <FC8BB192-7CDA-491F-9F9A-C4D3588BE2E5@ucsd.edu>

Dear all,

yes, high pass filtering is best. Ideally, you can in addition to  
removing the mean also detrend your continuous data prior to filtering  
("EEG.data = detrend(EEG.data')';" (this will remove any linear trend  
in each of the channels).
Also, it is important to be careful with filtering. EEGLAB function  
perform by default acausal filtering that preserve the phase  
information (using the filtfilt function that filters forward in time  
and then backward in time). However, this can make very low frequency  
components leak into the baseline period for event-related data. To  
fix this problem, there is now a new option for the non-linear  
filering function iirfilt.m to use causal filtering (filtering only  
forward in time). The phase absolute information can be distorted  
though. It is a trade-off and it will depend on what you want to do  
with your data. There is no perfect solution to my knowledge.

Arno

On May 26, 2010, at 11:44 AM, Achim, Andr? wrote:

> Note that, for this purpose, it is not necessary to calculate the  
> exact mean of each channel across all the recording. Estimating the  
> channel means over 10-20 seconds should do well enough. The high- 
> pass filter will do the rest.
>
> Andr? Achim
>
> -----Message d'origine-----
> De : eeglablist-bounces at sccn.ucsd.edu [mailto:eeglablist-bounces at sccn.ucsd.edu 
> ] De la part de Bradley Voytek
> Envoy? : 23 mai 2010 17:33
> ? : seba
> Cc : eeglablist
> Objet : Re: [Eeglablist] DC shifts in biosemi files: first do a  
> baselinecorrection and then a high pass filter
>
> Seba:
>
> You're correct. The best way is to simply remove the (true) DC offset
> before filtering. To do this you must subtract the mean channel
> voltage from each channel. Make sure you convert the channel datatype
> to "double" first, since BDF data are natively in "single" format.
>
> See this correspondence:
> http://sccn.ucsd.edu/pipermail/eeglablist/2008/002229.html
>
> ::brad
>
> On Thu, May 20, 2010 at 03:33, seba <Sebastian.Korb at unige.ch> wrote:
>> Dear all,
>>
>> as many of you know and have discussed on this list before (see for  
>> example post by Brad, september 12th, 2008),
>> Biosemi .bdf files look very weird when you open them for the first  
>> time in Matlab (with or without Eeglab).
>> In fact the amplitudes are completely shifted and out of range  
>> ("because BioSemi uses some of its bit depth to record the DC  
>> offsets", as Brad says).
>> And the solution for this problem was said to be simple: just apply  
>> a high-pass filter (e.g. at 0.1 Hz) and everything gets back into  
>> the normal range again.
>>
>> However, as I found out, if you do so for example with a  
>> butterworth filter, you get a huge artifact which forces you to  
>> throw away parts of your data in the beginning.
>> Actually, when the amplitude shift is huge, and even more so if I  
>> use a filter with a high order, it looks like the ENTIRE EEG (!!?)  
>> is polluted by this artificially introduced artifact.
>>
>> Thus, I would like to remind all people working with Biosemi files  
>> to FIRST do a baseline correction, and only then, apply their  
>> filters.
>> Doing so does not (at least on my data) result in this filter  
>> artifact.
>>
>> Just so that you see it yourself, you can try this little script in  
>> Matlab,
>> in which I just construct and then shift an artificial signal, and  
>> where the signal becomes really ugly after filtering:
>>
>>
>> %%%%%%%%%%%%%%%
>> clear all
>> clc
>>
>> x=-10*pi:0.001:10*pi;
>> y=sin(x)+1000000; %introduce an amplitude shift
>>
>> %%filter with a 5th order butter. In this case I the sampling  
>> frequency is
>> %%defined as 1024
>> [z,p,k] = butter(10, .1/(1024/2), 'high');
>> [sos, g] = zp2sos(z,p,k);
>> h1 = dfilt.df2sos(sos,g);
>> hfvt = fvtool(h1, 'FrequencyScale', 'log'); %plot
>> eeg_filt = filter(h1, y(:,:)); %apply the filter
>>
>> %plot the
>> figure; subplot(2,1,1);
>> plot(y(1:end));
>> subplot(2,1,2);
>> plot(eeg_filt(1:end));
>>
>>
>>
>> Sebastian Korb, Ph.D. Student
>> Swiss Center for Affective Sciences
>> University of Geneva, Switzerland
>> Tel.: +41223799812
>> Email: Sebastian.Korb at unige.ch
>> http://www.affective-sciences.org/user/34
>>
>>
>>
>>
>>
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime"  
>> to eeglablist-request at sccn.ucsd.edu
>>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu




From brian.murphy at unitn.it  Thu May 27 05:45:50 2010
From: brian.murphy at unitn.it (Brian Murphy)
Date: Thu, 27 May 2010 14:45:50 +0200
Subject: [Eeglablist] PhD position in Computational Neurolinguistics,
 U of Trento, deadline 8th June
In-Reply-To: <4BC45A2D.603@unitn.it>
References: <4BC45A2D.603@unitn.it>
Message-ID: <4BFE697E.5090204@unitn.it>

[apologies for any cross-postings]

Please see the posting below with details of a fully funded PhD position 
in computational neurolinguistics at the Centre for Mind/Brain Sciences, 
University of Trento, Italy.

This is a challenging interdisciplinary project, using machine learning 
methods to integrate data from neural recordings, from web-scale 
language corpora, and from behavioural experiments, and so to arrive at 
a better understanding of word semantics. The project may involve 
co-operation with international partners (Carnegie Mellon U, Cambridge 
University, Tokyo Tech).

Programming ability is essential, and further experience with one or 
more of the following desirable:
    - theoretical linguistics/semantics
    - psychology of concepts/semantic memory
    - computational linguistics
    - computational neuroscience
    - signal processing
    - neuroimaging technologies (EEG, MEG, fMRI)
    - experimental psychology

The deadline for applications is the 8th of June. CIMeC is an 
interdisciplinary centre for neurocognitive research led by Alfonso 
Caramazza. The centre is located in Rovereto, in the heart of the 
Italian Dolomite Alps. The scholarship includes full fees and living 
allowance.

-------------------

PHD POSITION IN COMPUTATIONAL NEUROLINGUISTICS

A PhD position in Computational Neurolinguistics (Conceptual
Representations) will be available at the Language Interaction and
Computation Lab (CLIC)  starting in late 2010 or early 2011.
      http://clic.cimec.unitn.it/

The successful candidate will work as part of a larger project whose
objective is to combine empirical data of different types (corpus
co-occurrence patterns, elicitation experiments, neuroimaging data) to
arrive at a better understanding of the organisation of conceptual
knowledge in the mind and brain. Your task will be to continue ongoing
work which uses machine learning and data-mining methods to extract
conceptual representations from recordings of neural activity (EEG,
MEG and fMRI). Ideally the candidate should have knowledge of both
computational linguistic techniques (particularly distributional
models of meaning), and experimental design (elicitation/behavioural).
Programming skills are a must, and experience with neuroimaging
techniques, machine learning and signal processing would be a plus.

The Language Interaction and Computation Lab (CLIC) is a unit of the
University of Trento's Centre for Mind/Brain Sciences
(www.cimec.unitn.it) or CIMEC: an interdisciplinary Centre for the
research in brain and cognition including neuroscientists,
psychologists, (computational) linguists, computational
neuroscientists, and physicists. CLIC consists of researchers from the
Departments of Computer Science and Cognitive Science carrying out
research on a range of topics, including concept acquisition and
information extraction from very large multi-modal corpora, combining
brain data and data from corpora to study cognition, and methods of
theoretical linguistics.

The normal course of PhD study in Italy is three years, with 
possibility of extension, and includes courses at the CIMeC Doctoral
School (portale.unitn.it/drcimec/). There are no tuition fees, and the
position includes a fellowship to cover living expenses.

For additional information please contact Brian Murphy
(brian.murphy at unitn.it) or Massimo Poesio (massimo.poesio at unitn.it)
directly, attaching your CV.

-- Brian Murphy Post-Doctoral Researcher Language, Interaction and 
Computation Lab Centre for Mind/Brain Sciences University of Trento 
http://clic.cimec.unitn.it/brian/



From raniaeldeeb at gmail.com  Wed May 26 13:32:59 2010
From: raniaeldeeb at gmail.com (Rania El Deeb)
Date: Wed, 26 May 2010 23:32:59 +0300
Subject: [Eeglablist] out of memory
In-Reply-To: <740C7575D2240A46AD365B1DDD27AF724F3EDE9FB3@bkrexmb3>
References: <Acr3X6fZBz/b2OHZTfeNv7CnuY9frA==>
	<740C7575D2240A46AD365B1DDD27AF724F3EDE9FB3@bkrexmb3>
Message-ID: <AANLkTinC9A9E63R8ODbb0N-HATXPtZdoaww9yVhusx1e@mail.gmail.com>

Dear Mr Martin,

Why don't you try to increase the virtual memory to several GBytes from your
hard disk?

You can do that from Control Panel->System->Advanced->Performance(Settings
button)->Advanced->Virtual Memory(change button)

I have tried it and it worked for much larger data and workspace .. Note
that you have to restart the computer for changes to take effect.

You may also need to clear large variables from the workspace once you don't
need them again, by using:
clear var_name

Hope this helps


On Wed, May 19, 2010 at 5:29 PM, Schecklmann Martin <
Martin.Schecklmann at medbo.de> wrote:

>  Dear colleagues,
>
> I'm a new user of EEGLAB and I think I have some problems with the RAM. I'm
> working with MatLab 7.10.0 (R2010a) and EEGLAB v8.0.3.5b. I'm using Windows
> XP with 3.21 GB RAM. When i want to load a dataset of a colleague in set
> file format (set file 5 MB und the fdt file with 400 MB) i get the message
> "out of memory" in a "fread Out error" window. The "memory" command in
> MatLab shows the following informations:
> _______________________________________________________
>
> Maximum possible array:              83 MB (8.720e+007 bytes) *
> Memory available for all arrays:    510 MB (5.343e+008 bytes) **
> Memory used by MATLAB:              288 MB (3.019e+008 bytes)
> Physical Memory (RAM):             3292 MB (3.451e+009 bytes)
>
> *  Limited by contiguous virtual address space available.
> ** Limited by virtual address space available.
> _______________________________________________________
>
> It looks that MatLab gets not enough memory. I already used older EEGLAB
> version, but that do not work as well. Did you have any advice for me?
>
> Thanks a lot,
>
> best regards,
>
> Martin
>
>
>
> ------------------------------
> Medizinische Einrichtungen des Bezirks Oberpfalz GmbH (www.medbo.de)
> Gesch?ftsf?hrer: Kurt H?upl
> Aufsichtsratvorsitzender: Bezirkstagspr?sident Franz L?ffler
> Sitz: Universit?tsstra?e 84, 93053 Regensburg
> Registergericht: Regensburg HR-Nr.: B9977
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 


>>>Success is getting what you want. Happiness is wanting what you get.<<<
--
SLPLS_Rania
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100526/ff9a0568/attachment.html>

From sklein at berkeley.edu  Sat May 29 02:40:37 2010
From: sklein at berkeley.edu (Stanley Klein)
Date: Sat, 29 May 2010 02:40:37 -0700
Subject: [Eeglablist] improving coherence estimates
Message-ID: <AANLkTimwn0hxFnjfRzwByXFHO3EnVNO0wNEDA7dV6nOk@mail.gmail.com>

Dear EEGlab,
   In three days we'll be presenting a poster at a brain-computer
interface (BCI) conference in Monterey, CA on the topic of how to
reduce noise in coherence estimates for BCI purposes. I would
like to ask EEGlab whether the two approaches of the poster have
already been published or discussed. I'm not familiar with prior
work on it and I would really like to know before I make a fool of
myself.

1) The usefulness of removing the ERP before calculating coherence.
Our simulations are the main topic of the poster.

2) The usefulness of using Cauchy wavelets as filters for time-frequency
analysis when one wants high resolution in time (like 1 to 1.5 cycles).
These filters have rapid falloff at low temporal frequencies so they are
appropriate for the 1/f nature of EEG noise.

Again, I'd be grateful for any leads to articles on either of this items.
And I look forward to seeing some of you in Monterey.

thanks,
Stan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100529/989bd26e/attachment.html>

From nickbedo at yahoo.com  Sun May 30 23:04:31 2010
From: nickbedo at yahoo.com (Nick Bedo)
Date: Sun, 30 May 2010 23:04:31 -0700 (PDT)
Subject: [Eeglablist] Changing data scroll timeline to display pre-stim times
Message-ID: <621748.76108.qm@web62002.mail.re1.yahoo.com>

I feel like this is an elementary question, but I can't quite figure it out.

I'm loading up datasets which were originally recorded as [-200 1800], but when I scroll through the data, it plots as [0 2000].  All of the data are present, but the timeline is shifted.  How can I reflect the true timing values?

Thanks,
Nick


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100530/65b719de/attachment.html>

From ethan at cfin.dk  Mon May 31 14:07:05 2010
From: ethan at cfin.dk (Ethan Weed)
Date: Mon, 31 May 2010 23:07:05 +0200
Subject: [Eeglablist] Grand Average overlays
Message-ID: <AANLkTilXebh_1L0TBLZGvwkM3_zFAZZ87LGYRDvJX2ek@mail.gmail.com>

Dear list,

I have been playing around with EEGLAB and like it a lot - but I have one
frustration that seems like it could be easily solved, if only I could
figure out how:

I would like to overlay plots of grand averages from several different
conditions, and I can find no straightforward way to do this. The
pop_comperp method only allows me to plot the grand average from one
condition. And although I can extract the averaged data from the variable
erp1, generated by pop_comperp, I am unable to plot it with the nice touches
of topoplot. Surely there must be something obvious I'm overlooking?

Best regards,
Ethan

-- 
Ethan Weed, M.A.
PhD student, Dept. of Linguistics
University of Aarhus
--------------------------------------------------
Cognitive Neuroscience Research Unit
Hammel Neurocenter
--------------------------------------------------
Interacting Minds Project
Center for Functionally Integrative Neuroscience
DNC, Aarhus Sygehus
N?rrebrogade 44
Building 10G
8000 Aarhus C
http://www.cfin.au.dk
http://www.interacting-minds.net
ethan at cfin.dk
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100531/5e7e8c97/attachment.html>

From arno at ucsd.edu  Tue Jun  1 16:56:35 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Tue, 1 Jun 2010 16:56:35 -0700
Subject: [Eeglablist] Grand Average overlays
In-Reply-To: <AANLkTilXebh_1L0TBLZGvwkM3_zFAZZ87LGYRDvJX2ek@mail.gmail.com>
References: <AANLkTilXebh_1L0TBLZGvwkM3_zFAZZ87LGYRDvJX2ek@mail.gmail.com>
Message-ID: <46E6002A-29F6-4172-918C-D9CA148C7D88@ucsd.edu>

Dear Ethan,

you can overlay conditions both using pop_comperp and in STUDY. In  
pop_comperp you can only overlay 2 conditions. In STUDY, you can  
overlay as many as you want. See the two following links from the  
tutorial below:

http://sccn.ucsd.edu/wiki/Chapter_07:_Selecting_Data_Epochs_and_Comparing

http://sccn.ucsd.edu/wiki/Chapter_04:_Processing_multiple_subjects_in_studies#STUDY_data_visualization_tools

Best,

Arno



From arno at ucsd.edu  Tue Jun  1 17:02:48 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Tue, 1 Jun 2010 17:02:48 -0700
Subject: [Eeglablist] Changing data scroll timeline to display pre-stim
	times
In-Reply-To: <621748.76108.qm@web62002.mail.re1.yahoo.com>
References: <621748.76108.qm@web62002.mail.re1.yahoo.com>
Message-ID: <8D32AE33-417F-4E87-8572-459EC4E937FC@ucsd.edu>

Dear Nick,

In the menu, Edit > Dataset Info, you may change the lower time limit  
in the "Start time..." edit box.
Hope this helps,

Arno

On May 30, 2010, at 11:04 PM, Nick Bedo wrote:

> I feel like this is an elementary question, but I can't quite figure  
> it out.
>
> I'm loading up datasets which were originally recorded as [-200  
> 1800], but when I scroll through the data, it plots as [0 2000].   
> All of the data are present, but the timeline is shifted.  How can I  
> reflect the true timing values?
>
> Thanks,
> Nick

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100601/f307e42d/attachment.html>

From j.martinovic at liverpool.ac.uk  Tue Jun  1 07:50:20 2010
From: j.martinovic at liverpool.ac.uk (Jasna Martinovic)
Date: Tue, 01 Jun 2010 15:50:20 +0100
Subject: [Eeglablist] phd studentship at the university of aberdeen
Message-ID: <4C051E2C.1090006@liverpool.ac.uk>

*Interactions between bottom-up and top-down biases in the processing of 
colour*

PhD studentship from the College of Life Sciences and Medicine of the 
University of Aberdeen, UK

Supervisors: Dr Jasna Martinovic and Professor Arash Sahraie

Starting date: October 2010

The efficiency of attentional selection depends on two factors: the 
saliency of the stimulus (bottom-up processing) and the adopted 
perceptual set (top-down processing). Relations between these stimulus 
and task-driven factors are still poorly understood. Most studies 
describe additive and independent effects of top-down and bottom-up 
factors. However, it seems that in difficult tasks, whith stimuli 
competing for limited perceptual resources, top-down and bottom-up 
processes may interact. In real-life situations, the visual system is 
habitually presented with the difficult task of selecting stimuli from 
diverse and dynamic scenes. Therefore it is reasonable to assume that 
interactions between bottom-up and top-down biases could be the norm, 
rather than an exception. The project will be focused on determining the 
relations between bottom-up and top-down factors when attending to 
different features in complex dynamic scenes, focusing on luminance and 
colour. The main methods will be psychophysics and 
electroencephalography (EEG), with both event-related and steady-state 
evoked potentials being examined.

The project is based at the School of Psychology of the University of 
Aberdeen. The school has state-of-the-art EEG facilities, including two 
64-electrode Biosemi ActiveTwo EEG systems. One of these systems will 
soon be upgraded to 128 electrodes. Vision Research Laboratories within 
the School are equipped with multiple psychophysical workstations and a 
high-resolution eye tracking system (EyeLink 1000). The successful 
candidate will join an ambitious group of vision scientists comprising 
of academic and research staff and students.

/ /

Interested candidates should have a background in psychology, 
engineering, mathematics or a related field. Candidates must be eligible 
for UK/EU fee status and should hold a First or Upper Second Class 
Honours degree, a Masters degree or an equivalent qualification. 
Programming skills (e.g. Matlab, C) and previous experience with 
electrophysiology, psychophysics and signal processing methods would be 
beneficial, but are not essential since training in these methods will 
be provided.

The student will receive an annual stipend ? in 2010/2011, this will be 
?13,590. ?1000 to cover the costs of attending scientific meetings and 
training workshops will also be made available.

Aberdeen is the third biggest city in Scotland, situated in the scenic 
North East area. It has a vibrant and large student community and is 
home to dynamic, home-grown music and art scenes. The city is graced by 
the Aberdeen Art Gallery, His Majesty?s Theatre, Maritime Museum and 
several local history museums. Aberdeen lies in an area of great natural 
beauty, with the Royal Deeside being about an hour?s drive away.

To apply, send an application form with a covering letter, an up-to-date 
CV (no longer than 2 sides of A4) and names of two referees by email to 
m.pignotti at abdn.ac.uk. <mailto:m.pignotti at abdn.ac.uk.> The application 
form can be downloaded from 
http://www.abdn.ac.uk/sras/word_docs/pgapp.doc The deadline for 
applications is 20 June 2010. If you are interested in the project and 
have any questions please contact Dr Jasna Martinovic 
(j.martinovic at abdn.ac.uk <mailto:j.martinovic at abdn.ac.uk>).

---

Dr Jasna Martinovic
School of Psychology
University of Aberdeen
William Guild Building
Aberdeen
AB24 2UB

tel: 01224 272240
email: j.martinovic @ abdn.ac.uk
web: http://www.abdn.ac.uk/~psy527/dept/




From kevin_spencer at hms.harvard.edu  Tue Jun  1 08:47:45 2010
From: kevin_spencer at hms.harvard.edu (Spencer, Kevin M.)
Date: Tue, 1 Jun 2010 11:47:45 -0400
Subject: [Eeglablist] improving coherence estimates
In-Reply-To: <AANLkTimwn0hxFnjfRzwByXFHO3EnVNO0wNEDA7dV6nOk@mail.gmail.com>
References: <AANLkTimwn0hxFnjfRzwByXFHO3EnVNO0wNEDA7dV6nOk@mail.gmail.com>
Message-ID: <C5D74C5905A21845B94A6D13332EDC4D4198B6E13E@ITCCRMAIL02.MED.HARVARD.EDU>

Dear Stan,

Regarding your first question, you should check out Truccolo et al., Clinical Neurophysiology 2002 113:206-226. In this study they demonstrated that subtraction of the ERP from single trials actually leaves residual activity on the single trials which distorts estimates of single-trial activity. This is highly relevant for both coherence and single-electrode analyses.

While subtracting the ERP from the single trials before subsequent analyses is not a good idea, if you're doing time-frequency analyses, you can subtract evoked power/coherence from total power/coherence (average of single-trial power/coherence).

Kevin
--------------------------------------------------------------------------------------------------
Kevin M. Spencer, Ph.D.
Director, Neural Dynamics Laboratory (http://ndl.hms.harvard.edu<http://ndl.hms.harvard.edu/>)
Research Health Scientist, VA Boston Healthcare System
Assistant Professor of Psychiatry, Harvard Medical School
--------------------------------------------------------------------------------------------------

________________________________
From: eeglablist-bounces at sccn.ucsd.edu [eeglablist-bounces at sccn.ucsd.edu] On Behalf Of Stanley Klein [sklein at berkeley.edu]
Sent: Saturday, May 29, 2010 5:40 AM
To: eeglablist at sccn.ucsd.edu
Subject: [Eeglablist] improving coherence estimates

Dear EEGlab,
   In three days we'll be presenting a poster at a brain-computer
interface (BCI) conference in Monterey, CA on the topic of how to
reduce noise in coherence estimates for BCI purposes. I would
like to ask EEGlab whether the two approaches of the poster have
already been published or discussed. I'm not familiar with prior
work on it and I would really like to know before I make a fool of
myself.

1) The usefulness of removing the ERP before calculating coherence.
Our simulations are the main topic of the poster.

2) The usefulness of using Cauchy wavelets as filters for time-frequency
analysis when one wants high resolution in time (like 1 to 1.5 cycles).
These filters have rapid falloff at low temporal frequencies so they are
appropriate for the 1/f nature of EEG noise.

Again, I'd be grateful for any leads to articles on either of this items.
And I look forward to seeing some of you in Monterey.

thanks,
Stan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100601/a9656897/attachment.html>

From ethan at cfin.dk  Wed Jun  2 02:57:58 2010
From: ethan at cfin.dk (Ethan Weed)
Date: Wed, 2 Jun 2010 11:57:58 +0200
Subject: [Eeglablist] Grand Average overlays
In-Reply-To: <46E6002A-29F6-4172-918C-D9CA148C7D88@ucsd.edu>
References: <AANLkTilXebh_1L0TBLZGvwkM3_zFAZZ87LGYRDvJX2ek@mail.gmail.com>
	<46E6002A-29F6-4172-918C-D9CA148C7D88@ucsd.edu>
Message-ID: <AANLkTimX5fdb-WjoXSyo3JrLeDZdgWa-1T541KAt20rw@mail.gmail.com>

Dear Arnaud,

Perfect! Thanks very much for pointing me in the right direction.

-Ethan

On Wed, Jun 2, 2010 at 1:56 AM, Arnaud Delorme <arno at ucsd.edu> wrote:

> Dear Ethan,
>
> you can overlay conditions both using pop_comperp and in STUDY. In
> pop_comperp you can only overlay 2 conditions. In STUDY, you can overlay as
> many as you want. See the two following links from the tutorial below:
>
> http://sccn.ucsd.edu/wiki/Chapter_07:_Selecting_Data_Epochs_and_Comparing
>
>
> http://sccn.ucsd.edu/wiki/Chapter_04:_Processing_multiple_subjects_in_studies#STUDY_data_visualization_tools
>
> Best,
>
> Arno
>
>


-- 
Ethan Weed, M.A.
PhD student, Dept. of Linguistics
University of Aarhus
--------------------------------------------------
Cognitive Neuroscience Research Unit
Hammel Neurocenter
--------------------------------------------------
Interacting Minds Project
Center for Functionally Integrative Neuroscience
DNC, Aarhus Sygehus
N?rrebrogade 44
Building 10G
8000 Aarhus C
http://www.cfin.au.dk
http://www.interacting-minds.net
ethan at cfin.dk
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100602/def041a2/attachment.html>

From tehuberpro at gmail.com  Wed Jun  2 06:42:25 2010
From: tehuberpro at gmail.com (ondrej lassak)
Date: Wed, 2 Jun 2010 15:42:25 +0200
Subject: [Eeglablist] artifacts in time freq plots
In-Reply-To: <20100512024638.73434.qmail@web3709.mail.tnz.yahoo.co.jp>
References: <m2p436e576d1005070853o909f741chff76234cd3467ec9@mail.gmail.com>
	<20100512024638.73434.qmail@web3709.mail.tnz.yahoo.co.jp>
Message-ID: <AANLkTikfySQjugo_CRWgEnM4WEuEsJeV9lD27xXxno6H@mail.gmail.com>

Hey guys

Arnaud is right (thx for the code analysis) its just a numerical
artifact, which is impossible to occur on a real life EEG, so Makoto
you can chill :D

This time I added to the ideal sinusoid pseudo white noise and since
it adds non 0 variance the numerical artifact cannot occure and one
get sweet line in the TF plot even when using base subtraction

try this code to see it for yourself:

fs=2048%samples/s
t=linspace(0,8,8*2048);%samples
f=6%Hz


x=.05*sin(2*pi*f*t);
x1=.05*sin(2*pi*f*t*2);
x=[x(1:8*2048/2) x1(1:8*2048/2)];
x=x+.01*rand(1,length(t))


figure; [ersp,itc,powbase,times,freqs,erspboot,itcboot] = newtimef(x,
8*2048, [-500 8000], 2048, 0,'baseline',[NaN],'basenorm','off',
'maxfreq' ,19,'nfreqs',150,'padratio', 32, 'scale', 'abs');
set(gcf,'Name',('original signal no baseline adjustment'))

figure; [ersp,itc,powbase,times,freqs,erspboot,itcboot] = newtimef(x,
8*2048, [-500 8000], 2048, 0,'baseline',[5000 6000],'basenorm','off',
'maxfreq' ,19,'nfreqs',150,'padratio', 32, 'scale', 'abs');
set(gcf,'Name','baseline adjusted [5000 6000] ms')



figure; [ersp,itc,powbase,times,freqs,erspboot,itcboot] = newtimef(x,
8*2048, [-500 8000], 2048, 0,'baseline',[2000 5000],'basenorm','off',
'maxfreq' ,19,'nfreqs',150,'padratio', 32, 'scale', 'abs');
set(gcf,'Name','baseline adjusted [2000 5000] ms')





On Wed, May 12, 2010 at 4:46 AM, Makoto Miyakoshi
<mataothefifth at yahoo.co.jp> wrote:
> Dear Ondrej,
>
> I replicated your results. This is shocking actually.
> I also found that the problem is probably in baseline subtraction because
> when I set 'baseline', NaN, the result seems right.
>
> Makoto
>
>
> --- ondrej lassak <tehuberpro at gmail.com> wrote:
>
>> I fed the TF analysis single sinusoid 6 and 12Hz and the TF plot
>> shows
>> multiple specral lines (more than two).
>> How can one rely on the TF when it introduces such massive artifacts
>> both in
>> pure FFT spectrogram and Wavelet scalogram?
>> Or am I doing something wrong? When only one freq during the whole
>> time span
>> is present the TF plots look like really bad moira and the presence
>> of the
>> freq is apparent only from the summation over time (left from the
>> main
>> plot).
>>
>>
>> The matlab report with function calls and resulting pictures is
>> attached
>> below (no scripts embedded in the html).
>> > _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
>


From agmabaib at gmail.com  Wed Jun  2 11:24:07 2010
From: agmabaib at gmail.com (Agustin Ibanez)
Date: Wed, 2 Jun 2010 15:24:07 -0300
Subject: [Eeglablist] Invitation to Publish in Social Neuroscience Journal
Message-ID: <AANLkTimVhgj1w_-Ck8_1kWOB0e-V04awa_HMnMmCFqOM@mail.gmail.com>

Dear Coleages
I would like to invite you to contribute with a paper for a special issue of
Social Neuroscience journal (Impact factor: 4.352; see
http://www.psypress.com/social-neuroscience-1747-0919) on Neuropsychiatric
disorders, which will be guest-edited by Dr. Facundo Manes. Given your
expertise in this field, I?d like to ask whether you might be interested in
contributing with a manuscript looking at any aspect of social neuroscience
in neuropsychiatric populations.



Please find a formal letter from Professor Manes attached. Because there is
a lot of interest in this issue, if you think you'll be submitting a
manuscript for this special issue, I would appreciate it if you could let me
know in advance so we can get a better feel for how many manuscripts will
have to be handled for editorial purposes (peer-review, space available,
etc.).



All the best,

Agustin


-- 
Agust?n Iba?ez, PhD
Director
Laboratory of Experimental Psychology & Neuroscience
Institute of Cognitive Neurology (INECO) & CONICET
Castex 3293 (CP 1425)
Buenos Aires, Argentina
Phone/Fax: +54 (11) 4807-4748
aibanez at neurologiacognitiva.org
http://www.neurologiacognitiva.org/
www.ineco.org.ar

Professor of Cognitive Neurosciences
Favaloro University
www.fundacionfavaloro.org/IN_neurociencias.htm

Affiliated Researcher,
Center for Social and Cognitive Neuroscience
The University of Chicago
5848 S. University Avenue
Chicago, Illinois 60637
773.702.8403, USA
http://ccsn.uchicago.edu/

Associate Researcher
Cognitive Neuroscience Laboratory
Faculty of Psychology
Universidad Diego Portales. Chile
Vergara 275, Santiago
+562 6762540/ 2518
agustin.ibanez at udp.cl
http://neuro.udp.cl/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100602/1900950f/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Soc Neurosci special issue.pdf
Type: application/pdf
Size: 67399 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100602/1900950f/attachment.pdf>

From mataothefifth at yahoo.co.jp  Thu Jun  3 19:14:54 2010
From: mataothefifth at yahoo.co.jp (Makoto Miyakoshi)
Date: Fri, 4 Jun 2010 11:14:54 +0900 (JST)
Subject: [Eeglablist] artifacts in time freq plots
In-Reply-To: <AANLkTikfySQjugo_CRWgEnM4WEuEsJeV9lD27xXxno6H@mail.gmail.com>
Message-ID: <20100604021454.60480.qmail@web3706.mail.tnz.yahoo.co.jp>

Dear Ondrej and Arno,

Yes. When I read your command line I actually missed that you used the
optional input 'basenorm' which I have never used. I should have been more
careful to check the input parameters before seeing the results. I
apologize that my initial comment was misleading too.

One thing I also notice is that you used 'nfreqs' and 'padratio' togetehr
but they are mutually exclusive and nfreqs is used when both are input.

This is suggestion for Arno, but is it possible in future to show what is
actually calculated in what way when presenting the figure, especially when
optional parameters are used? I know newtimef shows input parameters and
calculation processes, but I mean for the sake of drawing attention of new
users (and careless user like me) it may be helpful to present additional
lines like 'Caution: basenorm is on; the shown result is devided by
deviation of baseline period' or 'Caution: nfreqs and padratio are both
input; nfreqs is used and padratio is neglected'. I know these rules are
written in the help lines, but just for enabling a quick check.

Makoto


--- ondrej lassak <tehuberpro at gmail.com> wrote:

> Hey guys
> 
> Arnaud is right (thx for the code analysis) its just a numerical
> artifact, which is impossible to occur on a real life EEG, so Makoto
> you can chill :D
> 
> This time I added to the ideal sinusoid pseudo white noise and since
> it adds non 0 variance the numerical artifact cannot occure and one
> get sweet line in the TF plot even when using base subtraction
> 
> try this code to see it for yourself:
> 
> fs=2048%samples/s
> t=linspace(0,8,8*2048);%samples
> f=6%Hz
> 
> 
> x=.05*sin(2*pi*f*t);
> x1=.05*sin(2*pi*f*t*2);
> x=[x(1:8*2048/2) x1(1:8*2048/2)];
> x=x+.01*rand(1,length(t))
> 
> 
> figure; [ersp,itc,powbase,times,freqs,erspboot,itcboot] = newtimef(x,
> 8*2048, [-500 8000], 2048, 0,'baseline',[NaN],'basenorm','off',
> 'maxfreq' ,19,'nfreqs',150,'padratio', 32, 'scale', 'abs');
> set(gcf,'Name',('original signal no baseline adjustment'))
> 
> figure; [ersp,itc,powbase,times,freqs,erspboot,itcboot] = newtimef(x,
> 8*2048, [-500 8000], 2048, 0,'baseline',[5000 6000],'basenorm','off',
> 'maxfreq' ,19,'nfreqs',150,'padratio', 32, 'scale', 'abs');
> set(gcf,'Name','baseline adjusted [5000 6000] ms')
> 
> 
> 
> figure; [ersp,itc,powbase,times,freqs,erspboot,itcboot] = newtimef(x,
> 8*2048, [-500 8000], 2048, 0,'baseline',[2000 5000],'basenorm','off',
> 'maxfreq' ,19,'nfreqs',150,'padratio', 32, 'scale', 'abs');
> set(gcf,'Name','baseline adjusted [2000 5000] ms')
> 
> 
> 
> 
> 
> On Wed, May 12, 2010 at 4:46 AM, Makoto Miyakoshi
> <mataothefifth at yahoo.co.jp> wrote:
> > Dear Ondrej,
> >
> > I replicated your results. This is shocking actually.
> > I also found that the problem is probably in baseline subtraction
> because
> > when I set 'baseline', NaN, the result seems right.
> >
> > Makoto
> >
> >
> > --- ondrej lassak <tehuberpro at gmail.com> wrote:
> >
> >> I fed the TF analysis single sinusoid 6 and 12Hz and the TF plot
> >> shows
> >> multiple specral lines (more than two).
> >> How can one rely on the TF when it introduces such massive
> artifacts
> >> both in
> >> pure FFT spectrogram and Wavelet scalogram?
> >> Or am I doing something wrong? When only one freq during the whole
> >> time span
> >> is present the TF plots look like really bad moira and the
> presence
> >> of the
> >> freq is apparent only from the summation over time (left from the
> >> main
> >> plot).
> >>
> >>
> >> The matlab report with function calls and resulting pictures is
> >> attached
> >> below (no scripts embedded in the html).
> >> > _______________________________________________
> >> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> >> To unsubscribe, send an empty email to
> >> eeglablist-unsubscribe at sccn.ucsd.edu
> >> For digest mode, send an email with the subject "set digest mime"
> to
> > eeglablist-request at sccn.ucsd.edu
> >
> > _______________________________________________
> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> > To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> > For digest mode, send an email with the subject "set digest mime"
> to eeglablist-request at sccn.ucsd.edu
> >
> 



From hayder978 at yahoo.com  Sat Jun  5 01:49:43 2010
From: hayder978 at yahoo.com (Hayder Hussein)
Date: Sat, 5 Jun 2010 01:49:43 -0700 (PDT)
Subject: [Eeglablist] Coherence in meditation-seeking kind help from experts
	in EEGLAB
Message-ID: <932611.65115.qm@web38804.mail.mud.yahoo.com>

Dear Professors,

I really appreciate your help and spending time to reply me.
It is life or death!
I will do my thesis depending on this analysis for huge data (repeating the same as below).

Please I am seeking the help of your experience in using EEGLAB.
It is easy that may be I have a mistake in this, it will cost me too much to analyze a lot of data which may be the way of analysis is wrong from the very beginning, so I am asking your kind advice.

My research interest is to check the significant coherence changes between channels or components during meditation.

I will list the settings of EEGLAB for the analysis of the EEG signals took during rest and during meditation. It is too easy that I have falled in some mistake in this, which you the experts can easily see the flaws so please correct me if I am wrong.
I am not sure especially with the parameters between (((brackets)))

Now as a pilot study, I will analyze just one person's data.

1. EEG readings (20  channels,sampling rate 1000 sample/sec) during rest (mind wandering) for 200 seconds.

2. then EEG readings during meditation for 200 seconds, so the total EEG reading is 400 seconds.

3. (((Let us say that the baseline is the 1st 200 seconds, let us divide the meditation (200 seconds) into 40 epochs, each epoch last for 5 seconds.)))

4. The baseline for all the epochs is the same (the 1st 200seconds) which are the rest state.

5.So the dataset info should looks like:

Channels per frame         20
Frames per epoch           5000
Epochs                     (((40)))
Events                     (((None)))
Sampling rate              1000
Epoch start (sec)          (((200  )))
Epoch End                 ((( 205)))  (means the 1st epoch,((( the second will start    at 205 and ends in 210 ?))))

6. I should specify the pop_newcrossf() as the following:

First channel                          1
Second channel                         2
Epoch time range   [min max] (msec)    ((([200000    205000])))
Wavelet cycles                         3 0.5
Linear coherence or phase coherence   (((?)))
Bootstrap significance                 0.01
optional arguments                 'padratio',2,'baseline',[0 200000]


My understanding that this will make EEGLAB calculating the coherence between ch1 and ch2 for 40 epochs during meditation (each epoch 5 sec.) and using the baseline(1st 200sec. during rest) to normalize the coherence and calculate the significant coherence. Is this correct?

I am relatively new in this area, I apologize if my questions seems naive.

Haider Al-Wasiti, M.D.
M.Sc. student - Biomedical Engineering/UPM
wasiti at mutiara.upm.edu.my
hayder at wasiti.net
www.wasiti.net


      


From hayder978 at yahoo.com  Sat Jun  5 01:49:18 2010
From: hayder978 at yahoo.com (Hayder Hussein)
Date: Sat, 5 Jun 2010 01:49:18 -0700 (PDT)
Subject: [Eeglablist] Events and Epochs
Message-ID: <398220.87388.qm@web38805.mail.mud.yahoo.com>

Dear All,


I am confused about the events, is it right that every 1 epoch should contain 1 event? (either in the middle so the 1st part of the epoch will be the baseline and the second part will be the area of interest for the ERSP or coherence to be calculated).

Because if every epoch can contain more than 1 event so why we not make 1 epoch for every 1 event (it seems the events counting useless in this case as it is the same as the epoch count in the EEG reading).

Thank you in advance


      


From arno at ucsd.edu  Mon Jun  7 06:55:52 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Mon, 7 Jun 2010 15:55:52 +0200
Subject: [Eeglablist] Events and Epochs
In-Reply-To: <398220.87388.qm@web38805.mail.mud.yahoo.com>
References: <398220.87388.qm@web38805.mail.mud.yahoo.com>
Message-ID: <F43968BD-EBDB-4D19-80B7-3307638662BC@ucsd.edu>

Dear Hayder,


> I am confused about the events, is it right that every 1 epoch  
> should contain 1 event? (either in the middle so the 1st part of the  
> epoch will be the baseline and the second part will be the area of  
> interest for the ERSP or coherence to be calculated).


each epoch can contain multiple events.
Hope this helps,

Arno




From arno at ucsd.edu  Mon Jun  7 08:04:18 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Mon, 7 Jun 2010 17:04:18 +0200
Subject: [Eeglablist] artifacts in time freq plots
In-Reply-To: <20100604021454.60480.qmail@web3706.mail.tnz.yahoo.co.jp>
References: <20100604021454.60480.qmail@web3706.mail.tnz.yahoo.co.jp>
Message-ID: <CC61D89F-DF59-44E9-AA86-3199081473C3@ucsd.edu>

Dear Makoto,

yes, this is a good suggestion. We will implement it.
Thanks a lot,

Arno

On Jun 3, 2010, at 7:14 PM, Makoto Miyakoshi wrote:

> Dear Ondrej and Arno,
>
> Yes. When I read your command line I actually missed that you used the
> optional input 'basenorm' which I have never used. I should have  
> been more
> careful to check the input parameters before seeing the results. I
> apologize that my initial comment was misleading too.
>
> One thing I also notice is that you used 'nfreqs' and 'padratio'  
> togetehr
> but they are mutually exclusive and nfreqs is used when both are  
> input.
>
> This is suggestion for Arno, but is it possible in future to show  
> what is
> actually calculated in what way when presenting the figure,  
> especially when
> optional parameters are used? I know newtimef shows input parameters  
> and
> calculation processes, but I mean for the sake of drawing attention  
> of new
> users (and careless user like me) it may be helpful to present  
> additional
> lines like 'Caution: basenorm is on; the shown result is devided by
> deviation of baseline period' or 'Caution: nfreqs and padratio are  
> both
> input; nfreqs is used and padratio is neglected'. I know these rules  
> are
> written in the help lines, but just for enabling a quick check.
>
> Makoto
>
>
> --- ondrej lassak <tehuberpro at gmail.com> wrote:
>
>> Hey guys
>>
>> Arnaud is right (thx for the code analysis) its just a numerical
>> artifact, which is impossible to occur on a real life EEG, so Makoto
>> you can chill :D
>>
>> This time I added to the ideal sinusoid pseudo white noise and since
>> it adds non 0 variance the numerical artifact cannot occure and one
>> get sweet line in the TF plot even when using base subtraction
>>
>> try this code to see it for yourself:
>>
>> fs=2048%samples/s
>> t=linspace(0,8,8*2048);%samples
>> f=6%Hz
>>
>>
>> x=.05*sin(2*pi*f*t);
>> x1=.05*sin(2*pi*f*t*2);
>> x=[x(1:8*2048/2) x1(1:8*2048/2)];
>> x=x+.01*rand(1,length(t))
>>
>>
>> figure; [ersp,itc,powbase,times,freqs,erspboot,itcboot] = newtimef(x,
>> 8*2048, [-500 8000], 2048, 0,'baseline',[NaN],'basenorm','off',
>> 'maxfreq' ,19,'nfreqs',150,'padratio', 32, 'scale', 'abs');
>> set(gcf,'Name',('original signal no baseline adjustment'))
>>
>> figure; [ersp,itc,powbase,times,freqs,erspboot,itcboot] = newtimef(x,
>> 8*2048, [-500 8000], 2048, 0,'baseline',[5000 6000],'basenorm','off',
>> 'maxfreq' ,19,'nfreqs',150,'padratio', 32, 'scale', 'abs');
>> set(gcf,'Name','baseline adjusted [5000 6000] ms')
>>
>>
>>
>> figure; [ersp,itc,powbase,times,freqs,erspboot,itcboot] = newtimef(x,
>> 8*2048, [-500 8000], 2048, 0,'baseline',[2000 5000],'basenorm','off',
>> 'maxfreq' ,19,'nfreqs',150,'padratio', 32, 'scale', 'abs');
>> set(gcf,'Name','baseline adjusted [2000 5000] ms')
>>
>>
>>
>>
>>
>> On Wed, May 12, 2010 at 4:46 AM, Makoto Miyakoshi
>> <mataothefifth at yahoo.co.jp> wrote:
>>> Dear Ondrej,
>>>
>>> I replicated your results. This is shocking actually.
>>> I also found that the problem is probably in baseline subtraction
>> because
>>> when I set 'baseline', NaN, the result seems right.
>>>
>>> Makoto
>>>
>>>
>>> --- ondrej lassak <tehuberpro at gmail.com> wrote:
>>>
>>>> I fed the TF analysis single sinusoid 6 and 12Hz and the TF plot
>>>> shows
>>>> multiple specral lines (more than two).
>>>> How can one rely on the TF when it introduces such massive
>> artifacts
>>>> both in
>>>> pure FFT spectrogram and Wavelet scalogram?
>>>> Or am I doing something wrong? When only one freq during the whole
>>>> time span
>>>> is present the TF plots look like really bad moira and the
>> presence
>>>> of the
>>>> freq is apparent only from the summation over time (left from the
>>>> main
>>>> plot).
>>>>
>>>>
>>>> The matlab report with function calls and resulting pictures is
>>>> attached
>>>> below (no scripts embedded in the html).
>>>>> _______________________________________________
>>>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>>>> To unsubscribe, send an empty email to
>>>> eeglablist-unsubscribe at sccn.ucsd.edu
>>>> For digest mode, send an email with the subject "set digest mime"
>> to
>>> eeglablist-request at sccn.ucsd.edu
>>>
>>> _______________________________________________
>>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>>> To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>>> For digest mode, send an email with the subject "set digest mime"
>> to eeglablist-request at sccn.ucsd.edu
>>>
>>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu



From aluo at neurosky.com  Mon Jun  7 16:24:04 2010
From: aluo at neurosky.com (An Luo)
Date: Mon, 7 Jun 2010 16:24:04 -0700
Subject: [Eeglablist] Job positions at NeuroSky
Message-ID: <AANLkTil-0V7ZSCI5AaaIcJlrUO2dDSLonVKtREuxbA5t@mail.gmail.com>

Positions Available at NeuroSky

NeuroSky is the worldwide leader in bringing EEG technology to the
consumer mass market. Our well-known partners have utilized our
technology to create some of the most exciting products of the year,
including Mattel?s Mindflex and Uncle Milton?s Force Trainer. And this
is only the beginning.

Our Brain-Computer Interface technologies are available to developers
and researchers across a range of different industries. Our
advancements breathe new life into existing products and enable the
birth of new innovations. We?ve partnered with some of the most
prestigious academic institutions to develop the next wave of
solutions.

The NeuroSky Technology Development team is currently considering
applications in the following areas. If interested, please submit your
resume or CV to tom at neurosky.com.

1. Algorithm Development Scientist

The scientist/engineer will be responsible for running experiments
using EEG devices and other biosensors, and developing algorithms and
classifiers based on analysis of the recorded data.

Preferred Experience:
Experience with various classifiers and machine learning techniques
Experience with experiment design and analysis
Knowledge of typical EEG signals and artifacts
Knowledge of sleep and corresponding EEG signals
Ph.D. in signal processing or related areas preferred

2. Algorithm Development Scientist

The scientist?s primary responsibility will be in supporting external
EEG experiments using NeuroSky EEG devices.  These might be university
groups or NeuroSky?s commercial partners.  The job will be located in
Hong Kong.  Additionally, as time permits, the scientist may create
and run his or her own EEG experiments with a focus on creating useful
algorithms that estimate a person?s brain state.

Preferred Experience:
Experience with experiment design and analysis
Knowledge of typical EEG signals and artifacts
Ph.D. in cognitive science, neuroscience, signal processing or related
areas preferred
Experience with various classifiers and machine learning techniques


3. Circuit Design Engineer

The engineer will be responsible for implementing new product ideas in
human-computer interaction by creating PCB-based designs using new
various sensors. This includes designing sensor-level circuitry,
testing, and documentation of results.

Preferred Experience:
Experience with analog design at PCB level
Experience with sensor signal conditioning circuits
Knowledge of C programming, especially with embedded processors
Experience with bench top circuit testing

-- 
An Luo,  Ph.D.
Research Scientist
phone: (646) 209-7943
NeuroSky
125 S. Market St. Suite 900
San Jose, CA 95113

Twitter:  http://twitter.com/neuroSky_bci
Facebook:  http://www.facebook.com/NeuroSkyBCI
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
This email may contain confidential and privileged material for the
sole use of the intended recipient(s).
Any review, use, distribution or disclosure by others is strictly prohibited.
If you are not the intended recipient, please contact the sender by
reply email and delete all copies of this message.
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////



From kjyoder at gmail.com  Wed Jun  9 13:10:32 2010
From: kjyoder at gmail.com (Keith Yoder)
Date: Wed, 9 Jun 2010 16:10:32 -0400
Subject: [Eeglablist] ICA problem - using PCA to reduce number of
	'channels' for ICA
Message-ID: <AANLkTilgnxMnwrBvhIEIg6wFm0IQHGQzerRWkJ89H0di@mail.gmail.com>

I have a related question:

We are recording using 128-channel BioSemi cap, plus four ocular leads and
two mastoid leads.  All of the data are referenced to the right mastoid
(leaving EEG.data with 133 rows).  We would like to use ICA, but have found
that 133 components is too many (e.g. a CNV which is clear in the electrodes
does not fall out as a component, but can be isolated with PCA using 5
components).

Have other users had success using PCA to pre-process the data before
passing them to ICA, so as to reduce the number of components identified by
the ICA algorithm?

A possible solution that I have been trying to implement would use PCA to
identify a more reasonable number of components (e.g. 64) to then pass to
ICA (all using runica):

% after using runica with " ,'pca',64 " to define 64 PCAs
% manually prepare component activations for ICA as if they were
% data from EEG.data (as laid out in pop_runica)
>> tmpdata = reshape(EEG.icaact(:,:,:),64,EEG.pnts*EEG.trials);
>> tmpdata = tmpdata - repmat(mean(tmpdata,2), [1 size(tmpdata,2)]);
>> [EEG.icaweights, EEG.icasphere] = runica(tmpdata, 'lrate', 0.001,
'interupt','on');

However, this call to runica returns the component weights and sphering for
the ICA decomposition of the PCA-defined components, rather than ICA-defined
weights and sphering for 64 ICs for our 133 channels. Has anyone had success
with a technique along similar lines?

Alternatively, I could simply remove electrodes from EEG.data until I am
left with 64 channels, but I am wary of throwing out data entirely.

We are running EEGLAB 7.2.9.20b in 64-bit Matlab 7.10.0 (R2010a) in Mac OS X
10.5.7.

Many thanks,
Keith
--
Keith Yoder
Research Aide
Laboratory for the Neuroscience of Autism
Cornell University
Ithaca, NY
574.215.9678


On Mon, Feb 1, 2010 at 5:26 PM, Scott Makeig <smakeig at gmail.com> wrote:

> Joe is correct that ICA will not converge if the rank of the data matrix is
> less than the number of channels. The runica/binica algorithms are supposed
> to test the rank of the input data. If two channels are identical, or if
> some subset of n channels are otherwise interdependent, then the rank will
> be less than the number of channels and PCA reduction should be applied to
> remove the redundancy and allow the ICA decomposition to converge.
>
> Arno -- There was a problem with the Matlab rank() function on 64-bit
> machines, I believe. Has this been solved and Is the auto rank detection ->
> PCA option currently implemented in runica/binica?  Perhaps we could add a
> 'toy' rank() function pre-test (e.g. finding the rank of a small full-rank
> matrix to detect if rank() is working...) ? If so, run the rank test; if
> not, then warn the user or build a work-around rank function that will work
> properly?
>
> Scott
>
>
> On Mon, Feb 1, 2010 at 6:56 AM, Joseph Dien <jdien07 at mac.com> wrote:
>
>> When you say "so long", how long do you mean?  While ICA is not by its
>> nature a fast procedure, certain datasets can take much longer than usual.
>>  For example, I find that if two channels are perfectly correlated (1 or -1)
>> then an ICA run will take much longer.  This can happen if the data is mean
>> mastoid referenced and both channels are explicitly included in the data
>> because they will have a perfect -1 correlation (see Dien, 1998 for
>> reference issues).  It can also happen if a channel is shorted out during
>> acquisition and the reference channel is explicitly included because then
>> they will have a perfect correlation.  Also if two channels are shorted
>> together during the data acquisition they will be perfectly correlated with
>> each other.  My EP Toolkit (
>> https://sourceforge.net/projects/erppcatoolkit/) has code for dealing
>> with these situations so you might want to look into it.  It implements an
>> automated artifact correction routine that relies on EEGlab's runICA code,
>> among!
>>  other things.
>>
>> Cheers!
>>
>> Joe
>>
>>
>>
>> On Jan 29, 2010, at 4:49 AM, peng wang wrote:
>>
>> > Hi there,
>> >
>> >    I am using ICA to remove blinks via EEGLab. My dataset has 122
>> channels, and it takes so long to compute 122 components.
>> >    (1) So I tried to use the option "ncomps" (say, 24) to reduce the
>> number of components. However, an error message appears after computing:
>> "Matrix dimensions must agree".
>> >
>> >    (2) Then I tried fastICA instead as following,
>> >
>> > ==================
>> >  sz = size(EEG.data);
>> >  nchans = sz(1);
>> >  npts = sz(2);
>> >  ntrials = sz(3);
>> >  clear sz;
>> >  nICs = 24;
>> >  data = reshape(EEG.data,nchans,npts*ntrials);
>> >  [ica,V,W] = fastica(data,'numOfIC',nICs,'approach','symm');
>> >  EEG.icasphere = eye(nchans);
>> >  EEG.icaact = single(reshape(ica,nICs,npts,ntrials));
>> >  EEG.icawinv = V;
>> >  EEG.icaweights = W;
>> >  EEG = eeg_checkset( EEG );
>> >  clear V W ica data;
>> >
>> >  EEG = pop_saveset( EEG, 'filename','test_raw_ica');
>> > ==================
>> >
>> >    Everything seems fine. But when I reject the blink component via GUI
>> of eeglab and load the data again, Something strange happens. It seems the
>> amplitude of EEG.data become much smaller, about in  -1~1 range.  Thus I
>> wonder whether there was some normalization behind, and how can I correct
>> it? The problem would not repeat if I choose the number of components same
>> as channels event in fast ICA (e.g. change to "nICs = nchans" in the above
>> code).
>> >
>> >    Thank you for your help.
>> >
>> > best
>> > Peng
>> > _______________________________________________
>> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> > To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> > For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>>
>>
>>
>> --------------------------------------------------------------------------------
>>
>> Joseph Dien,
>> Senior Research Scientist
>> University of Maryland
>> 7005 52nd Avenue
>> College Park, MD 20742-0025
>>
>> E-mail: jdien07 at mac.com
>> Phone: 301-226-8848
>> Fax: 301-226-8811
>> http://homepage.mac.com/jdien07/
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>>
>
>
>
> --
> Scott Makeig, Research Scientist and Director, Swartz Center for
> Computational Neuroscience, Institute for Neural Computation, University of
> California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100609/c50856ba/attachment.html>

From krysta.chauncey at tufts.edu  Wed Jun  9 19:20:29 2010
From: krysta.chauncey at tufts.edu (Krysta Chauncey)
Date: Wed, 9 Jun 2010 22:20:29 -0400
Subject: [Eeglablist] datasets automatically saving as matlab.mat?
Message-ID: <AANLkTin9sby8Y0ujvChplpq0Rfxi7wo9vxN5MhVtO6WJ@mail.gmail.com>

All of a sudden this matlab session, whenever I try to save a dataset, I get
this message:

"Note that your memory options for saving datasets does not correspond
to the format of the datasets on disk (ignoring memory options)
Saving dataset...

Saving to: matlab.mat

Done."

What does that mean, how did it happen, and how do I fix it?

cheers,
Krysta

----------------------------
Krysta Chauncey, Ph.D.
Brain-Computer Interface Project
Human-Computer Interaction Lab
Computer Science Dept, Tufts University
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100609/dc1f2196/attachment.html>

From smakeig at gmail.com  Thu Jun 10 12:49:45 2010
From: smakeig at gmail.com (Scott Makeig)
Date: Thu, 10 Jun 2010 12:49:45 -0700
Subject: [Eeglablist] ICA problem - using PCA to reduce number of
	'channels' for ICA
In-Reply-To: <AANLkTilgnxMnwrBvhIEIg6wFm0IQHGQzerRWkJ89H0di@mail.gmail.com>
References: <AANLkTilgnxMnwrBvhIEIg6wFm0IQHGQzerRWkJ89H0di@mail.gmail.com>
Message-ID: <AANLkTikxB2tI6JcrZNvA0wyOHRIwv2R3efBHA9GxtjST@mail.gmail.com>

Keith -

>> [wts,sph] = runica(somedata,'pca',64)

does run first PCA, selects the largest 64 PC subspace, runs ICA (as its
commandline output should attest), and will return the appropriate matrices.
I do not have experience decomposing a CNV, and in fact routinely highpass
the data about 1 Hz to better resolve the 3-250 Hz activities. PCA would
accumulate much of the CNV in early components simply because it is
numerically large. If ICA separates the CNV into several component
processes, this might indeed reflect changes in the scalp distribution of
the CNV during its time course.

Scott Makeig

On Wed, Jun 9, 2010 at 1:10 PM, Keith Yoder <kjyoder at gmail.com> wrote:

> I have a related question:
>
> We are recording using 128-channel BioSemi cap, plus four ocular leads and
> two mastoid leads.  All of the data are referenced to the right mastoid
> (leaving EEG.data with 133 rows).  We would like to use ICA, but have found
> that 133 components is too many (e.g. a CNV which is clear in the electrodes
> does not fall out as a component, but can be isolated with PCA using 5
> components).
>
> Have other users had success using PCA to pre-process the data before
> passing them to ICA, so as to reduce the number of components identified by
> the ICA algorithm?
>
> A possible solution that I have been trying to implement would use PCA to
> identify a more reasonable number of components (e.g. 64) to then pass to
> ICA (all using runica):
>
> % after using runica with " ,'pca',64 " to define 64 PCAs
> % manually prepare component activations for ICA as if they were
> % data from EEG.data (as laid out in pop_runica)
> >> tmpdata = reshape(EEG.icaact(:,:,:),64,EEG.pnts*EEG.trials);
> >> tmpdata = tmpdata - repmat(mean(tmpdata,2), [1 size(tmpdata,2)]);
> >> [EEG.icaweights, EEG.icasphere] = runica(tmpdata, 'lrate', 0.001,
> 'interupt','on');
>
> However, this call to runica returns the component weights and sphering for
> the ICA decomposition of the PCA-defined components, rather than ICA-defined
> weights and sphering for 64 ICs for our 133 channels. Has anyone had success
> with a technique along similar lines?
>
> Alternatively, I could simply remove electrodes from EEG.data until I am
> left with 64 channels, but I am wary of throwing out data entirely.
>
> We are running EEGLAB 7.2.9.20b in 64-bit Matlab 7.10.0 (R2010a) in Mac OS
> X 10.5.7.
>
> Many thanks,
> Keith
> --
> Keith Yoder
> Research Aide
> Laboratory for the Neuroscience of Autism
> Cornell University
> Ithaca, NY
> 574.215.9678
>
>
> On Mon, Feb 1, 2010 at 5:26 PM, Scott Makeig <smakeig at gmail.com> wrote:
>
>> Joe is correct that ICA will not converge if the rank of the data matrix
>> is less than the number of channels. The runica/binica algorithms are
>> supposed to test the rank of the input data. If two channels are identical,
>> or if some subset of n channels are otherwise interdependent, then the rank
>> will be less than the number of channels and PCA reduction should be applied
>> to remove the redundancy and allow the ICA decomposition to converge.
>>
>> Arno -- There was a problem with the Matlab rank() function on 64-bit
>> machines, I believe. Has this been solved and Is the auto rank detection ->
>> PCA option currently implemented in runica/binica?  Perhaps we could add a
>> 'toy' rank() function pre-test (e.g. finding the rank of a small full-rank
>> matrix to detect if rank() is working...) ? If so, run the rank test; if
>> not, then warn the user or build a work-around rank function that will work
>> properly?
>>
>> Scott
>>
>>
>> On Mon, Feb 1, 2010 at 6:56 AM, Joseph Dien <jdien07 at mac.com> wrote:
>>
>>> When you say "so long", how long do you mean?  While ICA is not by its
>>> nature a fast procedure, certain datasets can take much longer than usual.
>>>  For example, I find that if two channels are perfectly correlated (1 or -1)
>>> then an ICA run will take much longer.  This can happen if the data is mean
>>> mastoid referenced and both channels are explicitly included in the data
>>> because they will have a perfect -1 correlation (see Dien, 1998 for
>>> reference issues).  It can also happen if a channel is shorted out during
>>> acquisition and the reference channel is explicitly included because then
>>> they will have a perfect correlation.  Also if two channels are shorted
>>> together during the data acquisition they will be perfectly correlated with
>>> each other.  My EP Toolkit (
>>> https://sourceforge.net/projects/erppcatoolkit/) has code for dealing
>>> with these situations so you might want to look into it.  It implements an
>>> automated artifact correction routine that relies on EEGlab's runICA code,
>>> among!
>>>  other things.
>>>
>>> Cheers!
>>>
>>> Joe
>>>
>>>
>>>
>>> On Jan 29, 2010, at 4:49 AM, peng wang wrote:
>>>
>>> > Hi there,
>>> >
>>> >    I am using ICA to remove blinks via EEGLab. My dataset has 122
>>> channels, and it takes so long to compute 122 components.
>>> >    (1) So I tried to use the option "ncomps" (say, 24) to reduce the
>>> number of components. However, an error message appears after computing:
>>> "Matrix dimensions must agree".
>>> >
>>> >    (2) Then I tried fastICA instead as following,
>>> >
>>> > ==================
>>> >  sz = size(EEG.data);
>>> >  nchans = sz(1);
>>> >  npts = sz(2);
>>> >  ntrials = sz(3);
>>> >  clear sz;
>>> >  nICs = 24;
>>> >  data = reshape(EEG.data,nchans,npts*ntrials);
>>> >  [ica,V,W] = fastica(data,'numOfIC',nICs,'approach','symm');
>>> >  EEG.icasphere = eye(nchans);
>>> >  EEG.icaact = single(reshape(ica,nICs,npts,ntrials));
>>> >  EEG.icawinv = V;
>>> >  EEG.icaweights = W;
>>> >  EEG = eeg_checkset( EEG );
>>> >  clear V W ica data;
>>> >
>>> >  EEG = pop_saveset( EEG, 'filename','test_raw_ica');
>>> > ==================
>>> >
>>> >    Everything seems fine. But when I reject the blink component via GUI
>>> of eeglab and load the data again, Something strange happens. It seems the
>>> amplitude of EEG.data become much smaller, about in  -1~1 range.  Thus I
>>> wonder whether there was some normalization behind, and how can I correct
>>> it? The problem would not repeat if I choose the number of components same
>>> as channels event in fast ICA (e.g. change to "nICs = nchans" in the above
>>> code).
>>> >
>>> >    Thank you for your help.
>>> >
>>> > best
>>> > Peng
>>> > _______________________________________________
>>> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>>> > To unsubscribe, send an empty email to
>>> eeglablist-unsubscribe at sccn.ucsd.edu
>>> > For digest mode, send an email with the subject "set digest mime" to
>>> eeglablist-request at sccn.ucsd.edu
>>>
>>>
>>>
>>> --------------------------------------------------------------------------------
>>>
>>> Joseph Dien,
>>> Senior Research Scientist
>>> University of Maryland
>>> 7005 52nd Avenue
>>> College Park, MD 20742-0025
>>>
>>> E-mail: jdien07 at mac.com
>>> Phone: 301-226-8848
>>> Fax: 301-226-8811
>>> http://homepage.mac.com/jdien07/
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>>> To unsubscribe, send an empty email to
>>> eeglablist-unsubscribe at sccn.ucsd.edu
>>> For digest mode, send an email with the subject "set digest mime" to
>>> eeglablist-request at sccn.ucsd.edu
>>>
>>
>>
>>
>> --
>> Scott Makeig, Research Scientist and Director, Swartz Center for
>> Computational Neuroscience, Institute for Neural Computation, University of
>> California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott<http://sccn.ucsd.edu/%7Escott>
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>>
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Scott Makeig, Research Scientist and Director, Swartz Center for
Computational Neuroscience, Institute for Neural Computation, University of
California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100610/a4d9cbaa/attachment.html>

From dnemrodo at uwaterloo.ca  Fri Jun 11 08:16:49 2010
From: dnemrodo at uwaterloo.ca (Dan Nemrodov)
Date: Fri, 11 Jun 2010 11:16:49 -0400
Subject: [Eeglablist] Clustering options
Message-ID: <AANLkTimCv1UmYR-H9_bbIX7W5MtoXu_Gm2fWHw_BidGN@mail.gmail.com>

Hello,
I would like to perform analysis of my data on ICs. To accomplish that I run
cluster analysis. However, my results are unstable, since the same method
with the same number of clusters yields each time quite different clusters.
I wonder if anyone has any experience with that and what would be the best
method to set the number of the expected clusters?
Cheers,
Dan Nemrodov
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100611/cfdc0acb/attachment.html>

From HINNES-BROWN at bionicear.org  Thu Jun 10 17:51:55 2010
From: HINNES-BROWN at bionicear.org (Hamish INNES-BROWN)
Date: Fri, 11 Jun 2010 10:51:55 +1000
Subject: [Eeglablist] ERPLAB toolbox
Message-ID: <18C7138F7DC13F498AE19062AC987443028148F0DE@franklin.medoto.unimelb.edu.au>

Hi all, I was just wondering if anyone is using the "ERPLAB toolbox"

http://www.erpinfo.org/erplab/erplab-toolbox/view

It seems like it would be really useful to do standard ERP analyses of your data in addition to any EEGLAB type analysis, but it seems to have vanished, or not appeared.



Hamish Innes-Brown
Music & Pitch Project
Bionic Ear Institute
t: +61 3 9667 7529
f: +61 3 9667 7518
e: hinnes-brown at bionicear.org




From arno at ucsd.edu  Sat Jun 12 08:43:09 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Sat, 12 Jun 2010 17:43:09 +0200
Subject: [Eeglablist] ICA problem - using PCA to reduce number of
	'channels' for ICA
In-Reply-To: <AANLkTilgnxMnwrBvhIEIg6wFm0IQHGQzerRWkJ89H0di@mail.gmail.com>
References: <AANLkTilgnxMnwrBvhIEIg6wFm0IQHGQzerRWkJ89H0di@mail.gmail.com>
Message-ID: <713BA908-A1B8-4429-B717-54CBE67DB56B@ucsd.edu>

Dear Keith,

we have tried both (removing channels or using PCA) and both work well. PCA 64 is fine. You may just use 'pca', 64 when calling the "Run ICA" menu. Concerning the size of the weight and sphere matrix. If you use 64 PCA dimensions, your sphere matrix is going to be 133x133 and the weight matrix is going to be 64*133 so this fine (since ICA_activities = weights x sphere x data). You might wonder why we have two matrices (sphere and weights) instead of a single (weights x sphere) matrix, and to tell the truth, I wonder about that sometimes as well because this confuses users and, in the end, very rarely does anybody look at the sphere and weights matrices separately. But this is the way we originally implemented it and we have sticked to this original implementation for now.

Importantly, always remember than using PCA to decrease the number of dimension is not a good idea and should be avoided when possible. The reason is that when you apply PCA, you get 133 components and you decide to only consider the first 64 ones and ignore the  other ones. Although these first 64 components account for most of the data, the remaining components may account for the linear projection of brain source onto multiple data channels. Therefore using PCA paradoxically introduces some non-linear noise (paradoxically because PCA is a linear algorithm). In practice this does not seem to be critical but theoretically removing channels is sometimes preferable to using PCA.

Best regards,

A. Delorme

On Jun 9, 2010, at 10:10 PM, Keith Yoder wrote:

> I have a related question:
> 
> We are recording using 128-channel BioSemi cap, plus four ocular leads and two mastoid leads.  All of the data are referenced to the right mastoid (leaving EEG.data with 133 rows).  We would like to use ICA, but have found that 133 components is too many (e.g. a CNV which is clear in the electrodes does not fall out as a component, but can be isolated with PCA using 5 components).  
> 
> Have other users had success using PCA to pre-process the data before passing them to ICA, so as to reduce the number of components identified by the ICA algorithm?  
> 
> A possible solution that I have been trying to implement would use PCA to identify a more reasonable number of components (e.g. 64) to then pass to ICA (all using runica):
> 
> % after using runica with " ,'pca',64 " to define 64 PCAs
> % manually prepare component activations for ICA as if they were
> % data from EEG.data (as laid out in pop_runica)
> >> tmpdata = reshape(EEG.icaact(:,:,:),64,EEG.pnts*EEG.trials);
> >> tmpdata = tmpdata - repmat(mean(tmpdata,2), [1 size(tmpdata,2)]);
> >> [EEG.icaweights, EEG.icasphere] = runica(tmpdata, 'lrate', 0.001, 'interupt','on');
> 
> However, this call to runica returns the component weights and sphering for the ICA decomposition of the PCA-defined components, rather than ICA-defined weights and sphering for 64 ICs for our 133 channels. Has anyone had success with a technique along similar lines? 
> 
> Alternatively, I could simply remove electrodes from EEG.data until I am left with 64 channels, but I am wary of throwing out data entirely.
> 
> We are running EEGLAB 7.2.9.20b in 64-bit Matlab 7.10.0 (R2010a) in Mac OS X 10.5.7.
> 
> Many thanks,
> Keith
> --
> Keith Yoder
> Research Aide
> Laboratory for the Neuroscience of Autism
> Cornell University
> Ithaca, NY
> 574.215.9678
> 
> 
> On Mon, Feb 1, 2010 at 5:26 PM, Scott Makeig <smakeig at gmail.com> wrote:
> Joe is correct that ICA will not converge if the rank of the data matrix is less than the number of channels. The runica/binica algorithms are supposed to test the rank of the input data. If two channels are identical, or if some subset of n channels are otherwise interdependent, then the rank will be less than the number of channels and PCA reduction should be applied to remove the redundancy and allow the ICA decomposition to converge. 
> 
> Arno -- There was a problem with the Matlab rank() function on 64-bit machines, I believe. Has this been solved and Is the auto rank detection -> PCA option currently implemented in runica/binica?  Perhaps we could add a 'toy' rank() function pre-test (e.g. finding the rank of a small full-rank matrix to detect if rank() is working...) ? If so, run the rank test; if not, then warn the user or build a work-around rank function that will work properly?
> 
> Scott
> 
> 
> On Mon, Feb 1, 2010 at 6:56 AM, Joseph Dien <jdien07 at mac.com> wrote:
> When you say "so long", how long do you mean?  While ICA is not by its nature a fast procedure, certain datasets can take much longer than usual.  For example, I find that if two channels are perfectly correlated (1 or -1) then an ICA run will take much longer.  This can happen if the data is mean mastoid referenced and both channels are explicitly included in the data because they will have a perfect -1 correlation (see Dien, 1998 for reference issues).  It can also happen if a channel is shorted out during acquisition and the reference channel is explicitly included because then they will have a perfect correlation.  Also if two channels are shorted together during the data acquisition they will be perfectly correlated with each other.  My EP Toolkit (https://sourceforge.net/projects/erppcatoolkit/) has code for dealing with these situations so you might want to look into it.  It implements an automated artifact correction routine that relies on EEGlab's runICA code, among!
>  other things.
> 
> Cheers!
> 
> Joe
> 
> 
> 
> On Jan 29, 2010, at 4:49 AM, peng wang wrote:
> 
> > Hi there,
> >
> >    I am using ICA to remove blinks via EEGLab. My dataset has 122 channels, and it takes so long to compute 122 components.
> >    (1) So I tried to use the option "ncomps" (say, 24) to reduce the number of components. However, an error message appears after computing: "Matrix dimensions must agree".
> >
> >    (2) Then I tried fastICA instead as following,
> >
> > ==================
> >  sz = size(EEG.data);
> >  nchans = sz(1);
> >  npts = sz(2);
> >  ntrials = sz(3);
> >  clear sz;
> >  nICs = 24;
> >  data = reshape(EEG.data,nchans,npts*ntrials);
> >  [ica,V,W] = fastica(data,'numOfIC',nICs,'approach','symm');
> >  EEG.icasphere = eye(nchans);
> >  EEG.icaact = single(reshape(ica,nICs,npts,ntrials));
> >  EEG.icawinv = V;
> >  EEG.icaweights = W;
> >  EEG = eeg_checkset( EEG );
> >  clear V W ica data;
> >
> >  EEG = pop_saveset( EEG, 'filename','test_raw_ica');
> > ==================
> >
> >    Everything seems fine. But when I reject the blink component via GUI of eeglab and load the data again, Something strange happens. It seems the amplitude of EEG.data become much smaller, about in  -1~1 range.  Thus I wonder whether there was some normalization behind, and how can I correct it? The problem would not repeat if I choose the number of components same as channels event in fast ICA (e.g. change to "nICs = nchans" in the above code).
> >
> >    Thank you for your help.
> >
> > best
> > Peng
> > _______________________________________________
> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> > To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> > For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
> 
> 
> --------------------------------------------------------------------------------
> 
> Joseph Dien,
> Senior Research Scientist
> University of Maryland
> 7005 52nd Avenue
> College Park, MD 20742-0025
> 
> E-mail: jdien07 at mac.com
> Phone: 301-226-8848
> Fax: 301-226-8811
> http://homepage.mac.com/jdien07/
> 
> 
> 
> 
> 
> 
> 
> 
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
> 
> 
> 
> -- 
> Scott Makeig, Research Scientist and Director, Swartz Center for Computational Neuroscience, Institute for Neural Computation, University of California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100612/fd620bb8/attachment.html>

From arno at ucsd.edu  Sat Jun 12 08:48:34 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Sat, 12 Jun 2010 17:48:34 +0200
Subject: [Eeglablist] datasets automatically saving as matlab.mat?
In-Reply-To: <AANLkTin9sby8Y0ujvChplpq0Rfxi7wo9vxN5MhVtO6WJ@mail.gmail.com>
References: <AANLkTin9sby8Y0ujvChplpq0Rfxi7wo9vxN5MhVtO6WJ@mail.gmail.com>
Message-ID: <B9C60A70-2266-4955-B3A1-AEB76EDC2C2B@ucsd.edu>

Dear Krysta,

you are probably using the menu "Save current dataset(s)" without having saved it prior to that. If you use "Save current dataset as" it will be fine. We will fix that so that the function pops a window to enter the file name.

Best regards,

A. Delorme


On Jun 10, 2010, at 4:20 AM, Krysta Chauncey wrote:

> All of a sudden this matlab session, whenever I try to save a dataset, I get this message:
> 
> "Note that your memory options for saving datasets does not correspond
> to the format of the datasets on disk (ignoring memory options)
> Saving dataset...
> 
> Saving to: matlab.mat
> 
> Done."
> 
> What does that mean, how did it happen, and how do I fix it?
> 
> cheers,
> Krysta
> 
> ----------------------------
> Krysta Chauncey, Ph.D.
> Brain-Computer Interface Project
> Human-Computer Interaction Lab
> Computer Science Dept, Tufts University
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu




From jkhartshorne at gmail.com  Sat Jun 12 13:25:37 2010
From: jkhartshorne at gmail.com (Joshua Hartshorne)
Date: Sat, 12 Jun 2010 16:25:37 -0400
Subject: [Eeglablist] rejecting channels for individual epochs
Message-ID: <AANLkTinLcaOJj8a10u9rXnooYFz5Gx8CLPq3PKh1l51y@mail.gmail.com>

When looking for artifacts, I notice that I often have epochs marked as bad
(by one method or another) due truly abnormal activity in only a single
electrode. If this happens only once or twice throughout the experiment for
that electrode, the channel is unlikely to be marked as bad.

I hate to toss out an entire trial due to a single electrode. Is there a way
to eliminate that channel only from that epoch and then interpolate?

Josh
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100612/f3242bf5/attachment.html>

From sjluck at ucdavis.edu  Mon Jun 14 09:48:44 2010
From: sjluck at ucdavis.edu (Steve Luck)
Date: Mon, 14 Jun 2010 09:48:44 -0700
Subject: [Eeglablist] ERPLAB toolbox
In-Reply-To: <mailman.0.1276282802.10010.eeglablist@sccn.ucsd.edu>
References: <mailman.0.1276282802.10010.eeglablist@sccn.ucsd.edu>
Message-ID: <7CAB95DC-4346-4A46-A8F0-6397EB5D7041@ucdavis.edu>

ERPLAB Toolbox will be coming soon.  We've delayed the release of the public beta to make sure that the core of the toolbox is strong, but we're nearly ready to release it.

> 
> From: Hamish INNES-BROWN <HINNES-BROWN at bionicear.org>
> Date: June 10, 2010 5:51:55 PM PDT
> To: "eeglablist at sccn.ucsd.edu" <eeglablist at sccn.ucsd.edu>
> Subject: [Eeglablist] ERPLAB toolbox
> 
> 
> Hi all, I was just wondering if anyone is using the "ERPLAB toolbox"
> 
> http://www.erpinfo.org/erplab/erplab-toolbox/view
> 
> It seems like it would be really useful to do standard ERP analyses of your data in addition to any EEGLAB type analysis, but it seems to have vanished, or not appeared.
> 
> 
> 
> Hamish Innes-Brown
> Music & Pitch Project
> Bionic Ear Institute
> t: +61 3 9667 7529
> f: +61 3 9667 7518
> e: hinnes-brown at bionicear.org
> 
> 
> 
> 
> 
> _______________________________________________
> eeglablist mailing list eeglablist at sccn.ucsd.edu
> Eeglablist page: http://www.sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsub at sccn.ucsd.edu
> To switch to non-digest mode, send an empty email to eeglablist-nodigest at sccn.ucsd.edu

--------------------------------------------------------------------
Steven J. Luck, Ph.D.
Interim Director, Center for Mind & Brain
Professor, Department of Psychology
University of California, Davis
Room 127
267 Cousteau Place
Davis, CA 95618
(530) 297-4424
sjluck at ucdavis.edu
Web: http://mindbrain.ucdavis.edu/people/sjluck
Calendar: http://www.google.com/calendar/embed?src=stevenjluck%40gmail.com&ctz=America/Los_Angeles
--------------------------------------------------------------------






-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100614/fbb0ccea/attachment.html>

From imponderabilion at gmail.com  Mon Jun 14 10:30:16 2010
From: imponderabilion at gmail.com (=?ISO-8859-2?Q?Miko=B3aj_Magnuski?=)
Date: Mon, 14 Jun 2010 19:30:16 +0200
Subject: [Eeglablist] Problems with creating STUDY file (function name must
	be a string)
Message-ID: <AANLkTikKpm0bEIpo0Hx0B6zXGt0lU6Io_KDxj1RWRBj7@mail.gmail.com>

Hello,

I am a fresh and happy user of EEGLAB v8.0.3.5b running on MATLAB 7.0.1 on
Windows XP.
I have recently encountered a problem that keeps me away from completing eeg
analysis.

When I try to create a new study (using all loaded datasets) a following
warning pops up:

'Function name must be a string'

All sets have channel locations, are divided into epochs and before trying
to create a study eeglab was working correct.

Do you have any idea what could have gone wrong and what are the possible
ways of correcting it?

I would be glad to give additional info, I just don't know what sort of
information would be relevant.

Miko?aj Magnuski
student of cognitive neuroscience
Warsaw School of Social Sciences and Humanities
Poland
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100614/7a465a66/attachment.html>

From leppert at eye-square.de  Mon Jun 14 03:38:17 2010
From: leppert at eye-square.de (Philipp Leppert)
Date: Mon, 14 Jun 2010 12:38:17 +0200
Subject: [Eeglablist] Computing component time/frequency transforms FFT?
In-Reply-To: <AANLkTinLcaOJj8a10u9rXnooYFz5Gx8CLPq3PKh1l51y@mail.gmail.com>
References: <AANLkTinLcaOJj8a10u9rXnooYFz5Gx8CLPq3PKh1l51y@mail.gmail.com>
Message-ID: <6A4478EC-86D3-4C2B-A914-649C257D7CF3@eye-square.de>

Hi,

I am looking for simple frequency bands FFT visualization and exports in eeglab, without localization.
Do I have to specify the the different bands like alpha myself?

Thank you for your help
Philipp



Am 12.06.2010 um 22:25 schrieb Joshua Hartshorne:

> When looking for artifacts, I notice that I often have epochs marked as bad (by one method or another) due truly abnormal activity in only a single electrode. If this happens only once or twice throughout the experiment for that electrode, the channel is unlikely to be marked as bad. 
> 
> I hate to toss out an entire trial due to a single electrode. Is there a way to eliminate that channel only from that epoch and then interpolate?
> 
> Josh
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu


EVENTS:

Eye Tracking BOF Event, 6:00-9:30 pm, June 22, Santa Clara, California:
http://www.baychi.org/
Listen to Andreas Thoelke's Talk on the i? Visualizer System!

........................................................................

... PHILIPP LEPPERT
    Director Planning
    
... eye square GmbH
    Schlesische Str. 29-30 (F)
    D-10997 Berlin
    Germany

... Fon +49 30 698144-22
    Fax +49 30 698144-10

... http://www.eye-square.com
    leppert at eye-square.com

........................................................................


Geschaeftsfuehrer/Managing Partner: Sabrina Duda, Michael Schiessl
Steuernummer/Tax No.: 29/506/02764
USt.Id.Nr./VAT No.: DE210454419
Registergericht/Court of Registry: Charlottenburg HRB 76686




From brice.rebsamen at gmail.com  Tue Jun 15 01:05:02 2010
From: brice.rebsamen at gmail.com (brice rebsamen)
Date: Tue, 15 Jun 2010 16:05:02 +0800
Subject: [Eeglablist] Decade of the Mind (DOM) 2010
Message-ID: <AANLkTinOzuuyWsmyPwjNTKQl3SnoI9jAsPZcOSm3yHZp@mail.gmail.com>

Dear fellow eeglab users

The Decade of the Mind (DOM) project is an international initiative to
advance our scientific understanding of how the mind and complex
behaviors are related to the activity of human brains. This year the
annual meeting will be held in Singapore, 18-20 Oct 2010.

The following speakers are invited to give a talk: J. Olds, J.L.
McClelland, R. Gallistel, D. Purves, K. Mogi, N. Thakor, J. Weng, C.
Guan, H. Ishiguro, C. Koch, M. Spitzer, C.A. Czeisler and N.
Birbaumer.

More information and registration here: http://dom-6.org

Regards
Brice Rebsamen


From palmeida at fpce.up.pt  Wed Jun 16 11:56:28 2010
From: palmeida at fpce.up.pt (Pedro R. Almeida)
Date: Wed, 16 Jun 2010 16:56:28 -0200 (CEST)
Subject: [Eeglablist] Number of subjects on study
Message-ID: <1509.172.16.52.169.1276700188.squirrel@webmail.fpce.up.pt>

Hi,

We?re just starting to use EEGLAB to process a few files from two groups of
subjects. Everything has worked fine up to the point when we had to plot channel
grandaveraged ERPs and get channel statistics for the ERP timecourse. It seems
EEGLAB won?t accept more than 37 subjects per study. As soon as we include a
38th subject and ask for a channel ERP plot we get the following message:


Reading erp data...??? Index exceeds matrix dimensions.

Error in ==> std_readerp>std_readerpsub at 485
    if isfield(erpstruct, 'labels'), chanlab{k} = erpstruct.labels{k}; end;

Error in ==> std_readerp at 220
                    if strcmpi(dtype, 'erp') alldata{c, g} = std_readerpsub(
ALLEEG, setinds{c,g}(:), allinds{c,g}(:), opt.timerange)';

Error in ==> std_erpplot at 322
        [STUDY erpdata alltimes] = std_readerp(STUDY, ALLEEG, 'channels',
opt.channels, 'timerange', opt.timerange, ...

Error in ==> pop_chanplot at 302
            eval(a); STUDY.history =  sprintf('%s\n%s',  STUDY.history, a);



We've searched the mailing list and found no such problem, so we must be getting
something basic wrong.
We?re not getting any further problems with other plots or measures on the
channel or components analysis. Any help with this would be much appreciated.

Thank you,

Pedro Almeida



-- 

Pedro R. Almeida

Neuropsychophysiology Laboratory
Faculty of Psychology and Education Sciences
University of Porto
Rua Dr. Manuel Pereira da Silva
4200-392 Porto (Portugal)
Tel. +351 226079700 (ext. 301)
Site: http://www.fpce.up.pt/labpsi/
Email: palmeida at fpce.up.pt



-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.



From krysta.chauncey at tufts.edu  Fri Jun 18 14:21:29 2010
From: krysta.chauncey at tufts.edu (Krysta Chauncey)
Date: Fri, 18 Jun 2010 17:21:29 -0400
Subject: [Eeglablist] checkset converts data to single precision if not
	matlab r11 r12 or r13?
Message-ID: <AANLkTikqOwzVm9GlajWK-G-qR1ZYbqdCQmorA7epZfrj@mail.gmail.com>

Looks like checkset is converting data to single precision if the Matlab
version used isn't R11-13; is this on purpose?  Looks like filtfilt won't
take single-precision data, from some of its early parameter-checking.

-K

----------------------------
Krysta Chauncey, Ph.D.
Brain-Computer Interface Project
Human-Computer Interaction Lab
Computer Science Dept, Tufts University
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100618/5d0f4766/attachment.html>

From d.arisi at ospedale.cremona.it  Sat Jun 19 01:32:48 2010
From: d.arisi at ospedale.cremona.it (d.arisi at ospedale.cremona.it)
Date: Sat, 19 Jun 2010 10:32:48 +0200 (CEST)
Subject: [Eeglablist] basic question
Message-ID: <7479.81.88.234.182.1276936368.squirrel@webmail.ospedale.cremona.it>

I need to run ICA on epoched EEGs of different subjects
In each subject 2 different visual stimuli were presented, in random
order, during the same EEG session (named "condition 1" and "condition
2"); epochs corresponding to each type of condition were grouped (off
line) in 2 datasets for each subject
>From EEGLAB tutorial I understand that a correct ICA decomposition must be
done concatenating all epochs ("condition 1" and "condition 2") for each
subjects and not running ICA on every separated dataset
Am I right? some suggestion?

regards
Daniele Arisi







From clemens.brunner at TUGraz.at  Wed Jun 16 08:01:25 2010
From: clemens.brunner at TUGraz.at (Clemens Brunner)
Date: Wed, 16 Jun 2010 17:01:25 +0200
Subject: [Eeglablist] Announcement: SigViewer 0.4.0 released
Message-ID: <05D5498A-ECC4-4EF2-A1B9-B6B3463E52CF@TUGraz.at>

SigViewer 0.4.0 has been released. You can get binaries for Windows, Mac OS X, and Ubuntu 10.04 (32bit and 64bit) at:

http://sigviewer.sourceforge.net/download.html

Feedback, wishes, and bug reports are highly welcome. We are also looking for people who would like to join the development team.

In case you have never heard of SigViewer, here's a short description from the project website:

SigViewer is a powerful viewing application for biosignals, originally designed to display electroencephalographic (EEG) data. SigViewer now supports several biosignal data formats through libbiosig4c++ (for example GDF, EDF, CNT, EEG, and many more). Besides displaying biosignals, SigViewer supports creating annotations to select artifacts or specific events.

SigViewer is written in standard C++ using the platform-independent graphical user interface (GUI) toolkit Qt 4. SigViewer is available for all three major platforms (namely Windows, Mac OS X, and Linux) using only one common source tree. Moreover, it does not depend on any proprietary tools such as Matlab or the Microsoft Foundation Classes (MFC), making it a truly free open-source program licensed under the GPL.

SigViewer is developed by the Institute for Knowledge Discovery, Graz University of Technology (http://bci.tugraz.at/).

Clemens


From alenarto at ucla.edu  Wed Jun 16 16:00:12 2010
From: alenarto at ucla.edu (Agatha Lenartowicz)
Date: Wed, 16 Jun 2010 16:00:12 -0700
Subject: [Eeglablist] extracting cluster specs for each subject
In-Reply-To: <AANLkTimdVFlel6vJHI-F4L2B15UvrGYfDkDCDdI2b7Eh@mail.gmail.com>
References: <4beafb87.a225e30a.41c3.2824@mx.google.com>
	<AANLkTimdVFlel6vJHI-F4L2B15UvrGYfDkDCDdI2b7Eh@mail.gmail.com>
Message-ID: <7857EBEE-EE95-44AA-85F2-B5D97104F0B4@ucla.edu>

Hi all ~ I would like to extract the subject specific dat contributing to each cluster solution (from an MP clustering applied to ICs). Can you point me to the appropriate data structure or set of files where these live?

Many thanks
agatha


On May 14, 2010, at 7:21 AM, Jason Ralph wrote:

> We have the same problem.  
> 
> There are 2 solutions: 
> 
> Solution #1
> 
> 1. Export .raw file. 
> 
> 2. Create a waveform tool to export events. Select relative time.
> 
> 3. In Excel (or write a script), convert NetStation's funky time format to seconds.milliseconds.  
> 4. Import .raw file into EEGLAB
> 5. Import events into EEGLAB (import events from matlab array or text file)  Your formatted event text file must have the fields type and latency, but you can add any additional fields desired.  
> 
> If you need help, we have written scripts to semi-automate this process.
> 
> 
> Solution # 2
> 
> 1.  Purchase Amp Server Pro SDK from EGI for $8,000.  ASPSDK allows you to capture EGI's data without using NetStation
> 2.  Acquire data directly into BCI2000 or Matlab  itself.
> 3.  Never use NetStation again!!!!
> 
> 
> We've been using solution 1 for some time, but are seriously considering solution 2.  
> 
> 
> 2010/5/12 Pawe? Augustynowicz <pawel.augustynowicz at gmail.com>
> Dear All,
> 
>  
> I have a short question about import data from EGI simple binary file format with events. By default, EEGLAB imports EGI?s simple binary with events, but in only one dimension. I mean, that I can see only the name of the event. Additional data (event fields) are not imported. Does anyone have an idea of how to attach this data do EEGLAB? I assume that this information is written in additional channels. But EEGLAB deletes all of these channels except one with the basic info.
> 
> I tried to insert this data manually, by there is another problem. I have a few epochs in original NetStation recording. These epochs are essential because of the experiment design (experiment is too long and we have to do breaks). When I export this data from NetStation as simple binary, this file is being broken into files representing epochs in original recording. I can import this bunch of files with pop_readsegegi command. This function does the job well, except one little bug: NetStation counts time from the beginning of the experiment. Pauses during the recording are treated as the beginning of epochs. In EEGLAB all epochs have continuous time, so there is no possibility to attach events exported from NetStation, because the time after the first epoch is different.
> 
>  
> Does anyone ran on these problems and managed to resolve it?
> 
>  
> Regards,
> 
> Pawel Augustynowicz
> 
> The John Paul II Catholic University of Lublin
> 
>  
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
> 
> 
> 
> -- 
> Jason Ralph
> CogWorks Laboratory
> Cognitive Science Department
> Rensselaer Polytechnic Institute
> 110 8th Street, Troy, NY 12180
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100616/374779a3/attachment.html>

From arno at ucsd.edu  Mon Jun 21 07:05:31 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Mon, 21 Jun 2010 07:05:31 -0700
Subject: [Eeglablist] basic question
In-Reply-To: <7479.81.88.234.182.1276936368.squirrel@webmail.ospedale.cremona.it>
References: <7479.81.88.234.182.1276936368.squirrel@webmail.ospedale.cremona.it>
Message-ID: <DD362FBE-DE09-43C3-8D67-F36EFA2DECF1@ucsd.edu>

Dear Daniele,

this is correct. ICA should be run using the concatenated data from both conditions. The reason is that otherwise it is going to be harder to compare the ICA components in the two conditions or remove the ICA artifacts in the two conditions (if you run ICA on each dataset separately, someone may always argue that the difference you observe between the two conditions arises because you have selected or removed different ICA components in each condition).

Hope this helps,

Arno

On Jun 19, 2010, at 1:32 AM, d.arisi at ospedale.cremona.it wrote:

> I need to run ICA on epoched EEGs of different subjects
> In each subject 2 different visual stimuli were presented, in random
> order, during the same EEG session (named "condition 1" and "condition
> 2"); epochs corresponding to each type of condition were grouped (off
> line) in 2 datasets for each subject
>> From EEGLAB tutorial I understand that a correct ICA decomposition must be
> done concatenating all epochs ("condition 1" and "condition 2") for each
> subjects and not running ICA on every separated dataset
> Am I right? some suggestion?
> 
> regards
> Daniele Arisi
> 
> 
> 
> 
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu




From arno at ucsd.edu  Mon Jun 21 07:20:17 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Mon, 21 Jun 2010 07:20:17 -0700
Subject: [Eeglablist] checkset converts data to single precision if not
	matlab r11 r12 or r13?
In-Reply-To: <AANLkTikqOwzVm9GlajWK-G-qR1ZYbqdCQmorA7epZfrj@mail.gmail.com>
References: <AANLkTikqOwzVm9GlajWK-G-qR1ZYbqdCQmorA7epZfrj@mail.gmail.com>
Message-ID: <2D3F5B3F-D983-431C-986B-5C7672ADAD49@ucsd.edu>

Dear Krysta,


> Looks like checkset is converting data to single precision if the Matlab version used isn't R11-13; is this on purpose?


yes, this is on purpose to save memory, although you can force EEGLAB to use double precision by checking the appropriate checkbox in the memory menu. Also EEGLAB does not support version 11 (5.3) to 13 (6.5) any more, only version 7.x (2004 onward).


>  Looks like filtfilt won't take single-precision data, from some of its early parameter-checking.


Yes, we have heard about this but have not encountered this problem ourselves (except as a bug in Matlab 7.0.0 but I think a more recent version of Matlab is having this problem as well). We will change the function so that it converts data to double precision prior to running filtfilt. For now, you may simply toggle the memory option prior to filtering.

Arno


From g.rousselet at psy.gla.ac.uk  Mon Jun 21 09:24:43 2010
From: g.rousselet at psy.gla.ac.uk (Guillaume Rousselet)
Date: Mon, 21 Jun 2010 17:24:43 +0100
Subject: [Eeglablist] Frontiers in Perception Science: Call for
	participation in upcoming Special Topic
Message-ID: <3F37CD03-CE30-4CEE-A128-A9D9CDCD1437@psy.gla.ac.uk>

Hosting Journal: Frontiers in Perception Science

Topic Title: Single-trial analyses of behavioural and neuroimaging  
data in perception and decision-making.

Host Editors: Guillaume A. Rousselet, Cyril R. Pernet, Paul Sajda

Description:

The cognitive psychology of perception and decision-making is at a  
cross-road. Most studies still employ categorical designs, a priori  
classified stimuli and perform statistical evaluations across  
subjects. However, a shift has been observed in recent years towards  
parametric designs in which the information content of stimuli is  
systematically manipulated to study the single-trial dynamics of  
behaviour (reaction times, eye movements) and brain activity (EEG,  
MEG, fMRI). By using the information contained in the variance of  
individual trials, the single-trial approach goes beyond the activity  
of the average brain:  it reveals the specificity of information  
processing in individual subjects, across tasks and stimulus space,  
revealing both inter-individual commonalties and differences. This  
special issue provides theoretical and empirical support for the study  
of single-trial data.

Topics of particular interest include:


1.    description of the richness of information in single-trials and  
how it can be successfully extracted;
2.    statistical issues related to measures of central tendency,  
control for multiple comparisons, multivariate approaches,  
hierarchical modelling and characterization of individual differences;
3.    how manipulation of the stimulus space can allow for a direct  
mapping of stimulus properties onto brain activity to infer dynamics  
of information processing and information content of brain states;
4.    how results from different brain imaging techniques can be  
integrated at the single-trial level.


Abstract Submission Deadline: September 01, 2010


Article Submission Deadline: January 03, 2011


Link to Special Topic: http://www.frontiersin.org/psychology/perceptionscience/specialtopics/98/


The publishing fee for contributors amounts to ?900, and a further  
reduction to ?720 for Frontiers Associate Editors.

Like all research published with Frontiers, the articles will be  
freely available to all of our readers on our website, and all Special  
Topic articles receive a DOI and are indexed in the National Institute  
of Health?s electronic depository of full text articles, PubMed  
Central, and many other international archives (Google Scholar,  
Directory of Open Access Journal, Psych INFO). Frontiers is also in  
the process of being archived in Thompson Scientific (ISI), and Web of  
Science.

Frontiers has also announced that in the very near future, all  
published Special Topics will be available to view and download in an  
eBook format!

For more information, you may refer to the Frontiers? Special Topic  
page, where you can get further guidance and browse past Special Topics.
<< http://www.frontiersin.org/specialtopicspage/ >>

With best regards,

Guillaume Rousselet
Associate Editor, Frontiers in Perception Science
www.frontiersin.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100621/542fc845/attachment.html>

From andrew.jahner at gmail.com  Mon Jun 21 10:51:50 2010
From: andrew.jahner at gmail.com (Andrew Jahn)
Date: Mon, 21 Jun 2010 13:51:50 -0400
Subject: [Eeglablist] Compatibility with MEG Data
Message-ID: <AANLkTinyuz6DG_fI4RYR-gFMyae3RtwEfODAVRwXNU0A@mail.gmail.com>

Hi all,

I was wondering whether EEGlab could read in and analyze MEG data.  Does it
have to be in a specific file format, or is .meg4 enough?


Thanks,

-Andrew
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100621/33c3daa6/attachment.html>

From arno at ucsd.edu  Mon Jun 21 10:27:30 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Mon, 21 Jun 2010 10:27:30 -0700
Subject: [Eeglablist] Number of subjects on study
In-Reply-To: <1509.172.16.52.169.1276700188.squirrel@webmail.fpce.up.pt>
References: <1509.172.16.52.169.1276700188.squirrel@webmail.fpce.up.pt>
Message-ID: <3098FCAD-C0B3-4AB1-9C16-62042550A5E5@ucsd.edu>

Dear Almeida,

there is no reason that EEGLAB should not be able to process more than 37 subjects.
We are processing studies with more than 300 subjects. 

Your problem arises because you have precomputed some files using 37 subjects and then try to do it again using 38 subjects. EEGLAB seems to have a problem with recognizing your precomputed files. The easy way around is to check the "recompute" box in the pre-computing graphic interface. We are working on fixing the problem.

Thanks for your feedback.
Arno

On Jun 16, 2010, at 11:56 AM, Pedro R. Almeida wrote:

> Hi,
> 
> We?re just starting to use EEGLAB to process a few files from two groups of
> subjects. Everything has worked fine up to the point when we had to plot channel
> grandaveraged ERPs and get channel statistics for the ERP timecourse. It seems
> EEGLAB won?t accept more than 37 subjects per study. As soon as we include a
> 38th subject and ask for a channel ERP plot we get the following message:
> 
> 
> Reading erp data...??? Index exceeds matrix dimensions.
> 
> Error in ==> std_readerp>std_readerpsub at 485
>    if isfield(erpstruct, 'labels'), chanlab{k} = erpstruct.labels{k}; end;
> 
> Error in ==> std_readerp at 220
>                    if strcmpi(dtype, 'erp') alldata{c, g} = std_readerpsub(
> ALLEEG, setinds{c,g}(:), allinds{c,g}(:), opt.timerange)';
> 
> Error in ==> std_erpplot at 322
>        [STUDY erpdata alltimes] = std_readerp(STUDY, ALLEEG, 'channels',
> opt.channels, 'timerange', opt.timerange, ...
> 
> Error in ==> pop_chanplot at 302
>            eval(a); STUDY.history =  sprintf('%s\n%s',  STUDY.history, a);
> 
> 
> 
> We've searched the mailing list and found no such problem, so we must be getting
> something basic wrong.
> We?re not getting any further problems with other plots or measures on the
> channel or components analysis. Any help with this would be much appreciated.
> 
> Thank you,
> 
> Pedro Almeida
> 
> 
> 
> -- 
> 
> Pedro R. Almeida
> 
> Neuropsychophysiology Laboratory
> Faculty of Psychology and Education Sciences
> University of Porto
> Rua Dr. Manuel Pereira da Silva
> 4200-392 Porto (Portugal)
> Tel. +351 226079700 (ext. 301)
> Site: http://www.fpce.up.pt/labpsi/
> Email: palmeida at fpce.up.pt
> 
> 
> 
> -- 
> This message has been scanned for viruses and
> dangerous content by MailScanner, and is
> believed to be clean.
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu




From arno at ucsd.edu  Tue Jun 22 14:53:02 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Tue, 22 Jun 2010 14:53:02 -0700
Subject: [Eeglablist] =?windows-1252?q?POSTDOCTORAL_POSITION_-_INSERM=92s_?=
 =?windows-1252?q?_=91Cognitive_Neuroimaging_Unit=27?=
Message-ID: <5972F55E-BBF8-4810-B19D-1666AC4909BB@ucsd.edu>

POSTDOCTORAL POSITION - Applications are invited for a postdoc position supervised by Ghislaine Dehaene-Lambertz to work on consciousness in infants using EEG. The team is part of INSERM?s  ?Cognitive Neuroimaging Unit' (http://www.unicog.org <http://www.unicog.org/> , director : Stanislas Dehaene) at NeuroSpin (director : Denis LeBihan) in the greater Paris region. NeuroSpin is a newly opened outstanding interdisciplinary research environment that houses several research laboratories and combines expertise in cognitive neuroscience and neuropsychology, magneto-electrophysiology, high field MR imaging and imaging data analysis. The project is part of an European community project to study consciousness in adults, monkeys, infants and comatose patients. The postdoc will program and analyse EEG experiments (subliminal presentation, stimulus collision, etc..) in infants and discuss the results with the other teams involved. Applicants should have a PhD degree in Neuroscience, Medicine, Psychology, or related areas. Prior experience with skills in EEG/MEG analysis is welcome. Salary will be commensurate with experience within the salary scale of the French public research organisations (~2100 euros per month). The position is funded for one to five years, and should be started during autumn 2010. Applications will be considered until the position is filled. For further information or to submit an application (including the names of two referees) please contact Ghislaine Dehaene-Lambertz, email: ghislaine.dehaene at cea.fr
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100622/0d04cdb2/attachment.html>

From andrew.jahner at gmail.com  Thu Jun 24 11:18:18 2010
From: andrew.jahner at gmail.com (Andrew Jahn)
Date: Thu, 24 Jun 2010 14:18:18 -0400
Subject: [Eeglablist] Opening MEG Files
Message-ID: <AANLkTin8M9aTlpvu7xZO-4RCkGz6H8uJnjWmZnPeTwaU@mail.gmail.com>

Hi,

We have been using BESA to process MEG data, but we are just now trying to
also look at it in EEGlab.  When opening up a MEG file, it appears to want a
.ds folder.  There are two .ds folders for each run in each subject's
directory, which contain a .acq, .hc, .ist, .meg4, .newds, .res4, and .trig
file.  When we try to read one of these into EEGlab, we get the following
error:

"Output argument "markers" (and maybe others) not assiged during call to
"/path/to/ctf_read_marker_file""

Is there a specific way that the .ds folder needs to be set up?  There is
another .meg4 file in each subject's folder which is much larger, and is the
file we have been reading into BESA.  However, EEGlab seems to want an
entire folder, and we were unsure whether it needs to contain specific files
for EEGlab to work.


Thanks,

-Andy
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100624/a1412f64/attachment.html>

From Alejo.Keuroghlanian at iit.it  Mon Jun 28 08:40:41 2010
From: Alejo.Keuroghlanian at iit.it (Alejo Keuroghlanian)
Date: Mon, 28 Jun 2010 17:40:41 +0200
Subject: [Eeglablist] How to keep track of epochs marked for rejection?
Message-ID: <FC14DB6243548F4683D9E3D7E9A51C381C8C16EE83@IITMXMBX.iit.local>

Dear all,

suppose I'm working with the GUI, with a data set consisting of 20 epochs

- I choose "Tools > Reject data epochs > reject data (all methods)", and by visual inspection I mark trials #1 and #3 for rejection

- then I press "Close (keep marks)"

- I clear the data set from memory

- then I load it again... 

Are these marks stored somewhere in the data set/EEG structure? I have tried this but the marks disappear... I would like to know which were the epochs that I rejected eventually. 

Thank you in advance. Cheers,

Alejo


From saim_rasheed at hotmail.com  Tue Jun 29 07:31:35 2010
From: saim_rasheed at hotmail.com (Saim Rasheed)
Date: Tue, 29 Jun 2010 20:31:35 +0600
Subject: [Eeglablist] Numerical Values in STUDY Structure
Message-ID: <SNT119-W63F22C0721A6DD0F8D18A6EACB0@phx.gbl>


 

Hi,

 

I am working with STUDY structure and computed ERSP, ITC and. I can plot them by a single click and analyse them.

Where I can find the numerical values for ERSP and ITC ? Where these values are stored?

 

I need them for further computations. Please help.

 

Thanks.

 

Saim
 		 	   		  
_________________________________________________________________
Your E-mail and More On-the-Go. Get Windows Live Hotmail Free.
https://signup.live.com/signup.aspx?id=60969
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100629/7f058417/attachment.html>

From julie at sccn.ucsd.edu  Tue Jun 29 10:46:52 2010
From: julie at sccn.ucsd.edu (Julie Onton)
Date: Tue, 29 Jun 2010 10:46:52 -0700 (PDT)
Subject: [Eeglablist] Numerical Values in STUDY Structure
In-Reply-To: <SNT119-W63F22C0721A6DD0F8D18A6EACB0@phx.gbl>
References: <SNT119-W63F22C0721A6DD0F8D18A6EACB0@phx.gbl>
Message-ID: <2769.159.71.184.228.1277833612.squirrel@sccn.ucsd.edu>

Good question.

Below is a script to help you find, load and plot your raw data.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% LOAD raw data:------------------------
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
basedir = '/home/you/whereyourdatais/';

subjs =
{'S01','S02','S03','S04','S05','S06','S07','S08','S09','S10','S11','S12','S13'};

subj = 3; % who do you want to plot?
cond = 1; % what condition number?

% for example, load ERSP:
load_string =
[basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icaersp'];

% OR ITC
load_string =
[basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icaitc'];

% OR spectra:
load_string =
[basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icaspec'];

% OR topomaps:
load_string =
[basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icatopo'];

% load the raw data into the variable 'RAWdata':
RAWdata = load('-mat',load_string);

% Mask an ERSP using calculated bootstrap limits (if you calculated them):

ic = 8; % choose IC to plot

oneic = ['RAWdata.comp',int2str(ic),'_ersp'];
oneic = eval(oneic);
% in case you want to see the removed baseline:
onebase = ['RAWdata.comp',int2str(ic),'_erspbase'];
onebase = eval(onebase);
% load the bootstrap significance limits:
oneboot = ['RAWdata.comp',int2str(ic),'_erspboot'];
oneboot = eval(oneboot);

maskERSP = oneic;
% zero out non-sig values:
maskERSP(find(oneic > repmat(oneboot(:,1),[1 size(oneic,2)])& oneic <
repmat(oneboot(:,2),[1 size(oneic,2)]))) = 0;

clim = 4; % set +/- color limits
figure; imagesc(RAWdata.times,RAWdata.freqs,maskERSP,[-clim clim]);
set(gca,'ydir','norm');
title(['Subj ',int2str(subj),'; IC ',int2str(ic),'; ',STUDY.condition{cond}]);
cbar;


Hope this helps,  Julie

-- 
Julie Onton, PhD
http://sccn.ucsd.edu/~julie

>
>
>
> Hi,
>
>
>
> I am working with STUDY structure and computed ERSP, ITC and. I can plot them
> by a single click and analyse them.
>
> Where I can find the numerical values for ERSP and ITC ? Where these values
> are stored?
>
>
>
> I need them for further computations. Please help.
>
>
>
> Thanks.
>
>
>
> Saim
>
> _________________________________________________________________
> Your E-mail and More On-the-Go. Get Windows Live Hotmail Free.
> https://signup.live.com/signup.aspx?id=60969_______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu



From d.steines at gmail.com  Tue Jun 29 14:23:25 2010
From: d.steines at gmail.com (David Steines)
Date: Tue, 29 Jun 2010 16:23:25 -0500
Subject: [Eeglablist] How to keep track of epochs marked for rejection?
In-Reply-To: <FC14DB6243548F4683D9E3D7E9A51C381C8C16EE83@IITMXMBX.iit.local>
References: <FC14DB6243548F4683D9E3D7E9A51C381C8C16EE83@IITMXMBX.iit.local>
Message-ID: <AANLkTikZkuAvzE3IHUK8D09O-Uy3Okezgut5JGxFhwpc@mail.gmail.com>

Hello Alejo,

I was looking into this problem yesterday, and the only record that I could
find was in EEG.history. If you open the history parameter, you will find a
call to pop_rejepoch( EEG, [indices of rejected epochs] ,0). The second
parameter lists the indices of the epochs that you rejected.

I'm not aware of any other parameter that holds this information, but if
there is one, I'd be interested to learn what it is as well.

Hope this helps,

David



On Mon, Jun 28, 2010 at 10:40 AM, Alejo Keuroghlanian <
Alejo.Keuroghlanian at iit.it> wrote:

> Dear all,
>
> suppose I'm working with the GUI, with a data set consisting of 20 epochs
>
> - I choose "Tools > Reject data epochs > reject data (all methods)", and by
> visual inspection I mark trials #1 and #3 for rejection
>
> - then I press "Close (keep marks)"
>
> - I clear the data set from memory
>
> - then I load it again...
>
> Are these marks stored somewhere in the data set/EEG structure? I have
> tried this but the marks disappear... I would like to know which were the
> epochs that I rejected eventually.
>
> Thank you in advance. Cheers,
>
> Alejo
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100629/f240a228/attachment.html>

From julie at sccn.ucsd.edu  Tue Jun 29 16:21:21 2010
From: julie at sccn.ucsd.edu (Julie Onton)
Date: Tue, 29 Jun 2010 16:21:21 -0700 (PDT)
Subject: [Eeglablist] How to keep track of epochs marked for rejection?
In-Reply-To: <AANLkTikZkuAvzE3IHUK8D09O-Uy3Okezgut5JGxFhwpc@mail.gmail.com>
References: <FC14DB6243548F4683D9E3D7E9A51C381C8C16EE83@IITMXMBX.iit.local>
	<AANLkTikZkuAvzE3IHUK8D09O-Uy3Okezgut5JGxFhwpc@mail.gmail.com>
Message-ID: <4197.159.71.184.228.1277853682.squirrel@sccn.ucsd.edu>

This is the correct structure element where reject marks are stored, but you
have to save the dataset in order for these marks to be preserved if the
dataset is cleared.

Julie

-- 
Julie Onton, PhD
http://sccn.ucsd.edu/~julie

> Hello Alejo,
>
> I was looking into this problem yesterday, and the only record that I could
> find was in EEG.history. If you open the history parameter, you will find a
> call to pop_rejepoch( EEG, [indices of rejected epochs] ,0). The second
> parameter lists the indices of the epochs that you rejected.
>
> I'm not aware of any other parameter that holds this information, but if
> there is one, I'd be interested to learn what it is as well.
>
> Hope this helps,
>
> David
>
>
>
> On Mon, Jun 28, 2010 at 10:40 AM, Alejo Keuroghlanian <
> Alejo.Keuroghlanian at iit.it> wrote:
>
>> Dear all,
>>
>> suppose I'm working with the GUI, with a data set consisting of 20 epochs
>>
>> - I choose "Tools > Reject data epochs > reject data (all methods)", and by
>> visual inspection I mark trials #1 and #3 for rejection
>>
>> - then I press "Close (keep marks)"
>>
>> - I clear the data set from memory
>>
>> - then I load it again...
>>
>> Are these marks stored somewhere in the data set/EEG structure? I have
>> tried this but the marks disappear... I would like to know which were the
>> epochs that I rejected eventually.
>>
>> Thank you in advance. Cheers,
>>
>> Alejo
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu



From arno at ucsd.edu  Tue Jun 29 17:52:09 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Tue, 29 Jun 2010 17:52:09 -0700
Subject: [Eeglablist] Numerical Values in STUDY Structure
In-Reply-To: <2769.159.71.184.228.1277833612.squirrel@sccn.ucsd.edu>
References: <SNT119-W63F22C0721A6DD0F8D18A6EACB0@phx.gbl>
	<2769.159.71.184.228.1277833612.squirrel@sccn.ucsd.edu>
Message-ID: <F6FEC69A-873D-41E7-853D-AEB358661734@ucsd.edu>

Another simple solution is to add additional output to the STUDY plotting functions after copying them from the history.
For instance, after creating or loading an EEGLAB STUDY, then precomputing the ERP for data channel, you may use the channel plotting interface to plot the ERP for electrode CZ (for instance). Then, the EEGLAB history contains the line

STUDY = std_erpplot(STUDY, ALLEEG, 'channels', { 'cz' }); 

Then, looking at the std_erpplot function help, you may add additional outputs

[STUDY erp erptimes ] = std_erpplot(STUDY, ALLEEG, 'channels', { 'cz' }); 

Note that the "erp" output will be a cell array containing an array for the ERP of each condition/group. If one subject only is present, each array will be a vector of values for the ERP of this subject in a given condition. If several subjects are present, each column of the array will contain the ERP for a given subject.

Arno

On Jun 29, 2010, at 10:46 AM, Julie Onton wrote:

> Good question.
> 
> Below is a script to help you find, load and plot your raw data.
> 
> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
> % LOAD raw data:------------------------
> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
> basedir = '/home/you/whereyourdatais/';
> 
> subjs =
> {'S01','S02','S03','S04','S05','S06','S07','S08','S09','S10','S11','S12','S13'};
> 
> subj = 3; % who do you want to plot?
> cond = 1; % what condition number?
> 
> % for example, load ERSP:
> load_string =
> [basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icaersp'];
> 
> % OR ITC
> load_string =
> [basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icaitc'];
> 
> % OR spectra:
> load_string =
> [basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icaspec'];
> 
> % OR topomaps:
> load_string =
> [basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icatopo'];
> 
> % load the raw data into the variable 'RAWdata':
> RAWdata = load('-mat',load_string);
> 
> % Mask an ERSP using calculated bootstrap limits (if you calculated them):
> 
> ic = 8; % choose IC to plot
> 
> oneic = ['RAWdata.comp',int2str(ic),'_ersp'];
> oneic = eval(oneic);
> % in case you want to see the removed baseline:
> onebase = ['RAWdata.comp',int2str(ic),'_erspbase'];
> onebase = eval(onebase);
> % load the bootstrap significance limits:
> oneboot = ['RAWdata.comp',int2str(ic),'_erspboot'];
> oneboot = eval(oneboot);
> 
> maskERSP = oneic;
> % zero out non-sig values:
> maskERSP(find(oneic > repmat(oneboot(:,1),[1 size(oneic,2)])& oneic <
> repmat(oneboot(:,2),[1 size(oneic,2)]))) = 0;
> 
> clim = 4; % set +/- color limits
> figure; imagesc(RAWdata.times,RAWdata.freqs,maskERSP,[-clim clim]);
> set(gca,'ydir','norm');
> title(['Subj ',int2str(subj),'; IC ',int2str(ic),'; ',STUDY.condition{cond}]);
> cbar;
> 
> 
> Hope this helps,  Julie
> 
> -- 
> Julie Onton, PhD
> http://sccn.ucsd.edu/~julie
> 
>> 
>> 
>> 
>> Hi,
>> 
>> 
>> 
>> I am working with STUDY structure and computed ERSP, ITC and. I can plot them
>> by a single click and analyse them.
>> 
>> Where I can find the numerical values for ERSP and ITC ? Where these values
>> are stored?
>> 
>> 
>> 
>> I need them for further computations. Please help.
>> 
>> 
>> 
>> Thanks.
>> 
>> 
>> 
>> Saim
>> 
>> _________________________________________________________________
>> Your E-mail and More On-the-Go. Get Windows Live Hotmail Free.
>> https://signup.live.com/signup.aspx?id=60969_______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu




From arno at ucsd.edu  Tue Jun 29 17:55:19 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Tue, 29 Jun 2010 17:55:19 -0700
Subject: [Eeglablist] How to keep track of epochs marked for rejection?
In-Reply-To: <FC14DB6243548F4683D9E3D7E9A51C381C8C16EE83@IITMXMBX.iit.local>
References: <FC14DB6243548F4683D9E3D7E9A51C381C8C16EE83@IITMXMBX.iit.local>
Message-ID: <26DFF8F7-1CAE-4910-872D-935033508104@ucsd.edu>

Dear Alejo,

you have to save the dataset before clearing it from memory. Then the marks will be stored along with it. You should be able see them again going back to the same menu.
In the EEG structure, these rejection are stored in the EEG.reject field (which contains subfields with marked epochs and electrodes for each rejection method). 

Hope this helps,

Arno

On Jun 28, 2010, at 8:40 AM, Alejo Keuroghlanian wrote:

> Dear all,
> 
> suppose I'm working with the GUI, with a data set consisting of 20 epochs
> 
> - I choose "Tools > Reject data epochs > reject data (all methods)", and by visual inspection I mark trials #1 and #3 for rejection
> 
> - then I press "Close (keep marks)"
> 
> - I clear the data set from memory
> 
> - then I load it again... 
> 
> Are these marks stored somewhere in the data set/EEG structure? I have tried this but the marks disappear... I would like to know which were the epochs that I rejected eventually. 
> 
> Thank you in advance. Cheers,
> 
> Alejo




From saim_rasheed at hotmail.com  Wed Jun 30 09:26:02 2010
From: saim_rasheed at hotmail.com (Saim Rasheed)
Date: Wed, 30 Jun 2010 22:26:02 +0600
Subject: [Eeglablist] Numerical Values in STUDY Structure
In-Reply-To: <F6FEC69A-873D-41E7-853D-AEB358661734@ucsd.edu>
References: <SNT119-W63F22C0721A6DD0F8D18A6EACB0@phx.gbl>,
	<2769.159.71.184.228.1277833612.squirrel@sccn.ucsd.edu>,
	<F6FEC69A-873D-41E7-853D-AEB358661734@ucsd.edu>
Message-ID: <SNT119-W10F26BCEF0F578891237DFEACC0@phx.gbl>



Thankyou very much Julie and Arno,
 
Your help really worked. I have got the ERSP and itc data into Matlab workspace. I have got another problem.
I have recorded the data for three different visual stimuli conditions from 4 electrode positions at 256 samples per second. There are certain number of time-locked trials, e.g. 60, 58 55 for three conditions respectively. Each trial consists of 3 seconds of data containing 768 data points, which were automatically processed into 200 time points and 100 frequency points from 3Hz to 128Hz. Now I can see in workspace, 100*200 variable each for all electrode positions as chan1_ersp, chan2_ersp, chan3_ersp and chan4_ersp. Trial information is bit confused for me now. Is it lost now? Is it possible to compute ERSP and ITC trial by trial because I need to process a single-trial for each visual condition.
Thanks
 
Saim

 

> Subject: Re: [Eeglablist] Numerical Values in STUDY Structure
> From: arno at ucsd.edu
> Date: Tue, 29 Jun 2010 17:52:09 -0700
> To: saim_rasheed at hotmail.com; eeglablist at sccn.ucsd.edu
> 
> Another simple solution is to add additional output to the STUDY plotting functions after copying them from the history.
> For instance, after creating or loading an EEGLAB STUDY, then precomputing the ERP for data channel, you may use the channel plotting interface to plot the ERP for electrode CZ (for instance). Then, the EEGLAB history contains the line
> 
> STUDY = std_erpplot(STUDY, ALLEEG, 'channels', { 'cz' }); 
> 
> Then, looking at the std_erpplot function help, you may add additional outputs
> 
> [STUDY erp erptimes ] = std_erpplot(STUDY, ALLEEG, 'channels', { 'cz' }); 
> 
> Note that the "erp" output will be a cell array containing an array for the ERP of each condition/group. If one subject only is present, each array will be a vector of values for the ERP of this subject in a given condition. If several subjects are present, each column of the array will contain the ERP for a given subject.
> 
> Arno
> 
> On Jun 29, 2010, at 10:46 AM, Julie Onton wrote:
> 
> > Good question.
> > 
> > Below is a script to help you find, load and plot your raw data.
> > 
> > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
> > % LOAD raw data:------------------------
> > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
> > basedir = '/home/you/whereyourdatais/';
> > 
> > subjs =
> > {'S01','S02','S03','S04','S05','S06','S07','S08','S09','S10','S11','S12','S13'};
> > 
> > subj = 3; % who do you want to plot?
> > cond = 1; % what condition number?
> > 
> > % for example, load ERSP:
> > load_string =
> > [basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icaersp'];
> > 
> > % OR ITC
> > load_string =
> > [basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icaitc'];
> > 
> > % OR spectra:
> > load_string =
> > [basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icaspec'];
> > 
> > % OR topomaps:
> > load_string =
> > [basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icatopo'];
> > 
> > % load the raw data into the variable 'RAWdata':
> > RAWdata = load('-mat',load_string);
> > 
> > % Mask an ERSP using calculated bootstrap limits (if you calculated them):
> > 
> > ic = 8; % choose IC to plot
> > 
> > oneic = ['RAWdata.comp',int2str(ic),'_ersp'];
> > oneic = eval(oneic);
> > % in case you want to see the removed baseline:
> > onebase = ['RAWdata.comp',int2str(ic),'_erspbase'];
> > onebase = eval(onebase);
> > % load the bootstrap significance limits:
> > oneboot = ['RAWdata.comp',int2str(ic),'_erspboot'];
> > oneboot = eval(oneboot);
> > 
> > maskERSP = oneic;
> > % zero out non-sig values:
> > maskERSP(find(oneic > repmat(oneboot(:,1),[1 size(oneic,2)])& oneic <
> > repmat(oneboot(:,2),[1 size(oneic,2)]))) = 0;
> > 
> > clim = 4; % set +/- color limits
> > figure; imagesc(RAWdata.times,RAWdata.freqs,maskERSP,[-clim clim]);
> > set(gca,'ydir','norm');
> > title(['Subj ',int2str(subj),'; IC ',int2str(ic),'; ',STUDY.condition{cond}]);
> > cbar;
> > 
> > 
> > Hope this helps, Julie
> > 
> > -- 
> > Julie Onton, PhD
> > http://sccn.ucsd.edu/~julie
> > 
> >> 
> >> 
> >> 
> >> Hi,
> >> 
> >> 
> >> 
> >> I am working with STUDY structure and computed ERSP, ITC and. I can plot them
> >> by a single click and analyse them.
> >> 
> >> Where I can find the numerical values for ERSP and ITC ? Where these values
> >> are stored?
> >> 
> >> 
> >> 
> >> I need them for further computations. Please help.
> >> 
> >> 
> >> 
> >> Thanks.
> >> 
> >> 
> >> 
> >> Saim
> >> 
> >> _________________________________________________________________
> >> Your E-mail and More On-the-Go. Get Windows Live Hotmail Free.
> >> https://signup.live.com/signup.aspx?id=60969_______________________________________________
> >> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> >> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> >> For digest mode, send an email with the subject "set digest mime" to
> >> eeglablist-request at sccn.ucsd.edu
> > 
> > _______________________________________________
> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> > To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> > For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
> 
 		 	   		  
_________________________________________________________________
Hotmail: Trusted email with powerful SPAM protection.
https://signup.live.com/signup.aspx?id=60969
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100630/6f414be6/attachment.html>

From julie at sccn.ucsd.edu  Wed Jun 30 11:34:10 2010
From: julie at sccn.ucsd.edu (Julie Onton)
Date: Wed, 30 Jun 2010 11:34:10 -0700 (PDT)
Subject: [Eeglablist] Numerical Values in STUDY Structure
In-Reply-To: <SNT119-W10F26BCEF0F578891237DFEACC0@phx.gbl>
References: <SNT119-W63F22C0721A6DD0F8D18A6EACB0@phx.gbl>,
	<2769.159.71.184.228.1277833612.squirrel@sccn.ucsd.edu>,
	<F6FEC69A-873D-41E7-853D-AEB358661734@ucsd.edu>
	<SNT119-W10F26BCEF0F578891237DFEACC0@phx.gbl>
Message-ID: <2213.159.71.184.228.1277922850.squirrel@sccn.ucsd.edu>

There is an option in the precompute menu to save single-trial data, but you
will have to recompute if you did not check this the first time around
(default is NOT to save single trials because this occupies a remarkable
amount of disk space). ITC, of course, is a value that is computed across
trials.

Julie

-- 
Julie Onton, PhD
http://sccn.ucsd.edu/~julie

>
>
> Thankyou very much Julie and Arno,
>
> Your help really worked. I have got the ERSP and itc data into Matlab
> workspace. I have got another problem.
> I have recorded the data for three different visual stimuli conditions from 4
> electrode positions at 256 samples per second. There are certain number of
> time-locked trials, e.g. 60, 58 55 for three conditions respectively. Each
> trial consists of 3 seconds of data containing 768 data points, which were
> automatically processed into 200 time points and 100 frequency points from 3Hz
> to 128Hz. Now I can see in workspace, 100*200 variable each for all electrode
> positions as chan1_ersp, chan2_ersp, chan3_ersp and chan4_ersp. Trial
> information is bit confused for me now. Is it lost now? Is it possible to
> compute ERSP and ITC trial by trial because I need to process a single-trial
> for each visual condition.
> Thanks
>
> Saim
>
>
>
>> Subject: Re: [Eeglablist] Numerical Values in STUDY Structure
>> From: arno at ucsd.edu
>> Date: Tue, 29 Jun 2010 17:52:09 -0700
>> To: saim_rasheed at hotmail.com; eeglablist at sccn.ucsd.edu
>>
>> Another simple solution is to add additional output to the STUDY plotting
>> functions after copying them from the history.
>> For instance, after creating or loading an EEGLAB STUDY, then precomputing
>> the ERP for data channel, you may use the channel plotting interface to plot
>> the ERP for electrode CZ (for instance). Then, the EEGLAB history contains
>> the line
>>
>> STUDY = std_erpplot(STUDY, ALLEEG, 'channels', { 'cz' });
>>
>> Then, looking at the std_erpplot function help, you may add additional
>> outputs
>>
>> [STUDY erp erptimes ] = std_erpplot(STUDY, ALLEEG, 'channels', { 'cz' });
>>
>> Note that the "erp" output will be a cell array containing an array for the
>> ERP of each condition/group. If one subject only is present, each array will
>> be a vector of values for the ERP of this subject in a given condition. If
>> several subjects are present, each column of the array will contain the ERP
>> for a given subject.
>>
>> Arno
>>
>> On Jun 29, 2010, at 10:46 AM, Julie Onton wrote:
>>
>> > Good question.
>> >
>> > Below is a script to help you find, load and plot your raw data.
>> >
>> > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>> > % LOAD raw data:------------------------
>> > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>> > basedir = '/home/you/whereyourdatais/';
>> >
>> > subjs =
>> > {'S01','S02','S03','S04','S05','S06','S07','S08','S09','S10','S11','S12','S13'};
>> >
>> > subj = 3; % who do you want to plot?
>> > cond = 1; % what condition number?
>> >
>> > % for example, load ERSP:
>> > load_string =
>> > [basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icaersp'];
>> >
>> > % OR ITC
>> > load_string =
>> > [basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icaitc'];
>> >
>> > % OR spectra:
>> > load_string =
>> > [basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icaspec'];
>> >
>> > % OR topomaps:
>> > load_string =
>> > [basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icatopo'];
>> >
>> > % load the raw data into the variable 'RAWdata':
>> > RAWdata = load('-mat',load_string);
>> >
>> > % Mask an ERSP using calculated bootstrap limits (if you calculated them):
>> >
>> > ic = 8; % choose IC to plot
>> >
>> > oneic = ['RAWdata.comp',int2str(ic),'_ersp'];
>> > oneic = eval(oneic);
>> > % in case you want to see the removed baseline:
>> > onebase = ['RAWdata.comp',int2str(ic),'_erspbase'];
>> > onebase = eval(onebase);
>> > % load the bootstrap significance limits:
>> > oneboot = ['RAWdata.comp',int2str(ic),'_erspboot'];
>> > oneboot = eval(oneboot);
>> >
>> > maskERSP = oneic;
>> > % zero out non-sig values:
>> > maskERSP(find(oneic > repmat(oneboot(:,1),[1 size(oneic,2)])& oneic <
>> > repmat(oneboot(:,2),[1 size(oneic,2)]))) = 0;
>> >
>> > clim = 4; % set +/- color limits
>> > figure; imagesc(RAWdata.times,RAWdata.freqs,maskERSP,[-clim clim]);
>> > set(gca,'ydir','norm');
>> > title(['Subj ',int2str(subj),'; IC ',int2str(ic),';
>> ',STUDY.condition{cond}]);
>> > cbar;
>> >
>> >
>> > Hope this helps, Julie
>> >
>> > --
>> > Julie Onton, PhD
>> > http://sccn.ucsd.edu/~julie
>> >
>> >>
>> >>
>> >>
>> >> Hi,
>> >>
>> >>
>> >>
>> >> I am working with STUDY structure and computed ERSP, ITC and. I can plot
>> them
>> >> by a single click and analyse them.
>> >>
>> >> Where I can find the numerical values for ERSP and ITC ? Where these
>> values
>> >> are stored?
>> >>
>> >>
>> >>
>> >> I need them for further computations. Please help.
>> >>
>> >>
>> >>
>> >> Thanks.
>> >>
>> >>
>> >>
>> >> Saim
>> >>
>> >> _________________________________________________________________
>> >> Your E-mail and More On-the-Go. Get Windows Live Hotmail Free.
>> >> https://signup.live.com/signup.aspx?id=60969_______________________________________________
>> >> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> >> To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> >> For digest mode, send an email with the subject "set digest mime" to
>> >> eeglablist-request at sccn.ucsd.edu
>> >
>> > _______________________________________________
>> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> > To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> > For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>>
>
> _________________________________________________________________
> Hotmail: Trusted email with powerful SPAM protection.
> https://signup.live.com/signup.aspx?id=60969



From arno at ucsd.edu  Wed Jun 30 16:58:58 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Wed, 30 Jun 2010 16:58:58 -0700
Subject: [Eeglablist] Numerical Values in STUDY Structure
In-Reply-To: <SNT119-W10F26BCEF0F578891237DFEACC0@phx.gbl>
References: <SNT119-W63F22C0721A6DD0F8D18A6EACB0@phx.gbl>,
	<2769.159.71.184.228.1277833612.squirrel@sccn.ucsd.edu>,
	<F6FEC69A-873D-41E7-853D-AEB358661734@ucsd.edu>
	<SNT119-W10F26BCEF0F578891237DFEACC0@phx.gbl>
Message-ID: <9F2C5D05-8E94-4D4A-8AC5-8970FCCCA49A@ucsd.edu>

Dear Saim,

yes, it is possible to compute and save trial information. Simply check the checkbox in the STUDY graphic interface to save single trials or use 'savetrials', 'on' from the command line. Trials will be saved in the 3rd dimension of the arrays you are mentioning.

Best regards,

A. Delorme

On Jun 30, 2010, at 9:26 AM, Saim Rasheed wrote:

> 
> Thankyou very much Julie and Arno,
>  
> Your help really worked. I have got the ERSP and itc data into Matlab workspace. I have got another problem.
> I have recorded the data for three different visual stimuli conditions from 4 electrode positions at 256 samples per second. There are certain number of time-locked trials, e.g. 60, 58 55 for three conditions respectively. Each trial consists of 3 seconds of data containing 768 data points, which were automatically processed into 200 time points and 100 frequency points from 3Hz to 128Hz. Now I can see in workspace, 100*200 variable each for all electrode positions as chan1_ersp, chan2_ersp, chan3_ersp and chan4_ersp. Trial information is bit confused for me now. Is it lost now? Is it possible to compute ERSP and ITC trial by trial because I need to process a single-trial for each visual condition.
> Thanks
>  
> Saim
> 
>  
> 
> > Subject: Re: [Eeglablist] Numerical Values in STUDY Structure
> > From: arno at ucsd.edu
> > Date: Tue, 29 Jun 2010 17:52:09 -0700
> > To: saim_rasheed at hotmail.com; eeglablist at sccn.ucsd.edu
> > 
> > Another simple solution is to add additional output to the STUDY plotting functions after copying them from the history.
> > For instance, after creating or loading an EEGLAB STUDY, then precomputing the ERP for data channel, you may use the channel plotting interface to plot the ERP for electrode CZ (for instance). Then, the EEGLAB history contains the line
> > 
> > STUDY = std_erpplot(STUDY, ALLEEG, 'channels', { 'cz' }); 
> > 
> > Then, looking at the std_erpplot function help, you may add additional outputs
> > 
> > [STUDY erp erptimes ] = std_erpplot(STUDY, ALLEEG, 'channels', { 'cz' }); 
> > 
> > Note that the "erp" output will be a cell array containing an array for the ERP of each condition/group. If one subject only is present, each array will be a vector of values for the ERP of this subject in a given condition. If several subjects are present, each column of the array will contain the ERP for a given subject.
> > 
> > Arno
> > 
> > On Jun 29, 2010, at 10:46 AM, Julie Onton wrote:
> > 
> > > Good question.
> > > 
> > > Below is a script to help you find, load and plot your raw data.
> > > 
> > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
> > > % LOAD raw data:------------------------
> > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
> > > basedir = '/home/you/whereyourdatais/';
> > > 
> > > subjs =
> > > {'S01','S02','S03','S04','S05','S06','S07','S08','S09','S10','S11','S12','S13'};
> > > 
> > > subj = 3; % who do you want to plot?
> > > cond = 1; % what condition number?
> > > 
> > > % for example, load ERSP:
> > > load_string =
> > > [basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icaersp'];
> > > 
> > > % OR ITC
> > > load_string =
> > > [basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icaitc'];
> > > 
> > > % OR spectra:
> > > load_string =
> > > [basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icaspec'];
> > > 
> > > % OR topomaps:
> > > load_string =
> > > [basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icatopo'];
> > > 
> > > % load the raw data into the variable 'RAWdata':
> > > RAWdata = load('-mat',load_string);
> > > 
> > > % Mask an ERSP using calculated bootstrap limits (if you calculated them):
> > > 
> > > ic = 8; % choose IC to plot
> > > 
> > > oneic = ['RAWdata.comp',int2str(ic),'_ersp'];
> > > oneic = eval(oneic);
> > > % in case you want to see the removed baseline:
> > > onebase = ['RAWdata.comp',int2str(ic),'_erspbase'];
> > > onebase = eval(onebase);
> > > % load the bootstrap significance limits:
> > > oneboot = ['RAWdata.comp',int2str(ic),'_erspboot'];
> > > oneboot = eval(oneboot);
> > > 
> > > maskERSP = oneic;
> > > % zero out non-sig values:
> > > maskERSP(find(oneic > repmat(oneboot(:,1),[1 size(oneic,2)])& oneic <
> > > repmat(oneboot(:,2),[1 size(oneic,2)]))) = 0;
> > > 
> > > clim = 4; % set +/- color limits
> > > figure; imagesc(RAWdata.times,RAWdata.freqs,maskERSP,[-clim clim]);
> > > set(gca,'ydir','norm');
> > > title(['Subj ',int2str(subj),'; IC ',int2str(ic),'; ',STUDY.condition{cond}]);
> > > cbar;
> > > 
> > > 
> > > Hope this helps, Julie
> > > 
> > > -- 
> > > Julie Onton, PhD
> > > http://sccn.ucsd.edu/~julie
> > > 
> > >> 
> > >> 
> > >> 
> > >> Hi,
> > >> 
> > >> 
> > >> 
> > >> I am working with STUDY structure and computed ERSP, ITC and. I can plot them
> > >> by a single click and analyse them.
> > >> 
> > >> Where I can find the numerical values for ERSP and ITC ? Where these values
> > >> are stored?
> > >> 
> > >> 
> > >> 
> > >> I need them for further computations. Please help.
> > >> 
> > >> 
> > >> 
> > >> Thanks.
> > >> 
> > >> 
> > >> 
> > >> Saim
> > >> 
> > >> _________________________________________________________________
> > >> Your E-mail and More On-the-Go. Get Windows Live Hotmail Free.
> > >> https://signup.live.com/signup.aspx?id=60969_______________________________________________
> > >> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> > >> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> > >> For digest mode, send an email with the subject "set digest mime" to
> > >> eeglablist-request at sccn.ucsd.edu
> > > 
> > > _______________________________________________
> > > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> > > To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> > > For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
> > 
> 
> Hotmail: Trusted email with powerful SPAM protection. Sign up now.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100630/1b0823ff/attachment.html>

From arno at ucsd.edu  Wed Jun 30 17:10:45 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Wed, 30 Jun 2010 17:10:45 -0700
Subject: [Eeglablist] Opening MEG Files
In-Reply-To: <AANLkTin8M9aTlpvu7xZO-4RCkGz6H8uJnjWmZnPeTwaU@mail.gmail.com>
References: <AANLkTin8M9aTlpvu7xZO-4RCkGz6H8uJnjWmZnPeTwaU@mail.gmail.com>
Message-ID: <BE21256C-5F05-4471-9732-9A16F050D41C@ucsd.edu>

Dear Andrew,

maybe you do not have a marker file (events) and this is why this function fails. Try also the "Import form other formats using FILEIO". This will probably work. Please also submit a bug report at 

http://sccn.ucsd.edu/eeglab/bugzilla

and upload a small folder and we will look at the problem.
Best regards,

Arno




From saim_rasheed at hotmail.com  Thu Jul  1 08:18:47 2010
From: saim_rasheed at hotmail.com (Saim Rasheed)
Date: Thu, 1 Jul 2010 21:18:47 +0600
Subject: [Eeglablist] Numerical Values in STUDY Structure
In-Reply-To: <2213.159.71.184.228.1277922850.squirrel@sccn.ucsd.edu>
References: <SNT119-W63F22C0721A6DD0F8D18A6EACB0@phx.gbl>,
	<2769.159.71.184.228.1277833612.squirrel@sccn.ucsd.edu>,
	<F6FEC69A-873D-41E7-853D-AEB358661734@ucsd.edu>,
	<SNT119-W10F26BCEF0F578891237DFEACC0@phx.gbl>,
	<2213.159.71.184.228.1277922850.squirrel@sccn.ucsd.edu>
Message-ID: <SNT119-W54B8F896F7FB94C445CEEAEACD0@phx.gbl>


Thanks a lot, I got the trial by trial data from '*.dattimef' files. In '.ersp' files individual trials are not available, Are these values averaged across all the trial values?
In '.datitc' values are visible as complex numbers in Matlab workspace, I write the data into excell file, there it write only real part, not the complex part ?
I have to use the data (after conversion into another format) into LIBSVM for classification of my visual conditions. I dont know, How important is the complex part for ITC computations?

Saim 

> Date: Wed, 30 Jun 2010 11:34:10 -0700
> Subject: RE: [Eeglablist] Numerical Values in STUDY Structure
> From: julie at sccn.ucsd.edu
> To: saim_rasheed at hotmail.com
> CC: arno at ucsd.edu; eeglablist at sccn.ucsd.edu
> 
> There is an option in the precompute menu to save single-trial data, but you
> will have to recompute if you did not check this the first time around
> (default is NOT to save single trials because this occupies a remarkable
> amount of disk space). ITC, of course, is a value that is computed across
> trials.
> 
> Julie
> 
> -- 
> Julie Onton, PhD
> http://sccn.ucsd.edu/~julie
> 
> >
> >
> > Thankyou very much Julie and Arno,
> >
> > Your help really worked. I have got the ERSP and itc data into Matlab
> > workspace. I have got another problem.
> > I have recorded the data for three different visual stimuli conditions from 4
> > electrode positions at 256 samples per second. There are certain number of
> > time-locked trials, e.g. 60, 58 55 for three conditions respectively. Each
> > trial consists of 3 seconds of data containing 768 data points, which were
> > automatically processed into 200 time points and 100 frequency points from 3Hz
> > to 128Hz. Now I can see in workspace, 100*200 variable each for all electrode
> > positions as chan1_ersp, chan2_ersp, chan3_ersp and chan4_ersp. Trial
> > information is bit confused for me now. Is it lost now? Is it possible to
> > compute ERSP and ITC trial by trial because I need to process a single-trial
> > for each visual condition.
> > Thanks
> >
> > Saim
> >
> >
> >
> >> Subject: Re: [Eeglablist] Numerical Values in STUDY Structure
> >> From: arno at ucsd.edu
> >> Date: Tue, 29 Jun 2010 17:52:09 -0700
> >> To: saim_rasheed at hotmail.com; eeglablist at sccn.ucsd.edu
> >>
> >> Another simple solution is to add additional output to the STUDY plotting
> >> functions after copying them from the history.
> >> For instance, after creating or loading an EEGLAB STUDY, then precomputing
> >> the ERP for data channel, you may use the channel plotting interface to plot
> >> the ERP for electrode CZ (for instance). Then, the EEGLAB history contains
> >> the line
> >>
> >> STUDY = std_erpplot(STUDY, ALLEEG, 'channels', { 'cz' });
> >>
> >> Then, looking at the std_erpplot function help, you may add additional
> >> outputs
> >>
> >> [STUDY erp erptimes ] = std_erpplot(STUDY, ALLEEG, 'channels', { 'cz' });
> >>
> >> Note that the "erp" output will be a cell array containing an array for the
> >> ERP of each condition/group. If one subject only is present, each array will
> >> be a vector of values for the ERP of this subject in a given condition. If
> >> several subjects are present, each column of the array will contain the ERP
> >> for a given subject.
> >>
> >> Arno
> >>
> >> On Jun 29, 2010, at 10:46 AM, Julie Onton wrote:
> >>
> >> > Good question.
> >> >
> >> > Below is a script to help you find, load and plot your raw data.
> >> >
> >> > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
> >> > % LOAD raw data:------------------------
> >> > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
> >> > basedir = '/home/you/whereyourdatais/';
> >> >
> >> > subjs =
> >> > {'S01','S02','S03','S04','S05','S06','S07','S08','S09','S10','S11','S12','S13'};
> >> >
> >> > subj = 3; % who do you want to plot?
> >> > cond = 1; % what condition number?
> >> >
> >> > % for example, load ERSP:
> >> > load_string =
> >> > [basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icaersp'];
> >> >
> >> > % OR ITC
> >> > load_string =
> >> > [basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icaitc'];
> >> >
> >> > % OR spectra:
> >> > load_string =
> >> > [basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icaspec'];
> >> >
> >> > % OR topomaps:
> >> > load_string =
> >> > [basedir,subjs{subj},'\',STUDY.datasetinfo(1).filename(1:end-4),'.icatopo'];
> >> >
> >> > % load the raw data into the variable 'RAWdata':
> >> > RAWdata = load('-mat',load_string);
> >> >
> >> > % Mask an ERSP using calculated bootstrap limits (if you calculated them):
> >> >
> >> > ic = 8; % choose IC to plot
> >> >
> >> > oneic = ['RAWdata.comp',int2str(ic),'_ersp'];
> >> > oneic = eval(oneic);
> >> > % in case you want to see the removed baseline:
> >> > onebase = ['RAWdata.comp',int2str(ic),'_erspbase'];
> >> > onebase = eval(onebase);
> >> > % load the bootstrap significance limits:
> >> > oneboot = ['RAWdata.comp',int2str(ic),'_erspboot'];
> >> > oneboot = eval(oneboot);
> >> >
> >> > maskERSP = oneic;
> >> > % zero out non-sig values:
> >> > maskERSP(find(oneic > repmat(oneboot(:,1),[1 size(oneic,2)])& oneic <
> >> > repmat(oneboot(:,2),[1 size(oneic,2)]))) = 0;
> >> >
> >> > clim = 4; % set +/- color limits
> >> > figure; imagesc(RAWdata.times,RAWdata.freqs,maskERSP,[-clim clim]);
> >> > set(gca,'ydir','norm');
> >> > title(['Subj ',int2str(subj),'; IC ',int2str(ic),';
> >> ',STUDY.condition{cond}]);
> >> > cbar;
> >> >
> >> >
> >> > Hope this helps, Julie
> >> >
> >> > --
> >> > Julie Onton, PhD
> >> > http://sccn.ucsd.edu/~julie
> >> >
> >> >>
> >> >>
> >> >>
> >> >> Hi,
> >> >>
> >> >>
> >> >>
> >> >> I am working with STUDY structure and computed ERSP, ITC and. I can plot
> >> them
> >> >> by a single click and analyse them.
> >> >>
> >> >> Where I can find the numerical values for ERSP and ITC ? Where these
> >> values
> >> >> are stored?
> >> >>
> >> >>
> >> >>
> >> >> I need them for further computations. Please help.
> >> >>
> >> >>
> >> >>
> >> >> Thanks.
> >> >>
> >> >>
> >> >>
> >> >> Saim
> >> >>
> >> >> _________________________________________________________________
> >> >> Your E-mail and More On-the-Go. Get Windows Live Hotmail Free.
> >> >> https://signup.live.com/signup.aspx?id=60969_______________________________________________
> >> >> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> >> >> To unsubscribe, send an empty email to
> >> eeglablist-unsubscribe at sccn.ucsd.edu
> >> >> For digest mode, send an email with the subject "set digest mime" to
> >> >> eeglablist-request at sccn.ucsd.edu
> >> >
> >> > _______________________________________________
> >> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> >> > To unsubscribe, send an empty email to
> >> eeglablist-unsubscribe at sccn.ucsd.edu
> >> > For digest mode, send an email with the subject "set digest mime" to
> >> eeglablist-request at sccn.ucsd.edu
> >>
> >
> > _________________________________________________________________
> > Hotmail: Trusted email with powerful SPAM protection.
> > https://signup.live.com/signup.aspx?id=60969
> 

 		 	   		  
_________________________________________________________________
Your E-mail and More On-the-Go. Get Windows Live Hotmail Free.
https://signup.live.com/signup.aspx?id=60969
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100701/f099fa60/attachment.html>

From lpxcr at nottingham.ac.uk  Fri Jul  2 02:25:04 2010
From: lpxcr at nottingham.ac.uk (Retzler Christopher)
Date: Fri, 2 Jul 2010 10:25:04 +0100
Subject: [Eeglablist] Different baseline correction for different types of
	event?
Message-ID: <E91C1CF22361DA45990CFA4823F79919FEFDFE@VUIEXCHC.ad.nottingham.ac.uk>

Hi
 
I'm currently re-anlayzing some data, I have been epoching using the following
 
 

EEG = pop_epoch( EEG, {  'cho+'  'cho-'  'chx '  'cig+'  'cig-'  'cix '  }, [-2  2], 'newname', 'epochs', 'epochinfo', 'yes'); 

 

[ALLEEG EEG CURRENTSET] = pop_newset(ALLEEG, EEG, 1, 'overwrite', 'on', 'gui', 'off');

    

% baseline correct using -200 to 0 as baseline 

EEG = pop_rmbase(EEG,[-200 0]);

 

 

Is it possible to apply seperate baseline corrections to two types of events? So for 'chx' and 'cix' events I would like to baseline correct at -1800 - -1600 as these are pre-response epochs. For the other events I would like to use  -200 - 0ms as the baseline (these are post-reponse). Ideally I would like to process these events together so that I can run the rest of pre-processing on the entire data set (i.e. ICA artifact rejection etc) and not have to do it twice.

 

Many thanks in advance.

 
Chris Retzler
 
PhD student
School of Psychology 
University of Nottingham 
University Park 
Nottingham 
NG7 2RD. 
This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system:
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100702/00dd7829/attachment.html>

From Dorothy.Bishop at psy.ox.ac.uk  Sat Jul  3 01:54:08 2010
From: Dorothy.Bishop at psy.ox.ac.uk (Dorothy Bishop)
Date: Sat, 03 Jul 2010 09:54:08 +0100
Subject: [Eeglablist] Simple intro to ERSP
Message-ID: <4C2F08C1.A716.003A.1@psy.ox.ac.uk>

To try and ensure I understand time-frequency analysis, I am writing a simplified introduction that assumes little background knowledge.
I have adopted the approach of using simplified simulated signals to allow users to explore what happens with different command parameters.
A draft, just covering aspects of ERSP can be found on http://bishoptechbits.blogspot.com/

I hope it might be useful to others, but it is a case of the blind leading the blind, and there may be errors in it.
Comments and corrections would be very welcome - I'd be especially grateful if someone knowledgeable would look at it and let me know if I've got things wrong.

PS
One reason I did this was because I wanted to compare results on ERSP in a mismatch paradigm when you
a) subtracted condition 1 from condition 2 and then did ERSP on the difference waveform
vs 
b) Did ERSP on condition 1 and condition 2, and subtracted the resulting ERSPs.

My simulations suggest (a) is more sensitive, but I'd be interested in other views on this.


Dorothy Bishop
Professor of Developmental Neuropsychology
Department of Experimental Psychology
University of Oxford
OX1 3UD
Website: http://psyweb.psy.ox.ac.uk/oscci/
Blog: http://deevybee.blogspot.com/
tel: +44 (0)1865 271369
fax: +44 (0)1865 281255
 





From lutobu at gmail.com  Sun Jul  4 17:32:44 2010
From: lutobu at gmail.com (ludwing torres)
Date: Mon, 5 Jul 2010 02:32:44 +0200
Subject: [Eeglablist] help to export field matrix
Message-ID: <AANLkTikOs9uYBQ_LZ3yEwbtC5uFSYe_KMuw9hfCIq-t4@mail.gmail.com>

Hi, I need to know how to export the field matrix corresponding to the
espherical model of the head, to the matlab workspace ; in order to compute
an inverse and a forward problem. thanks.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100705/e81fa194/attachment.html>

From jdesjardins at brocku.ca  Sat Jul  3 10:24:40 2010
From: jdesjardins at brocku.ca (James Desjardins)
Date: Sat, 03 Jul 2010 13:24:40 -0400
Subject: [Eeglablist] Different baseline correction for different
	types	of event?
In-Reply-To: <E91C1CF22361DA45990CFA4823F79919FEFDFE@VUIEXCHC.ad.nottingham.ac.uk>
References: <E91C1CF22361DA45990CFA4823F79919FEFDFE@VUIEXCHC.ad.nottingham.ac.uk>
Message-ID: <20100703132440.hb4p203ooogocs80@webmail.brocku.ca>

Hi Chris,

I do not know of a way to apply a condition specific baseline  
correction in a single command. I have performed a similar  
manipulation by segmenting the data into to two files and then  
appending them together into a single file using "pop_mergeset". Then  
continue further processing on the single appended dataset.

If your data is already segmented into a single file (baseline  
corrected to -200ms to 0ms) and you would like to change the baseline  
period of your "chx","cix" condition type epochs to -1600ms to  
-1400ms, a script like the following should work:


% Create a vector of baseline times plus zero point (ms) and convert  
it to data point indices...
ms_vals=[-1800,-1600,0];
pnt_vals=round(ms_vals/(1000/EEG.srate)-((EEG.xmin*1000)/(1000/EEG.srate)-1));


% Obtain indices for epochs of the condition type to be changed...
j=0;
for i=1:length(EEG.event);
     trial=EEG.event(i).epoch;
     if  
(strcmp(EEG.event(i).type,'chx')||strcmp(EEG.event(i).type,'cix'))&& ...
             EEG.event(i).latency==pnt_vals(3)+EEG.pnts*(trial-1);
         j=j+1;ind(j)=trial;
     end
end

% Calculate baseline values for the epochs of the condition type to be  
changed...
bl=mean(EEG.data(:,pnt_vals(1):pnt_vals(2),ind),2);

% Subtract baseline values from the epochs of the condition type to be  
changed...
for i=1:EEG.nbchan;
     for j=1:EEG.pnts;
         EEG.data(i,j,ind)=EEG.data(i,j,ind)-bl(i,1,:);
     end
end



James Desjardins
Technician, MA Student
Department of Psychology, Behavioural Neuroscience
Cognitive and Affective Neuroscience Lab
Brock University
St. Catharines, ON, Canada




Quoting Retzler Christopher <lpxcr at nottingham.ac.uk>:

> Hi
>
> I'm currently re-anlayzing some data, I have been epoching using the  
>  following
>
>
>
> EEG = pop_epoch( EEG, {  'cho+'  'cho-'  'chx '  'cig+'  'cig-'    
> 'cix '  }, [-2  2], 'newname', 'epochs', 'epochinfo', 'yes');
>
>
>
> [ALLEEG EEG CURRENTSET] = pop_newset(ALLEEG, EEG, 1, 'overwrite',   
> 'on', 'gui', 'off');
>
>
>
> % baseline correct using -200 to 0 as baseline
>
> EEG = pop_rmbase(EEG,[-200 0]);
>
>
>
>
>
> Is it possible to apply seperate baseline corrections to two types   
> of events? So for 'chx' and 'cix' events I would like to baseline   
> correct at -1800 - -1600 as these are pre-response epochs. For the   
> other events I would like to use  -200 - 0ms as the baseline (these   
> are post-reponse). Ideally I would like to process these events   
> together so that I can run the rest of pre-processing on the entire   
> data set (i.e. ICA artifact rejection etc) and not have to do it   
> twice.
>
>
>
> Many thanks in advance.
>
>
> Chris Retzler
>
> PhD student
> School of Psychology
> University of Nottingham
> University Park
> Nottingham
> NG7 2RD.
> This message has been checked for viruses but the contents of an attachment
> may still contain software viruses which could damage your computer system:
> you are advised to perform your own checks. Email communications with the
> University of Nottingham may be monitored as permitted by UK legislation.







From jan.kremlacek at lfhk.cuni.cz  Tue Jul  6 07:07:19 2010
From: jan.kremlacek at lfhk.cuni.cz (Kremlacek, Jan)
Date: Tue, 6 Jul 2010 16:07:19 +0200
Subject: [Eeglablist] Simple intro to ERSP
References: <4C2F08C1.A716.003A.1@psy.ox.ac.uk>
Message-ID: <CEBF6AD5AA01724D84D3395EE8FB1727175D8A@tempexbe.lfhk.cuni.cz>

Dear Dorothy,

thank you for a nice tutorial. I would like to make a note to a problem
of the ERSP of  difference vs. difference of ERSPs. 

In your example you are using a fixed phase, which advances the first
approach (the ERSP of difference) as you showed, but in a case that the
induced activity is not phase aligned, then the second approach
(difference of ERSPs) should bring you more sensitive results.

With kind regards,
Jan

-----Original Message-----
From: eeglablist-bounces at sccn.ucsd.edu
[mailto:eeglablist-bounces at sccn.ucsd.edu] On Behalf Of Dorothy Bishop
Sent: Saturday, July 03, 2010 10:54 AM
To: eeglablist at sccn.ucsd.edu
Subject: [Eeglablist] Simple intro to ERSP

To try and ensure I understand time-frequency analysis, I am writing a
simplified introduction that assumes little background knowledge.
I have adopted the approach of using simplified simulated signals to
allow users to explore what happens with different command parameters.
A draft, just covering aspects of ERSP can be found on
http://bishoptechbits.blogspot.com/

I hope it might be useful to others, but it is a case of the blind
leading the blind, and there may be errors in it.
Comments and corrections would be very welcome - I'd be especially
grateful if someone knowledgeable would look at it and let me know if
I've got things wrong.

PS
One reason I did this was because I wanted to compare results on ERSP in
a mismatch paradigm when you
a) subtracted condition 1 from condition 2 and then did ERSP on the
difference waveform
vs 
b) Did ERSP on condition 1 and condition 2, and subtracted the resulting
ERSPs.

My simulations suggest (a) is more sensitive, but I'd be interested in
other views on this.


Dorothy Bishop
Professor of Developmental Neuropsychology
Department of Experimental Psychology
University of Oxford
OX1 3UD
Website: http://psyweb.psy.ox.ac.uk/oscci/
Blog: http://deevybee.blogspot.com/
tel: +44 (0)1865 271369
fax: +44 (0)1865 281255
 



_______________________________________________
Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
To unsubscribe, send an empty email to
eeglablist-unsubscribe at sccn.ucsd.edu
For digest mode, send an email with the subject "set digest mime" to
eeglablist-request at sccn.ucsd.edu



From bornalikundu at gmail.com  Wed Jul  7 12:28:28 2010
From: bornalikundu at gmail.com (Bornali Kundu)
Date: Wed, 7 Jul 2010 14:28:28 -0500
Subject: [Eeglablist] Analyzing ERSPs
Message-ID: <AANLkTikDqY58Tw-lDcoPKgSZzJI0bIw-9MrswaGbfar9@mail.gmail.com>

Hello,

This question is not directly related to using EEGLAB, but has to do with
analyzing ERSPs.  What is an acceptable way to statistically compare ERSPs
within subjects over multiple sessions?  I have 10 subjects, where each
repeated the same task during 2 separate sessions.  I want to compare ERSP
amplitude over multiple sessions for each subject with ERSP amplitude over
many subjects within one session.  Does it make sense to run some type of
ANOVA at every time and frequency point or a MANOVA?  Or is there a better
way to compare two ERSPs taking into consideration the "pattern" of activity
over time and frequency (perhaps using a cluster-based permutation
strategy)?  Any thoughts would be greatly appreciated.

Thanks!

Bornali
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100707/c962338e/attachment.html>

From g.rousselet at psy.gla.ac.uk  Thu Jul  8 08:44:20 2010
From: g.rousselet at psy.gla.ac.uk (Guillaume Rousselet)
Date: Thu, 8 Jul 2010 16:44:20 +0100
Subject: [Eeglablist]  Analyzing ERSPs
In-Reply-To: <AANLkTikDqY58Tw-lDcoPKgSZzJI0bIw-9MrswaGbfar9@mail.gmail.com>
References: <AANLkTikDqY58Tw-lDcoPKgSZzJI0bIw-9MrswaGbfar9@mail.gmail.com>
Message-ID: <9E5EB922-A2C5-42EF-98D7-A842B1BCF6A8@psy.gla.ac.uk>

Hey Bornali,

You could do different analyses to address different questions.

[1] Whatever you do it is a good idea to use cluster based statistics,  
with bootstrap or permutation.

[2] If you're main interest is to show the reliability of the ERSP  
responses within subject, you can run single-trial analyses to compare  
session 1 and session 2 in each subject. Reliability should result in  
small if any differences. You may want to centre your data first to  
remove mean session effects if you are interested in the reliability  
of the effects of different conditions. If your design includes  
different conditions, you could run, at each time and each electrode,  
an ANOVA including a session factor, and correct for multiple  
comparisons using a cluster test.

[3] If you're main interest is to show the reliability of the ERSP  
responses across subjects, you can run group comparisons of the two  
sessions, using the mean ERSP from each subject (thus after removing  
the single-trial variance associated with each subject).

[4] You could perform [3] after fitting a model on each subject, and  
perform a pairwise group comparisons on the beta coefficients of the  
model, similarly to what people do in fMRI. This would tell you if you  
have significant session effect across subjects. Of course this  
analysis will not reveal session effects that are not consistent in  
space and time across subjects - the typical problem with group  
statistics.

Best wishes,

Guillaume



On 7 Jul 2010, at 20:28, Bornali Kundu wrote:

> Hello,
>
> This question is not directly related to using EEGLAB, but has to do  
> with analyzing ERSPs.  What is an acceptable way to statistically  
> compare ERSPs within subjects over multiple sessions?  I have 10  
> subjects, where each repeated the same task during 2 separate  
> sessions.  I want to compare ERSP amplitude over multiple sessions  
> for each subject with ERSP amplitude over many subjects within one  
> session.  Does it make sense to run some type of ANOVA at every time  
> and frequency point or a MANOVA?  Or is there a better way to  
> compare two ERSPs taking into consideration the "pattern" of  
> activity over time and frequency (perhaps using a cluster-based  
> permutation strategy)?  Any thoughts would be greatly appreciated.
>
> Thanks!
>
> Bornali
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu


************************************************************************************
Guillaume A. Rousselet, Ph.D., lecturer

School of Psychology
Institute of Neuroscience & Psychology
Centre for Cognitive Neuroimaging (CCNi)

The University of Glasgow, charity number SC004401

http://www.psy.gla.ac.uk/staff/index.php?id=GAR01

Email: g.rousselet at psy.gla.ac.uk
Fax. +44 (0)141 330 4606
Tel. +44 (0)141 330 6652
Cell +44 (0)791 779 7833


"Corporations have only one duty: to promote their own and their  
owners' interests. They have no capacity, and their executives no  
authority, to act out of a genuine sense of responsibility to society,  
to avoid causing harm to people and the environment, or to work to  
advance the public good in ways that are unrelated to their own self- 
interest. Deregulation thus rests upon the suspect premise that  
corporations will respect social and environmental interests without  
being compelled by government to do so. No one would seriously suggest  
that individuals should regulate themselves, that laws against murder,  
assault, and theft are unnecessary because people are socially  
responsible. Yet oddly, we are asked to believe that corporate persons  
- institutional psychopaths who lack any sense of moral conviction and  
who have the power and motivation to cause harm and devastation in the  
world - should be left free to govern themselves."

	Joel Bakan - The Corporation - 2004
************************************************************************************



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100708/cf4ab809/attachment.html>

From mschuber at mail.upb.de  Thu Jul  8 01:28:23 2010
From: mschuber at mail.upb.de (Michael Schubert)
Date: Thu, 08 Jul 2010 10:28:23 +0200
Subject: [Eeglablist] channel location file
Message-ID: <4C358C27.7080903@mail.upb.de>

Dear all,

are there any channel location files for the Neuroscan QuickCap40 (40 
channels) available? On the Neuroscan ftp I only found "dave.zip" which 
contains 32 and 64 channels + I get an error when I read those in, anyway.

And a second question:
I'm playing around with ICA to reject ocular artifacts from continuous 
data. Up to now I could find only few papers about the validity of 
ICA-corrected data. I'm asking myself that question because very often I 
get components representing the ocular artifacts, but they also seem to 
contain eeg-information, especially when I use the fastica algorithm. 
This results in quiet different looking signals in sections where no 
artifacts occur - mainly in frontal channels of course.

Any help is highly appreciated.

Thanks!

Michael


From nickbedo at yahoo.com  Fri Jul  9 18:46:54 2010
From: nickbedo at yahoo.com (Nick Bedo)
Date: Fri, 9 Jul 2010 18:46:54 -0700 (PDT)
Subject: [Eeglablist] Difference between ITC and Channel Cross-Coherence?
Message-ID: <34986.61502.qm@web62005.mail.re1.yahoo.com>

Hi everyone,

As I have collected more data, I have started using the Study structure to 
compile all participants' epoched datasets.  When I was looking through 
individual participants' datasets, I was using Channel Cross-Coherence (via 
pop_newcrossf) to get coherence values.  Now that everything is compiled into a 
Study, I see the ITC option.  How does ITC differ from the Cross-Coherence 
option that I was using before?

Thanks in advance,
Nick


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100709/f53adf1f/attachment.html>

From milltyl at gmail.com  Fri Jul  9 12:35:37 2010
From: milltyl at gmail.com (Tyler Miller)
Date: Fri, 9 Jul 2010 14:35:37 -0500
Subject: [Eeglablist] Spike exporting to eeglab
Message-ID: <AANLkTimP5wwGbkxN5GcEJeAdGarPuBiu6o2hWa8ZR8fh@mail.gmail.com>

Hello All,

Has anyone used Spike to collect neurodata and then transfer to eeglab?

I'm using Spike (version 7) to collect several channels of EEG/EMG/EDA/EKG
data and 4 channels of event and marker information. Once the data is
collected there is a option to export as a .mat file.

I assume the export procedure is working properly in Spike, however when I
attempt to open the file in eeglab, I get a "pop_edit setname ()" error
message. Any ideas why I would get this error message? From the archives
I've noticed some threads about "headers," is this my problem?

I'm new to eeglab and would appreciate any help. Thanks!

Respectfully,
Tyler

-- 
Tyler Miller
Department of Psychology
Texas A&M University
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100709/dfd703b6/attachment.html>

From Dorothy.Bishop at psy.ox.ac.uk  Sat Jul 10 09:48:37 2010
From: Dorothy.Bishop at psy.ox.ac.uk (Dorothy Bishop)
Date: Sat, 10 Jul 2010 17:48:37 +0100
Subject: [Eeglablist] plotphase in time-frequency analysis
Message-ID: <4C38B277.A716.003A.1@psy.ox.ac.uk>

Thanks to those who have commented on the blog I did on ERSP.
I am writing a bit more, but have a query about phase of ITC.

I've created simple simulated waveforms, made by summing sine waves with some variation from trial to trial of phase,  and adding a bit of noise, and then setting phase to be constant just for part of the waveform, t1 to t2.
 I was surprised to see that where I would have expected a block of positive ITC (ie in the interval t1-t2), I had instead a sequence of blocks of +ve and -ve ITC.
Visually there is an abrupt transition from blocks of red to blocks of  blue.

I can get the output to look as I expect it by setting 'plotphase' to 'off', and I notice that in some sample scripts, this option is set.

Does this mean that the polarity of ITC is irrelevant? I had thought that +ve meant phase was positively correlated from trial to trial, and -ve mean it was negatively correlated, but i then started to wonder whether the circular nature of phase meant that when something is totally in phase it might be equally likely to come out as + or -? 
Yours a bit confusedly.
thanks






Dorothy Bishop
Professor of Developmental Neuropsychology
Department of Experimental Psychology
University of Oxford
OX1 3UD
Website: http://psyweb.psy.ox.ac.uk/oscci/
Blog: http://deevybee.blogspot.com/
tel: +44 (0)1865 271369
fax: +44 (0)1865 281255
 





From balazs at cogpsyphy.hu  Mon Jul 12 06:43:45 2010
From: balazs at cogpsyphy.hu (Laszlo Balazs)
Date: Mon, 12 Jul 2010 15:43:45 +0200
Subject: [Eeglablist] channel location file
In-Reply-To: <4C358C27.7080903@mail.upb.de>
References: <4C358C27.7080903@mail.upb.de>
Message-ID: <4C3B1C11.50205@cogpsyphy.hu>

Hello Michael,
To your second question: We have just started to try ICA based EOG 
correction and I have the same concern about throwing out some baby with 
the bath water. Once I have tested the regression based method provided 
by Neuroscan. I chose a few subjects with frequent blinks but still high 
enough number of clean sweeps. Then I compared uncorrected and corrected 
sweeps by visual inspection and also the averaged ERP to targets in a 
visual oddball (blink free epochs only).  I have seen marked decrease in 
both EEG and ERP amplitude over Fp1/2. We plan to repeat this test soon 
including ICA and hope it will outperform the regression method. I 
wonder if there are papers or other ideas for validation tests around?
Kind regards,
Laszlo

> Dear all,
>
> are there any channel location files for the Neuroscan QuickCap40 (40 
> channels) available? On the Neuroscan ftp I only found "dave.zip" which 
> contains 32 and 64 channels + I get an error when I read those in, anyway.
>
> And a second question:
> I'm playing around with ICA to reject ocular artifacts from continuous 
> data. Up to now I could find only few papers about the validity of 
> ICA-corrected data. I'm asking myself that question because very often I 
> get components representing the ocular artifacts, but they also seem to 
> contain eeg-information, especially when I use the fastica algorithm. 
> This results in quiet different looking signals in sections where no 
> artifacts occur - mainly in frontal channels of course.
>
> Any help is highly appreciated.
>
> Thanks!
>
> Michael
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
>
>
> __________ Information from ESET NOD32 Antivirus, version of virus signature database 5271 (20100712) __________
>
> The message was checked by ESET NOD32 Antivirus.
>
> http://www.eset.com
>
>
>   


-- 

----------------------------------------------------------------------------
     Laszlo Balazs, Ph.D.          /      dr. Bal?zs L?szl?
Institute for Psychology HAS       /   MTA Pszichol?giai Kutat?it?zet
Budapest, P O B 398, H-1394 Hungary / Tel:+36(1)354-2410  Fax:+36(1)354-2416



__________ Information from ESET NOD32 Antivirus, version of virus signature database 5271 (20100712) __________

The message was checked by ESET NOD32 Antivirus.

http://www.eset.com




From Dorothy.Bishop at psy.ox.ac.uk  Sun Jul 11 02:16:11 2010
From: Dorothy.Bishop at psy.ox.ac.uk (Dorothy Bishop)
Date: Sun, 11 Jul 2010 10:16:11 +0100
Subject: [Eeglablist] 2nd instalment of time-freq analysis for beginners,
	by a	 nonexpert
Message-ID: <4C3999EB.A716.003A.1@psy.ox.ac.uk>

http://bishoptechbits.blogspot.com/



Dorothy Bishop
Professor of Developmental Neuropsychology
Department of Experimental Psychology
University of Oxford
OX1 3UD
Website: http://psyweb.psy.ox.ac.uk/oscci/
Blog: http://deevybee.blogspot.com/
tel: +44 (0)1865 271369
fax: +44 (0)1865 281255
 





From jdesjardins at brocku.ca  Sun Jul 11 10:57:34 2010
From: jdesjardins at brocku.ca (James Desjardins)
Date: Sun, 11 Jul 2010 13:57:34 -0400
Subject: [Eeglablist] plotphase in time-frequency analysis
In-Reply-To: <4C38B277.A716.003A.1@psy.ox.ac.uk>
References: <4C38B277.A716.003A.1@psy.ox.ac.uk>
Message-ID: <20100711135734.hmlmf0yu9wos840o@webmail.brocku.ca>

Hi Dorothy,

You are correct, the ITC should not be negative. The colour in the ITC  
plot when you have the "plotphase" option set to "on" only indicates  
the sign of the phase angle at each time point and frequency in the  
figure (it looks like the newtimef option name has been changed to  
"plotphasesign" to be less ambiguous). This option renders the ITC  
colour bar a bit misleading, but notice how the blue to red colour  
shifts mimic the phase angle shifts in the ERP plotted below the ITC  
figure.

Pasting the code below to your tutorial code following your call to  
newtimef will display a series of six compass plots that describe the  
data in the ITC figure.


%Store single trial data from TF decomposition in vector tfdata...
[ersp,itc,powbase,times,freqs,erspboot,itcboot,tfdata]=newtimef(  
mysig',2000,[-500 1500],1000,cyc,'plotphasesign','on');

text(52,20,'Complex wave');

%plot compass figure for 9.9Hz at times -21, 185, and 220 respectively...
figure;compass(tfdata(4,40,:));text(52,20,'TFDATA, t=-21ms, fq=9.9Hz');
figure;compass(tfdata(4,64,:));text(52,20,'TFDATA, t=185ms, fq=9.9Hz');
figure;compass(tfdata(4,68,:));text(52,20,'TFDATA, t=220ms, fq=9.9Hz');

%Set absolute value of complex coefficients = 1...
itcdata=tfdata./abs(tfdata);

figure;compass(itcdata(4,40,:))
hold on;compass(mean(itcdata(4,40,:),3),'r')

figure;compass(itcdata(4,64,:))
hold on;compass(mean(itcdata(4,64,:),3),'r')

figure;compass(itcdata(4,68,:))
hold on;compass(mean(itcdata(4,68,:),3),'r')
%-----------------------------------------------------------------------



The first three compass plots illustrate the complex coefficients  
produced by the wavelet decomposition called from the newtimef  
function. Each compass figure produces an arrow emanating from the  
origin of the circle for each trial at a specific time point and  
frequency. All of the compass plots present values for 9.9Hz (freqs=4)  
and the first three plots illustrate time point -21ms, 185ms and then  
220ms respectively. The length of each line in these figures  
represents the amplitude of the signal at that time and frequency  
(information used for ERSP) while the direction from the origin of the  
circle represents the phase angle for that time and frequency  
(information used for ITC).

ITC is calculated by setting the absolute value (length of the arrow  
in compass plots) of the complex coefficients to 1 and then  
calculating the average coefficient. This is illustrated in the second  
set of three compass plots. These three compass plots are identical to  
the first three except that the length of the blue arrows for each  
trial are fixed at 1. The red line in each plot represents the average  
coefficient. Notice how the length of this line corresponds to the  
values in ITC plots (unambiguously when "plotphasesign" is set to  
"off"), and that the length of this line is determined only by the  
degree to which the phase angles cluster together across trials at a  
give time point and frequency. The color shift in the ITC plots when  
"plotphasesign" is set to "on" only refers to the direction of the red  
line in these compass plots.

I hope that this is helpful. I appreciate the work that you are doing.

James Desjardins
Technician, MA Student
Department of Psychology, Behavioural Neuroscience
Cognitive and Affective Neuroscience Lab
Brock University
St. Catharines, ON, Canada


Quoting Dorothy Bishop <Dorothy.Bishop at psy.ox.ac.uk>:

> Thanks to those who have commented on the blog I did on ERSP.
> I am writing a bit more, but have a query about phase of ITC.
>
> I've created simple simulated waveforms, made by summing sine waves   
> with some variation from trial to trial of phase,  and adding a bit   
> of noise, and then setting phase to be constant just for part of the  
>  waveform, t1 to t2.
>  I was surprised to see that where I would have expected a block of   
> positive ITC (ie in the interval t1-t2), I had instead a sequence of  
>  blocks of +ve and -ve ITC.
> Visually there is an abrupt transition from blocks of red to blocks of  blue.
>
> I can get the output to look as I expect it by setting 'plotphase'   
> to 'off', and I notice that in some sample scripts, this option is   
> set.
>
> Does this mean that the polarity of ITC is irrelevant? I had thought  
>  that +ve meant phase was positively correlated from trial to trial,  
>  and -ve mean it was negatively correlated, but i then started to   
> wonder whether the circular nature of phase meant that when   
> something is totally in phase it might be equally likely to come out  
>  as + or -?
> Yours a bit confusedly.
> thanks
>
>
>
>
>
>
> Dorothy Bishop
> Professor of Developmental Neuropsychology
> Department of Experimental Psychology
> University of Oxford
> OX1 3UD
> Website: http://psyweb.psy.ox.ac.uk/oscci/
> Blog: http://deevybee.blogspot.com/
> tel: +44 (0)1865 271369
> fax: +44 (0)1865 281255
>
>
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to  
>  eeglablist-request at sccn.ucsd.edu
>






From jdesjardins at brocku.ca  Sun Jul 11 11:14:45 2010
From: jdesjardins at brocku.ca (James Desjardins)
Date: Sun, 11 Jul 2010 14:14:45 -0400
Subject: [Eeglablist] Difference between ITC and Channel	Cross-Coherence?
In-Reply-To: <34986.61502.qm@web62005.mail.re1.yahoo.com>
References: <34986.61502.qm@web62005.mail.re1.yahoo.com>
Message-ID: <20100711141445.0i7yrlz9eock4cs0@webmail.brocku.ca>

Hi Nick,

Cross-coherence and ITC both relate to the consistency of phase angle  
from trial to trial at each time point and frequency. The only  
difference is that ITC represents the consistency of the phase angle  
FOR A SINGLE SIGNAL across trials for each time point and frequency,  
while channel or IC cross-coherence represents the consistency of the  
phase angle DIFFERENCE BETWEEN TWO SIGNALS across trials for each time  
point and frequency.


James Desjardins
Technician, MA Student
Department of Psychology, Behavioural Neuroscience
Cognitive and Affective Neuroscience Lab
Brock University
St. Catharines, ON, Canada


Quoting Nick Bedo <nickbedo at yahoo.com>:

> Hi everyone,
>
> As I have collected more data, I have started using the Study structure to
> compile all participants' epoched datasets.  When I was looking through
> individual participants' datasets, I was using Channel Cross-Coherence (via
> pop_newcrossf) to get coherence values.  Now that everything is   
> compiled into a
> Study, I see the ITC option.  How does ITC differ from the Cross-Coherence
> option that I was using before?
>
> Thanks in advance,
> Nick
>
>
>






From Gerd.Waldhauser at psychology.lu.se  Tue Jul 13 02:24:18 2010
From: Gerd.Waldhauser at psychology.lu.se (Gerd Waldhauser)
Date: Tue, 13 Jul 2010 11:24:18 +0200
Subject: [Eeglablist] channel location file
In-Reply-To: <4C358C27.7080903@mail.upb.de>
References: <4C358C27.7080903@mail.upb.de>
Message-ID: <1AE83D2C4D186C4A88F447CA64120D80ED4B19C0FC@UWEXMBX01.uw.lu.se>

Hi Michael,

I guess you can always export your channel locations after opening your continuous or epoched file in Neuroscan (Edit > Channel Layout > Export positions). EEGLAB is able to read this file. However, this only works if you have defined your positions in Neuroscan before. To do this without a file you may want to try out, for example, the electrode location detection according to the 10-20 system in the Channel Layout interface in Neuroscan before exporting. However, you will have to check that it gets the locations right. No definite answer, but hope this helps a bit.

Best,
Gerd

-----Ursprungligt meddelande-----
Fr?n: Michael Schubert [mailto:mschuber at mail.upb.de] 
Skickat: den 8 juli 2010 10:28
Till: eeglablist at sccn.ucsd.edu
?mne: [Eeglablist] channel location file

Dear all,

are there any channel location files for the Neuroscan QuickCap40 (40 
channels) available? On the Neuroscan ftp I only found "dave.zip" which 
contains 32 and 64 channels + I get an error when I read those in, anyway.

And a second question:
I'm playing around with ICA to reject ocular artifacts from continuous 
data. Up to now I could find only few papers about the validity of 
ICA-corrected data. I'm asking myself that question because very often I 
get components representing the ocular artifacts, but they also seem to 
contain eeg-information, especially when I use the fastica algorithm. 
This results in quiet different looking signals in sections where no 
artifacts occur - mainly in frontal channels of course.

Any help is highly appreciated.

Thanks!

Michael




From r.oostenveld at fcdonders.ru.nl  Tue Jul 13 00:56:58 2010
From: r.oostenveld at fcdonders.ru.nl (Robert Oostenveld)
Date: Tue, 13 Jul 2010 09:56:58 +0200
Subject: [Eeglablist] Call for Papers - special issue - Academic Software
	Applications for Electromagnetic Brain Mapping Using MEG and EEG
Message-ID: <F08B152A-C64F-4878-9C65-A896AA45DC5A@fcdonders.ru.nl>

Dear EEGLAB users and software developers,

[[appologies for multiple postings]]

Let me draw your attention to the following call for papers. Please  
distribute this call for papers to anyone you think may be interested  
in submitting, and to relevant email discusison lists.

best regards,
Robert Oostenveld, PhD
Senior Researcher & MEG Physicist
Donders Institute for Brain, Cognition and Behaviour
Radboud University Nijmegen

--------------------------------------------------
http://www.hindawi.com/journals/cin/si/ebm.html
--------------------------------------------------

Academic Software Applications for Electromagnetic Brain Mapping Using  
MEG and EEG
Call for Papers

The field of Magnetoencephalography (MEG) and Electroencephalography  
(EEG) source imaging is maturing rapidly. This scientific growth is  
accompanied by a variety of complementary and/or concurrent software  
solutions from the academic world.

The objective of this CIN Special Issue is to help the neuroimaging  
obtain an overview of state-of-the-art academic software applications  
for MEG/EEG data analysis, how they differ and interact, and of  
upcoming methodological trends and technical developments; the topics  
to be covered include, but are not limited to, academic software  
solutions for:

MEG and EEG data acquisition
Data preprocessing, that is, filtering, artifact detection, rejection  
or correction, trial sorting, averaging
Segmentation and geometrical modeling of head tissues
Computational electromagnetics for MEG/EEG forward modeling
MEG/EEG source analysis
Statistical appraisal and inference: confidence intervals on measures  
and hypothesis testing
Identification and evaluation of evoked, induced event-related brain  
responses and ongoing brain activity
Time-frequency decompositions, advanced spectral analysis, time series  
modeling
Estimation of functional and effective connectivity
Authors should provide detailed information regarding their software  
toolbox or application by addressing the following topics: open source  
software (yes/no), i/o file formats available, operating system,  
Matlab required (yes/no), interoperability with other software, and so  
forth.

Further, the software needs to be available for download free of  
charge at the time of manuscript submission, with sufficient  
documentation provided online to be able to reproduce the data  
analysis featured in the manuscript.

Before submission authors should carefully read over the journal's  
Author Guidelines, which are located athttp://www.hindawi.com/journals/cin/guidelines.html 
. Prospective authors should submit an electronic copy of their  
complete manuscript through the journal Manuscript Tracking System at http://mts.hindawi.com/ 
  according to the following timetable:

Manuscript Due	September 1, 2010
First Round of Reviews	December 1, 2010
Publication Date	March 1, 2011

Lead Guest Editor
Sylvain Baillet, Department of Neurology & Biophysics, Medical College  
of Wisconsin, WI, USA

Guest Editors
Karl Friston, Wellcome Trust Centre for Neuroimaging, London, UK
Robert Oostenveld, Donders Centre for Cognitive Neuroimaging Radboud  
University Nijmegen, The Netherlands






From Kris.Baetens at vub.ac.be  Thu Jul 15 04:24:35 2010
From: Kris.Baetens at vub.ac.be (Kris Baetens)
Date: Thu, 15 Jul 2010 13:24:35 +0200
Subject: [Eeglablist] ICA as artefact correction method - dilemma
Message-ID: <47294c3eeff325dee@wm-srv.ulb.ac.be>

Dear all,

I have been messing for some time with filter/ICA issues. I would be very grateful if anybody could shed some light on the matter. 

In a number of experiments, I have used sentences as stimulus material. We collected ERP responses to the final word of the last of a series of sentences and are interested in N400 and P300-like effects. Participants were instructed by means of an icon to ?do their blinking? as much as possible during short pauses of a few seconds that followed about 2 seconds after each final sentence. We have used a DC amplifier with an average recording reference. 

Regardless of whether I use FIR or IIR filters, the higher my high-pass filter cut-off, the more drift I get in the participant ERP averages following the final sentences. That is, if I use a 6th order two-way Butterworth filter with half-amplitude cut-off of 0.01 Hz, for example, there is no particular drift in the ERP following the critical end sentences, whereas a similar filter with a 0.3Hz cut-off results in drifts that go from 0 to 30 ?V over the course of a one second in participant averages.  
These drifts are outspoken in the vertical EOG channel but in the frontal channels as well. Considering the fact that many trials are followed by eye blinks (+/-2 or three seconds after the time lock), it seems obvious that the drift is a result of the eye blinks and the filtering applied to them. 
However, the ?normal? drift left in the trials (taking all channels into account) is much higher when I use a 0.01Hz high pass than when I use a 03Hz high pass, as one would expect.  

I'm wrestling a bit with the following dilemma:
-I have seen that when I use an adequate high-pass filter (0.5 or 1Hz) I get a very nice decomposition of my data, enabling the precise removal of eye blink activity, jaw muscle activation etcetera. 
However, when using such filters, I get enormous drifts in the frontal channels, as explained above (and somehow, this doesn?t  attract too much ?attention? of the ICA algorithm, still enabling a proper decomposition). Also, I am concerned that using such filters in classical ERP research might cause some problems (cf. Prof. Luck?s book), especially when the ERP components of interest are rather big slow ones like the N400 and P300. 
-On the other hand, when using a high-pass filter in the range of 0.01 ? 0.1Hz (as is recommended by many), the ICA algorithm fails to decompose the data well. I can still get rid of some substantial EOG activity, but no real proper correction.

My questions are the following:
-Given the fact that the ICA algorithm works well only when one uses high pass filters in the range of 0.5-1Hz, and that using such filters is most often advised against by people working in classical ERP research, is ICA really utilizable in classical ERP-grand-average-style research as a method of eye blink correction?
-Is it generally a bad idea to instruct participants to do their blinking at a fixed moment that starts a few seconds after your time-lock stimulus?
-What sort of distortions or invalid conclusions could possibly arise from using high-threshold high pass filtering (i.e., 0.5Hz 6th order Butterworth) when one applies it to all conditions, on a grand average ERP-level? 
-What sort of high-pass filter would you advise in general for DC recordings? 

Many thanks in any case,

Kris Baetens
Ph.D. fellow of the Research Foundation - Flanders (FWO) 
Dept. Experimental and Applied Psychology
Faculty of Psychology and Educational Sciences
Vrije Universiteit Brussel
Pleinlaan 2, 1050 Elsene 
+32 2 629 23 31




From l.garcia.d at gmail.com  Thu Jul 15 06:55:51 2010
From: l.garcia.d at gmail.com (Luis Garcia Dominguez)
Date: Thu, 15 Jul 2010 09:55:51 -0400
Subject: [Eeglablist] DEL2MAP
Message-ID: <AANLkTiktr-QejR1pCSzJuKIiLo4XNFlK5L_tkNKCNa4a@mail.gmail.com>

Hello everybody,

I am using the DEL2MAP function to convert the current eeg montage to
laplacian, but noticed the function is pretty slow when fed with data
matrices, in my computer a segment of 32-by-200 runs in about 10 sec.
The bottleneck seems to be in the function GDATAV4 that is called by
GRIDDATA, in particular a double FOR-cicle inside that function. Have
any of you consider the vectorization of that function or any other
way to improve speed?

Thanks


From nickbedo at yahoo.com  Thu Jul 15 11:24:59 2010
From: nickbedo at yahoo.com (Nick Bedo)
Date: Thu, 15 Jul 2010 11:24:59 -0700 (PDT)
Subject: [Eeglablist] Re-referencing -- do I need to?
Message-ID: <573320.97298.qm@web62001.mail.re1.yahoo.com>

Hi everyone,


My recordings were referenced to both mastoids, but I don't have a channel that 
contains those data.  Is re-referencing advised/necessary in my case?  I was 
playing around with re-referencing to the average, and it really warps the 
waveform amplitudes compared to the ERPs calculated with our original methods. 
 Any input would be helpful.

Thanks in advance,
Nick


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100715/b34980e4/attachment.html>

From milltyl at gmail.com  Thu Jul 15 15:41:28 2010
From: milltyl at gmail.com (Tyler Miller)
Date: Thu, 15 Jul 2010 17:41:28 -0500
Subject: [Eeglablist] Spike exporting to eeglab
In-Reply-To: <AANLkTim01dwmy0ehrDvhuKyRv5o4jko1HgmD29b7RUmh@mail.gmail.com>
References: <AANLkTimP5wwGbkxN5GcEJeAdGarPuBiu6o2hWa8ZR8fh@mail.gmail.com>
	<AANLkTim01dwmy0ehrDvhuKyRv5o4jko1HgmD29b7RUmh@mail.gmail.com>
Message-ID: <AANLkTikkLU7ap2fp78ipVdvQCpFLRwfupAHhdOpgEAXY@mail.gmail.com>

Thank you for your suggestion Ehsan, but can you elaborate? I'm starting to
think the the problem I'm having isn't importing the file to Matlab, but
rather my problem is getting eeglab to open it from Matlab.

Using the following point-and-click commands in eeglab: menu --> Import data
--> from ASCII/float file or Matlab array, a menu opens that allows me to
browse and specify the file I need opened. If I browse for the file and
select it, the first dropdown box in the menu automatically changes to
"Matlab .mat file." To me, this means eeglab is recognizing the file.

If all I do is load the file and select "Ok," I get the 'pop_editset()
erorr: cannot read .mat file 'file path''

Are the channel locations required to open the file data set?

Respectfully,
Tyler Miller



> I am using MATLAB SON library written by Malcolm Lidierth to import the
> spike data into MATLAB environment. Once you able to do this, I think you
> will be able to use the imported files in EEGlab as well.
> Attached is the manual for this library.
>
> Have fun,
>

Ehsan Negahbani
Ph.D. Student
School of Engineering
Faculty of Science & Engineering
The University of Waikato
Private Bag 3105
Hamilton 3240, NZ
enegahbani at gmail.com
ehsann at waikato.ac.nz

>
>
> On Sat, Jul 10, 2010 at 7:35 AM, Tyler Miller <milltyl at gmail.com> wrote:
>
>> Hello All,
>>
>> Has anyone used Spike to collect neurodata and then transfer to eeglab?
>>
>> I'm using Spike (version 7) to collect several channels of EEG/EMG/EDA/EKG
>> data and 4 channels of event and marker information. Once the data is
>> collected there is a option to export as a .mat file.
>>
>> I assume the export procedure is working properly in Spike, however when I
>> attempt to open the file in eeglab, I get a "pop_edit setname ()" error
>> message. Any ideas why I would get this error message? From the archives
>> I've noticed some threads about "headers," is this my problem?
>>
>> I'm new to eeglab and would appreciate any help. Thanks!
>>
>> Respectfully,
>> Tyler
>>
>> --
>> Tyler Miller
>> Department of Psychology
>> Texas A&M University
>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100715/3d1c21ba/attachment.html>

From shackman at wisc.edu  Thu Jul 15 12:56:08 2010
From: shackman at wisc.edu (Alexander J. Shackman)
Date: Thu, 15 Jul 2010 14:56:08 -0500
Subject: [Eeglablist] ICA as artefact correction method - dilemma
In-Reply-To: <47294c3eeff325dee@wm-srv.ulb.ac.be>
References: <47294c3eeff325dee@wm-srv.ulb.ac.be>
Message-ID: <AANLkTinlSnt1aZnJmsYp9kc67LTG3AcIGJDBNyyRk4y_@mail.gmail.com>

hi,

this issue cropped up in a paper we recently submitted. a reviewer made an
interesting suggestion that may address your concern:

"A hp filter of approx. 1 Hz clearly helps to reduce the complexity of the
data, as it reduces non-stationarity introduced by channel drift
(stationarity is one of the ICA assumptions). I am not aware of any
published 'definitive tests' demonstrating that hp filtering helps to
improve ICA quality and reliability, but in my experience (and the
experience of many other ICA users) this clearly is the case. A hp filter of
1 or 2 Hz would however clearly ruin many ERP components. To bypass this
problem, ICA training could be done on hp filtered data and the resulting
weights be applied to the non-filtered data. As far as I remember, one of
the Onton ICA reviews discusses this strategy and the effect of hp
filtering."

i would be very interested in knowing if anyone has experience with this
strategy.

best wishes,
alex

On Thu, Jul 15, 2010 at 6:24 AM, Kris Baetens <Kris.Baetens at vub.ac.be>wrote:

> Dear all,
>
> I have been messing for some time with filter/ICA issues. I would be very
> grateful if anybody could shed some light on the matter.
>
> In a number of experiments, I have used sentences as stimulus material. We
> collected ERP responses to the final word of the last of a series of
> sentences and are interested in N400 and P300-like effects. Participants
> were instructed by means of an icon to ?do their blinking? as much as
> possible during short pauses of a few seconds that followed about 2 seconds
> after each final sentence. We have used a DC amplifier with an average
> recording reference.
>
> Regardless of whether I use FIR or IIR filters, the higher my high-pass
> filter cut-off, the more drift I get in the participant ERP averages
> following the final sentences. That is, if I use a 6th order two-way
> Butterworth filter with half-amplitude cut-off of 0.01 Hz, for example,
> there is no particular drift in the ERP following the critical end
> sentences, whereas a similar filter with a 0.3Hz cut-off results in drifts
> that go from 0 to 30 ?V over the course of a one second in participant
> averages.
> These drifts are outspoken in the vertical EOG channel but in the frontal
> channels as well. Considering the fact that many trials are followed by eye
> blinks (+/-2 or three seconds after the time lock), it seems obvious that
> the drift is a result of the eye blinks and the filtering applied to them.
> However, the ?normal? drift left in the trials (taking all channels into
> account) is much higher when I use a 0.01Hz high pass than when I use a 03Hz
> high pass, as one would expect.
>
> I'm wrestling a bit with the following dilemma:
> -I have seen that when I use an adequate high-pass filter (0.5 or 1Hz) I
> get a very nice decomposition of my data, enabling the precise removal of
> eye blink activity, jaw muscle activation etcetera.
> However, when using such filters, I get enormous drifts in the frontal
> channels, as explained above (and somehow, this doesn?t  attract too much
> ?attention? of the ICA algorithm, still enabling a proper decomposition).
> Also, I am concerned that using such filters in classical ERP research might
> cause some problems (cf. Prof. Luck?s book), especially when the ERP
> components of interest are rather big slow ones like the N400 and P300.
> -On the other hand, when using a high-pass filter in the range of 0.01 ?
> 0.1Hz (as is recommended by many), the ICA algorithm fails to decompose the
> data well. I can still get rid of some substantial EOG activity, but no real
> proper correction.
>
> My questions are the following:
> -Given the fact that the ICA algorithm works well only when one uses high
> pass filters in the range of 0.5-1Hz, and that using such filters is most
> often advised against by people working in classical ERP research, is ICA
> really utilizable in classical ERP-grand-average-style research as a method
> of eye blink correction?
> -Is it generally a bad idea to instruct participants to do their blinking
> at a fixed moment that starts a few seconds after your time-lock stimulus?
> -What sort of distortions or invalid conclusions could possibly arise from
> using high-threshold high pass filtering (i.e., 0.5Hz 6th order Butterworth)
> when one applies it to all conditions, on a grand average ERP-level?
> -What sort of high-pass filter would you advise in general for DC
> recordings?
>
> Many thanks in any case,
>
> Kris Baetens
> Ph.D. fellow of the Research Foundation - Flanders (FWO)
> Dept. Experimental and Applied Psychology
> Faculty of Psychology and Educational Sciences
> Vrije Universiteit Brussel
> Pleinlaan 2, 1050 Elsene
> +32 2 629 23 31
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Alexander J. Shackman, Ph.D.
Wisconsin Psychiatric Institute & Clinics and
Department of Psychology
University of Wisconsin-Madison
1202 West Johnson Street
Madison, Wisconsin 53706

Telephone: +1 (608) 358-5025
Fax: +1 (608) 265-2875
Email: shackman at wisc.edu
http://psyphz.psych.wisc.edu/~shackman
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100715/76b5e67b/attachment.html>

From dgroppe at cogsci.ucsd.edu  Thu Jul 15 20:20:47 2010
From: dgroppe at cogsci.ucsd.edu (David Groppe)
Date: Thu, 15 Jul 2010 20:20:47 -0700
Subject: [Eeglablist] Re-referencing -- do I need to?
In-Reply-To: <573320.97298.qm@web62001.mail.re1.yahoo.com>
References: <573320.97298.qm@web62001.mail.re1.yahoo.com>
Message-ID: <AANLkTimxY9ZjaD-_MrVJbjLW-SWdkIbq2AiwB923SXY8@mail.gmail.com>

Hi Nick,
  I highly recommend reading the bit in Steve Luck's book on referencing:
http://mitpress.mit.edu/catalog/item/default.asp?ttype=2&tid=10677

It does a great job covering the pros and cons of different common
references (including the average reference and bimastoid).
The rest of the book is well worth reading too.
      -D


On Thu, Jul 15, 2010 at 11:24 AM, Nick Bedo <nickbedo at yahoo.com> wrote:
> Hi everyone,
>
> My recordings were referenced to both mastoids, but I don't have a channel
> that contains those data. ?Is re-referencing advised/necessary in my case?
> ?I was playing around with re-referencing to the average, and it really
> warps the waveform amplitudes compared to the ERPs calculated with our
> original methods. ?Any input would be helpful.
> Thanks in advance,
> Nick
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



From lutobu at gmail.com  Thu Jul 15 20:58:00 2010
From: lutobu at gmail.com (ludwing torres)
Date: Fri, 16 Jul 2010 05:58:00 +0200
Subject: [Eeglablist] How can I specify nore than two a dipoles to do a
	dipole simulation and a custom timecourse
Message-ID: <AANLkTikqzb_v8b_q_rBLdY_GiESpu_29dLvPpVoCNi3o@mail.gmail.com>

Hi everyone, I'm starting to use the field trip toolbox and I've read the
example in the link below, to compute forward simulated data and apply
dipole fitting, but the example describes how to do it for two dipoles only:
http://fieldtrip.fcdonders.nl/example/compute_forward_simulated_data_and_apply_a_dipole_fit
my question is how to do it with more than two dipoles. I've tried to aply
the same scheme for 128 dipoles
I've done this for the dipole moments part:
>> for i=1:128
moms(i,3*(i-1)+1)=1;
end
>> moms=moms';
>>cfg.dip.mom=moms;
>> for i=1:128
si(i,:)=sin(i*3*time*2*pi);
end
cfg.dip.signal={si};
and I'm getting the next error message in the dipolefitting part:
>> dip1 = dipolefitting(cfg, avg3);
the input is timelock data with 30 channels and 250 timebins
using headmodel specified in the configuration
using electrodes specified in the configuration
selected 30 channels
selected 250 topographies
Warning: not enough channels to perform a dipole fit
> In dipolefitting at 438
??? Error using ==> dipolefitting at 471
inconsistent number of dipoles in configuration
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100716/ec6d3096/attachment.html>

From silvia.corbera at yale.edu  Fri Jul 16 11:40:25 2010
From: silvia.corbera at yale.edu (Silvia Corbera)
Date: Fri, 16 Jul 2010 13:40:25 -0500
Subject: [Eeglablist] mean amplitudes
Message-ID: <4C40A799.8010508@yale.edu>

Dear everybody,

I am trying to calculate the mean amplitudes of several ERP components 
in my average files and I don't know how to do it.
How do you calculate mean amplitudes in each electrode? I know you can 
go to the plot and see the pics and the values of the pics but I don't 
know how to get the mean average of specific time interval.

Thank you so much,
s

-- 
Silvia Corbera, Ph.D.
Post-Doctorate Associate                                                                                                                    
Yale School of Medicine                                                                                                            
Department of Psychiatry                                                                                                  
Connecticut Mental Health Center                                                                                                  
34 Park Street, New Haven CT 06519                                                                                         
(203) 974 7862                                                                                                     
silvia.corbera at yale.edu




From Dorothy.Bishop at psy.ox.ac.uk  Mon Jul 19 00:34:46 2010
From: Dorothy.Bishop at psy.ox.ac.uk (Dorothy Bishop)
Date: Mon, 19 Jul 2010 08:34:46 +0100
Subject: [Eeglablist] 3rd instalment of simplified intro to time-frequency
	analysis
Message-ID: <4C440E27.A716.003A.1@psy.ox.ac.uk>

Is now available on
http://bishoptechbits.blogspot.com/2010/07/simplified-introduction-part-3.html 



Dorothy Bishop
Professor of Developmental Neuropsychology
Department of Experimental Psychology
University of Oxford
OX1 3UD
Website: http://psyweb.psy.ox.ac.uk/oscci/
Blog: http://deevybee.blogspot.com/
tel: +44 (0)1865 271369
fax: +44 (0)1865 281255
 





From Gregor.Volberg at psychologie.uni-regensburg.de  Sun Jul 18 21:56:38 2010
From: Gregor.Volberg at psychologie.uni-regensburg.de (Gregor Volberg)
Date: Mon, 19 Jul 2010 06:56:38 +0200
Subject: [Eeglablist] Antw:  DEL2MAP
In-Reply-To: <AANLkTiktr-QejR1pCSzJuKIiLo4XNFlK5L_tkNKCNa4a@mail.gmail.com>
References: <AANLkTiktr-QejR1pCSzJuKIiLo4XNFlK5L_tkNKCNa4a@mail.gmail.com>
Message-ID: <4C43F7260200005700006E92@gwsmtp1.uni-regensburg.de>

Hi Luis,

you might want to have a look ath the CSD toolbox by J?rgen Kayser for that
purpose,
http://psychophysiology.cpmc.columbia.edu/Software/CSDtoolbox/index.html. It
is easy to use and much faster than del2map.

Cheers,
Gregor





-- 
Dr. rer. nat. Gregor Volberg <gregor.volberg at psychologie.uni-regensburg.de> (
mailto:gregor.volberg at psychologie.uni-regensburg.de )
University of Regensburg
Institute for Experimental Psychology
93040 Regensburg, Germany
Tel: +49 941 943 3862 
Fax: +49 941 943 3233
http://www.psychologie.uni-regensburg.de/Greenlee/team/volberg/volberg.html
>>> Luis Garcia Dominguez <l.garcia.d at gmail.com> 15.07.2010 15:55 >>>
Hello everybody,

I am using the DEL2MAP function to convert the current eeg montage to
laplacian, but noticed the function is pretty slow when fed with data
matrices, in my computer a segment of 32-by-200 runs in about 10 sec.
The bottleneck seems to be in the function GDATAV4 that is called by
GRIDDATA, in particular a double FOR-cicle inside that function. Have
any of you consider the vectorization of that function or any other
way to improve speed?

Thanks
_______________________________________________
Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html 
To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu 
For digest mode, send an email with the subject "set digest mime" to
eeglablist-request at sccn.ucsd.edu


From boazsadeh at gmail.com  Sat Jul 17 02:37:09 2010
From: boazsadeh at gmail.com (boaz sadeh)
Date: Sat, 17 Jul 2010 12:37:09 +0300
Subject: [Eeglablist] Reversing data back to generic format with the BVA
	import/export plugin
Message-ID: <AANLkTinUm-E8zEAhB0j6T2Rff0siYRswyZBJBOZR5NY3@mail.gmail.com>

Dear mailing list members,

I have a question regarding the BVA import/export eeglab plugin. When I
reverse .set files back to generic data format (like those
originally produced by BrainProducts systems) with the latest version of the
plugin that I know of (bvaio 1.57), data are converted fine but stimulus
triggers are missing. As a matter of fact a single trigger manages to appear
in the new file created, among the few dozens of triggers that are present
in the .set file that I depart from. This single trigger that is converted
properly is always the one before last trigger. In all datasets I tried,
this phenomenon repeats itself: only the one before last trigger
survives. This looks quite strange to me...

if any of you knows the problem I encounter I will be happy to hear it and
learn the way to reverse the data properly.

Many thanks,

Boaz
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100717/e7497219/attachment.html>

From yhe at rotman-baycrest.on.ca  Tue Jul 20 12:28:44 2010
From: yhe at rotman-baycrest.on.ca (Yu He)
Date: Tue, 20 Jul 2010 15:28:44 -0400
Subject: [Eeglablist] time-frequency plots of all channels
Message-ID: <4C45F8EC.8040908@rotman-baycrest.on.ca>

  Deal all,

I am  new to EEGlab. With the script 'newtimef' you will get a plot for 
one channel each time.
Could be possible to put the plots of all channels onto one figure?

Does anyone has a script for this multichannel plotting of time 
frequency analysis they would be will to share?

Thanks very much in advance for any response.

Yu
Rotman Research Institute
Toronto



From bross at rotman-baycrest.on.ca  Wed Jul 21 15:24:55 2010
From: bross at rotman-baycrest.on.ca (Bernhard Ross)
Date: Wed, 21 Jul 2010 18:24:55 -0400
Subject: [Eeglablist] MEG programmer position available in Toronto
Message-ID: <4C4773B7.4060204@rotman-baycrest.on.ca>


A full time position of a MEG Programmer is available at the Rotman 
Research Institute in Toronto, Canada.

PRIMARY RESPONSIBILITIES

The MEG programmer will be responsible for:
-- Performing all steps of advanced MEG data analysis;
-- Supporting scientists and students with their data analysis;
-- Developing new software tools;
-- Development of advanced data analysis tools which would include 
improving MEG source analysis approaches;
-- Integrating imaging modalities and integrating experimental 
procedures of MEG with eye tracking and TMS;
-- Training students and post-doctoral fellows in using MEG data 
acquisition and data analysis
-- Maintaining the MEG equipment

QUALIFICATIONS:
-- M.Sc. degree in Physics, Biomedical Engineering, Electrical 
Engineering, Computer Sciences or other relevant disciplines.
-- Advanced programming skills primarily for MATLAB, shell scripting, 
and LINUX administration are essential, as well as a strong motivation 
to improve and extend the skills in these fields
-- Familiarity with principles of signal processing in time and 
frequency domain, statistical methods of data analysis
-- Previous experience with MEG/EEG data acquisition and analysis or 
medical imaging analysis (e.g. fMRI) is highly desirable.


-- 
Bernhard Ross Ph.D.
Dept. of Medical Biophysics, University of Toronto
Baycrest Centre, Rotman Research Institute
3560 Bathurst Street
Toronto, ON, Canada M6A 2E1
phone: +1 416 785 2500 ext 2690
fax:   +1 416 785 2862
e-mail: <bross at rotman-baycrest.on.ca>
WEB:    <http://www.rotman-baycrest.on.ca/index.php?section=215>
WEB:    <http://www.baycrest.org/>


From liam.kilmartin at nuigalway.ie  Wed Jul 21 15:41:16 2010
From: liam.kilmartin at nuigalway.ie (Kilmartin, Liam)
Date: Wed, 21 Jul 2010 23:41:16 +0100
Subject: [Eeglablist] EEG signal preprocessing prior to coherence analysis
Message-ID: <02E3C88A420BA84A85B775AEB360779A034638E5@EVS1.ac.nuigalway.ie>

Hi,
 
I am relatively new to EEG signal analysis though I do have significant experience in signal processing in other application spaces. I plan to undertake a coherence analysis using ERLCOH (using newcrossf) on a common average referenced 64 channel EEG database. This study will examine the response of subjects to visual stimuli. The raw EEG signals have been ocular corrected and epoched in advance of my analysis (with trial rejection due to excessive residual artefacts). My concerns now related to what form of additional pre-processing (if any) would be expected prior to a linear coherence analysis specifically to deal with the volume conduction problem.
 
The specific questions that I have which I would appreciate any comments on:
 
(a) Does the use of a common average reference during acquisition in any way significantly address the volume conduction problem in the context of a coherence analysis ?
 
(b) Some apparently basic approaches that I have come across in the literature which attempt to address the issue of volume conduction resulting in an over-estimation of "real" coherence include:
 
    (i) Considering only the imaginary part of coherence to provide a lower bound estimate for "real" coherence
    (ii) Ignoring coherence magnitude values which have coherence phase angles close to 0 degrees (as these are likely due to volume conduction)
 
Intuitively it would appear to me that techniques such as these are likely to under-estimate "real" coherence (admittedly perhaps not to the same degree which ignoring them will tend to over-estimate coherence). Do any of these techniques offer acceptable solutions to the problem ?
 
(c) More complex approaches which I have also encountered in the literature involve source estimation approaches or the application of  laplacian surface models prior to coherence analysis. Would such approaches be viewed as the "gold standard" approach and hence would the use of such techniques be an  expectation of most reviewers of any future publications in this area ?
 
I appreciate that there may be no definitive answer to this question and which approach is best may be a matter of opinion, however any comments on these issues would be much appreciated.
 
Regards,
 
Liam
 
Liam Kilmartin
College of Engineering and Informatics,
National University of Ireland, Galway,
IRELAND 
 



From finnigan.simon at gmail.com  Thu Jul 22 00:17:22 2010
From: finnigan.simon at gmail.com (Simon Finnigan)
Date: Thu, 22 Jul 2010 17:17:22 +1000
Subject: [Eeglablist] EEG data from Compumedics system
Message-ID: <AANLkTinWPwvFHgXD3CQTdX9VawXx8mDXysEk-OgHY41X@mail.gmail.com>

hi,

does anyone have any experience with opening/analysing [in EEGLAB] data from
a Compumedics [clinical] EEG system? If so please let me know and I'd like
to follow up with a specific question or two.
[I believe ProFusion would be the standard commercial software routinely
used in clinical settings for viewing such files.]

cheers, Simon
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100722/57a0834e/attachment.html>

From ehong at mprc.umaryland.edu  Thu Jul 22 08:36:59 2010
From: ehong at mprc.umaryland.edu (Elliot Hong)
Date: Thu, 22 Jul 2010 11:36:59 -0400
Subject: [Eeglablist] Electrophysiology position at Baltimore, Maryland
Message-ID: <sc482d7a.059@MPRC.UMARYLAND.EDU>

Open to applicant with good signal processing and scripting skills. Some human data collection requirement. Post-doc in neuroscience, psychology, computer, engineering or related fields. Will consider other background and degree with skills and motivation. Interest in electrophysiology experiments for translational, genetics or clinical applications is a plus. fMRI, TMS, eye movement for genetics and psychosis studies are some of the other active areas in the lab. Can sponsor J-1 or H-1 for qualified applicant. Independent studies will be encouraged. If interested, please contact Elliot Hong (Tel: 410 402 6828; ehong at mprc.umaryland.edu) with a resume and a note.
 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100722/2beed056/attachment.html>

From awainselboim at yahoo.com.ar  Thu Jul 22 07:19:23 2010
From: awainselboim at yahoo.com.ar (Alejandro Wainselboim)
Date: Thu, 22 Jul 2010 07:19:23 -0700 (PDT)
Subject: [Eeglablist] Variables from ERSP analysis
Message-ID: <958060.47260.qm@web32802.mail.mud.yahoo.com>





Dear all, 

?????????????
I have recently performed an ERSP analysis with EEGLAB of a study composed with
data from 24 participants that underwent 2 different experimental conditions
with a 19 channel eeg recording system. Epoch time limits were -200ms to +2500
ms. The newtimef parameters that I used were: 





 'cycles',
[0.5 0.1], 'padratio', [1],? 'alpha', [NaN],? 'freqscale',
['linear]', 'scale', ['abs'], 'nfreqs', [100], 'freqs', [1.1 20]

?

According to this I would have 100
frequencies and 200 time points analyzed.

?

I see that the *.DATERSP files that are
generated after running the analysis for each participant include a chan_ersp
and a chan_erspbase variable for each channel. The? chan_ersp
variables are matrices composed of 100 rows and 200 columns, which I
understand represent the values in that channel? for each of the 100
frequencies? (rows) at each of the analyzed time points (columns). I would
like to know ?in what scale should I interpret these values??

?

On the other hand, the chan_erspbase variables
are matrices composed of 1 row and 100 columns. I understand that the values
obtained in this case represent the mean? power? for each frequency
(column) calculated along the entire baseline period (from -200 to 0 ms in this
case). ?Is this interpretation correct? ?In what scale should I? read the
obtained values of the chan_erspbase matrices?

?

Thank you in advance for your help, 

? ? ? ? ? ?
? ? ? ? ? ? ? ? ? ? ?
? ? ? ? ? ? ? ?? Alejandro
Wainselboim

? ? ? ? ? ?
? ? ? ? ? ? ? ? ? ? ?
? ? ? ? ? ? ? ? ?? Behavioral
Biology??

????????????????????????????????????????????????????
IBYME-CONICET

??????????????????????????????????????????????????
Vuelta de Obligado 2490

??????????????????????????????????????????????????
Buenos Aires, Argentina




      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100722/0f51bb7c/attachment.html>

From smakeig at gmail.com  Sun Jul 25 17:55:52 2010
From: smakeig at gmail.com (Scott Makeig)
Date: Sun, 25 Jul 2010 17:55:52 -0700
Subject: [Eeglablist] Postdoctoral research and lab manager positions at
	SCCN / UCSD and at Aberdeen, Maryland
Message-ID: <AANLkTi=whr1_ACBZNhixvg+-D6kZ97NKmMesfYE3nfEx@mail.gmail.com>

**
*

Several positions are now available at the Swartz Center, UCSD and at ARL
Aberdeen, Maryland in a large  collaborative project on cognitive monitoring
using concurrent EEG, behavioral, and environmental data. Please forward
this message to qualified potential applicants.


Contact Scott Makeig    (smakeig at ucsd.edu)
***

*
*

*Mobile Brain/body Imaging laboratory manager*: A senior scientific research
associate will work with faculty researchers to develop and supervise two
innovative, spatially adjacent laboratories for conducting new experiments
in Mobile Brain/body Imaging (MoBI) combining concurrent high-density EEG,
body motion capture, eye tracking, scene recording, and psychophysiology
measurements. The lab manager will develop hardware and facilities plans
with PIs and experimenters, will coordinate purchase and integration of
equipment and software, manage lab scheduling, archiving and preprocessing
of the data, and will supervise laboratory technicians who schedule and
conduct experiments and maintain laboratory supplies.

*Postdoctoral researchers *: Four postdoctoral positions are now available
at the Swartz Center for Computational Neuroscience under a basic research
project, "Cognition and Neuroergonomics," to develop neurocognitive
monitoring based on simultaneous high-density EEG, body motion capture, eye
tracking, and psychophysiological data recording and analysis for and with
the Army Research Laboratory, Aberdeen, Maryland.

*1. Statistical analysis of multimodal brain/body imaging data. *This
researcher will work with Scott Makeig and a multi-disciplinary team of
Swartz Center, UCSD, and other university and ARL researchers to develop,
implement, and apply *methods to fuse and mine multimodal mobile brain/body
imaging (MoBI) data for information about arousal, distribution of
attention, evaluation of events, motivation, and behavioral intent* of
operators working in information-rich, enclosed control environments.
Methods to be employed include sparse and complete independent component
analysis (ICA), related non-stationary factor analysis, and distributional
data modeling. Requirements include a PhD in mathematics, statistics, signal
processing, or related field with a concentration in modeling and
statistical signal processing and machine learning / data mining methods.

*2. Brain dynamics supporting active vision. *This researcher will work with
Scott Makeig and a multi-disciplinary team of Swartz Center, UCSD, other
university and ARL researchers to develop, implement, and apply methods for
fusing task, sensory, and eye tracking information with simultaneously
recorded EEG and body motion capture data to *better describe the
macroscopic EEG brain dynamics supporting active vision*, with a goal of
developing basic principles underlying use of concurrent EEG and eye
tracking data for monitoring operators working in information-rich, enclosed
control environments. Requirements include a PhD in cognitive science,
biomedical engineering, or related field with a concentration or significant
experience in modeling, signal processing, and statistical data mining
methods.

*3. Brain dynamics supporting motivated behavior. *This researcher will work
with Scott Makeig and a multi-disciplinary team of Swartz Center, UCSD, and
other university and ARL researchers to develop, implement, and apply
methods for fusing task, sensory, body motion capture, and electromyographic
(EMG) data with concurrent eye tracking, psychophysiological and
high-density EEG to *better model the task-motivated motor behavior of
operators and the macroscopic EEG brain dynamics supporting that behavior*,
with a goal of developing basic principles underlying use of concurrent EEG
and body motion data for monitoring the cognition of operators working in
information-rich, enclosed control environments. Requirements include a PhD
in biomechanics, biomedical engineering, neuroscience, or related field with
a concentration or in depth experience in modeling and statistical signal
processing and machine learning methods.

*4. EEG brain dynamics supporting active cognition. *This researcher will
work with Scott Makeig and a multi-disciplinary team of Swartz Center, UCSD,
and other university and ARL researchers to develop, implement, and apply
methods for fusing task, sensory, and high-density EEG information with
simultaneously recorded eye tracking and body motion capture data to *better
describe the non-stationary distributed EEG brain dynamics supporting task
performance*, with a goal of developing basic principles for cognitive
monitoring of operators working in information-rich, enclosed control
environments. Requirements include a PhD in cognitive science, engineering,
or related field with a concentration or in depth experience in modeling and
statistical signal processing / machine learning methods.

*Three researchers based primarily at Army Research Laboratories, Aberdeen,
Maryland, *with allowance for up to 3 months per year of collaborative
research at UCSD or other project partners in San Antonio Texas, Ann Arbor
Michigan, Taiwan, Osnabruck Germany and/or elsewhere. Funding for these
positions will be available for 5-10 years, with transitions to civil
service employment possible. Candidates must have a PhD in a relevant
discipline and should be US citizens. These researchers will collaborate in
developing and implementing the advanced methods of data recording and
analysis with project groups at UCSD and other universities, while
collaborating with ARL researchers to design, perform, and analyze parallel
basic and applied research studies. Areas of research may include
programming of numerical, parallel GPU-based analysis and real-time
interactive data collection software, cognitive science or bioengineering
with an emphasis on data modeling and machine learning, or a related area.
Applications from both new PhD and more senior researchers will be
considered.


-- 
Scott Makeig, Research Scientist and Director, Swartz Center for
Computational Neuroscience, Institute for Neural Computation, University of
California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100725/cb162104/attachment.html>

From smakeig at ucsd.edu  Sun Jul 25 18:01:48 2010
From: smakeig at ucsd.edu (Scott Makeig)
Date: Sun, 25 Jul 2010 18:01:48 -0700
Subject: [Eeglablist] Postdoctoral research and lab manager positions at
	SCCN, La Jolla CA 	and Aberdeen, MD
Message-ID: <AANLkTimiJLF8TozhHQfWVwxkyPACCSM4vcN7saJ9JfJi@mail.gmail.com>

*

Several positions are now available at the Swartz Center, UCSD and at ARL
Aberdeen, Maryland in a large  collaborative project on cognitive monitoring
using concurrent EEG, behavioral, and environmental data. Please forward
this message to qualified potential applicants.


Contact Scott Makeig    (smakeig at ucsd.edu)
***

*
*

*Mobile Brain/body Imaging laboratory manager*: A senior scientific research
associate will work with faculty researchers to develop and supervise two
innovative, spatially adjacent laboratories for conducting new experiments
in Mobile Brain/body Imaging (MoBI) combining concurrent high-density EEG,
body motion capture, eye tracking, scene recording, and psychophysiology
measurements. The lab manager will develop hardware and facilities plans
with PIs and experimenters, will coordinate purchase and integration of
equipment and software, manage lab scheduling, archiving and preprocessing
of the data, and will supervise laboratory technicians who schedule and
conduct experiments and maintain laboratory supplies.

*Postdoctoral researchers *: Four postdoctoral positions are now available
at the Swartz Center for Computational Neuroscience under a basic research
project, "Cognition and Neuroergonomics," to develop neurocognitive
monitoring based on simultaneous high-density EEG, body motion capture, eye
tracking, and psychophysiological data recording and analysis for and with
the Army Research Laboratory, Aberdeen, Maryland.

*1. Statistical analysis of multimodal brain/body imaging data. *This
researcher will work with Scott Makeig and a multi-disciplinary team of
Swartz Center, UCSD, and other university and ARL researchers to develop,
implement, and apply *methods to fuse and mine multimodal mobile brain/body
imaging (MoBI) data for information about arousal, distribution of
attention, evaluation of events, motivation, and behavioral intent* of
operators working in information-rich, enclosed control environments.
Methods to be employed include sparse and complete independent component
analysis (ICA), related non-stationary factor analysis, and distributional
data modeling. Requirements include a PhD in mathematics, statistics, signal
processing, or related field with a concentration in modeling and
statistical signal processing and machine learning / data mining methods.

*2. Brain dynamics supporting active vision. *This researcher will work with
Scott Makeig and a multi-disciplinary team of Swartz Center, UCSD, other
university and ARL researchers to develop, implement, and apply methods for
fusing task, sensory, and eye tracking information with simultaneously
recorded EEG and body motion capture data to *better describe the
macroscopic EEG brain dynamics supporting active vision*, with a goal of
developing basic principles underlying use of concurrent EEG and eye
tracking data for monitoring operators working in information-rich, enclosed
control environments. Requirements include a PhD in cognitive science,
biomedical engineering, or related field with a concentration or significant
experience in modeling, signal processing, and statistical data mining
methods.

*3. Brain dynamics supporting motivated behavior. *This researcher will work
with Scott Makeig and a multi-disciplinary team of Swartz Center, UCSD, and
other university and ARL researchers to develop, implement, and apply
methods for fusing task, sensory, body motion capture, and electromyographic
(EMG) data with concurrent eye tracking, psychophysiological and
high-density EEG to *better model the task-motivated motor behavior of
operators and the macroscopic EEG brain dynamics supporting that behavior*,
with a goal of developing basic principles underlying use of concurrent EEG
and body motion data for monitoring the cognition of operators working in
information-rich, enclosed control environments. Requirements include a PhD
in biomechanics, biomedical engineering, neuroscience, or related field with
a concentration or in depth experience in modeling and statistical signal
processing and machine learning methods.

*4. EEG brain dynamics supporting active cognition. *This researcher will
work with Scott Makeig and a multi-disciplinary team of Swartz Center, UCSD,
and other university and ARL researchers to develop, implement, and apply
methods for fusing task, sensory, and high-density EEG information with
simultaneously recorded eye tracking and body motion capture data to *better
describe the non-stationary distributed EEG brain dynamics supporting task
performance*, with a goal of developing basic principles for cognitive
monitoring of operators working in information-rich, enclosed control
environments. Requirements include a PhD in cognitive science, engineering,
or related field with a concentration or in depth experience in modeling and
statistical signal processing / machine learning methods.

*Three researchers based primarily at Army Research Laboratories, Aberdeen,
Maryland, *with allowance for up to 3 months per year of collaborative
research at UCSD or other project partners in San Antonio Texas, Ann Arbor
Michigan, Taiwan, Osnabruck Germany and/or elsewhere. Funding for these
positions will be available for 5-10 years, with transitions to civil
service employment possible. Candidates must have a PhD in a relevant
discipline and should be US citizens. These researchers will collaborate in
developing and implementing the advanced methods of data recording and
analysis with project groups at UCSD and other universities, while
collaborating with ARL researchers to design, perform, and analyze parallel
basic and applied research studies. Areas of research may include
programming of numerical, parallel GPU-based analysis and real-time
interactive data collection software, cognitive science or bioengineering
with an emphasis on data modeling and machine learning, or a related area.
Applications from both new PhD and more senior researchers will be
considered.

-- 
Scott Makeig, Research Scientist and Director, Swartz Center for
Computational Neuroscience, Institute for Neural Computation, University of
California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100725/3c73d1d6/attachment.html>

From sheraz.khan at polytechnique.edu  Mon Jul 26 11:08:14 2010
From: sheraz.khan at polytechnique.edu (Sheraz Khan)
Date: Mon, 26 Jul 2010 11:08:14 -0700 (PDT)
Subject: [Eeglablist] MEG POST DOCTORAL Position available at the Martinos
	Center for Biomedical Imaging/Harvard Medical School
Message-ID: <860147.58097.qm@web45306.mail.sp1.yahoo.com>

Post doctoral position available at the Martinos Center for ?Biomedical
Imaging, Massachusetts General Hospital/ Harvard Medical School, for

a person with a PhD in MEG/EEG data analysis/methodology/signal
processing. Background in specific aspects of neuroscience or cognitive
science is

not required. The position will involve investigating sensory
processing, cortical connectivity, and other processing ?abnormalities
in autism,

using primarily MEG data with potential ?cross-linking with MRI. There
is a range of possible projects to choose ?from within these categories.





The position requires working closely with other members of the
Martinos ?center (in particular Dr. Matti H?m?l?inen), as well as with
other lab members, to develop, explore, improve and apply ?various data
analysis methodologies. Multiple paradigms are routinely used in the
lab, each offering opportunities for

different approaches and research directions.



The position will involve running the MEG and MRI sessions with
children ?and adolescents, both healthy and with autism spectrum
disorders. ?Recruitment of subjects is carried out by other members
(research ?coordinators) in the lab.



A minimum commitment of two years is required. Salary will be competitive and commensurate with experience.



For more information or to apply, please email CV and cover letter to ?Tal Kenet, at tal at nmr.mgh.harvard.edu


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100726/5ee9122c/attachment.html>

From jdien07 at mac.com  Mon Jul 26 17:24:07 2010
From: jdien07 at mac.com (Joseph Dien)
Date: Mon, 26 Jul 2010 20:24:07 -0400
Subject: [Eeglablist] Re-referencing -- do I need to?
In-Reply-To: <573320.97298.qm@web62001.mail.re1.yahoo.com>
References: <573320.97298.qm@web62001.mail.re1.yahoo.com>
Message-ID: <41A6EA56-5DD4-4AFE-9BE8-FAF3FBB96214@mac.com>

You might find my paper on the average reference to be of interest too.

Dien, J. (1998). Issues in the application of the average reference: Review, critiques, and recommendations. Behavior Research Methods, Instruments, and Computers, 30(1), 34-43.

Short answer is that not having the channel with the mastoid voltages (really, the channel with their differences) just means that you lost the information in those two channels.  It doesn't really affect the decision on whether to shift to a different reference scheme.

Average reference will definitely give you a different result.  Whether it is warping the waveforms compared to the mastoids reference or whether the mastoids reference is warping the waveforms compared to the average reference depends a lot on your montage is relative.

The big questions are how many electrodes did you use and how well did they sample the scalp surface?
Also, how do you plan to analyze and interpret the resulting data?

Cheers!

Joe



On Jul 15, 2010, at 2:24 PM, Nick Bedo wrote:

> Hi everyone,
> 
> My recordings were referenced to both mastoids, but I don't have a channel that contains those data.  Is re-referencing advised/necessary in my case?  I was playing around with re-referencing to the average, and it really warps the waveform amplitudes compared to the ERPs calculated with our original methods.  Any input would be helpful.
> 
> Thanks in advance,
> Nick
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu


--------------------------------------------------------------------------------

Joseph Dien,
Senior Research Scientist
Center for Advanced Study of Language
University of Maryland 
7005 52nd Avenue
College Park, MD 20742-0025

E-mail: jdien07 at mac.com
Phone: 301-226-8848
Fax: 301-226-8811
http://homepage.mac.com/jdien07/










From leiding at ou.edu  Tue Jul 27 15:15:39 2010
From: leiding at ou.edu (Ding, Lei)
Date: Tue, 27 Jul 2010 22:15:39 +0000
Subject: [Eeglablist] Job Opening
Message-ID: <10185504C20C3D4CBDE71837AB615D361769BAA0@it-lightning.sooner.net.ou.edu>

Postdoctoral Research Associates

The Computational Imaging Laboratory in the School of Electrical and Computer Engineering at the University of Oklahoma, Norman invites applications for postdoctoral research associate positions. Candidates should hold a PhD or equivalent in biomedical engineering, electrical engineering, physics, or a related field. Experience in EEG/MEG forward/inverse modeling/imaging/mapping, bioelectromagnetic modeling and imaging, image segmentation and reconstruction, or biomedical signal and image processing is desired.

The Laboratory has the following ongoing research projects: functional electrophysiological neuroimaging (MEG/EEG), source localization, magnetic resonance image reconstruction, human-computer interaction, and neuroergonomics. The Laboratory has strong collaborations with investigators in various departments at the University of Oklahoma Health Science Center, and at the United Hospital, Minnesota, Federal Air Administration, and the Laureate Institute for Brain Research, Tulsa, Oklahoma, among others.

To apply, e-mail CV (pdf file) to Prof. Lei Ding at leiding at ou.edu<mailto:leiding at ou.edu> For further information about the Laboratory, visit http://faculty-staff.ou.edu/D/Lei.Ding-1/index.htm.
The University of Oklahoma, Norman is an equal opportunity employer and educator.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100727/b8307257/attachment.html>

From leiding at ou.edu  Tue Jul 27 15:16:48 2010
From: leiding at ou.edu (Ding, Lei)
Date: Tue, 27 Jul 2010 22:16:48 +0000
Subject: [Eeglablist] PhD in a biomedical engineering lab
Message-ID: <10185504C20C3D4CBDE71837AB615D361769BAB3@it-lightning.sooner.net.ou.edu>

Positions Available PhD research assistant position at the Computational Imaging Lab of the School of Electrical and Computer Engineering in the University of Oklahoma in the following areas:

  *   Functional Electrophysiological Neuroimaging
  *   Bioelectromagnetism
  *   Mathematical Modeling and Computational Electrophysiology
  *   Numerical, Theoretical, and Experimental Studies of Human Electrophysiological Phenomena
  *   Human-Computer Interaction
  *   Neuroergonomics
Interested candidates with BS and MS degrees are welcome. Please e-mail pdf version CV and contact Prof. Lei Ding (http://www.ou.edu/coe/ece/Faculty/faculty_directory/dr_ding.html) at leiding at ou.edu<mailto:leiding at ou.edu>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100727/db02f8b5/attachment.html>

From hannu.loimo at helsinki.fi  Wed Jul 28 00:09:03 2010
From: hannu.loimo at helsinki.fi (Hannu Loimo)
Date: Wed, 28 Jul 2010 10:09:03 +0300
Subject: [Eeglablist] How to do data decomposition with ICA?
In-Reply-To: <AANLkTinNKY0vZDhLZ2=gBE7uKk-70HLGV4tta+OSWveR@mail.gmail.com>
References: <AANLkTinNKY0vZDhLZ2=gBE7uKk-70HLGV4tta+OSWveR@mail.gmail.com>
Message-ID: <AANLkTi=DfSc0t7KNga3-bSsPR+aN4Xa5-CRbXsvwLi=f@mail.gmail.com>

Hi,

We are using ICA to remove CI-artefact from our data. ICA seems to find
pretty easily at least some of the artefact induced by CI-devise. There are
8 events in our experiment and two of those events (duration and gap
deviants) produce different form of CI-artefact in time domain. All events
are made out of syllable sequence ta-ta-ta. Gap deviant means that second
syllable starts 100ms later than in other events. Duration deviant means
that the whole ta-ta-ta sequence lasts 50ms longer. Presumably CI-artefacts
rising from different events have same kind of spatial distribution and
presumably most significant difference is how these artefact peaks are
located in time. So basically there are three differend shapes of
CI-artefact present in time domain: one for gap deviant, one for duration
deviant and one for other six stimuli. After doing ICA for whole dataset
(all events) CI-artefacts emerging from different events seem to be
contained in one component and not many. My question is:

Is it a problem to do ICA for whole dataset or shoud ICA be done for
individual events separatelly? Size of dataset is 2000 epochs and separating
events into different dataset would end up datasets as small as 150-200
epochs (too few for ICA?). So far we have been doing ICA for whole datasets,
then removed CI-artefact components and after that separated data into
different events. What I'm asking is does it affect the final outcome of our
separate event ERP composition if you remove CI-artefact components that are
calculated according to all events or is it same if you do ICA for single
events and then remove components?

thanks,

Hannu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100728/8ccc667b/attachment.html>

From smakeig at gmail.com  Wed Jul 28 10:02:07 2010
From: smakeig at gmail.com (Scott Makeig)
Date: Wed, 28 Jul 2010 10:02:07 -0700
Subject: [Eeglablist] How to do data decomposition with ICA?
In-Reply-To: <AANLkTi=DfSc0t7KNga3-bSsPR+aN4Xa5-CRbXsvwLi=f@mail.gmail.com>
References: <AANLkTinNKY0vZDhLZ2=gBE7uKk-70HLGV4tta+OSWveR@mail.gmail.com>
	<AANLkTi=DfSc0t7KNga3-bSsPR+aN4Xa5-CRbXsvwLi=f@mail.gmail.com>
Message-ID: <AANLkTimAQ2Xen2emEDHoSLR2Y2p8EfgpfZ5QzH3TZtwY@mail.gmail.com>

Hannu  -  The  'CI' ( = ??) artifact has the same scalp projection in all
conditions, then ICA will separate it into a single component (assuming its
activity is not linearly related to other, spatially-varying phenomena).
To most cleanly separate the artifact from the rest of the data, it is best
to decompose all the data at once. Of course, if the other sources in the
different experimental blocks are quite different (say, as an extreme
example, if the subject in the three blocks was respectively awake, asleep,
and having an epileptic seizure), then separate decompositions followed by
artifact component comparison and matching might be more effective.

Note that ICA (instantaneous ICA, including infomax, runica, binica, Amica)
does not pay attention to the time waveforms at all, just the collection of
scalp maps for all the time points (in no particular order).

Scott Makeig

On Wed, Jul 28, 2010 at 12:09 AM, Hannu Loimo <hannu.loimo at helsinki.fi>wrote:

>
> Hi,
>
> We are using ICA to remove CI-artefact from our data. ICA seems to find
> pretty easily at least some of the artefact induced by CI-devise. There are
> 8 events in our experiment and two of those events (duration and gap
> deviants) produce different form of CI-artefact in time domain. All events
> are made out of syllable sequence ta-ta-ta. Gap deviant means that second
> syllable starts 100ms later than in other events. Duration deviant means
> that the whole ta-ta-ta sequence lasts 50ms longer. Presumably CI-artefacts
> rising from different events have same kind of spatial distribution and
> presumably most significant difference is how these artefact peaks are
> located in time. So basically there are three differend shapes of
> CI-artefact present in time domain: one for gap deviant, one for duration
> deviant and one for other six stimuli. After doing ICA for whole dataset
> (all events) CI-artefacts emerging from different events seem to be
> contained in one component and not many. My question is:
>
> Is it a problem to do ICA for whole dataset or shoud ICA be done for
> individual events separatelly? Size of dataset is 2000 epochs and separating
> events into different dataset would end up datasets as small as 150-200
> epochs (too few for ICA?). So far we have been doing ICA for whole datasets,
> then removed CI-artefact components and after that separated data into
> different events. What I'm asking is does it affect the final outcome of our
> separate event ERP composition if you remove CI-artefact components that are
> calculated according to all events or is it same if you do ICA for single
> events and then remove components?
>
> thanks,
>
> Hannu
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Scott Makeig, Research Scientist and Director, Swartz Center for
Computational Neuroscience, Institute for Neural Computation, University of
California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100728/678fbe26/attachment.html>

From shackman at wisc.edu  Wed Jul 28 09:33:55 2010
From: shackman at wisc.edu (Alexander J. Shackman)
Date: Wed, 28 Jul 2010 11:33:55 -0500
Subject: [Eeglablist] searching the listserv archives
Message-ID: <AANLkTi=+-DvLB8pATx2eOOY+Y+m01x+xuhtOfUUL1q2E@mail.gmail.com>

Is there any way to run key word searches on the EEGLAB listserv archives?

Thanks much,
Alex

-- 
Alexander J. Shackman, Ph.D.
Wisconsin Psychiatric Institute & Clinics and
Department of Psychology
University of Wisconsin-Madison
1202 West Johnson Street
Madison, Wisconsin 53706

Telephone: +1 (608) 358-5025
Fax: +1 (608) 265-2875
Email: shackman at wisc.edu
http://psyphz.psych.wisc.edu/~shackman
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100728/6491e587/attachment.html>

From shackman at wisc.edu  Wed Jul 28 17:31:12 2010
From: shackman at wisc.edu (Alexander J. Shackman)
Date: Wed, 28 Jul 2010 19:31:12 -0500
Subject: [Eeglablist] searching the listserv archives
In-Reply-To: <347F33792DC2447F856C921A22F0AB1B@mine>
References: <AANLkTi=+-DvLB8pATx2eOOY+Y+m01x+xuhtOfUUL1q2E@mail.gmail.com>
	<347F33792DC2447F856C921A22F0AB1B@mine>
Message-ID: <AANLkTikqi9nGo1zyB7M7Xyhf4V=Tcf7gpZiwYDK1qKYo@mail.gmail.com>

excellent suggestion - thanks!

alex

On Wed, Jul 28, 2010 at 5:39 PM, Philip Michael Zeman <pzeman at alumni.uvic.ca
> wrote:

>  use google.  Type "[Eeglablist]" and your search term
>
>
> the square brackets should help narrow the search to the EEGlab list
>
> Good luck
>
> Phil
>
> ----- Original Message -----
> *From:* Alexander J. Shackman <shackman at wisc.edu>
> *To:* eeglablist at sccn.ucsd.edu
> *Sent:* Wednesday, July 28, 2010 9:33 AM
> *Subject:* [Eeglablist] searching the listserv archives
>
> Is there any way to run key word searches on the EEGLAB listserv archives?
>
> Thanks much,
> Alex
>
> --
> Alexander J. Shackman, Ph.D.
> Wisconsin Psychiatric Institute & Clinics and
> Department of Psychology
> University of Wisconsin-Madison
> 1202 West Johnson Street
> Madison, Wisconsin 53706
>
> Telephone: +1 (608) 358-5025
> Fax: +1 (608) 265-2875
> Email: shackman at wisc.edu
> http://psyphz.psych.wisc.edu/~shackman<http://psyphz.psych.wisc.edu/%7Eshackman>
>
> ------------------------------
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
>


-- 
Alexander J. Shackman, Ph.D.
Wisconsin Psychiatric Institute & Clinics and
Department of Psychology
University of Wisconsin-Madison
1202 West Johnson Street
Madison, Wisconsin 53706

Telephone: +1 (608) 358-5025
Fax: +1 (608) 265-2875
Email: shackman at wisc.edu
http://psyphz.psych.wisc.edu/~shackman
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100728/e6b6cc29/attachment.html>

From bradley.voytek at gmail.com  Wed Jul 28 18:31:18 2010
From: bradley.voytek at gmail.com (Bradley Voytek)
Date: Wed, 28 Jul 2010 18:31:18 -0700
Subject: [Eeglablist] searching the listserv archives
In-Reply-To: <AANLkTikqi9nGo1zyB7M7Xyhf4V=Tcf7gpZiwYDK1qKYo@mail.gmail.com>
References: <AANLkTi=+-DvLB8pATx2eOOY+Y+m01x+xuhtOfUUL1q2E@mail.gmail.com>
	<347F33792DC2447F856C921A22F0AB1B@mine>
	<AANLkTikqi9nGo1zyB7M7Xyhf4V=Tcf7gpZiwYDK1qKYo@mail.gmail.com>
Message-ID: <AANLkTikC1RmR2_bvWJXkJ07cN+xrGRh-ypE+y7cVKQQ0@mail.gmail.com>

Or append:

site:sccn.ucsd.edu/pipermail/eeglablist/

to your Google search. This limits the searches to the eeglab listserv.

e.g.:

biosemi means site:sccn.ucsd.edu/pipermail/eeglablist/

::brad

On Wed, Jul 28, 2010 at 17:31, Alexander J. Shackman <shackman at wisc.edu> wrote:
> excellent suggestion - thanks!
>
> alex
>
> On Wed, Jul 28, 2010 at 5:39 PM, Philip Michael Zeman
> <pzeman at alumni.uvic.ca> wrote:
>>
>> use google.? Type "[Eeglablist]" and your search term
>>
>>
>> the square brackets should help narrow the search to the EEGlab list
>>
>> Good luck
>>
>> Phil
>>
>> ----- Original Message -----
>> From: Alexander J. Shackman
>> To:?eeglablist at sccn.ucsd.edu
>> Sent: Wednesday, July 28, 2010 9:33 AM
>> Subject: [Eeglablist] searching the listserv archives
>> Is there any way to run key word searches on the EEGLAB listserv archives?
>>
>> Thanks much,
>> Alex
>>
>> --
>> Alexander J. Shackman, Ph.D.
>> Wisconsin Psychiatric Institute & Clinics and
>> Department of Psychology
>> University of Wisconsin-Madison
>> 1202 West Johnson Street
>> Madison, Wisconsin 53706
>>
>> Telephone: +1 (608) 358-5025
>> Fax: +1 (608) 265-2875
>> Email: shackman at wisc.edu
>> http://psyphz.psych.wisc.edu/~shackman
>>
>> ________________________________
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>
>
> --
> Alexander J. Shackman, Ph.D.
> Wisconsin Psychiatric Institute & Clinics and
> Department of Psychology
> University of Wisconsin-Madison
> 1202 West Johnson Street
> Madison, Wisconsin 53706
>
> Telephone: +1 (608) 358-5025
> Fax: +1 (608) 265-2875
> Email: shackman at wisc.edu
> http://psyphz.psych.wisc.edu/~shackman
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



From lutobu at gmail.com  Wed Jul 28 14:04:21 2010
From: lutobu at gmail.com (ludwing torres)
Date: Wed, 28 Jul 2010 23:04:21 +0200
Subject: [Eeglablist] View functions code and importing channel locations
	from .mat file
Message-ID: <AANLkTinJy0B_wP5XYbZcCqb3cjSbodmBfK7KgKS8fxTC@mail.gmail.com>

Hi all.
Is there any way to view the code of this functions?
pop_multifit
pop_dipfit_settings
I see that these are the functions that eeglab uses to compute the dipole
fitting of many sources, from de data given by the electrodes in the scalp

Another question:
?Is there any way to  import  channel locations from a .mat file?
?Any way to import head model properties from it too?

Thanks for your help.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100728/7f2a114b/attachment.html>

From sara.graziadio at newcastle.ac.uk  Thu Jul 29 03:27:57 2010
From: sara.graziadio at newcastle.ac.uk (Sara Graziadio)
Date: Thu, 29 Jul 2010 11:27:57 +0100
Subject: [Eeglablist] high density EEG,
	source localization and electrode positions
Message-ID: <327AA03572E1D141AAD8769367EF12DE3B9084EC94@EXSAN01.campus.ncl.ac.uk>

Hello everybody,
We are setting up an ERP study with high density EEG (128 electrodes). We would like to localize sources from the ERP components and possibly to relate these data with fMRI activations. We would have MRI data of the subjects. We are now in the process of considering the possibility to use the Polhemus to digitalize the electrode positions and we found out that there are no papers (at least that I can find) that compare the accuracy of source localization with high density EEG adding the information about the positions of all the 128 electrodes, or of just a part of them (for example 30 evenly spread on the scalp and including some of the more external positions), or just using the standard 10-10 system locations. Does anybody of you have any thoughts about it? Have you had ever tried to compare the 3 options I have described? 
Thank you very much

Best wishes

Sara Graziadio, PhD
Research Associate
Newcastle University






From g.rousselet at psy.gla.ac.uk  Thu Jul 29 13:12:15 2010
From: g.rousselet at psy.gla.ac.uk (Guillaume Rousselet)
Date: Thu, 29 Jul 2010 21:12:15 +0100
Subject: [Eeglablist] high density EEG,
	source localization and electrode positions
In-Reply-To: <327AA03572E1D141AAD8769367EF12DE3B9084EC94@EXSAN01.campus.ncl.ac.uk>
References: <327AA03572E1D141AAD8769367EF12DE3B9084EC94@EXSAN01.campus.ncl.ac.uk>
Message-ID: <30A96331-CC72-4A09-BEDF-6C0E64CDA89D>

Sara, this article and maybe more recent ones from this group might  
answer your question:

http://www.ncbi.nlm.nih.gov/pubmed/15351361

Best

Guillaume


On 29 Jul 2010, at 11:27, Sara Graziadio wrote:

> Hello everybody,
> We are setting up an ERP study with high density EEG (128  
> electrodes). We would like to localize sources from the ERP  
> components and possibly to relate these data with fMRI activations.  
> We would have MRI data of the subjects. We are now in the process of  
> considering the possibility to use the Polhemus to digitalize the  
> electrode positions and we found out that there are no papers (at  
> least that I can find) that compare the accuracy of source  
> localization with high density EEG adding the information about the  
> positions of all the 128 electrodes, or of just a part of them (for  
> example 30 evenly spread on the scalp and including some of the more  
> external positions), or just using the standard 10-10 system  
> locations. Does anybody of you have any thoughts about it? Have you  
> had ever tried to compare the 3 options I have described?
> Thank you very much
>
> Best wishes
>
> Sara Graziadio, PhD
> Research Associate
> Newcastle University
>
>
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu


************************************************************************************
Guillaume A. Rousselet, Ph.D., lecturer

School of Psychology
Institute of Neuroscience & Psychology
Centre for Cognitive Neuroimaging (CCNi)

The University of Glasgow, charity number SC004401

http://www.psy.gla.ac.uk/staff/index.php?id=GAR01

Email: g.rousselet at psy.gla.ac.uk
Fax. +44 (0)141 330 4606
Tel. +44 (0)141 330 6652
Cell +44 (0)791 779 7833


"For reasons I wish I understood, the spectacle of sync strikes a  
chord in us, somewhere deep in our souls. It's a wonderful and  
terrifying thing. Unlike many other phenomena, the witnessing of it  
touches people at a primal level. Maybe we instinctively realize that  
if we ever find the source of spontaneous order, we will have  
discovered the secret of the universe."

	Steven Strogatz - Sync - 2003
************************************************************************************




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100729/db5add94/attachment.html>

From smakeig at gmail.com  Thu Jul 29 11:03:43 2010
From: smakeig at gmail.com (Scott Makeig)
Date: Thu, 29 Jul 2010 11:03:43 -0700
Subject: [Eeglablist] high density EEG,
	source localization and electrode 	positions
In-Reply-To: <327AA03572E1D141AAD8769367EF12DE3B9084EC94@EXSAN01.campus.ncl.ac.uk>
References: <327AA03572E1D141AAD8769367EF12DE3B9084EC94@EXSAN01.campus.ncl.ac.uk>
Message-ID: <AANLkTi=4j8VFV4bW=_mQ_xWN4pqUDYOHfuZ5hxxesoCB@mail.gmail.com>

Sara -

Measuring the positions of a subset of the electrodes is an interesting idea
(for speed). We are exploring electrode position measurement now in the lab,
and will consider trying this ourselves. You should be aware that Polhemus'
magnetic systems are inaccurate if any magnetic metal is nearby. There are
also ultrasound and camera-based systems available (at various prices). I
believe it should be possible to develop low-cost simple camera-based
software, but the development process would in itself cost time and money.
I'd be interested in hearing others' experience with electrode position
measurement.

Scott Makeig

PS Zeynep Akalin Acar's NFT can build an individualized BEM head model from
electrode positions, with or without MR (though with MR is better, of
course). She is currently preparing a simulation paper on the consequences
of not using MR for source localization -- or using inaccurate electrode
positions on a standard head model, as is most common in EEG labs today.

PPS Nima Bigdely-Shamlo has contributed an EEGLAB function that can find
maximally uniformly-spaced subsets of an electrode montage.


On Thu, Jul 29, 2010 at 3:27 AM, Sara Graziadio <
sara.graziadio at newcastle.ac.uk> wrote:

> Hello everybody,
> We are setting up an ERP study with high density EEG (128 electrodes). We
> would like to localize sources from the ERP components and possibly to
> relate these data with fMRI activations. We would have MRI data of the
> subjects. We are now in the process of considering the possibility to use
> the Polhemus to digitalize the electrode positions and we found out that
> there are no papers (at least that I can find) that compare the accuracy of
> source localization with high density EEG adding the information about the
> positions of all the 128 electrodes, or of just a part of them (for example
> 30 evenly spread on the scalp and including some of the more external
> positions), or just using the standard 10-10 system locations. Does anybody
> of you have any thoughts about it? Have you had ever tried to compare the 3
> options I have described?
> Thank you very much
>
> Best wishes
>
> Sara Graziadio, PhD
> Research Associate
> Newcastle University
>
>
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Scott Makeig, Research Scientist and Director, Swartz Center for
Computational Neuroscience, Institute for Neural Computation, University of
California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100729/3074eb2e/attachment.html>

From hannu.loimo at helsinki.fi  Fri Jul 30 05:06:37 2010
From: hannu.loimo at helsinki.fi (Hannu Loimo)
Date: Fri, 30 Jul 2010 15:06:37 +0300
Subject: [Eeglablist] How to do data decomposition with ICA?
In-Reply-To: <AANLkTimAQ2Xen2emEDHoSLR2Y2p8EfgpfZ5QzH3TZtwY@mail.gmail.com>
References: <AANLkTinNKY0vZDhLZ2=gBE7uKk-70HLGV4tta+OSWveR@mail.gmail.com>
	<AANLkTi=DfSc0t7KNga3-bSsPR+aN4Xa5-CRbXsvwLi=f@mail.gmail.com>
	<AANLkTimAQ2Xen2emEDHoSLR2Y2p8EfgpfZ5QzH3TZtwY@mail.gmail.com>
Message-ID: <AANLkTimWAUOV3R1hYXm5mNgvTqYD+DyPwJrqDiG3Eop+@mail.gmail.com>

Thanks for your anserw Scott

By CI I ment cochlear implant.

Hannu

On Wed, Jul 28, 2010 at 8:02 PM, Scott Makeig <smakeig at gmail.com> wrote:

> Hannu  -  The  'CI' ( = ??) artifact has the same scalp projection in all
> conditions, then ICA will separate it into a single component (assuming its
> activity is not linearly related to other, spatially-varying phenomena).
> To most cleanly separate the artifact from the rest of the data, it is best
> to decompose all the data at once. Of course, if the other sources in the
> different experimental blocks are quite different (say, as an extreme
> example, if the subject in the three blocks was respectively awake, asleep,
> and having an epileptic seizure), then separate decompositions followed by
> artifact component comparison and matching might be more effective.
>
> Note that ICA (instantaneous ICA, including infomax, runica, binica, Amica)
> does not pay attention to the time waveforms at all, just the collection of
> scalp maps for all the time points (in no particular order).
>
> Scott Makeig
>
> On Wed, Jul 28, 2010 at 12:09 AM, Hannu Loimo <hannu.loimo at helsinki.fi>wrote:
>
>>
>> Hi,
>>
>> We are using ICA to remove CI-artefact from our data. ICA seems to find
>> pretty easily at least some of the artefact induced by CI-devise. There are
>> 8 events in our experiment and two of those events (duration and gap
>> deviants) produce different form of CI-artefact in time domain. All events
>> are made out of syllable sequence ta-ta-ta. Gap deviant means that second
>> syllable starts 100ms later than in other events. Duration deviant means
>> that the whole ta-ta-ta sequence lasts 50ms longer. Presumably CI-artefacts
>> rising from different events have same kind of spatial distribution and
>> presumably most significant difference is how these artefact peaks are
>> located in time. So basically there are three differend shapes of
>> CI-artefact present in time domain: one for gap deviant, one for duration
>> deviant and one for other six stimuli. After doing ICA for whole dataset
>> (all events) CI-artefacts emerging from different events seem to be
>> contained in one component and not many. My question is:
>>
>> Is it a problem to do ICA for whole dataset or shoud ICA be done for
>> individual events separatelly? Size of dataset is 2000 epochs and separating
>> events into different dataset would end up datasets as small as 150-200
>> epochs (too few for ICA?). So far we have been doing ICA for whole datasets,
>> then removed CI-artefact components and after that separated data into
>> different events. What I'm asking is does it affect the final outcome of our
>> separate event ERP composition if you remove CI-artefact components that are
>> calculated according to all events or is it same if you do ICA for single
>> events and then remove components?
>>
>> thanks,
>>
>> Hannu
>>
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>>
>
>
>
> --
> Scott Makeig, Research Scientist and Director, Swartz Center for
> Computational Neuroscience, Institute for Neural Computation, University of
> California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott<http://sccn.ucsd.edu/%7Escott>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100730/d4c6772e/attachment.html>

From lindemann at cg.cs.tu-bs.de  Fri Jul 30 01:15:23 2010
From: lindemann at cg.cs.tu-bs.de (Lea Lindemann)
Date: Fri, 30 Jul 2010 10:15:23 +0200
Subject: [Eeglablist] Odd looking channel spectra
Message-ID: <4C528A1B.5020300@cg.cs.tu-bs.de>

Hi,

I'm new to EEGLAB and EEG and am trying to familiarize myself with the 
data processing tools.

When I visualize the spectra of my EEG channels with 'Plot > Channel 
spectra and maps' (default setting) the spectra look quite odd (see link 
below) although the EEG itself looks normal. It is plotted with 'Plot > 
Channel data (scroll)', 'Display > Remove DC offset' = on. What could be 
the reason for this?

I use EEGLAB v8.0.3.5b with Matlab 7.1.b The data were acquired using 
the BioSemi ActiveTwo system with 32 electrodes on the head + 4 
electrodes EOG and average mastoids as reference.

EEG:
ftp://europa.cg.cs.tu-bs.de/pub/public/people/lindemann/eeg.png

Spectra:
ftp://europa.cg.cs.tu-bs.de/pub/public/people/lindemann/spectrum.png

Best regards,
Lea


From sheraz.khan at polytechnique.edu  Fri Jul 30 10:21:04 2010
From: sheraz.khan at polytechnique.edu (Sheraz Khan)
Date: Fri, 30 Jul 2010 10:21:04 -0700 (PDT)
Subject: [Eeglablist] RESEARCH ASSISTANT / DATA ANALYST Position available
	at the Martinos Center for Biomedical Imaging/ Harvard
	Medical School
Message-ID: <629444.20281.qm@web45304.mail.sp1.yahoo.com>

RESEARCH ASSISTANT / DATA ANALYST Position available at the 
Martinos Center for Biomedical Imaging/ Harvard Medical School.

Research assistant / data analyst Position available at the 
Martinos Center for Biomedical Imaging, Massachusetts General Hospital, 
Charlestown, MA. This is an ideal position for someone with excellent 
programming / signal processing / analytical skills, who is interested 
in neuroscience in general and or in neurodevelopmental disorders in 
particular, and would like exposure to the field before deciding on how 
to proceed in their career.

PRINCIPAL DUTIES AND RESPONSIBILITIES

* Primary responsibilities will include:
-Collection of MEG/MRI data from healthy children and children with 
neurodevelopmental disorders such as autism (i.e.
working directly with subjects, mostly children, but also adults).
-Analysis of MEG data with opportunities also for MRI data analysis
-Programming stimuli for experiments
-working with a team to improve data analysis approaches and developing 
new analysis techniques
-Opportunities will be given to take an active part in influencing the 
directions of the research to those thinking about
graduate school in neuroscience.

* Secondary responsibilities will include:
-Keeping up with relevant literature in the field and occasionally lead 
the weekly journal club
-Contributions to grant and paper writing.
-Some contributions to general lab operations such as IRB and database 
maintenance

QUALIFICATIONS
A B.Sc. is required, M.Sc./M.Eng. is preferred. Relevant signal 
processing experience in an academic setting (e.g. thesis work) or 
non-academic job is a must. A two-year time commitment is preferred. The 
applicant must be facile with computers and programming. Experience with 
Matlab is optimal, but an excellent background in programming in general 
is sufficient. The applicant should be familiar with linux/unix 
operating systems, and a quick learner of complex software packages and 
new concepts. Ideally, the applicant would highly skilled in signal 
processing, be it from engineering, computer science, MEG/EEG, single 
unit work, or other backgrounds.

Contact
For more information or to apply, please email CV and cover letter to 
Tal Kenet, at tal at nmr.mgh.harvard.edu




      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100730/35cce5ca/attachment.html>

From nickbedo at yahoo.com  Fri Jul 30 13:55:43 2010
From: nickbedo at yahoo.com (Nick Bedo)
Date: Fri, 30 Jul 2010 13:55:43 -0700 (PDT)
Subject: [Eeglablist] Manually Rejecting Channels for STUDY
Message-ID: <926161.69597.qm@web62002.mail.re1.yahoo.com>

Hi everyone,

I am trying to manually reject a particular channel in a few datasets that are 
used in my STUDY structure. So far I have used the following:
EEG = pop_select(EEG,'nochannel',{'C3'});  % removes channel C3 from dataset

However, when I try to plot ERPs through the STUDY, EEGlab notices an imbalance 
of channels among the datasets and gives me an error.  Is there a way to bypass 
this or mark certain channels as 'do not include' or something like that?
 
Thanks,
Nick



      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100730/10186188/attachment.html>

From cmkarns at gmail.com  Fri Jul 30 17:00:00 2010
From: cmkarns at gmail.com (Christina Karns)
Date: Fri, 30 Jul 2010 17:00:00 -0700
Subject: [Eeglablist]  Odd looking channel spectra
Message-ID: <AANLkTim4j7+kGPYAncQTorDMvgKZ9nO61vRNdatkgEbg@mail.gmail.com>

Hi,

We posted something about this a while ago. I hope this link is helpful.
http://sccn.ucsd.edu/pipermail/eeglablist/2008/002229.html

--Christina Karns
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100730/d745a2e3/attachment.html>

From pascale.sandmann at uni-oldenburg.de  Sat Jul 31 04:06:06 2010
From: pascale.sandmann at uni-oldenburg.de (Pascale Sandmann)
Date: Sat, 31 Jul 2010 13:06:06 +0200 (CEST)
Subject: [Eeglablist]  How to do data decomposition with ICA?
In-Reply-To: <AANLkTimWAUOV3R1hYXm5mNgvTqYD+DyPwJrqDiG3Eop+@mail.gmail.com>
References: <AANLkTinNKY0vZDhLZ2=gBE7uKk-70HLGV4tta+OSWveR@mail.gmail.com>
	<AANLkTi=DfSc0t7KNga3-bSsPR+aN4Xa5-CRbXsvwLi=f@mail.gmail.com>
	<AANLkTimAQ2Xen2emEDHoSLR2Y2p8EfgpfZ5QzH3TZtwY@mail.gmail.com>
	<AANLkTimWAUOV3R1hYXm5mNgvTqYD+DyPwJrqDiG3Eop+@mail.gmail.com>
Message-ID: <49489.93.223.104.133.1280574366.squirrel@webmail.uni-oldenburg.de>

Dear Hannu,
Based on our experience with cochlear-implant (CI) artefacts, I can also
recommend you to do the ICA-based artefact reduction for all events at
once (i.e., to remove the CI-artefact components which are calculated for
all events of one dataset). In our last experiment (Sandmann et al., Clin
Neurophysiol 2010) we used 13 different events and we computed ICA for all
the events at once. Using this procedure, we could identify several
components representing the CI artefacts (typically around 8 components
out of 60 components) and CI artefacts could be successfully reduced.
Pascale


> Thanks for your anserw Scott
>
> By CI I ment cochlear implant.
>
> Hannu
>
> On Wed, Jul 28, 2010 at 8:02 PM, Scott Makeig <smakeig at gmail.com> wrote:
>
>> Hannu  -  The  'CI' ( = ??) artifact has the same scalp projection in
>> all
>> conditions, then ICA will separate it into a single component (assuming
>> its
>> activity is not linearly related to other, spatially-varying phenomena).
>> To most cleanly separate the artifact from the rest of the data, it is
>> best
>> to decompose all the data at once. Of course, if the other sources in
>> the
>> different experimental blocks are quite different (say, as an extreme
>> example, if the subject in the three blocks was respectively awake,
>> asleep,
>> and having an epileptic seizure), then separate decompositions followed
>> by
>> artifact component comparison and matching might be more effective.
>>
>> Note that ICA (instantaneous ICA, including infomax, runica, binica,
>> Amica)
>> does not pay attention to the time waveforms at all, just the collection
>> of
>> scalp maps for all the time points (in no particular order).
>>
>> Scott Makeig
>>
>> On Wed, Jul 28, 2010 at 12:09 AM, Hannu Loimo
>> <hannu.loimo at helsinki.fi>wrote:
>>
>>>
>>> Hi,
>>>
>>> We are using ICA to remove CI-artefact from our data. ICA seems to find
>>> pretty easily at least some of the artefact induced by CI-devise. There
>>> are
>>> 8 events in our experiment and two of those events (duration and gap
>>> deviants) produce different form of CI-artefact in time domain. All
>>> events
>>> are made out of syllable sequence ta-ta-ta. Gap deviant means that
>>> second
>>> syllable starts 100ms later than in other events. Duration deviant
>>> means
>>> that the whole ta-ta-ta sequence lasts 50ms longer. Presumably
>>> CI-artefacts
>>> rising from different events have same kind of spatial distribution and
>>> presumably most significant difference is how these artefact peaks are
>>> located in time. So basically there are three differend shapes of
>>> CI-artefact present in time domain: one for gap deviant, one for
>>> duration
>>> deviant and one for other six stimuli. After doing ICA for whole
>>> dataset
>>> (all events) CI-artefacts emerging from different events seem to be
>>> contained in one component and not many. My question is:
>>>
>>> Is it a problem to do ICA for whole dataset or shoud ICA be done for
>>> individual events separatelly? Size of dataset is 2000 epochs and
>>> separating
>>> events into different dataset would end up datasets as small as 150-200
>>> epochs (too few for ICA?). So far we have been doing ICA for whole
>>> datasets,
>>> then removed CI-artefact components and after that separated data into
>>> different events. What I'm asking is does it affect the final outcome
>>> of our
>>> separate event ERP composition if you remove CI-artefact components
>>> that are
>>> calculated according to all events or is it same if you do ICA for
>>> single
>>> events and then remove components?
>>>
>>> thanks,
>>>
>>> Hannu
>>>
>>>
>>> _______________________________________________
>>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>>> To unsubscribe, send an empty email to
>>> eeglablist-unsubscribe at sccn.ucsd.edu
>>> For digest mode, send an email with the subject "set digest mime" to
>>> eeglablist-request at sccn.ucsd.edu
>>>
>>
>>
>>
>> --
>> Scott Makeig, Research Scientist and Director, Swartz Center for
>> Computational Neuroscience, Institute for Neural Computation, University
>> of
>> California San Diego, La Jolla CA 92093-0961,
>> http://sccn.ucsd.edu/~scott<http://sccn.ucsd.edu/%7Escott>
>>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu


-- 
Dr. Pascale Sandmann
Department of Psychology
Neuropsychology Lab
Carl von Ossietzky University of Oldenburg
D-26111 Oldenburg
Germany

Office: A7 0-047a
Phone: +49-441-798-4945
Fax:   +49-441-798-5522
Email: pascale.sandmann at uni-oldenburg.de
http://www.psychologie.uni-oldenburg.de/46667.html




From sjluck at ucdavis.edu  Mon Aug  2 14:41:45 2010
From: sjluck at ucdavis.edu (Steve Luck)
Date: Mon, 2 Aug 2010 14:41:45 -0700
Subject: [Eeglablist] Research Assistant Positions in clinical cognitive
	neuroscience at the Maryland Psychiatric Research Center
Message-ID: <A85A5D0D-79C5-4AE3-ABFE-39A0CFECAA30@ucdavis.edu>

Research Assistant Positions in clinical cognitive neuroscience at the Maryland Psychiatric Research Center.

Up to three research assistant positions are available in the laboratory of Dr. James Gold to work on studies of the nature of cognitive and affective disturbances in schizophrenia.  We use behavioral, ERP, and fMRI methods.  Previous experience with experimental task programming, image analysis, or working with people with schizophrenia experience is desirable.   Ongoing NIMH supported projects include work on attention and working memory (in collaboration with Steve Luck),  cognitive measure development and validation with fMRI (in collaboration with Cam Carter, Deanna Barch, Angus MacDonald, Dan Ragland, Charan Ranganath), reward processing ( in collaboration with Michael Frank, James Waltz, Elliot Stein).   Positions offer opportunities to be involved with both basic cognitive neuroscience method development and applications with clinical populations.

Please send CV and letter of support to jgold at mprc.umaryland.edu.  Formal applications should also be submitted through U. Maryland Baltimore  HR website : http://www.hr.umaryland.edu/careers/.... for  Research Assistant, Clinical requisition #s 5447, 5445

--------------------------------------------------------------------
Steven J. Luck, Ph.D.
Director, Center for Mind & Brain
Professor, Department of Psychology
University of California, Davis
Room 127
267 Cousteau Place
Davis, CA 95618
(530) 297-4424
sjluck at ucdavis.edu
Web: http://mindbrain.ucdavis.edu/people/sjluck
Calendar: http://www.google.com/calendar/embed?src=stevenjluck%40gmail.com&ctz=America/Los_Angeles
--------------------------------------------------------------------







-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100802/a884afbc/attachment.html>

From mail at nikobusch.net  Mon Aug  2 10:38:00 2010
From: mail at nikobusch.net (Niko Busch)
Date: Mon, 2 Aug 2010 19:38:00 +0200 (MEST)
Subject: [Eeglablist] Removal of distortion due to overlapping ERP responses
	- ADJAR filter
Message-ID: <201008021738.o72Hc0kb022339@post.webmailer.de>

An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100802/6d3b8753/attachment.html>
-------------- next part --------------
Hi everyone,

I am looking for a way to remove the distortion from an ERP waveformthat is due to overlapping ERP responses to different events. I willemploy an experimental paradigm in which an experimental event A isfollowed by an event B after a variable SOA on each trial. My interestis in the ERP evoked by the Bs. However, the resulting ERP waveform -time-locked to event B - will be "contaminated" by the responses to theAs.

It seems that people use two different approaches to remove the contribution of the As: 
1. Include trials with A events only and then subtract the average of A from the average of A+B.
2. A convolution approach - the ADJAR method specified in this paper:

Woldorff MG. Distortion of ERP averages due to overlap from temporallyadjacent ERPs: analysis and correction. Psychophysiology. 1993Jan;30(1):98-119. 

Has anyone implemented ADJAR or something similar in Eeglab/Matlab andis willing to share the code? Any help would be very much appreciated!

Best,
Niko

From maarcc at gmail.com  Tue Aug  3 05:10:00 2010
From: maarcc at gmail.com (Marc)
Date: Tue, 3 Aug 2010 22:10:00 +1000
Subject: [Eeglablist] EEG amplifier and alternative input
Message-ID: <AANLkTimf5jPy9ov-gm3nci9VzQ9uZfnPfaCgx2mjQYGt@mail.gmail.com>

Hi.

This is not directly related to EEGLab. But I thought someone may be
able to give some pointers.

We're doing some experiments measuring scalp EEG of participants. One
of the variables we intend to record is the participant's voice reply.
We're wondering if it is possible to use the same EEG hardware
amplifier to record the voice reply? Have anyone tried that before?
That is, instead of connecting the electrodes to the one of the 64
channels of the amplifier, we connect a microphone to the input
channel. I assume we will be able to record as good a signal as any
audio amplifier? Is there any thing else to watch out for?

Thanks for any advise.

Marc.


From Ross.Fulham at newcastle.edu.au  Tue Aug  3 17:47:04 2010
From: Ross.Fulham at newcastle.edu.au (Ross Fulham)
Date: Wed, 04 Aug 2010 10:47:04 +1000
Subject: [Eeglablist] EEG amplifier and alternative input
Message-ID: <4C59452A0200001700121663@WINDOMPRD00.newcastle.edu.au>

Marc,

The first question is what do you intend doing with the voice response? Are you hoping to measure reaction time; Do you want to be able to play it back and understand what was said?

(a) EEG is typically recorded at 500-1000 Hz. This is way below the sampling rate used to record hi-fidelity voice/music (44000Hz). With 1000Hz sampling the recorded voice will be uninterpretable, though you might be able to rectify the signal; obtain the envelope of the voice waveform; and extract a reaction time measure. The most accurate way to do this is to use an electronic circuit to rectify the signal coming from the microphone, then low-pass filter it before feeding it into the EEG headbox.

(b) Connecting any inputs into the EEG headbox other than the electrode wires (1) causes safety issues if the device is not properly electrically isolated; (2) may introduce electrical artifacts into the EEG recordings.  If your microphone setup requires any type of electrical power, make sure it is battery powered and not connected to the mains supply.

(c) having subjects make a vocal response is going to introduce all manner of movement related artifacts into the EEG recordings, some of which will be time-locked to your stimuli.

If all you want is to measure voice related reaction times, there are devices available that are designed for this purpose and are connected to the stimulus generation system in the same way a push-button response is recorded.

Ross
 

>>> Marc <maarcc at gmail.com> 04/08/10 4:12 AM >>>
Hi.

This is not directly related to EEGLab. But I thought someone may be
able to give some pointers.

We're doing some experiments measuring scalp EEG of participants. One
of the variables we intend to record is the participant's voice reply.
We're wondering if it is possible to use the same EEG hardware
amplifier to record the voice reply? Have anyone tried that before?
That is, instead of connecting the electrodes to the one of the 64
channels of the amplifier, we connect a microphone to the input
channel. I assume we will be able to record as good a signal as any
audio amplifier? Is there any thing else to watch out for?

Thanks for any advise.

Marc.
_______________________________________________
Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu



From jefferiksen at comcast.net  Tue Aug  3 21:51:35 2010
From: jefferiksen at comcast.net (Jeff Eriksen)
Date: Tue, 3 Aug 2010 21:51:35 -0700
Subject: [Eeglablist] EEG amplifier and alternative input
In-Reply-To: <AANLkTimf5jPy9ov-gm3nci9VzQ9uZfnPfaCgx2mjQYGt@mail.gmail.com>
References: <AANLkTimf5jPy9ov-gm3nci9VzQ9uZfnPfaCgx2mjQYGt@mail.gmail.com>
Message-ID: <00ac01cb3390$b8145ac0$283d1040$@net>

Marc,

EEG recorded for ERPs is generally 0.1 - 100 Hz.
Human hearing is roughly 20-20,000 Hz.
Voice recording requires a substantial part of that, but I do not know
off-hand how high.
I would say you would probably get a poor voice recording, unless you have
higher end amplifiers that can sample at 5-10 KHz.

-Jeff Eriksen

-----Original Message-----
From: eeglablist-bounces at sccn.ucsd.edu
[mailto:eeglablist-bounces at sccn.ucsd.edu] On Behalf Of Marc
Sent: Tuesday, August 03, 2010 5:10 AM
To: eeglablist at sccn.ucsd.edu
Subject: [Eeglablist] EEG amplifier and alternative input

Hi.

This is not directly related to EEGLab. But I thought someone may be
able to give some pointers.

We're doing some experiments measuring scalp EEG of participants. One
of the variables we intend to record is the participant's voice reply.
We're wondering if it is possible to use the same EEG hardware
amplifier to record the voice reply? Have anyone tried that before?
That is, instead of connecting the electrodes to the one of the 64
channels of the amplifier, we connect a microphone to the input
channel. I assume we will be able to record as good a signal as any
audio amplifier? Is there any thing else to watch out for?

Thanks for any advise.

Marc.
_______________________________________________
Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
For digest mode, send an email with the subject "set digest mime" to
eeglablist-request at sccn.ucsd.edu



From fhd at sover.net  Wed Aug  4 08:23:57 2010
From: fhd at sover.net (Frank H Duffy)
Date: Wed, 04 Aug 2010 11:23:57 -0400
Subject: [Eeglablist] Removal of distortion due to overlapping ERP
 responses - ADJAR filter
In-Reply-To: <201008021738.o72Hc0kb022339@post.webmailer.de>
References: <201008021738.o72Hc0kb022339@post.webmailer.de>
Message-ID: <4C59860D.7080302@sover.net>

A key thing is to understand whether, on physiological grounds, there 
may be an interaction between the two conditions, i.e., A alters the B 
waveform (or the opposite)  as seen for the long latency AER if you 
closely pair auditory clicks.  Is the problem a simple arithmetic 
(result =A+B)  or is there likely to be a strong interaction effect 
between A and B?  Indeed some people estimate the degree of interaction 
by adding A alone and B alone signals  and compare to the A+B experiment.
Good luck.

Frank  fhd at sover.net


On 8/2/2010 1:38 PM, Niko Busch wrote:
> Hi everyone,
>
> I am looking for a way to remove the distortion from an ERP waveformthat is due to overlapping ERP responses to different events. I willemploy an experimental paradigm in which an experimental event A isfollowed by an event B after a variable SOA on each trial. My interestis in the ERP evoked by the Bs. However, the resulting ERP waveform -time-locked to event B - will be "contaminated" by the responses to theAs.
>
> It seems that people use two different approaches to remove the contribution of the As:
> 1. Include trials with A events only and then subtract the average of A from the average of A+B.
> 2. A convolution approach - the ADJAR method specified in this paper:
>
> Woldorff MG. Distortion of ERP averages due to overlap from temporallyadjacent ERPs: analysis and correction. Psychophysiology. 1993Jan;30(1):98-119.
>
> Has anyone implemented ADJAR or something similar in Eeglab/Matlab andis willing to share the code? Any help would be very much appreciated!
>
> Best,
> Niko
>    
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100804/7ac6b49d/attachment.html>

From bogdanm at allied-bionics.com  Tue Aug  3 10:56:55 2010
From: bogdanm at allied-bionics.com (B. Motoc)
Date: Tue, 3 Aug 2010 11:56:55 -0600
Subject: [Eeglablist] EEG amplifier and alternative input
In-Reply-To: <AANLkTimf5jPy9ov-gm3nci9VzQ9uZfnPfaCgx2mjQYGt@mail.gmail.com>
References: <AANLkTimf5jPy9ov-gm3nci9VzQ9uZfnPfaCgx2mjQYGt@mail.gmail.com>
Message-ID: <6A5C31F2AD6449BA911663D91EB20975@ABIMOBILE01>

Hi Marc,

It would be unusual to be able to successfully record voice over an EEG
channel (hardware and associated software). The main issue is that EEG is a
low frequency event (usually 0+ to 200 Hz) while voice is a high frequency
process (1000 to 10,000 Hz). Reducing a channel's bandwidth to under 1000 Hz
would totally compromise the capacity to transfer voice based information.
If you want to save response values (did the subject say yes or no or maybe
....) then you could find a way to encode these outcomes and shift them into
EEG frequencies and, this way, keep them with the rest.

Let me know more,
Bogdan

-----Original Message-----
From: eeglablist-bounces at sccn.ucsd.edu
[mailto:eeglablist-bounces at sccn.ucsd.edu] On Behalf Of Marc
Sent: Tuesday, August 03, 2010 6:10 AM
To: eeglablist at sccn.ucsd.edu
Subject: [Eeglablist] EEG amplifier and alternative input

Hi.

This is not directly related to EEGLab. But I thought someone may be
able to give some pointers.

We're doing some experiments measuring scalp EEG of participants. One
of the variables we intend to record is the participant's voice reply.
We're wondering if it is possible to use the same EEG hardware
amplifier to record the voice reply? Have anyone tried that before?
That is, instead of connecting the electrodes to the one of the 64
channels of the amplifier, we connect a microphone to the input
channel. I assume we will be able to record as good a signal as any
audio amplifier? Is there any thing else to watch out for?

Thanks for any advise.

Marc.
_______________________________________________
Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
For digest mode, send an email with the subject "set digest mime" to
eeglablist-request at sccn.ucsd.edu



From fhd at sover.net  Wed Aug  4 08:16:31 2010
From: fhd at sover.net (Frank H Duffy)
Date: Wed, 04 Aug 2010 11:16:31 -0400
Subject: [Eeglablist] EEG amplifier and alternative input
In-Reply-To: <AANLkTimf5jPy9ov-gm3nci9VzQ9uZfnPfaCgx2mjQYGt@mail.gmail.com>
References: <AANLkTimf5jPy9ov-gm3nci9VzQ9uZfnPfaCgx2mjQYGt@mail.gmail.com>
Message-ID: <4C59844F.4070805@sover.net>

There are two major issues, both related to the differences in EEG and 
voice spectral content:

1. EEG amplifiers typically record from 0.5 -100 Hz but may be able to 
open to 500Hz for EP data collection. However, voice typically requires, 
for telephone-like reproduction, a 500 Hz to 3000 Hz band pass. Thus the 
EEG amplifiers would likely reject and/or severly distort audio inputs.
2. As most EEG machines are digital these days, they time sample 
(digitize) EEG signals. As a minimum they must use a rate that is 
slightly more than twice the highest desired input signal, e.g., 200 H 
samplige rate for 100 Hz EEG.  For voice one would require a 
digitization rate of at least 2x 3000Hz; typically one wants a minimum 
of 10000 Hz for voice.  So the EEG devices typical sampling rates of 
200-1024 Hz would not reasonably sample voice signals.

So, no, voice over EEG amplifiers won't work. It would take a complex, 
parallel EEG and separate voice amplifier and A-D converter setups to 
match voice to EEG. This can be done. Indeed some EEG machines match 
both voice and video to EEG. However, there is a lot of engineering to 
make the EEG and voice/video follow the same clock time.

Best of luck.

Frank Duffy, MD  fhd at sover.net
For the above discussion also BSEE and ham (k1moq)
On 8/3/2010 8:10 AM, Marc wrote:
> Hi.
>
> This is not directly related to EEGLab. But I thought someone may be
> able to give some pointers.
>
> We're doing some experiments measuring scalp EEG of participants. One
> of the variables we intend to record is the participant's voice reply.
> We're wondering if it is possible to use the same EEG hardware
> amplifier to record the voice reply? Have anyone tried that before?
> That is, instead of connecting the electrodes to the one of the 64
> channels of the amplifier, we connect a microphone to the input
> channel. I assume we will be able to record as good a signal as any
> audio amplifier? Is there any thing else to watch out for?
>
> Thanks for any advise.
>
> Marc.
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
>
>    



From grega.repovs at psy.ff.uni-lj.si  Wed Aug  4 11:18:01 2010
From: grega.repovs at psy.ff.uni-lj.si (Grega Repovs)
Date: Wed, 4 Aug 2010 14:18:01 -0400
Subject: [Eeglablist] EEG amplifier and alternative input
In-Reply-To: <00ac01cb3390$b8145ac0$283d1040$@net>
References: <AANLkTimf5jPy9ov-gm3nci9VzQ9uZfnPfaCgx2mjQYGt@mail.gmail.com>
	<00ac01cb3390$b8145ac0$283d1040$@net>
Message-ID: <0A32A035-9158-4DEE-9819-D57346BB5DAC@psy.ff.uni-lj.si>

Hi all,

In one of our studies we have successfully recorded voice using microphone on one of the channels along the EEG signal. The information captured is not sufficient for playback, but it is good enough to be able to compute voice reaction times. If only a few words are used and are distinct enough, they might also be used to classify / identify the response. That fulfilled the needs for our study.

All the best,

Grega Repovs

PS If the aim is to compute reaction times, do take into account that some phonemes have more recognizable 'explosive' onset (eg b, p) while with others amplitude builds up slowly (eg s), which might lead to systematic differences in estimated word response onset. With a large set of trials with mixed starting phonemes, the mean should be fine, however when the same words are used for the same conditions it might lead to a spurious difference between conditions.

On 4 Aug 2010, at 00:51, "Jeff Eriksen" <jefferiksen at comcast.net> wrote:

> Marc,
> 
> EEG recorded for ERPs is generally 0.1 - 100 Hz.
> Human hearing is roughly 20-20,000 Hz.
> Voice recording requires a substantial part of that, but I do not know
> off-hand how high.
> I would say you would probably get a poor voice recording, unless you have
> higher end amplifiers that can sample at 5-10 KHz.
> 
> -Jeff Eriksen
> 
> -----Original Message-----
> From: eeglablist-bounces at sccn.ucsd.edu
> [mailto:eeglablist-bounces at sccn.ucsd.edu] On Behalf Of Marc
> Sent: Tuesday, August 03, 2010 5:10 AM
> To: eeglablist at sccn.ucsd.edu
> Subject: [Eeglablist] EEG amplifier and alternative input
> 
> Hi.
> 
> This is not directly related to EEGLab. But I thought someone may be
> able to give some pointers.
> 
> We're doing some experiments measuring scalp EEG of participants. One
> of the variables we intend to record is the participant's voice reply.
> We're wondering if it is possible to use the same EEG hardware
> amplifier to record the voice reply? Have anyone tried that before?
> That is, instead of connecting the electrodes to the one of the 64
> channels of the amplifier, we connect a microphone to the input
> channel. I assume we will be able to record as good a signal as any
> audio amplifier? Is there any thing else to watch out for?
> 
> Thanks for any advise.
> 
> Marc.
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
> 



From jstremblay at usouthal.edu  Wed Aug  4 11:44:28 2010
From: jstremblay at usouthal.edu (Jack Shelley-Tremblay)
Date: Wed, 04 Aug 2010 13:44:28 -0500
Subject: [Eeglablist] EEG amplifier and alternative input
Message-ID: <4C596EAC.252A.00F2.0@usouthal.edu>

Also, don't forget that you are introducing all of the artifact associated with

speaking (EMG, movement, etc...).

If you are trying to correlate EEG with ongoing speech you must

address the EEG artifact issues first, I would say.

Also, with any decent digital recording software, even shareware, you

will get a time code, and this could be sequenced to your EEG time code

quite easily. Probably would be better to record speech in a separate program, in

my opinion.


There are two major issues, both related to the differences in EEG and 
voice spectral content:

1. EEG amplifiers typically record from 0.5 -100 Hz but may be able to 
open to 500Hz for EP data collection. However, voice typically requires, 
for telephone-like reproduction, a 500 Hz to 3000 Hz band pass. Thus the 
EEG amplifiers would likely reject and/or severly distort audio inputs.
2. As most EEG machines are digital these days, they time sample 
(digitize) EEG signals. As a minimum they must use a rate that is 
slightly more than twice the highest desired input signal, e.g., 200 H 
samplige rate for 100 Hz EEG.  For voice one would require a 
digitization rate of at least 2x 3000Hz; typically one wants a minimum 
of 10000 Hz for voice.  So the EEG devices typical sampling rates of 
200-1024 Hz would not reasonably sample voice signals.

So, no, voice over EEG amplifiers won't work. It would take a complex, 
parallel EEG and separate voice amplifier and A-D converter setups to 
match voice to EEG. This can be done. Indeed some EEG machines match 
both voice and video to EEG. However, there is a lot of engineering to 
make the EEG and voice/video follow the same clock time.

Best of luck.

Frank Duffy, MD  fhd at sover.net 
For the above discussion also BSEE and ham (k1moq)
On 8/3/2010 8:10 AM, Marc wrote:
> Hi.
>
> This is not directly related to EEGLab. But I thought someone may be
> able to give some pointers.
>
> We're doing some experiments measuring scalp EEG of participants. One
> of the variables we intend to record is the participant's voice reply.
> We're wondering if it is possible to use the same EEG hardware
> amplifier to record the voice reply? Have anyone tried that before?
> That is, instead of connecting the electrodes to the one of the 64
> channels of the amplifier, we connect a microphone to the input
> channel. I assume we will be able to record as good a signal as any
> audio amplifier? Is there any thing else to watch out for?
>
> Thanks for any advise.
>
> Marc.
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html 
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu 
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu 
>
>    

_______________________________________________
Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html 
To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu 
For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu 


John Shelley-Tremblay, PhD
Associate Professor
Department of Psychology
307 University Blvd. North
University of South Alabama
Mobile, AL 36688-0002
251-460-6883

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: Jack Shelley-Tremblay.vcf
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100804/6c0632ed/attachment.ksh>

From dnewby at une.edu.au  Tue Aug 10 04:15:37 2010
From: dnewby at une.edu.au (dnewby at une.edu.au)
Date: Tue, 10 Aug 2010 21:15:37 +1000 (EST)
Subject: [Eeglablist] Channel locations
Message-ID: <51141.61.69.186.132.1281438937.squirrel@mail.une.edu.au>

Hi

This is a very basic question related to creating topographic maps in
EEGLAB using Neuroscan .avg file - I attach a .pdf with one of the maps
created with EEGLAB and the same one in Neuroscan where many channels
(e.g., T6) are out of boundaries for EEGLAB cap compared with the original
Neuroscan.

I tried to delete certain channels of no great interest (e.g., HEOR) and
when I tried to plot, received an error message saying that channel number
does not correspond to original 40 and hence plot was not created.

How could I "push" all original (Neuroscan) 40 channels to be within the
boundaries of EEGLAB cap without losing/deleting any channels?

Thanks for any advice.

Daiva




-------------- next part --------------
A non-text attachment was scrubbed...
Name: Channel locations.pdf
Type: application/pdf
Size: 130383 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100810/52209bb8/attachment.pdf>

From lutobu at gmail.com  Tue Aug 10 14:00:59 2010
From: lutobu at gmail.com (ludwing torres)
Date: Tue, 10 Aug 2010 23:00:59 +0200
Subject: [Eeglablist] help: source reconstruction with time course
Message-ID: <AANLkTike8uJa9_0xuEiupJQ-y3FXnRub-+ZvPPVpPu7R@mail.gmail.com>

Hello. Im trying to perform a source reconstruction from a 10-20 system of
19 channel locations, after I perform ICA, I go for dipole fitting using
dipfit, and I obtain exactly the same number of dipoles than the number of
channels, and I only get  the positions at a single time moment.

My question is: How can I  get the source reconstruction of the dipoles in
the entire timecourse of the signals of the channels, and why I am obtaining
the same number of sources: isn't it supposed to be greater the number of
sources than the number of scalp measures?

Another thing is: if I have the radius of the spherical head model, how can
I use this volume and conductance configurations rather than those that
already exist in eeglab?

thank you for all your replies
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100810/0d4621b5/attachment.html>

From anjoibanez at gmail.com  Wed Aug 11 03:15:16 2010
From: anjoibanez at gmail.com (=?ISO-8859-1?Q?Antonio_Ib=E1=F1ez?=)
Date: Wed, 11 Aug 2010 12:15:16 +0200
Subject: [Eeglablist] help: source reconstruction with time course
In-Reply-To: <AANLkTike8uJa9_0xuEiupJQ-y3FXnRub-+ZvPPVpPu7R@mail.gmail.com>
References: <AANLkTike8uJa9_0xuEiupJQ-y3FXnRub-+ZvPPVpPu7R@mail.gmail.com>
Message-ID: <AANLkTinBhVCiw==mknysxZaX0_01ybhGvcwqOpkTh6cB@mail.gmail.com>

Hello i think what you obtain is the best fitting dipoles for every
component in the entire time interval; that is, only one solution for the
entire interval. Even though you have the same number of components and
electrodes, you probably want to fit only those that explain more than 90%
of the variance in the data. If  I am not wrong, dipole models solutions
always give you a limited number of sources (max = number of electrodes). If
you want to obtain more sources you may consider a distributed model for the
inverse solution. For a discussion about different solutions, see Michel et
al. (2004). EEG source imaging. Clinical Neurophisiology, 115.

I believe that the spherical settings can be modified in 'Head model and
settings', selecting the 'custom model files' option as a model.

hope this helps,

Antonio

2010/8/10 ludwing torres <lutobu at gmail.com>

> Hello. Im trying to perform a source reconstruction from a 10-20 system of
> 19 channel locations, after I perform ICA, I go for dipole fitting using
> dipfit, and I obtain exactly the same number of dipoles than the number of
> channels, and I only get  the positions at a single time moment.
>
> My question is: How can I  get the source reconstruction of the dipoles in
> the entire timecourse of the signals of the channels, and why I am obtaining
> the same number of sources: isn't it supposed to be greater the number of
> sources than the number of scalp measures?
>
> Another thing is: if I have the radius of the spherical head model, how can
> I use this volume and conductance configurations rather than those that
> already exist in eeglab?
>
> thank you for all your replies
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100811/1af3f8bf/attachment.html>

From czelano at gmail.com  Fri Aug 13 10:55:40 2010
From: czelano at gmail.com (christina maria zelano)
Date: Fri, 13 Aug 2010 12:55:40 -0500
Subject: [Eeglablist] Can EEGLab import Bio-logic CEEGraph .EEG data?
In-Reply-To: <AANLkTimH3+TM13B1MOdkLK239Q6bLdGdk-cugagPtyAB@mail.gmail.com>
References: <AANLkTimH3+TM13B1MOdkLK239Q6bLdGdk-cugagPtyAB@mail.gmail.com>
Message-ID: <AANLkTimTSZj-pN6+7Ci4nmdeNQ=VrsawEP7ZW2fkL19d@mail.gmail.com>

Is it possible to import .EEG data from a Bio-logic CEEGraph system? When I
try to load the file using loadeeg.m, it does not work. It finds zero
channels. Any suggestions would be greatly appreciated.

Thanks!!

Christina
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100813/4dcbb00c/attachment.html>

From Alejo.Keuroghlanian at iit.it  Sun Aug 15 07:21:36 2010
From: Alejo.Keuroghlanian at iit.it (Alejo Keuroghlanian)
Date: Sun, 15 Aug 2010 16:21:36 +0200
Subject: [Eeglablist] Running binica under Windows XP: some problems.
Message-ID: <FC14DB6243548F4683D9E3D7E9A51C382ACC73CA24@IITMXMBX.iit.local>

Dear all,

I'm working under Windows XP, with Matlab version 7.5.0.342(R2007b), and EEGlab version 8.3.5b. 

I'm trying to use binica, but I'm getting some errors, and I don't understand why.

I have followed the instructions in the webpage: http://sccn.ucsd.edu/eeglab/binica/   There, I downloaded the files in the link "Windows PC (95, 98, NT, 2000, XP?. older version)". The files were 1ST_README.txt, ica.exe, and ica.sc. I put all these files in a folder included in the Matlab path. Then, I modified the icadefs.m file in such a way that 

ICABINARY = 'C:\Documents and Settings\Alejo\My Documents\eeglab8_0_3_5b\binica\ica_pc\ica.exe' 

Then, when I execute the command

>> [wts,sph] = binica('C:\Documents and Settings\Alejo\My Documents\MATLAB\acticap_19_03_10\stefano\stefano_cwalker_fcz_ref_wout_ft9_ft10_artrej_sr250.dat',62,300)

I get the following messages: 

binica: using source file 'C:\Documents and Settings\Alejo\My Documents\eeglab8_0_3_5b\eeglab8_0_3_5b\functions\sigprocfunc\binica.sc'
binica(): using binary ica file '?/C:\Documents and Settings\Alejo\My Documents\eeglab8_0_3_5b\binica\ica_pc\ica.exe'
binica(): no optional (flag, argument) pairs received.
scriptfile = binica975.sc

Running ica from script file binica975.sc
The system cannot find the file specified. 
??? Error using ==> floatread at 168
floatread() fopen() error.

Error in ==> binica at 414
    wts = floatread(weightsfile,[ncomps Inf],[],0);

------

I have also tried to modify the SC variable in the icadefs.m file in such a way that 

SC  = ['C:\Documents and Settings\Alejo\My Documents\eeglab8_0_3_5b\binica\ica_pc\ica.sc']; 

but still I get some other error messages:

binica: using source file 'C:\Documents and Settings\Alejo\My Documents\eeglab8_0_3_5b\binica\ica_pc\ica.sc'
binica(): using binary ica file 'C:\Documents and Settings\Alejo\My Documents\eeglab8_0_3_5b\binica\ica_pc\ica.exe'
??? Attempted to access s(2); index out of bounds because numel(s)=1.

Error in ==> binica>rmcomment at 451
     while s(n)~=symb % discard # comments

Error in ==> binica>read_sc at 504
     s = rmcomment(s,'#');

Error in ==> binica at 208
[flags,args] = read_sc(SC); % read flags and args in master SC file

-------

Anyway, I understand I'm getting something completely wrong, because I realize binica is building up an ad hoc .sc file each time I run it, so probably there is no point in modifying the SC variable in the icadefs.m file.

Is there a step-by-step tutorial about using binica under Windows, for dummies?

Thanks in advance.

Best,

Alejo  


From j.martinovic at liverpool.ac.uk  Mon Aug 16 09:57:56 2010
From: j.martinovic at liverpool.ac.uk (Jasna Martinovic)
Date: Mon, 16 Aug 2010 17:57:56 +0100
Subject: [Eeglablist] postdoctoral positions at the University of Leipzig
Message-ID: <4C696E14.60607@liverpool.ac.uk>

  Postdoctoral positions at the University of Leipzig, Germany, in the 
laboratory of Matthias M. Mueller

Two postdoctoral positions are available immediately in the lab of Prof 
M.M.Mueller at the Institute of Experimental Psychology, University of 
Leipzig. The employment is full-time and will be initially awarded for 
two years with the possibility of extension. The salary is commensurate 
with the applicant?s qualifications and previous work experience.

The research in Prof Mueller?s lab focuses on the neural mechanisms of 
attention and object representation. Topics include visual and 
somatosensory attention, attention in the elderly, effects of emotion 
and motivation on attentional processing, visual object representation 
and memory. The research facilities include a 128-channel EEG lab and a 
behavioural testing booth. Prof Mueller?s lab has a fine record in 
investigating brain responses related to visual attention and object 
representation.

Here are some representative publications from the recent years:
Andersen, S.K., Hillyard, S.A., M?ller, M.M.: Attention facilitates 
multiple stimulus features in parallel in human visual cortex. Current 
Biology, 2008, 18, 1006-1009.
Andersen, S. K., Fuchs, S., M?ller, M. M. (2009). Effects of 
feature-selective and spatial attention at different stages of visual 
processing. Journal of Cognitive Neuroscience, doi:10.1162/ jocn.2009.21328.
Adler, J., Giabbiconi, C. M., & M?ller, M.M. (2009). Shift of attention 
to the body location of distracters is mediated by perceptual load in 
sustained somatosensory attention. Biological Psychology, 81(2), 77-85.
Hindi Attar, C., M?ller, M. M., Andersen, S. K., B?chel, C., Rose, M. 
(2010). Emotional processing in a salient motion context: Integration of 
motion and emotion in both V5/hMT+ and the amygdala. Journal of 
Neuroscience, 30(15), 5204-5210
Martinovic, J., Gruber, T., M?ller, M.M.: Induced gamma-band responses 
predict recognition delays during object identification. Journal of 
Cognitive Neuroscience, 2007, 19, 921-934.

The postdoctoral researcher should be able to conduct independent 
research into any of the main research areas within the lab, using EEG 
and behavioural methods. Candidates should have a background in 
Psychology, Biology or Neuroscience and hold a PhD or be close to 
completion of a PhD. Previous neuroimaging experience is preferred ? in 
particular EEG or MEG. Familiarity with the use of Matlab for data 
analysis and experimental design would be advantageous. Furthermore, 
excellent writing and presentation skills in English are needed. 
Positions require teaching at undergraduate and graduate level (German 
and/or English; modules will be distributed according to 
expertise/background).

Applications with a letter of interest, CV, publication list and details 
of two referees should be sent to Prof M.M.Mueller at 
m.mueller at rz.uni-leipzig.de. The deadline for applications is 20th of 
September 2010. For more information about the Institute, visit 
www.uni-leipzig.de/~psyall2 (website works best in IE).



From zeynep at sccn.ucsd.edu  Mon Aug 16 14:13:00 2010
From: zeynep at sccn.ucsd.edu (Zeynep Akalin Acar)
Date: Mon, 16 Aug 2010 14:13:00 -0700
Subject: [Eeglablist] How to register electrodes to segmented head model
In-Reply-To: <48AA2366.6010308@nmsu.edu>
References: <48AA2366.6010308@nmsu.edu>
Message-ID: <AANLkTikG=A9_7UcjGMej3vEsObYe5tC8n7BT9PchS7Fq@mail.gmail.com>

Hi Hung,

You can use NFT for coregistration of EEG electrodes to a realistic
head surface. NFT can read ANALYZE format images, do segmentation and
create realistic surface meshes. It then allows you to register
digitizer coordinates to the surface mesh in two steps. First, a
manual coregistration is done using EEGLAB's function, then, an
automatic coregistration is performed minimizing the distances between
the electrodes and scalp surface.

For more information on NFT: http://sccn.ucsd.edu/nft/

Zeynep.

On Mon, Aug 18, 2008 at 6:35 PM, Hung Dang <hungptit at nmsu.edu> wrote:
> Dear all,
> I need to fit the Biosemi 128 electrode to the segmented ?volume image
> in ANALYZE ?7.5 format.
> I wonder if there is any free software could do ?that?
>
> Any suggestion would be very appreciated.
>
> Thanks a lot,
> Hung
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
>



From hecke at nld.ds.mpg.de  Tue Aug 17 06:15:29 2010
From: hecke at nld.ds.mpg.de (hecke)
Date: Tue, 17 Aug 2010 15:15:29 +0200
Subject: [Eeglablist] =?iso-8859-1?q?places_left=3A_Fall_Course_on_Computa?=
 =?iso-8859-1?q?tional_Neuroscience_in_G=F6ttingen=2C_Germany?=
Message-ID: <4C6A8B71.7040005@nld.ds.mpg.de>

*------------------------------*
* still a few places available *
*------------------------------*



Applications are invited for the eighth fall course on

           COMPUTATIONAL NEUROSCIENCE in G?ttingen, Germany

                     September 20th - 24th, 2010

                                    organized by Hecke Schrobsdorff


The course is intended to provide graduate students and young
researchers from all parts of neuroscience with working knowledge of
theoretical and computational methods in neuroscience and to acquaint
them with recent developments in this field.

The course includes tutorials and lectures of the following researchers:

    * Daniel A. Butts, University of Maryland, College Park
    * Sophie Deneve, Ecole Normale Superieur, Paris
    * Hansj?rg Scherberger, German Primate Center, G?ttingen
    * Elad Schneidman, Weizmann Institute of Science, Rehovot
    * Susanne Still, University of Hawaii

The course takes place at the Department of Nonlinear Dynamics of the
Max-Planck Institute for Dynamics and Self-Organization, Bunsenstr. 10,
 D-37073 G?ttingen. A course fee of 100 Euro includes participation in
the tutorials, study materials, and part of the social events. The
number of participants is limited to about 30. Course language is English.

To apply please fill out the application form at:
http://www.bccn-goettingen.de/events-1/cns-course
as soon as possible.

Best wishes and looking forward to seeing you in G?ttingen

Hecke


From yoh at psychology.rutgers.edu  Thu Aug 19 19:43:30 2010
From: yoh at psychology.rutgers.edu (Yaroslav Halchenko)
Date: Thu, 19 Aug 2010 22:43:30 -0400
Subject: [Eeglablist] EEGLAB test suite
Message-ID: <20100820024330.GS18647@onerussian.com>

Dear EEGLAB Team, contributors, and users,

Recently I've sent following request to Andreas who have contributed at
some point to deliver a test suite for EEGLAB in an attempt to make it
(partially) compatible with Octave.  Unfortunately I have got no reply,
so I have decided to ask here:

* is there anyone interested/continuing the endeavor of Andreas in
  making EEGLAB Octave-friendly?  Besides GUI aspect, upon quick
  look I see that bulk of functionality in EEGLAB fails even syntax
  compatibility due to excessive use of nested functions.  But, I guess,
  if that aspect would be kept in mind for future EEGLAB developments it
  could be easily made Octave compatible.  or may be there are more
  severe issues?

* was Andreas' test suite (or parts of it) ever incorporated
  anywhere else?

Thanks in advance for clarifications!

----- Forwarded message from Yaroslav Halchenko <yoh at dartmouth.edu> -----

Date: Tue, 10 Aug 2010 11:41:40 -0400
From: Yaroslav Halchenko <yoh at dartmouth.edu>
To: Andreas Romeyke <romeyke at cbs.mpg.de>
Cc: Michael Hanke <michael.hanke at gmail.com>
Subject: EEGLAB test suite

Hi Andreas,

Me and Michael are the team of NeuroDebian project [1] working
toward making Debian GNU/Linux a convenient platform for neuroscience
research [2].  As the next target we are planing to support a selection
of the most popular toolboxes for Matlab (e.g. SPM and EEGLAB), and one
of our accents is to pursue at least partial (non-GUI functionality)
compatibility with recent versions of Octave.  For such purpose
unittests (or any other testing framework) are indispensable.  We saw
that you had done an outstanding job developing a test suite for EEGLAB
[3].  Unfortunately SVN repository has moved so no links on wiki are
up-to-date and new location  needs an authentication so we could
not check either the test-suite partially or in full could be used with
current version of eeglab.

We wonder if we could obtain access to SVN and either you would mind
sharing your thoughts on current status of the test-suite and your
eeglab for octave endeavor. 

Thank you in advance for the reply!

[1] http://neuro.debian.net/#the-team
[2] http://neuro.debian.net/
[3] https://svnserv.cbs.mpg.de/trac/eeglab
-- 
Yaroslav O. Halchenko
Postdoctoral Fellow,   Department of Psychological and Brain Sciences
Dartmouth College, 419 Moore Hall, Hinman Box 6207, Hanover, NH 03755
Phone: +1 (603) 646-9834                       Fax: +1 (603) 646-1419
WWW:   http://www.linkedin.com/in/yarik        

----- End forwarded message -----

-- 
Yaroslav O. Halchenko
Postdoctoral Fellow,   Department of Psychological and Brain Sciences
Dartmouth College, 419 Moore Hall, Hinman Box 6207, Hanover, NH 03755
Phone: +1 (603) 646-9834                       Fax: +1 (603) 646-1419
WWW:   http://www.linkedin.com/in/yarik        



From yuanx041 at umn.edu  Thu Aug 19 20:26:06 2010
From: yuanx041 at umn.edu (Han Yuan)
Date: Thu, 19 Aug 2010 22:26:06 -0500
Subject: [Eeglablist] Release of eConnectome 1.0 full version
Message-ID: <002a01cb4017$6de4fca0$49aef5e0$@edu>

Dear Colleagues,

 

We are pleased to announce the release of the full-version eConnectome 1.0.
eConnectome is a free open-source MATLAB software package for imaging brain
functional connectivity from electrophysiological signals. It provides
interactive graphical interfaces for EEG/ECoG preprocessing, source
estimation, connectivity analysis and visualization. The beta-version of
eConnectome 1.0b was previously released on March 12, 2010. The current
release of full version 1.0 has been much enhanced and expanded. 

 

The eConnectome 1.0 allows the connectivity imaging from EEG and ECoG over
the sensor and source domains. The new features of the full version include
functions of adaptive connectivity analysis, processing of event-related
potential and event-related (de)synchronization, and individual anatomic
modeling. The visualization module has been enhanced to provide dynamic view
of continuous potential mapping and adaptive connectivity network. It now
supports exporting the images or movies into a variety of formats. Previous
bugs of the 1.0b version reported during the beta testing have also been
fixed. 

 

This package is designed for use by researchers in neuroscience, psychology,
cognitive science, clinical neurophysiology, neurology and other
disciplines. The graphical interface-based platform requires little
programming knowledge or experience with MATLAB.

 

eConnectome is developed by the Biomedical Functional Imaging and
Neuroengineering Laboratory at the University of Minnesota, directed by Dr.
Bin He. The visualization module is jointly developed with Drs. Fabio
Babiloni and Laura Astolfi at the University of Rome "La Sapienza". 

 

Free download of eConnectome and more information can be found at
http://econnectome.umn.edu/index.htm

 

Sincerely yours,

 

Bin He, PhD

Distinguished McKnight University Professor

Director, Biomedical Functional Imaging and Neuroengineering Laboratory

Director, Center for Neuroengineering

University of Minnesota

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100819/8548118c/attachment.html>

From arno at ucsd.edu  Fri Aug 20 09:05:49 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Fri, 20 Aug 2010 09:05:49 -0700
Subject: [Eeglablist] EEGLAB test suite
In-Reply-To: <20100820024330.GS18647@onerussian.com>
References: <20100820024330.GS18647@onerussian.com>
Message-ID: <389497E3-8DAC-4986-8856-45223A0DFABA@ucsd.edu>

Dear Yaroslav,


> * is there anyone interested/continuing the endeavor of Andreas in
>  making EEGLAB Octave-friendly?  Besides GUI aspect, upon quick
>  look I see that bulk of functionality in EEGLAB fails even syntax
>  compatibility due to excessive use of nested functions.  But, I guess,
>  if that aspect would be kept in mind for future EEGLAB developments it
>  could be easily made Octave compatible.  or may be there are more
>  severe issues?


We have not spent a lot of time trying to make EEGLAB Octave Friendly.
You may see this page though.

http://sccn.ucsd.edu/wiki/EEGLAB_and_Octave

Basically, it requires tremendous amount of work to adapt EEGLAB graphics to Octave. You could use Octave for computation only but then you would still need Matlab to visualize your results (so why not use Matlab in the first place). For some unknown reasons, Octave has been stuck with using the limited functionality and flexibility of GNUplot for graphics (which has evolved since 1986 but just cannot compare to the extended graphics capabilities of Matlab which is based on Java or to the graphic capabilities most other free software for that matter). Nevertheless, it could still make sense to use Octave on clusters of computers and do visualization on a single workstation (avoiding the need to purchase multiple Matlab licenses). If someone is motivated to make part of the EEGLAB Octave compatible, we will open the code for him/her.


> * was Andreas' test suite (or parts of it) ever incorporated
>  anywhere else?


Andreas Romeyke and Maxim Duester from Germany worked extensively in 2007 on an EEGLAB test suite (a series of functions - or test cases - to test EEGLAB). However, Andreas does not support the web site https://svnserv.cbs.mpg.de/trac/eeglab any more and they discontinued their efforts in 2007. However, we recently (2010) dig out their old code and incorporated their testsuite with our own EEGLAB test functions. There are now more than 600 test functions thats test almost all functions of EEGLAB (it does not mean that these functions work in all cases but it is a guarantee that at least they work - or do not crash - in some cases). If someone is interested in helping to maintain or develop further EEGLAB test cases, please let us know. Unlike EEGLAB, test cases are not under currently under SVN but we could easily create a repository.

Arno




From yoh at psychology.rutgers.edu  Fri Aug 20 11:43:38 2010
From: yoh at psychology.rutgers.edu (Yaroslav Halchenko)
Date: Fri, 20 Aug 2010 14:43:38 -0400
Subject: [Eeglablist] EEGLAB test suite
In-Reply-To: <389497E3-8DAC-4986-8856-45223A0DFABA@ucsd.edu>
References: <20100820024330.GS18647@onerussian.com>
	<389497E3-8DAC-4986-8856-45223A0DFABA@ucsd.edu>
Message-ID: <20100820184338.GM12007@onerussian.com>

Thank you Arnaud,

On Fri, 20 Aug 2010, Arnaud Delorme wrote:
> We have not spent a lot of time trying to make EEGLAB Octave Friendly.
> You may see this page though.
> http://sccn.ucsd.edu/wiki/EEGLAB_and_Octave
yeap -- was there ;-) I just wondered if there were any changes
since then behind the curtain ;)

> Nevertheless, it could still make sense to use Octave on clusters of
> computers and do visualization on a single workstation (avoiding the
> need to purchase multiple Matlab licenses).
That is one kind of basic usage scenario I also had in mind.  Few times
I have used EEGLAB for basic data conversion and preprocessing which,
theoretically, did not require any GUI/plotting per se.  

Additional future extension (if someone gets interested) could be 
exposing some interesting EEGLAB functionality to be used within other
processing pipeline interfaces, such as
http://nipy.sourceforge.net/nipype/ . For that no GUI/plotting would be
required from EEGLAB -- just I/O facilities and processing

> If someone is motivated to
> make part of the EEGLAB Octave compatible, we will open the code for
> him/her.
hm... do you mean
- providing commit rights
or
- exposing some additional code on top of what is in SVN already? ;)

> discontinued their efforts in 2007. However, we recently (2010) dig out
> their old code and incorporated their testsuite with our own EEGLAB test
> functions. There are now more than 600 test functions thats test almost
> all functions of EEGLAB (it does not mean that these functions work in
> all cases but it is a guarantee that at least they work - or do not
> crash - in some cases). If someone is interested in helping to maintain
> or develop further EEGLAB test cases, please let us know. Unlike EEGLAB,
> test cases are not under currently under SVN but we could easily create
> a repository.
That would be terrific and utterly useful. In my opinion, scientific
applications (especially for scripting languages/environments) MUST have
testing framework -- from unittests all the way to documentation.
In Python world it kinda becomes a convention that project comes with
unittests and there are doctests in addition which verify correctness of
the examples in documentation strings within the code and within the
documentation.  Makes it look nice and behave as expected on users'
systems.

so, why not just include this tests battery within eeglab itself?  I am
ok with separate repository, but I just wonder if it wouldn't be of
greater value and convenience to be within EEGLAB?

-- 
Yaroslav O. Halchenko
Postdoctoral Fellow,   Department of Psychological and Brain Sciences
Dartmouth College, 419 Moore Hall, Hinman Box 6207, Hanover, NH 03755
Phone: +1 (603) 646-9834                       Fax: +1 (603) 646-1419
WWW:   http://www.linkedin.com/in/yarik        


From landsness at wisc.edu  Sat Aug 21 11:57:53 2010
From: landsness at wisc.edu (Eric Landsness)
Date: Sat, 21 Aug 2010 13:57:53 -0500
Subject: [Eeglablist] does newcrossf fischer's z-transform the coherence
	values
Message-ID: <7220aed523fa0.4c6fdb61@wiscmail.wisc.edu>

I was wondering if the newcrossf function computes the fischer's z-tranform of the coherence values before averaging across trials?  The reason that I ask is that Lopes da Silva 1973 claim that the distribution of coherence values is not a normal distribution and therefore must be z-transformed first.

Thanks,
Eric


From datashare at sciencedb.net  Wed Aug 25 06:46:20 2010
From: datashare at sciencedb.net (Science DB datashare)
Date: Wed, 25 Aug 2010 15:46:20 +0200 (CEST)
Subject: [Eeglablist] Announcement : EEG/MEG database
Message-ID: <8181f0de27b09d9c6011e24e92bca9f4.squirrel@squirrel-webmail.surftown.com>



Dear EEGLAB users,

Data sharing is a great way to find new collaborators. We would like to
announce a our new database for EEG/MEG data on ScienceDB.net

The concept is simple: If you have data to share you create a new topic in
the EEG/MEG data share group?s forum, where you describe your data, and
how you want to share it.

Please go to : http://sciencedb.net/groups/eegmeg-data-share/

ScienceDB.net is a non-profit portal for the scientific community. Our
service is sponsored by ads and donations. If you have any suggestions or
comments, please mail us at datashare at sciencedb.net.


Kind regards,

Science DB
http://sciencedb.net/




From dgroppe at cogsci.ucsd.edu  Wed Aug 25 22:43:41 2010
From: dgroppe at cogsci.ucsd.edu (David Groppe)
Date: Wed, 25 Aug 2010 22:43:41 -0700
Subject: [Eeglablist] channel location file
In-Reply-To: <4C3B1C11.50205@cogpsyphy.hu>
References: <4C358C27.7080903@mail.upb.de>
	<4C3B1C11.50205@cogpsyphy.hu>
Message-ID: <AANLkTi=P3ZK6nGadaLvcrcc0f73vbEqiUpi8k1G1vJLs@mail.gmail.com>

Hi Michael & Lazlo,
    Because there are more EEG sources than EEG sensors, it is
impossible for ICA (or any other linear spatial filter) to perfectly
filter out EEG artifacts.  Thus just using ICA to correct for
artifacts will distort the data to some extent.  See the following for
the explanation:

Groppe, D.M., Makeig, S., &  Kutas, M. (2008) Independent component
analysis of event-related potentials. Cognitive Science Online, 6.1,
pp. 1-44. http://cogsci-online.ucsd.edu/6/6-1.pdf

That being said, the amount of distortion may be negligible.  The most
careful study of this I know of is a recent paper by Mognon et al. in
Psychophysiology:
http://onlinelibrary.wiley.com/doi/10.1111/j.1469-8986.2010.01061.x/abstract

They found that the distortion caused by ICA artifact correction
ranged from around 75% to 25% the magnitude of the ERP components of
interest.  Personally, I've used ICA artifact correction for several
years and have found that my main ERP effects of interest (e.g., N170,
P300, N400, P600, auditory N1) look and behave as they should.  I
would be a bit nervous though about using any type of artifact
correction to study frontal ERPs unless the components were rather
robust (like the auditory N1's projection to frontal channels).
    hope this helps,
       -David

David M. Groppe, Ph.D.
Department of Cognitive Science, 0515
University of California, San Diego
9500 Gilman Dr.
La Jolla, CA 92093-0515
dgroppe at cogsci.ucsd.edu



On Mon, Jul 12, 2010 at 6:43 AM, Laszlo Balazs <balazs at cogpsyphy.hu> wrote:
> Hello Michael,
> To your second question: We have just started to try ICA based EOG
> correction and I have the same concern about throwing out some baby with
> the bath water. Once I have tested the regression based method provided
> by Neuroscan. I chose a few subjects with frequent blinks but still high
> enough number of clean sweeps. Then I compared uncorrected and corrected
> sweeps by visual inspection and also the averaged ERP to targets in a
> visual oddball (blink free epochs only). ?I have seen marked decrease in
> both EEG and ERP amplitude over Fp1/2. We plan to repeat this test soon
> including ICA and hope it will outperform the regression method. I
> wonder if there are papers or other ideas for validation tests around?
> Kind regards,
> Laszlo
>
>> Dear all,
>>
>> are there any channel location files for the Neuroscan QuickCap40 (40
>> channels) available? On the Neuroscan ftp I only found "dave.zip" which
>> contains 32 and 64 channels + I get an error when I read those in, anyway.
>>
>> And a second question:
>> I'm playing around with ICA to reject ocular artifacts from continuous
>> data. Up to now I could find only few papers about the validity of
>> ICA-corrected data. I'm asking myself that question because very often I
>> get components representing the ocular artifacts, but they also seem to
>> contain eeg-information, especially when I use the fastica algorithm.
>> This results in quiet different looking signals in sections where no
>> artifacts occur - mainly in frontal channels of course.
>>
>> Any help is highly appreciated.
>>
>> Thanks!
>>
>> Michael
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
>>
>>
>> __________ Information from ESET NOD32 Antivirus, version of virus signature database 5271 (20100712) __________
>>
>> The message was checked by ESET NOD32 Antivirus.
>>
>> http://www.eset.com
>>
>>
>>
>
>
> --
>
> ----------------------------------------------------------------------------
> ? ? Laszlo Balazs, Ph.D. ? ? ? ? ?/ ? ? ?dr. Bal?zs L?szl?
> Institute for Psychology HAS ? ? ? / ? MTA Pszichol?giai Kutat?it?zet
> Budapest, P O B 398, H-1394 Hungary / Tel:+36(1)354-2410 ?Fax:+36(1)354-2416
>
>
>
> __________ Information from ESET NOD32 Antivirus, version of virus signature database 5271 (20100712) __________
>
> The message was checked by ESET NOD32 Antivirus.
>
> http://www.eset.com
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
>



From ievans at une.edu.au  Thu Sep  2 23:42:55 2010
From: ievans at une.edu.au (Evans, Ian D.)
Date: Fri, 3 Sep 2010 16:42:55 +1000
Subject: [Eeglablist] Eye movement artifact in ICA - missing coordinates for
	EOG electrodes
Message-ID: <0DDFFD83A96343F7A518B024FF3F9E6E@TerminaHB>

Greetings all! I'm trying to isolate eye movement artifact with ICA in an 
attempt to make EOG correction more accurate and hopefully remove less of 
the genuine underlying data that regression methods might otherwise remove, 
however I am having trouble tracking down a satisfactory scalp map that 
includes co-ordinates (Cartesian, 2D polar, 3D cylindrical) for the EOG 
electrodes on the face. Neither Neuroscan nor any setup or .loc files in 
EEGLab seem to provide these details, and all attempts to isolate eye 
movement in ICA without the location of the eye movement electrodes have 
been met with mocking laughter.

The recordings have already been made on a 64-channel cap with six EOG 
electrodes (one above and below each eye, another beside each eye for 
horizontal eye movement) using the traditional 10-20 system. All 
co-ordinates for the scalp are already at hand, just the facial electrodes 
are needed. If anyone can make such a coordinate map available it would be 
most helpful, thank you kindly!

Ian Evans
Cognitive and Affective Neuroscience
University of New England. 



From tej.tadi at epfl.ch  Sat Sep  4 11:48:03 2010
From: tej.tadi at epfl.ch (Tadi Tej)
Date: Sat, 4 Sep 2010 20:48:03 +0200
Subject: [Eeglablist] Paired t-tests on the STUDY data : frequency data
Message-ID: <59BC90F4F36B35478410592AB3F1426F46206AC1FB@REX2.intranet.epfl.ch>

Dear all,

I ran the paired t-tests in the STUDY option for two conditions in eeglab  with  a  p-value < 0.05 and  FDR.
 
I plotted all conditions on the same panel and i see:

1. the power map for the 1st condition and second condition only for the chosen frequency ? or do i see the difference of cond1-cond2?

2. and on the extreme right the map significant differences of the channels that pop up significantly different between the two conditions.

3. How do  i plot the scalp maps for a whole band ( eg: 8-14Hz) or can i modify this outside the study options? now i can plot this for one frequency.

4. How do i plot the scalp map for the difference ( cond1-cond2 and not p- value) between two conditions for a frequency range or a particular band like above.




Thanks a ton

Best,
Tej


From arno at ucsd.edu  Tue Sep  7 03:10:36 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Tue, 7 Sep 2010 18:10:36 +0800
Subject: [Eeglablist] Which EEGLAB version should I use
Message-ID: <666ACCFC-9130-454D-97E7-A3F9CF9C4387@ucsd.edu>

Some users have reported using different EEGLAB versions to perform different data processing (version 6.03b for one type of data processing and version 8.x for another type of data processing). If you are a regular EEGLAB user, it is always better to have the latest EEGLAB version (usually updated weekly). If you encounter a problem, simply submit a bug to the bug repository http://sccn.ucsd.edu/eeglab/bugzilla and it will be addressed (and fixed) usually within a week. 

If you do not want to download a new zip file every week or every two weeks, simply set up your computer to access the EEGLAB repository using SVN (and then you can simply run an update on the repository). For more information see http://sccn.ucsd.edu/wiki/How_to_download_EEGLAB 

Arno

ps: backward compatibility is our foremost concern. EEGLAB version 9 can still load datasets and run history scripts generated by EEGLAB 2.1 back in 2001!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100907/4a8d5645/attachment.html>

From petsol at gmail.com  Tue Sep  7 01:25:35 2010
From: petsol at gmail.com (=?ISO-8859-1?Q?P=E9ter_Solt=E9sz?=)
Date: Tue, 7 Sep 2010 10:25:35 +0200
Subject: [Eeglablist] How to run AMICA
Message-ID: <AANLkTin03sZLcJvs9dX1tHcbVb6yoPC291BM3xpeed4N@mail.gmail.com>

Dear eeglab members,

I'm trying to figure out how to run amica from eeglab. Is there a way,
or is it a detached function? I found that there are some traces in
the pop_runica function (v9), however I'm not able to get it running.

Thanks in advance...

P?ter



From ievans at une.edu.au  Sun Sep  5 07:15:41 2010
From: ievans at une.edu.au (Evans, Ian D.)
Date: Mon, 6 Sep 2010 00:15:41 +1000
Subject: [Eeglablist] Eye movement artifact in ICA - missing coordinates
	for EOG electrodes
In-Reply-To: <AANLkTimzqCf4j5L09akFoOJwLequ1foRUSNf3qj_9f_t@mail.gmail.com>
References: <0DDFFD83A96343F7A518B024FF3F9E6E@TerminaHB>
	<AANLkTimzqCf4j5L09akFoOJwLequ1foRUSNf3qj_9f_t@mail.gmail.com>
Message-ID: <829FE0A620CC496B82A656C672051443@TerminaHB>

Shall endeavor to find out this week, if there's time amongst the residential schools. But from what I can discern the EOG electrodes do seem to make a difference - when I ran an ICA on the eye calibration task used to measure artifact without any EOG electrodes in the mix, around 40 of the 68 components highlighted the F8 region as a strong source, and this in a task purely involving eye movement, no other processing involved other than possibly recognizing the shapes used to indicate where the subject should look. Tried running it again but included some estimated positions of EOG electrodes at the front, and there were still 40+ components all indicating activity at the frontal lobe, but now the activity was around the face instead of F8. If nothing else, the EOG information should assist ICA by isolating changes in multiple electrodes with correlate with the EOG data and rendering it completely independent from the underlying actual data.

As for labels, well I couldn't find any consistent EOG labeling systems for MATLAB, so why not start gathering labels? Let me know what labels you use for the eye channels and I can build the .loc file - David Groppe (many thanks again!) sent me a set of coordinates for the ocular electrodes, but there is no problem with having multiple names for the same set of coordinates - EEGLab detects what we're using anyway. Worth a shot.


Ian Evans
Cognitive and Affective Neuroscience
University of New England.
ievans at une.edu.au


From: Baris Demiral 
Sent: Sunday, September 05, 2010 8:08 PM
To: Evans, Ian D. 
Cc: eeglablist at sccn.ucsd.edu 
Subject: Re: [Eeglablist] Eye movement artifact in ICA - missing coordinates for EOG electrodes


Hi Ian,  


I also made a similar query few months ago. I use 64 electrode 10-20 system plus 4 EOGs. I asked whether there are specific labels for such EOG electrodes in the standard-10-5-Cap385.sfp so that I can map these electrodes on the fly. Are there labels for those, or can we decide and add those labels and coordinates as a community together?


Actually, I find something more important: How much really does this influence ICA related artifact detection and correction? Is there a huge change when you include horizontal and vertical EOGs in the ICA based ERP calculations? Are there others over there who compared  ICA based computations with and without including the EOG electrodes?


Baris


On Fri, Sep 3, 2010 at 7:42 AM, Evans, Ian D. <ievans at une.edu.au> wrote:

  Greetings all! I'm trying to isolate eye movement artifact with ICA in an
  attempt to make EOG correction more accurate and hopefully remove less of
  the genuine underlying data that regression methods might otherwise remove,
  however I am having trouble tracking down a satisfactory scalp map that
  includes co-ordinates (Cartesian, 2D polar, 3D cylindrical) for the EOG
  electrodes on the face. Neither Neuroscan nor any setup or .loc files in
  EEGLab seem to provide these details, and all attempts to isolate eye
  movement in ICA without the location of the eye movement electrodes have
  been met with mocking laughter.

  The recordings have already been made on a 64-channel cap with six EOG
  electrodes (one above and below each eye, another beside each eye for
  horizontal eye movement) using the traditional 10-20 system. All
  co-ordinates for the scalp are already at hand, just the facial electrodes
  are needed. If anyone can make such a coordinate map available it would be
  most helpful, thank you kindly!

  Ian Evans
  Cognitive and Affective Neuroscience
  University of New England.

  _______________________________________________
  Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
  To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
  For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu




-- 
SB Demiral, PhD.
Department of Psychology 
7 George Square
The University of Edinburgh
Edinburgh, EH8 9JZ
UK
Phone: +44 (0131) 6503063
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100906/1041473e/attachment.html>

From ievans at une.edu.au  Mon Sep  6 23:02:27 2010
From: ievans at une.edu.au (Evans, Ian D.)
Date: Tue, 7 Sep 2010 16:02:27 +1000
Subject: [Eeglablist] Eye movement artifact in ICA - missing coordinates
	for EOG electrodes
In-Reply-To: <3AA7D2B6-CAA9-45E9-9580-0E07AE40A9FE@aol.com>
References: <0DDFFD83A96343F7A518B024FF3F9E6E@TerminaHB>
	<3AA7D2B6-CAA9-45E9-9580-0E07AE40A9FE@aol.com>
Message-ID: <97423D25462C4DD8B26C0480F38460D6@TerminaHB>

I believe so - that is what I'm currently testing for. The current methods 
of removing eye movement artifacts involve regression analysis techniques 
which do filter out most of the artifactual potentials, leaving a largely 
pure EEG recording. It it my view (for now, at least) that ICA may be more 
accurate at detecting & thus filtering out EOG artifacts, leaving us with a 
much cleaner recording.

By measuring muscular activity around the eyes using the EOG electrodes 
real-time eye movement can be measured with a modicum of accuracy, at least 
enough accuracy to remove most artifacts. The challenge is measuring the 
effects on other electrodes, and this effect varies from person to person, 
and indeed recording to recording. This is why I start every EEG recording 
with a four-minute calibration task to measure & isolate types of eye 
movement (up, down, left, right, blink) so they can be corrected later.

But ultimately yes, with the right technique a simple subtraction of  EEG- 
associated EOG potentials from a recording is possible, and indeed 
encouraged. Current methods largely work well, but we can always do better.

Ian Evans
Cognitive & Affective Neuroscience
University of New England

--------------------------------------------------
From: "Bill Torch" <eyecomm at aol.com>
Sent: Tuesday, September 07, 2010 1:10 PM
To: "Evans, Ian D." <ievans at une.edu.au>
Subject: Re: [Eeglablist] Eye movement artifact in ICA - missing coordinates 
for EOG electrodes

> Ian, in response to your question, i have a question? If you had 
> quantitative EEG- associated EOG potentials that were directly tied to 
> actual real time eye movement coordinates, would you be able to subtract 
> these real time "artifactual" electrical potentials from the EEG
> get "pure" EEG?
>
> Dr. Bill Torch
> Medical Director, Neurodevelopmental & Neurodiagnostic  Ctr
>
>
> Sent from my iPhone
>
> On Sep 3, 2010, at 1:42 AM, "Evans, Ian D." <ievans at une.edu.au> wrote:
>
>> Greetings all! I'm trying to isolate eye movement artifact with ICA in an
>> attempt to make EOG correction more accurate and hopefully remove less of
>> the genuine underlying data that regression methods might otherwise 
>> remove,
>> however I am having trouble tracking down a satisfactory scalp map that
>> includes co-ordinates (Cartesian, 2D polar, 3D cylindrical) for the EOG
>> electrodes on the face. Neither Neuroscan nor any setup or .loc files in
>> EEGLab seem to provide these details, and all attempts to isolate eye
>> movement in ICA without the location of the eye movement electrodes have
>> been met with mocking laughter.
>>
>> The recordings have already been made on a 64-channel cap with six EOG
>> electrodes (one above and below each eye, another beside each eye for
>> horizontal eye movement) using the traditional 10-20 system. All
>> co-ordinates for the scalp are already at hand, just the facial 
>> electrodes
>> are needed. If anyone can make such a coordinate map available it would 
>> be
>> most helpful, thank you kindly!
>>
>> Ian Evans
>> Cognitive and Affective Neuroscience
>> University of New England.
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to 
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to 
>> eeglablist-request at sccn.ucsd.edu 



From Febo.Cincotti at uniroma1.it  Wed Sep  8 12:03:11 2010
From: Febo.Cincotti at uniroma1.it (Febo.Cincotti at uniroma1.it)
Date: Wed, 8 Sep 2010 21:03:11 +0200
Subject: [Eeglablist] 2nd TOBI Workshop - Rome, 2-3 December 2010
Message-ID: <OFE47E3744.2EC97AA6-ONC1257798.0068A989@Uniroma1.it>



TOBI Workshop II:

Translational issues in BCI development: user needs, ethics, and technology transfer

Fondazione Santa Lucia, Rome, Italy
Dec. 2-3, 2010

http://www.tobi-project.org/TOBI-workshop-2
--------------------------------

= Announcement =

The TOBI Project (Tools for Brain Interaction, http://www.tobi-project.org) is organizing its second workshop, which follows the one
held in Graz on February 2010.

The goal of the 2nd TOBI workshop is to draw the current and future scenarios involving themes of utmost relevance to fill the gap
between the promises of the neural engineering achievements and the clinical application reality in terms of BCIs as a daily use
assisted device and as add-on intervention in the rehabilitation protocols:

i.   user centered research and design
ii.  neuroethics
iii. technology transfer

The scientific program will consist of keynote talks, oral presentations, poster presentations, and round table.

A satellite session on clustering of EU-funded projects will take place (open to participants to EU-funded projects).

Partial list of speakers:

- Richard Frackowiak, Centre Hospitalier Universitaire Vaudois, Switzerland.
- Andrea K?bler, Julius-Maximilians Universit?t W?rzburg, Germany
- Donatella Mattia, Fondazione Santa Lucia, Rome, Italy
- Jos? del R. Mill?n, Ecole Polytechnique F?d?rale de Lausanne, Switzerland
- Klaus-R. M?ller, Technische Universit?t Berlin, Germany
- Guglielmo Tamburrini, Federico II University of Naples, Italy
- Paul Timmers, Head of Unit for ICT for Inclusion in the European Commission (to be confirmed).


= Call for papers =

Participants are invited to submit a 2-pages paper, which will be peer reviewed. Template and instruction for submission can be
found on the workshop's web site.

Papers can be accepted either for an oral or a poster presentation.

Accepted papers will be published on a special issue on the International Journal of Bioelectromagnetism (ISSN 1456-7865).

Authors of selected papers will be invited to submit an extended version for a special issue to be published on a prominent journal
of the field (to be announced).

Deadline for paper submission: October 11, 2010


= Registration =

Participants are required to register through the conference menagement system (link available on the workshop's web site).
Registration includes lunches, coffee breaks, and the social dinner.

Registration fees are:

     70 Euro by November 1
    120 Euro by November 19
    170 Euro onsite


= Important dates =

Paper submission:      11 October  2010
Early registration:     1 November 2010
Late registration:     19 November 2010
Workshop:             2-3 December 2010


= Venue =

The workshop will be held at the Congress Center of Fondazione Santa Lucia IRCCS, via Ardeatina 306, in the south-eastern part of
Rome, close to the Appia Antica Park.

Hotel rooms have been pre-booked downtown Rome (in the vicinity of Piazza Repubblica, conveniently linked the main city attractions
by public transportation).

Complimentary buses will transfer the participants (at the beginning and at the end of the sessions) between the workshop venue and
Piazza Repubblica.

The social program will include a visit to the archaeological attractions of the city, followed by a dinner downtown.
--------------------------------

= Info =

Official web page:

   http://www.tobi-project.org/TOBI-workshop-2

Send info requests to:

   tobiworkshop_AT_hsantalucia.it

Find a more detailed PDF version of the announcement at:

   http://www.tobi-project.org/sites/default/files/public/Workshop/Announcement_2ndTOBIws.pdf
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100908/22b3b231/attachment.html>

From demiral.007 at googlemail.com  Sun Sep  5 03:08:02 2010
From: demiral.007 at googlemail.com (Baris Demiral)
Date: Sun, 5 Sep 2010 11:08:02 +0100
Subject: [Eeglablist] Eye movement artifact in ICA - missing coordinates
 for EOG electrodes
In-Reply-To: <0DDFFD83A96343F7A518B024FF3F9E6E@TerminaHB>
References: <0DDFFD83A96343F7A518B024FF3F9E6E@TerminaHB>
Message-ID: <AANLkTimzqCf4j5L09akFoOJwLequ1foRUSNf3qj_9f_t@mail.gmail.com>

Hi Ian,

I also made a similar query few months ago. I use 64 electrode 10-20 system
plus 4 EOGs. I asked whether there are specific labels for such EOG
electrodes in the standard-10-5-Cap385.sfp so that I can map these
electrodes on the fly. Are there labels for those, or can we decide and add
those labels and coordinates as a community together?

Actually, I find something more important: How much really does this
influence ICA related artifact detection and correction? Is there a huge
change when you include horizontal and vertical EOGs in the ICA based ERP
calculations? Are there others over there who compared  ICA based
computations with and without including the EOG electrodes?

Baris

On Fri, Sep 3, 2010 at 7:42 AM, Evans, Ian D. <ievans at une.edu.au> wrote:

> Greetings all! I'm trying to isolate eye movement artifact with ICA in an
> attempt to make EOG correction more accurate and hopefully remove less of
> the genuine underlying data that regression methods might otherwise remove,
> however I am having trouble tracking down a satisfactory scalp map that
> includes co-ordinates (Cartesian, 2D polar, 3D cylindrical) for the EOG
> electrodes on the face. Neither Neuroscan nor any setup or .loc files in
> EEGLab seem to provide these details, and all attempts to isolate eye
> movement in ICA without the location of the eye movement electrodes have
> been met with mocking laughter.
>
> The recordings have already been made on a 64-channel cap with six EOG
> electrodes (one above and below each eye, another beside each eye for
> horizontal eye movement) using the traditional 10-20 system. All
> co-ordinates for the scalp are already at hand, just the facial electrodes
> are needed. If anyone can make such a coordinate map available it would be
> most helpful, thank you kindly!
>
> Ian Evans
> Cognitive and Affective Neuroscience
> University of New England.
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
SB Demiral, PhD.
Department of Psychology
7 George Square
The University of Edinburgh
Edinburgh, EH8 9JZ
UK
Phone: +44 (0131) 6503063
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100905/08eab478/attachment.html>

From ansboss at yahoo.com  Tue Sep  7 13:06:55 2010
From: ansboss at yahoo.com (jim king)
Date: Tue, 7 Sep 2010 13:06:55 -0700 (PDT)
Subject: [Eeglablist] eeglablist Digest, Vol 71, Issue 2
In-Reply-To: <mailman.0.1283886008.30882.eeglablist@sccn.ucsd.edu>
Message-ID: <613989.31300.qm@web111724.mail.gq1.yahoo.com>

Question:? is the ".edf" file extension (I believe this stands for "European Data Format") the most common file format for eeg machines?

JIm King, INACS, Inc.


--- On Tue, 9/7/10, eeglablist-request at sccn.ucsd.edu <eeglablist-request at sccn.ucsd.edu> wrote:

From: eeglablist-request at sccn.ucsd.edu <eeglablist-request at sccn.ucsd.edu>
Subject: eeglablist Digest, Vol 71, Issue 2
To: eeglablist at sccn.ucsd.edu
Date: Tuesday, September 7, 2010, 2:00 PM

Send eeglablist mailing list submissions to
??? eeglablist at sccn.ucsd.edu

To subscribe or unsubscribe via the World Wide Web, visit
??? http://sccn.ucsd.edu/mailman/listinfo/eeglablist
or, via email, send a message with subject or body 'help' to
??? eeglablist-request at sccn.ucsd.edu

You can reach the person managing the list at
??? eeglablist-owner at sccn.ucsd.edu

When replying, please edit your Subject line so it is more specific
than "Re: Contents of eeglablist digest..."
Today's Topics:

???1. Which EEGLAB version should I use (Arnaud Delorme)
_______________________________________________
eeglablist mailing list eeglablist at sccn.ucsd.edu
Eeglablist page: http://www.sccn.ucsd.edu/eeglab/eeglabmail.html
To unsubscribe, send an empty email to eeglablist-unsub at sccn.ucsd.edu
To switch to non-digest mode, send an empty email to eeglablist-nodigest at sccn.ucsd.edu


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100907/3751c6fb/attachment.html>

From arno at ucsd.edu  Wed Sep  8 16:21:07 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Thu, 9 Sep 2010 07:21:07 +0800
Subject: [Eeglablist] How to run AMICA
In-Reply-To: <AANLkTin03sZLcJvs9dX1tHcbVb6yoPC291BM3xpeed4N@mail.gmail.com>
References: <AANLkTin03sZLcJvs9dX1tHcbVb6yoPC291BM3xpeed4N@mail.gmail.com>
Message-ID: <3DC8CAF6-D3FA-487C-98EA-339E5C613DC0@ucsd.edu>

Dear Peter,

we are working on a conplex plugin for AMICA (AMICA is a new promising ICA algorithm programmed by Jason Palmer). For running it directly in Matlab, you may simply use Jason's Matlab function

http://sccn.ucsd.edu/~jason/amica/amica.m

then run the small script below after loading a dataset into EEGLAB

[EEG.icaweights EEG.icawinv] = amica(EEG.data(:,:));
EEG.icasphere = eye(size(EEG.icaweights));
[ALLEEG EEG CURRENTSET] = eeg_store(ALLEEG, EEG, CURRENTSET);
eeglab redraw;

Arno

On Sep 7, 2010, at 4:25 PM, P?ter Solt?sz wrote:

> Dear eeglab members,
> 
> I'm trying to figure out how to run amica from eeglab. Is there a way,
> or is it a detached function? I found that there are some traces in
> the pop_runica function (v9), however I'm not able to get it running.
> 
> Thanks in advance...
> 
> P?ter
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu




From arno at ucsd.edu  Thu Sep  9 01:05:46 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Thu, 9 Sep 2010 16:05:46 +0800
Subject: [Eeglablist] GPUs and EEGLAB
Message-ID: <CEC2FFFC-AD99-4C17-BB34-92D19794EAA6@ucsd.edu>

Here is a short analysis of using GPUs for performing EEG processing.
GPUs are Graphical Processing Units that allow performing general purpose computation

http://sccn.ucsd.edu/wiki/GPU_and_EEGLAB

Arno


From schalk at wadsworth.org  Fri Sep 10 05:33:59 2010
From: schalk at wadsworth.org (Gerwin Schalk)
Date: Fri, 10 Sep 2010 08:33:59 -0400
Subject: [Eeglablist] 2nd Intl. Workshop on Advances in Electrocorticography
Message-ID: <931C1C7D-6792-4BE0-986B-8A3F7F38D6CC@wadsworth.org>

Dear colleague,

This is to announce the 2nd Intl. Workshop on Advances in Electrocorticography, which will be held prior to the upcoming annual meeting of the Society for Neuroscience in San Diego on Thu, Nov 11 - Fri, Nov. 12, 2010 at the Marriott Hotel and Marina, 333 West Harbor Drive, San Diego, California. 

The first day of this workshop focuses on current clinical use of electrocorticography and emerging clinical opportunities. The second day focuses on principles of ECoG signals and emerging research.

We hope that you will be able to join us in this exciting event with world-class speakers. The brochure with the complete program and registration information is attached.

Gerwin Schalk and Anthony Ritaccio



IMPORTANT NOTICE: This e-mail and any attachments may contain
confidential or sensitive information which is, or may be, legally
privileged or otherwise protected by law from further disclosure.  It
is intended only for the addressee.  If you received this in error or
from someone who was not authorized to send it to you, please do not
distribute, copy or use it or any attachments.  Please notify the
sender immediately by reply e-mail and delete this from your
system. Thank you for your cooperation.


-------------- next part --------------
A non-text attachment was scrubbed...
Name: Brain Mapping 2010 Brochure final.pdf
Type: application/pdf
Size: 958984 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100910/b002f202/attachment.pdf>

From smakeig at gmail.com  Wed Sep  8 19:26:52 2010
From: smakeig at gmail.com (Scott Makeig)
Date: Wed, 8 Sep 2010 19:26:52 -0700
Subject: [Eeglablist] Eye movement artifact in ICA - missing coordinates
 for EOG electrodes
In-Reply-To: <829FE0A620CC496B82A656C672051443@TerminaHB>
References: <0DDFFD83A96343F7A518B024FF3F9E6E@TerminaHB>
	<AANLkTimzqCf4j5L09akFoOJwLequ1foRUSNf3qj_9f_t@mail.gmail.com>
	<829FE0A620CC496B82A656C672051443@TerminaHB>
Message-ID: <AANLkTikutdw9EeQQUQd1A-6tqxpBSgHkuvt+cgZCUvyY@mail.gmail.com>

If the EOG electrodes are bipolar -- that is, not linked to the same
reference electrode as the rest of the channels -- then their values in the
IC map are 'floating' with respect to the other channels (i.e. have some DC
offset) and their values in IC maps should be ignored when submitting the IC
maps to source localization (either formal or 'by eye') . But since activity
in the EOG channel should be linearly related to the other channels (if well
recorded), it may be included in ICA decomposition; the relevant source
activities will also contribute to the EOG channel.

If important or useful, it should be possible to estimate the true values
for the EOG electrodes in the IC maps by making a source model for activity
projected to the other common-reference channels, projecting it to the two
EOG electrode locations, and then minimizing the difference of their
projected difference from the decomposed IC EOG-channel value (i.e.,
removing the floating EOG-channel offset to make the resulting IC map as
smooth as possible).

Scott Makeig

On Sun, Sep 5, 2010 at 7:15 AM, Evans, Ian D. <ievans at une.edu.au> wrote:

>  Shall endeavor to find out this week, if there's time amongst the
> residential schools. But from what I can discern the EOG electrodes do seem
> to make a difference - when I ran an ICA on the eye calibration task used to
> measure artifact without any EOG electrodes in the mix, around 40 of the 68
> components highlighted the F8 region as a strong source, and this in a task
> purely involving eye movement, no other processing involved other than
> possibly recognizing the shapes used to indicate where the subject should
> look. Tried running it again but included some estimated positions of
> EOG electrodes at the front, and there were still 40+ components all
> indicating activity at the frontal lobe, but now the activity was around the
> face instead of F8. If nothing else, the EOG information should assist ICA
> by isolating changes in multiple electrodes with correlate with the EOG data
> and rendering it completely independent from the underlying actual data.
>
> As for labels, well I couldn't find any consistent EOG labeling systems for
> MATLAB, so why not start gathering labels? Let me know what labels you use
> for the eye channels and I can build the .loc file - David Groppe (many
> thanks again!) sent me a set of coordinates for the ocular electrodes, but
> there is no problem with having multiple names for the same set of
> coordinates - EEGLab detects what we're using anyway. Worth a shot.
>
>
> Ian Evans
> Cognitive and Affective Neuroscience
> University of New England.
> ievans at une.edu.au
>
>  *From:* Baris Demiral <demiral.007 at googlemail.com>
> *Sent:* Sunday, September 05, 2010 8:08 PM
> *To:* Evans, Ian D. <ievans at une.edu.au>
> *Cc:* eeglablist at sccn.ucsd.edu
> *Subject:* Re: [Eeglablist] Eye movement artifact in ICA - missing
> coordinates for EOG electrodes
>
> Hi Ian,
>
> I also made a similar query few months ago. I use 64 electrode 10-20 system
> plus 4 EOGs. I asked whether there are specific labels for such EOG
> electrodes in the standard-10-5-Cap385.sfp so that I can map these
> electrodes on the fly. Are there labels for those, or can we decide and add
> those labels and coordinates as a community together?
>
> Actually, I find something more important: How much really does this
> influence ICA related artifact detection and correction? Is there a huge
> change when you include horizontal and vertical EOGs in the ICA based ERP
> calculations? Are there others over there who compared  ICA based
> computations with and without including the EOG electrodes?
>
> Baris
>
> On Fri, Sep 3, 2010 at 7:42 AM, Evans, Ian D. <ievans at une.edu.au> wrote:
>
>> Greetings all! I'm trying to isolate eye movement artifact with ICA in an
>> attempt to make EOG correction more accurate and hopefully remove less of
>> the genuine underlying data that regression methods might otherwise
>> remove,
>> however I am having trouble tracking down a satisfactory scalp map that
>> includes co-ordinates (Cartesian, 2D polar, 3D cylindrical) for the EOG
>> electrodes on the face. Neither Neuroscan nor any setup or .loc files in
>> EEGLab seem to provide these details, and all attempts to isolate eye
>> movement in ICA without the location of the eye movement electrodes have
>> been met with mocking laughter.
>>
>> The recordings have already been made on a 64-channel cap with six EOG
>> electrodes (one above and below each eye, another beside each eye for
>> horizontal eye movement) using the traditional 10-20 system. All
>> co-ordinates for the scalp are already at hand, just the facial electrodes
>> are needed. If anyone can make such a coordinate map available it would be
>> most helpful, thank you kindly!
>>
>> Ian Evans
>> Cognitive and Affective Neuroscience
>> University of New England.
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>>
>
>
>
> --
> SB Demiral, PhD.
> Department of Psychology
> 7 George Square
> The University of Edinburgh
> Edinburgh, EH8 9JZ
> UK
> Phone: +44 (0131) 6503063
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Scott Makeig, Research Scientist and Director, Swartz Center for
Computational Neuroscience, Institute for Neural Computation, University of
California San Diego, La Jolla CA 92093-0559,
http://sccn.ucsd.edu/~scott<http://sccn.ucsd.edu/%7Escott>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100908/a2612aa0/attachment.html>

From demiral.007 at googlemail.com  Thu Sep  9 10:58:33 2010
From: demiral.007 at googlemail.com (Baris Demiral)
Date: Thu, 9 Sep 2010 18:58:33 +0100
Subject: [Eeglablist] Eye movement artifact in ICA - missing coordinates
 for EOG electrodes
In-Reply-To: <829FE0A620CC496B82A656C672051443@TerminaHB>
References: <0DDFFD83A96343F7A518B024FF3F9E6E@TerminaHB>
	<AANLkTimzqCf4j5L09akFoOJwLequ1foRUSNf3qj_9f_t@mail.gmail.com>
	<829FE0A620CC496B82A656C672051443@TerminaHB>
Message-ID: <AANLkTi=aYqwZP5DQHijneRVm8MQeBF80AU4E2s8p9F9F@mail.gmail.com>

Well, I acknowledge that the labels are arbitrary at this point, but I
propose to the eeglab community that we use the EOG labels as follows:

Horizontal:
H1:      EOG on the Left Temple (H indicating horizontal)
H2:      EOG on the Right Temple

Vertical:
VA1:      EOG Above the Left Eye (V indicating Vertical)
VB1:      EOG Below the Left Eye

VA2 and VB2 respectively on the right eye.

Mastoids (already agreed):
M1: Mastoid Left
M2: Mastoid Right


At least if I can have the coordinates of these positions (which some of you
already agreed upon) it will be fabulous. And I think I would only include
H1, H2, and VA1 and VA2  EOGs in my IC computations, because the VB1 and VB2
are generally prone to the local mouth, cheek and lip movements and may not
be contaminating the EEGs that much.

Regarding to your findings, Ian, I think adding EOG channels will lead to
the fallowing: Increase the number electrodes --> increase the number of of
ICs --> which will naturally increase the number of artefact containing ICs
--> this will bias some of the the artefactual ICs to shift towards the face
or anterior regions. It looks like adding up EOGs in ICA based
corrections/rejections is better to diagnose and differentiate very anterior
components such as ELAN or LAN form the eye movement related artifects. For
instance if you see an IC on F7 it is possible that it is a left anterior
negativity but not an artefact. I recently used ADJUST, and I am wondering
whether anyone using or developing ADJUST can help us to understand how
ADJUST will behave in this case. I am very much interested in artefact
correction.

Related to Scott's point: We use bipolar EOG electrodes in a BioSemi system
and when we import the file into eeglab we use one of the mastoid electrodes
as a reference. As far as I understand, bipolar electrodes referenced to the
same electrode as the other EEG electrodes will not cause any problem for
ICA based rejections in this case. Is this correct?

Thanks,

Baris



On Sun, Sep 5, 2010 at 3:15 PM, Evans, Ian D. <ievans at une.edu.au> wrote:

>  Shall endeavor to find out this week, if there's time amongst the
> residential schools. But from what I can discern the EOG electrodes do seem
> to make a difference - when I ran an ICA on the eye calibration task used to
> measure artifact without any EOG electrodes in the mix, around 40 of the 68
> components highlighted the F8 region as a strong source, and this in a task
> purely involving eye movement, no other processing involved other than
> possibly recognizing the shapes used to indicate where the subject should
> look. Tried running it again but included some estimated positions of
> EOG electrodes at the front, and there were still 40+ components all
> indicating activity at the frontal lobe, but now the activity was around the
> face instead of F8. If nothing else, the EOG information should assist ICA
> by isolating changes in multiple electrodes with correlate with the EOG data
> and rendering it completely independent from the underlying actual data.
>
> As for labels, well I couldn't find any consistent EOG labeling systems for
> MATLAB, so why not start gathering labels? Let me know what labels you use
> for the eye channels and I can build the .loc file - David Groppe (many
> thanks again!) sent me a set of coordinates for the ocular electrodes, but
> there is no problem with having multiple names for the same set of
> coordinates - EEGLab detects what we're using anyway. Worth a shot.
>
>
> Ian Evans
> Cognitive and Affective Neuroscience
> University of New England.
> ievans at une.edu.au
>
>  *From:* Baris Demiral <demiral.007 at googlemail.com>
> *Sent:* Sunday, September 05, 2010 8:08 PM
> *To:* Evans, Ian D. <ievans at une.edu.au>
> *Cc:* eeglablist at sccn.ucsd.edu
> *Subject:* Re: [Eeglablist] Eye movement artifact in ICA - missing
> coordinates for EOG electrodes
>
> Hi Ian,
>
> I also made a similar query few months ago. I use 64 electrode 10-20 system
> plus 4 EOGs. I asked whether there are specific labels for such EOG
> electrodes in the standard-10-5-Cap385.sfp so that I can map these
> electrodes on the fly. Are there labels for those, or can we decide and add
> those labels and coordinates as a community together?
>
> Actually, I find something more important: How much really does this
> influence ICA related artifact detection and correction? Is there a huge
> change when you include horizontal and vertical EOGs in the ICA based ERP
> calculations? Are there others over there who compared  ICA based
> computations with and without including the EOG electrodes?
>
> Baris
>
> On Fri, Sep 3, 2010 at 7:42 AM, Evans, Ian D. <ievans at une.edu.au> wrote:
>
>> Greetings all! I'm trying to isolate eye movement artifact with ICA in an
>> attempt to make EOG correction more accurate and hopefully remove less of
>> the genuine underlying data that regression methods might otherwise
>> remove,
>> however I am having trouble tracking down a satisfactory scalp map that
>> includes co-ordinates (Cartesian, 2D polar, 3D cylindrical) for the EOG
>> electrodes on the face. Neither Neuroscan nor any setup or .loc files in
>> EEGLab seem to provide these details, and all attempts to isolate eye
>> movement in ICA without the location of the eye movement electrodes have
>> been met with mocking laughter.
>>
>> The recordings have already been made on a 64-channel cap with six EOG
>> electrodes (one above and below each eye, another beside each eye for
>> horizontal eye movement) using the traditional 10-20 system. All
>> co-ordinates for the scalp are already at hand, just the facial electrodes
>> are needed. If anyone can make such a coordinate map available it would be
>> most helpful, thank you kindly!
>>
>> Ian Evans
>> Cognitive and Affective Neuroscience
>> University of New England.
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>>
>
>
>
> --
> SB Demiral, PhD.
> Department of Psychology
> 7 George Square
> The University of Edinburgh
> Edinburgh, EH8 9JZ
> UK
> Phone: +44 (0131) 6503063
>



-- 
SB Demiral, PhD.
Department of Psychology
7 George Square
The University of Edinburgh
Edinburgh, EH8 9JZ
UK
Phone: +44 (0131) 6503063
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100909/b7b95a9e/attachment.html>

From EJM9F at hscmail.mcc.virginia.edu  Mon Sep 13 13:26:35 2010
From: EJM9F at hscmail.mcc.virginia.edu (Modestino, Edward J *HS)
Date: Mon, 13 Sep 2010 16:26:35 -0400
Subject: [Eeglablist] ICA Debate
In-Reply-To: <926161.69597.qm@web62002.mail.re1.yahoo.com>
References: <926161.69597.qm@web62002.mail.re1.yahoo.com>
Message-ID: <CA7B140F2F20CA43AC79DF86B7357ABDD764290579@HSCSEMAIL32.hscs.virginia.edu>

We are having a heated debate here about EEGLAB and ICA.  When you use ICA to remove components in EEGLAB, does it remove these components only from the channels where the component is present or does it remove it from all channels?
Thanks
Ed

Edward Justin Modestino, Ph.D.
Postdoctoral Research Associate
Ray Westphal Neuroimaging Laboratory
Division of Perceptual Studies
Department of Psychiatry and Neurobehavioral Sciences
University of Virginia
Email: ejm9f at virginia.edu


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100913/3701d4b6/attachment.html>

From alenarto at ucla.edu  Mon Sep 13 10:32:38 2010
From: alenarto at ucla.edu (Agatha Lenartowicz)
Date: Mon, 13 Sep 2010 10:32:38 -0700
Subject: [Eeglablist] toolbar missing in GUI
Message-ID: <EC8D182A-84DE-41EE-883A-37548ED066EC@ucla.edu>

Hi all - 

As of recently the toolbar is missing in the EEGLAB GUI (so I can't load/analyze etc.). I expect this is a simple fix and some weird setting at my end. I usually script but in this case wanted to verify some things via the GUI.

This is occurring no matter which version of EEGLAB I start up (excluding the newest which I am not yet transitioned to). I have no idea if it's because of a Matlab update or conflicting file names etc... but if anyone has seen this and can help pin point me to the problem please let me know!

Thanks!
Agatha




From andreas.galka at googlemail.com  Tue Sep 14 14:23:13 2010
From: andreas.galka at googlemail.com (Andreas Galka)
Date: Tue, 14 Sep 2010 23:23:13 +0200
Subject: [Eeglablist] ICA Debate
In-Reply-To: <CA7B140F2F20CA43AC79DF86B7357ABDD764290579@HSCSEMAIL32.hscs.virginia.edu>
References: <926161.69597.qm@web62002.mail.re1.yahoo.com>
	<CA7B140F2F20CA43AC79DF86B7357ABDD764290579@HSCSEMAIL32.hscs.virginia.edu>
Message-ID: <AANLkTi=9FA7xbAG4n=+=otrpfuMthnxY+a65h-X3CJRp@mail.gmail.com>

> We are having a heated debate here about EEGLAB and ICA.? When you use ICA
> to remove components in EEGLAB, does it remove these components only from
> the channels where the component is present or does it remove it from all
> channels?

>From all channels, in my understanding. The reason is that the mixing
matrix of the
ICA data model will typically have no elements that are exactly zero.
But for channels
for which the components are virtually absent, the elements will be
very small (compared
to other elements), and the change of those channels under artefact
removal will be
hardly visible.

Greetings,
Andreas Galka



From pzeman at alumni.uvic.ca  Tue Sep 14 12:40:17 2010
From: pzeman at alumni.uvic.ca (Philip Michael Zeman)
Date: Tue, 14 Sep 2010 12:40:17 -0700
Subject: [Eeglablist] ICA Debate
In-Reply-To: <CA7B140F2F20CA43AC79DF86B7357ABDD764290579@HSCSEMAIL32.hscs.virginia.edu>
References: <926161.69597.qm@web62002.mail.re1.yahoo.com>
	<CA7B140F2F20CA43AC79DF86B7357ABDD764290579@HSCSEMAIL32.hscs.virginia.edu>
Message-ID: <53CD450FB6A94D8E98F3CE72801631D6@mine>

Hello Ed

 

I'll make my response in terms of the ICA (runica) algorithm in EEGLab and I will use a food example.

 

Using runICA (or a pie knife)

(1) disassemble the pie into pieces.  Each piece does not (except for special circumstances) equate to individual EEG channels. Sometimes it might look like only one or a couple channels are involved.  

(2) Removing any of the pieces removes the rejected piece from the pie entirely. Variance of varied amounts on each channel that relates to the component in question will be removed from all channels.

(3) when you re-assemble the remaining pieces into a new pie, you get a different pie. The new pie does not contain the pieces that were removed.

 

Hence, you can completely remove ocular artifacts from the EEG using this 3 step process.

 

For example:

 

Using runICA



(1) decompose you 32 channel EEG dataset into maximally statistically independent components using runICA. (The rank of the data should be 32.)

(2) reject components identified as ocular artifact

(3) re-combine the remaining components to create a new version of the 32 channel EEG dataset. (The rank of the data will now be less than 32.  If you removed 1 component, the rank will be 31.)

 

 The Special Circumstance above:



Artefacts such as "electrode pops" that are on single channels will have equivalent components that relate specifically to the channel with the 'pop' artefact.





 

Phil


=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Philip Michael Zeman B.Eng, Ph.D. (aka: International Data Analysis Guru Mercenary for Hire)
Applied Brain and Vision Sciences Inc.
Brain Function Analysis for Novel Paradigms and Serious Games
Analysis of Pharmaceutical Effects on Brain Function
http://www.abvsciences.com
Latest Brain Research Result:
http://www.spatialbrain.com
Email:   pzeman at alumni.uvic.ca
Phone: +1-250-589-4234
LinkedIn Profile: http://ca.linkedin.com/in/philipmichaelzeman
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=




  ----- Original Message ----- 
  From: Modestino, Edward J *HS 
  To: eeglablist at sccn.ucsd.edu 
  Sent: Monday, September 13, 2010 1:26 PM
  Subject: [Eeglablist] ICA Debate


  We are having a heated debate here about EEGLAB and ICA.  When you use ICA to remove components in EEGLAB, does it remove these components only from the channels where the component is present or does it remove it from all channels?

  Thanks

  Ed

   

  Edward Justin Modestino, Ph.D.

  Postdoctoral Research Associate

  Ray Westphal Neuroimaging Laboratory

  Division of Perceptual Studies

  Department of Psychiatry and Neurobehavioral Sciences

  University of Virginia

  Email: ejm9f at virginia.edu

   

   



------------------------------------------------------------------------------


  _______________________________________________
  Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
  To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
  For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100914/1dac2643/attachment.html>

From saim_rasheed at hotmail.com  Thu Sep 16 16:07:54 2010
From: saim_rasheed at hotmail.com (Saim Rasheed)
Date: Fri, 17 Sep 2010 05:07:54 +0600
Subject: [Eeglablist] ERP related error in STUDY
Message-ID: <SNT119-W48B309FFD7C2DE3269124CEA7A0@phx.gbl>


Hello,

 

I have created a STUDY set with three conditions and seven subjects. After precomputing channel measures, tried to plot ERPs, Spectra, ERSP and ITC. All is plotting well but ERP is giving the following error meesage:

***********************************************************

***********************************************************

Reading erp data...??? Index exceeds matrix dimensions.

Error in ==> std_readerp>std_readerpsub at 485
    if isfield(erpstruct, 'labels'), chanlab{k} = erpstruct.labels{k}; end;

Error in ==> std_readerp at 220
                    if strcmpi(dtype, 'erp') alldata{c, g} = std_readerpsub( ALLEEG,
                    setinds{c,g}(:), allinds{c,g}(:), opt.timerange)';

Error in ==> std_erpplot at 322
        [STUDY erpdata alltimes] = std_readerp(STUDY, ALLEEG, 'channels', opt.channels,
        'timerange', opt.timerange, ...

Error in ==> pop_chanplot at 323
                eval(a); STUDY.history =  sprintf('%s\n%s',  STUDY.history, a);
 
??? Error while evaluating uicontrol Callback

***********************************************************

***********************************************************

 

please help, thanks.

 

Saim
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100917/d1b601e3/attachment.html>

From simonshlomo.poil at cncr.vu.nl  Thu Sep 16 11:08:59 2010
From: simonshlomo.poil at cncr.vu.nl (Simon-Shlomo Poil)
Date: Thu, 16 Sep 2010 20:08:59 +0200
Subject: [Eeglablist] pop_iirfilt warning? Matrix is close to singular or
	badly scaled
Message-ID: <AANLkTik621djmd9B38FGLKHafOPBJk5SB1KFTr3NOHwM@mail.gmail.com>

Dear eeglab list,

I am using pop_iirfilt to highpass my eeg data at 0.5 Hz, and get the
following warning:

.Warning: Matrix is close to singular or badly scaled.
         Results may be inaccurate. RCOND = 6.130406e-017.
> In filtfilt at 81
  In iirfilt at 213
  In pop_iirfilt at 234

However, if I resample my data from 1000 Hz to 500 Hz the warning
disappears? At first, I thought this was due to some extreme values
being smoothed by the resampling; I, however, do not find any extreme
values in the data.

Does somebody have a clue why the iir filter does not work with a high
sampling frequency?


Best regards,
-- 
Simon-Shlomo Poil

Neuronal Oscillations and Cognition Group (NOC)
Department of Integrative Neurophysiology (INF)
Center for Neurogenomics and Cognitive Research (CNCR)
Neuroscience Campus Amsterdam
VU University Amsterdam
De Boelelaan 1085, Room B-435
1081 HV Amsterdam, The Netherlands

E-mail: simonshlomo.poil at cncr.vu.nl
Phone: +31 20 5989408
Webpage: http://www.poil.dk and http://www.cncr.nl


From guanw at rhpcs.mcmaster.ca  Thu Sep 16 07:51:38 2010
From: guanw at rhpcs.mcmaster.ca (Weiguang Guan)
Date: Thu, 16 Sep 2010 10:51:38 -0400 (EDT)
Subject: [Eeglablist] Do ICs have locations?
In-Reply-To: <alpine.LMD.2.00.1009161027340.19474@rhpcserv>
References: <alpine.LMD.2.00.1009161027340.19474@rhpcserv>
Message-ID: <alpine.LMD.2.00.1009161050420.19474@rhpcserv>

Hello,

I'm new to EEGLAB. I don't think the components after ICA have location 
information associated. If this is true, then how could EEGLAB plot ICs on 
a scalp?

Weiguang



From Ronald.Phlypo at ugent.be  Mon Sep 20 01:06:02 2010
From: Ronald.Phlypo at ugent.be (Ronald Phlypo)
Date: Mon, 20 Sep 2010 10:06:02 +0200
Subject: [Eeglablist] Do ICs have locations?
In-Reply-To: <alpine.LMD.2.00.1009161050420.19474@rhpcserv>
References: <alpine.LMD.2.00.1009161027340.19474@rhpcserv>
	<alpine.LMD.2.00.1009161050420.19474@rhpcserv>
Message-ID: <4C9715EA.207@ugent.be>

  Dear Weiguang,

The IC's are obtained through the decomposition of the observations y(t) 
as y(t) = A*x(t), where the x(t) are the estimates of the sources. The 
model could easily be rewritten using vector notation only as y(t) = Sum 
over i of a[i]*s[i], where the a[i] are the column vectors taken from A.

The a[i] thus clearly expose the spatial distribution of the s[i] over 
the scalp (the spatial distributions do not depend on the time and are 
supposed stationary in this model), up to a global scaling factor which 
includes a possible sign change. Moreover, taking a[i]^2 -- which stands 
for the quadratic of each of the elements in the column vector -- 
returns the power distribution of the i-th source up to a global scaling 
factor.

Hope this helps,

Ronald



Le 16/09/2010 16:51, Weiguang Guan a ?crit :
> Hello,
>
> I'm new to EEGLAB. I don't think the components after ICA have location
> information associated. If this is true, then how could EEGLAB plot ICs on
> a scalp?
>
> Weiguang
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu


From saim_rasheed at hotmail.com  Mon Sep 20 08:29:05 2010
From: saim_rasheed at hotmail.com (Saim Rasheed)
Date: Mon, 20 Sep 2010 21:29:05 +0600
Subject: [Eeglablist] Algorithms used for ERSP and ITC
Message-ID: <SNT119-W40F65DCF8F24EB856491A0EA7E0@phx.gbl>


 

Hello,

 

Can any one please let me know, what mathematical algorithms are used in EEGLAB to plot ERSP and ITC? (Any references)

 

Please respond. Thankyou

 

Saim


 
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100920/94f69289/attachment.html>

From arno at ucsd.edu  Mon Sep 20 19:37:14 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Mon, 20 Sep 2010 19:37:14 -0700
Subject: [Eeglablist] Algorithms used for ERSP and ITC
In-Reply-To: <SNT119-W40F65DCF8F24EB856491A0EA7E0@phx.gbl>
References: <SNT119-W40F65DCF8F24EB856491A0EA7E0@phx.gbl>
Message-ID: <3F937275-48E7-410F-AD24-C5F5E24FC3C7@ucsd.edu>

Dear Saim,

the reference for these algorithms as they are implemented in EEGLAB is

Delorme, A., Makeig, S. (2004) EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis. Journal of Neuroscience Methods, 134(1), 9-21. Author's PDF, Science direct.

Best regards,

Arno
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100920/6bf01e7a/attachment.html>

From guanw at rhpcs.mcmaster.ca  Mon Sep 20 13:28:31 2010
From: guanw at rhpcs.mcmaster.ca (Weiguang Guan)
Date: Mon, 20 Sep 2010 16:28:31 -0400 (EDT)
Subject: [Eeglablist] Do ICs have locations?
In-Reply-To: <4C9715EA.207@ugent.be>
References: <alpine.LMD.2.00.1009161027340.19474@rhpcserv>
	<alpine.LMD.2.00.1009161050420.19474@rhpcserv>
	<4C9715EA.207@ugent.be>
Message-ID: <alpine.LMD.2.00.1009201623250.30225@rhpcserv>

Hi Ronald,

Thank you for explaining ICA for me. So, each of ICA maps in 2D shows the 
percentage of contribution from each channel made to a particular IC.

Weiguang

On Mon, 20 Sep 2010, Ronald Phlypo wrote:

>  Dear Weiguang,
>
> The IC's are obtained through the decomposition of the observations y(t)
> as y(t) = A*x(t), where the x(t) are the estimates of the sources. The
> model could easily be rewritten using vector notation only as y(t) = Sum
> over i of a[i]*s[i], where the a[i] are the column vectors taken from A.
>
> The a[i] thus clearly expose the spatial distribution of the s[i] over
> the scalp (the spatial distributions do not depend on the time and are
> supposed stationary in this model), up to a global scaling factor which
> includes a possible sign change. Moreover, taking a[i]^2 -- which stands
> for the quadratic of each of the elements in the column vector --
> returns the power distribution of the i-th source up to a global scaling
> factor.
>
> Hope this helps,
>
> Ronald
>
>
>
> Le 16/09/2010 16:51, Weiguang Guan a ?crit :
>> Hello,
>>
>> I'm new to EEGLAB. I don't think the components after ICA have location
>> information associated. If this is true, then how could EEGLAB plot ICs on
>> a scalp?
>>
>> Weiguang
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
>

From kedar_d2 at yahoo.co.in  Mon Sep 20 23:10:41 2010
From: kedar_d2 at yahoo.co.in (Kedarnath Senapati)
Date: Tue, 21 Sep 2010 11:40:41 +0530 (IST)
Subject: [Eeglablist] what is the realistic model for EEG signal?
Message-ID: <743557.40083.qm@web8318.mail.in.yahoo.com>

Dear EEGLAB-Friends
?
Can any one tell me, which one is the realistic model for?EEG signal??The linear instantaneous mixture model? Or the convolutional mixture model? And which model is?assumed in EEGLAB to decompose the EEG signals into its Independent Components??Also which one is a good referenced research paper to understand the associated algorithm?
?
Any help is highly appreciated. Thank you so much in advance.
?
Sincerely Yours
?
Kedarnath Senapati

Kedarnath Senapati
Institute of Mathematics & Applications
Andharua
Bhubaneswar-751003
INDIA.

91 9749872091(cell).

Happy moments, praise God. 
Difficult moments, seek God. 
Quiet moments, worship God. 
Painful moments, trust God. 
Every moment, thank God.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100921/ff3dc4ce/attachment.html>

From rxy2009 at yahoo.com  Mon Sep 20 15:25:13 2010
From: rxy2009 at yahoo.com (Rajeev Yadav)
Date: Mon, 20 Sep 2010 15:25:13 -0700 (PDT)
Subject: [Eeglablist] Min. data length for ICA analysis?
Message-ID: <748865.13611.qm@web35903.mail.mud.yahoo.com>

Hi,


I have a basic question for ICA analysis. 
What should be the minimum length for data to perform ICA analysis? 

 Regards
Rajeev Yadav
~~~~~~~~~~~~~~~~~~~~~~~
"As many branches, so many trees."
~~~~~~~~~~~~~~~~~~~~~~~
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100920/b641ef3c/attachment.html>

From dgroppe at cogsci.ucsd.edu  Tue Sep 21 21:59:28 2010
From: dgroppe at cogsci.ucsd.edu (David Groppe)
Date: Tue, 21 Sep 2010 21:59:28 -0700
Subject: [Eeglablist] Min. data length for ICA analysis?
In-Reply-To: <748865.13611.qm@web35903.mail.mud.yahoo.com>
References: <748865.13611.qm@web35903.mail.mud.yahoo.com>
Message-ID: <AANLkTinG0QjgG2pCeTwPyjO+_ddduR0R_RfBtyN99u1V@mail.gmail.com>

Hi Rajeev,
      In the following article:

Onton,J., Wester?eld, M., Townsend, J.,Makeig, S. (2006) Imaging human
EEG dynamics using independent component analysis. Neurosci. Biobehav.
Rev. 30(6), 808?822.

Onton et al. recommend a minimum of 20 time points per channel^2 while
noting that more data than that generally helps.  By evaluating the
reliability of ICA, we've found that 20 time points per channel^2 is
indeed sufficient to produce a usable proportion of reliable ICs, but
that you can do much better with more data.  See Figure 7 of:

Groppe, D.M., Makeig, S., &  Kutas, M. (2009) Identifying reliable
independent components via split-half comparisons. NeuroImage, 45
pp.1199-1211.
(http://www.cogsci.ucsd.edu/~dgroppe/PUBLICATIONS/Groppe2009.pdf)

hope this helps,
     -David

On Mon, Sep 20, 2010 at 3:25 PM, Rajeev Yadav <rxy2009 at yahoo.com> wrote:
> Hi,
>
>
> I have a basic question for ICA analysis.
> What should be the minimum length for data to perform ICA analysis?
>
> Regards
> Rajeev Yadav
> ~~~~~~~~~~~~~~~~~~~~~~~
> "As many branches, so many trees."
> ~~~~~~~~~~~~~~~~~~~~~~~
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
David Groppe, Ph.D.
dgroppe at cogsci.ucsd.edu
http://www.cogsci.ucsd.edu/~dgroppe/



From dgroppe at cogsci.ucsd.edu  Tue Sep 21 22:34:08 2010
From: dgroppe at cogsci.ucsd.edu (David Groppe)
Date: Tue, 21 Sep 2010 22:34:08 -0700
Subject: [Eeglablist] ICA as artefact correction method - dilemma
In-Reply-To: <47294c3eeff325dee@wm-srv.ulb.ac.be>
References: <47294c3eeff325dee@wm-srv.ulb.ac.be>
Message-ID: <AANLkTin+gzXQytxPMxW9zcB5zm1r_CiCYg4y5yqsa=ZT@mail.gmail.com>

On Thu, Jul 15, 2010 at 4:24 AM, Kris Baetens <Kris.Baetens at vub.ac.be> wrote:
> Dear all,
>
> I have been messing for some time with filter/ICA issues. I would be very grateful if anybody could shed some light on the matter.
>
> In a number of experiments, I have used sentences as stimulus material. We collected ERP responses to the final word of the last of a series of sentences and are interested in N400 and P300-like effects. Participants were instructed by means of an icon to ?do their blinking? as much as possible during short pauses of a few seconds that followed about 2 seconds after each final sentence. We have used a DC amplifier with an average recording reference.
>
> Regardless of whether I use FIR or IIR filters, the higher my high-pass filter cut-off, the more drift I get in the participant ERP averages following the final sentences. That is, if I use a 6th order two-way Butterworth filter with half-amplitude cut-off of 0.01 Hz, for example, there is no particular drift in the ERP following the critical end sentences, whereas a similar filter with a 0.3Hz cut-off results in drifts that go from 0 to 30 ?V over the course of a one second in participant averages.
> These drifts are outspoken in the vertical EOG channel but in the frontal channels as well. Considering the fact that many trials are followed by eye blinks (+/-2 or three seconds after the time lock), it seems obvious that the drift is a result of the eye blinks and the filtering applied to them.
> However, the ?normal? drift left in the trials (taking all channels into account) is much higher when I use a 0.01Hz high pass than when I use a 03Hz high pass, as one would expect.
>
> I'm wrestling a bit with the following dilemma:
> -I have seen that when I use an adequate high-pass filter (0.5 or 1Hz) I get a very nice decomposition of my data, enabling the precise removal of eye blink activity, jaw muscle activation etcetera.
> However, when using such filters, I get enormous drifts in the frontal channels, as explained above (and somehow, this doesn?t ?attract too much ?attention? of the ICA algorithm, still enabling a proper decomposition). Also, I am concerned that using such filters in classical ERP research might cause some problems (cf. Prof. Luck?s book), especially when the ERP components of interest are rather big slow ones like the N400 and P300.
> -On the other hand, when using a high-pass filter in the range of 0.01 ? 0.1Hz (as is recommended by many), the ICA algorithm fails to decompose the data well. I can still get rid of some substantial EOG activity, but no real proper correction.
>
> My questions are the following:
> -Given the fact that the ICA algorithm works well only when one uses high pass filters in the range of 0.5-1Hz, and that using such filters is most often advised against by people working in classical ERP research, is ICA really utilizable in classical ERP-grand-average-style research as a method of eye blink correction?

Hi Kris,
  I've found that with EEG data broken up into 1-2 second epochs, ICA
can do a decent job removing EEG artifacts without high-pass filtering
(if you have sufficient data).  ICA does appear to perform better
though when you dampen low frequencies.  A simple way to do this is to
epoch your data and then remove the mean of each epoch.  We found that
doing this tends to massively improve the reliability of ICA's
results.

See Figure 8 of:
Groppe, D.M., Makeig, S., & ?Kutas, M. (2009) Identifying reliable
independent components via split-half comparisons. NeuroImage, 45
pp.1199-1211. (http://www.cogsci.ucsd.edu/~dgroppe/PUBLICATIONS/Groppe2009.pdf)

> -Is it generally a bad idea to instruct participants to do their blinking at a fixed moment that starts a few seconds after your time-lock stimulus?
> -What sort of distortions or invalid conclusions could possibly arise from using high-threshold high pass filtering (i.e., 0.5Hz 6th order Butterworth) when one applies it to all conditions, on a grand average ERP-level?

Luck nicely explains this in the book you mentioned.  The problem is
that high pass filters can induce oscillations that push effects far
forward or backward in time and can cause them to flip polarities.  If
you follow my recommendation and simply remove the mean of each epoch,
you won't be inducing any oscillatory activity, so general waveform
shape should be relatively preserved.
       hope this helps,
           -David

> -What sort of high-pass filter would you advise in general for DC recordings?
>
> Many thanks in any case,
>
> Kris Baetens
> Ph.D. fellow of the Research Foundation - Flanders (FWO)
> Dept. Experimental and Applied Psychology
> Faculty of Psychology and Educational Sciences
> Vrije Universiteit Brussel
> Pleinlaan 2, 1050 Elsene
> +32 2 629 23 31
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
>



-- 
David Groppe, Ph.D.
dgroppe at cogsci.ucsd.edu
http://www.cogsci.ucsd.edu/~dgroppe/



From balazs at cogpsyphy.hu  Tue Sep 21 10:50:52 2010
From: balazs at cogpsyphy.hu (balazs)
Date: Tue, 21 Sep 2010 19:50:52 +0200
Subject: [Eeglablist] Missing MEX file in EEProbe plugin 64 bit
Message-ID: <4C98F07C.3070209@cogpsyphy.hu>

Dear all,
It seems that EEProbe - the plugin that imports files recorded with ANT 
system - is not working under 64 bit OS (neither Win-7 nor Ubuntu). It 
screams for a missing MEX file. Is there anybody out there having this 
MEX or being able to compile it? Or shall I ask ANT?
Best,
Laszlo

-- 
      Laszlo Balazs, Ph.D.          /      dr. Bal?zs L?szl?
 Institute for Psychology HAS       /   MTA Pszichol?giai Kutat?int?zet
                 P O B 398, Budapest, Hungary, H-1394
                Tel:+36(1)354-2410 | Fax:+36(1)354-2416
                      http://www.cogpsyphy.hu/balazs



__________ Information from ESET NOD32 Antivirus, version of virus signature database 5467 (20100921) __________

The message was checked by ESET NOD32 Antivirus.

http://www.eset.com




From whelanrob at gmail.com  Wed Sep 22 11:23:53 2010
From: whelanrob at gmail.com (Robert Whelan)
Date: Wed, 22 Sep 2010 19:23:53 +0100
Subject: [Eeglablist] FASTER
Message-ID: <4C9A49B9.3070204@tcd.ie>



Dear EEGLAB list,

FASTER -- Fully Automated Statistical Thresholding for EEG artifact Rejection -- is a software suite that works in tandem with EEGLAB. FASTER is a fully automated, unsupervised method for processing of high density EEG data.//FASTER/  /has been peer-reviewed, it is free and the software is open source. At the moment there is a GUI that can be called from the Matlab command line and we are currently working on implementing FASTER as a plugin for EEGLAB.

The reference is: Nolan, H., Whelan, R.,&  Reilly, R.B. (2010). FASTER: Fully Automated Statistical Thresholding for EEG artifact Rejection./Journal of Neuroscience Methods, 192/, 152-162.

For more information, and to download the program, please visit:http://www.mee.tcd.ie/~neuraleng/Research/Faster  <http://www.mee.tcd.ie/%7Eneuraleng/Research/Faster>

Please email me robert.whelan at tcd.ie if you have any questions.

Robert, Hugh&  Richard

-- 
Robert Whelan, PhD
Senior Research Scientist

Trinity Centre for Bioengineering
Trinity College Dublin

Department of Neurology
St. Vincent's University Hospital
Elm Park, Dublin 4

webpage: http://www.mee.tcd.ie/~neuraleng/People/Robert

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100922/68ecfd44/attachment.html>

From arno at ucsd.edu  Wed Sep 22 21:28:48 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Wed, 22 Sep 2010 21:28:48 -0700
Subject: [Eeglablist] Missing MEX file in EEProbe plugin 64 bit
In-Reply-To: <4C98F07C.3070209@cogpsyphy.hu>
References: <4C98F07C.3070209@cogpsyphy.hu>
Message-ID: <02DFD724-8960-45F3-AAEB-BD944A0D72BA@ucsd.edu>

Dear Laszlo,

Yes, this is a problem. The best option is to contact ANT but last time I heard, they had not compiled the MEX file for 64-bit machines. Otherwise, open it under 32-bit and save it as an EEGLAB file. Then you will be able to process the EEGLAB file under your 64-bit machine.

Best,

Arno

On Sep 21, 2010, at 10:50 AM, balazs wrote:

> Dear all,
> It seems that EEProbe - the plugin that imports files recorded with ANT 
> system - is not working under 64 bit OS (neither Win-7 nor Ubuntu). It 
> screams for a missing MEX file. Is there anybody out there having this 
> MEX or being able to compile it? Or shall I ask ANT?
> Best,
> Laszlo
> 
> -- 
>      Laszlo Balazs, Ph.D.          /      dr. Bal?zs L?szl?
> Institute for Psychology HAS       /   MTA Pszichol?giai Kutat?int?zet
>                 P O B 398, Budapest, Hungary, H-1394
>                Tel:+36(1)354-2410 | Fax:+36(1)354-2416
>                      http://www.cogpsyphy.hu/balazs
> 
> 
> 
> __________ Information from ESET NOD32 Antivirus, version of virus signature database 5467 (20100921) __________
> 
> The message was checked by ESET NOD32 Antivirus.
> 
> http://www.eset.com
> 
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu




From jordicostafa at gmail.com  Wed Sep 22 17:43:00 2010
From: jordicostafa at gmail.com (Jordi Costa Faidella)
Date: Wed, 22 Sep 2010 20:43:00 -0400
Subject: [Eeglablist] Using ICA with interpolated channels
Message-ID: <326D1E69-8542-4C65-A2D8-6023B058DF17@gmail.com>

Dear EEGlab users,

this question arose when reading the new FASTER method, but I think is of general importance for all of us. Is it correct to perform an ICA on a dataset in which some of the channels have been interpolated?

thanks,

Jordi


From pzeman at alumni.uvic.ca  Wed Sep 22 12:18:38 2010
From: pzeman at alumni.uvic.ca (Philip Michael Zeman)
Date: Wed, 22 Sep 2010 12:18:38 -0700
Subject: [Eeglablist] FASTER
In-Reply-To: <4C9A49B9.3070204@tcd.ie>
References: <4C9A49B9.3070204@tcd.ie>
Message-ID: <41CE4CD789B54B8897668B9E4ED4A063@mine>

Robert

Fantastic algorithm and Paper.

Thanks for posting this.


Phil


=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Philip Michael Zeman B.Eng, Ph.D.
Applied Brain and Vision Sciences Inc.
Brain Function Analysis for Novel Paradigms and Serious Games
Analysis of Pharmaceutical Effects on Brain Function
http://www.abvsciences.com
Latest Brain Research Result:
http://www.spatialbrain.com
Email:   pzeman at alumni.uvic.ca
Phone: +1-250-589-4234
LinkedIn Profile: http://ca.linkedin.com/in/philipmichaelzeman
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
  ----- Original Message ----- 
  From: Robert Whelan 
  To: eeglablist at sccn.ucsd.edu 
  Sent: Wednesday, September 22, 2010 11:23 AM
  Subject: [Eeglablist] FASTER


Dear EEGLAB list,

FASTER -- Fully Automated Statistical Thresholding for EEG artifact Rejection -- is a software suite that works in tandem with EEGLAB. FASTER is a fully automated, unsupervised method for processing of high density EEG data. FASTER has been peer-reviewed, it is free and the software is open source. At the moment there is a GUI that can be called from the Matlab command line and we are currently working on implementing FASTER as a plugin for EEGLAB. 

The reference is: Nolan, H., Whelan, R., & Reilly, R.B. (2010). FASTER: Fully Automated Statistical Thresholding for EEG artifact Rejection. Journal of Neuroscience Methods, 192, 152-162.

For more information, and to download the program, please visit: http://www.mee.tcd.ie/~neuraleng/Research/Faster

Please email me robert.whelan at tcd.ie if you have any questions. 

Robert, Hugh & Richard-- 
Robert Whelan, PhD
Senior Research Scientist

Trinity Centre for Bioengineering
Trinity College Dublin

Department of Neurology
St. Vincent's University Hospital
Elm Park, Dublin 4

webpage: http://www.mee.tcd.ie/~neuraleng/People/Robert

------------------------------------------------------------------------------


  _______________________________________________
  Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
  To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
  For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100922/ffa91ffa/attachment.html>

From yoh at psychology.rutgers.edu  Wed Sep 22 16:43:39 2010
From: yoh at psychology.rutgers.edu (Yaroslav Halchenko)
Date: Wed, 22 Sep 2010 19:43:39 -0400
Subject: [Eeglablist] EEGLAB test suite
In-Reply-To: <20100820184338.GM12007@onerussian.com>
References: <20100820024330.GS18647@onerussian.com>
	<389497E3-8DAC-4986-8856-45223A0DFABA@ucsd.edu>
	<20100820184338.GM12007@onerussian.com>
Message-ID: <20100922234339.GV12007@onerussian.com>

On Fri, 20 Aug 2010, Yaroslav Halchenko wrote:
> > crash - in some cases). If someone is interested in helping to maintain
> > or develop further EEGLAB test cases, please let us know. Unlike EEGLAB,
> > test cases are not under currently under SVN but we could easily create
> > a repository.
> That would be terrific and utterly useful. In my opinion, scientific
> >...<
> so, why not just include this tests battery within eeglab itself?  I am
> ok with separate repository, but I just wonder if it wouldn't be of
> greater value and convenience to be within EEGLAB?

Since it might be that I was not clear enough -- yes, I am interested to
see the test cases ;)

-- 
Yaroslav O. Halchenko
Postdoctoral Fellow,   Department of Psychological and Brain Sciences
Dartmouth College, 419 Moore Hall, Hinman Box 6207, Hanover, NH 03755
Phone: +1 (603) 646-9834                       Fax: +1 (603) 646-1419
WWW:   http://www.linkedin.com/in/yarik        


From whelanrob at gmail.com  Tue Sep 28 03:04:12 2010
From: whelanrob at gmail.com (Robert Whelan)
Date: Tue, 28 Sep 2010 11:04:12 +0100
Subject: [Eeglablist] Re. Using ICA with interpolated channels
Message-ID: <AANLkTim0wGSCd44phn_WNXqJfNgo0LyOn5qYoboqOA1W@mail.gmail.com>

Jordi Costa Faidella wrote "Is it correct to perform an ICA on a dataset in
which some of the channels have been interpolated?"

This is an interesting question and we considered both orders (each order
has some advantages and disadvantages) for the FASTER method. Ultimately, we
decided to run interpolation first followed ICA. Here was our rationale:

As the EEGLAB manual recommends ? ?ICA works best when given a large amount
of basically similar and mostly clean data.? (see p.59). Therefore, an ICA
on a dataset in which some channels are noisy (perhaps with a lot of
non-stereotypic data due to a problem with the electrode) may decrease the
quality of the ICA (i.e., dissimilar activations are mixed into the ICs).

On the other hand, interpolating before ICA raises a couple of issues 1) it
reduces the dimensionality of the data and 2) introduces some non-linearity
into the data (if the interpolation method was not linear), which is
detrimental to the ICA solution. We dealt with Issue 1in FASTER by
restricting the maximum number of ICs to correspond with the reduced rank of
the data after interpolation.

The choice then was between reducing the quality of the ICA by introducing
noisy channels or reducing the quality of the ICA by the non-linearity
introduced due to spherical interpolation. Although ICA assumes linearity,
there is almost certainly some non-linearity in the signals recorded at the
scalp, and the non-linearity introduced by spherical interpolation is likely
only a small contributer to the overall non-linearity. In any case, based on
pilot testing we found that when the ICA was done with noisy channels
included (i.e., not interpolated out) the resulting components were less
useful than when the data were cleaner (i.e., the channels were
interpolated). As an aside, testing algorithms on real data proved much more
informative than testing on the simulated data, perhaps due to the inclusion
of non-stereotypic artefacts in the real data.

That said, we are certainly open to persuasion on this issue and/or
suggestions about how to quantify which order is better. Also, might there
be situations in which one order is superior to the other, perhaps depending
on the maximum number of ICs that can be generated?

If there is demand, we can also configure FASTER so that the user can select
the order of the processing steps. Email me directly robert.whelan at tcd.ie or
whelanrob at gmail.com if this is something that people might want or with any
other suggestions.

Best Regards,

Rob & Hugh

-- 
Robert Whelan, PhD
Senior Research Scientist

Trinity Centre for Bioengineering
Trinity College Dublin

Department of Neurology
St. Vincent's University Hospital
Elm Park, Dublin 4

webpage: http://www.mee.tcd.ie/~neuraleng/People/Robert
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100928/79c898dc/attachment.html>

From pzeman at alumni.uvic.ca  Tue Sep 28 09:03:42 2010
From: pzeman at alumni.uvic.ca (Philip Michael Zeman)
Date: Tue, 28 Sep 2010 09:03:42 -0700
Subject: [Eeglablist] Using ICA with interpolated channels
In-Reply-To: <326D1E69-8542-4C65-A2D8-6023B058DF17@gmail.com>
References: <326D1E69-8542-4C65-A2D8-6023B058DF17@gmail.com>
Message-ID: <082446FB6B234D8D9C596B4AA78AC9C5@mine>

Hello Jordi

there are a number of methods of interpolation and not all are equal. 
However, generally:

an interpolated channel is the linear combination of 2 or more other real 
EEG channels.

Hence, a new dimension of data is not being created by doing the 
interpolation.

Hence, the interpolated channel does not benefit the ICA process.

This said,

it is pemissible to interpolate channels of back-projected (scalp projected) 
components "after" applying ICA.


Phil


=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Philip Michael Zeman B.Eng, Ph.D.
Applied Brain and Vision Sciences Inc.
Brain Function Analysis for Novel Paradigms and Serious Games
Analysis of Pharmaceutical Effects on Brain Function
http://www.abvsciences.com
Latest Brain Research Result:
http://www.spatialbrain.com
Email:   pzeman at alumni.uvic.ca
Phone: +1-250-589-4234
LinkedIn Profile: http://ca.linkedin.com/in/philipmichaelzeman
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

----- Original Message ----- 
From: "Jordi Costa Faidella" <jordicostafa at gmail.com>
To: <eeglablist at sccn.ucsd.edu>
Sent: Wednesday, September 22, 2010 5:43 PM
Subject: [Eeglablist] Using ICA with interpolated channels


> Dear EEGlab users,
>
> this question arose when reading the new FASTER method, but I think is of 
> general importance for all of us. Is it correct to perform an ICA on a 
> dataset in which some of the channels have been interpolated?
>
> thanks,
>
> Jordi
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to 
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to 
> eeglablist-request at sccn.ucsd.edu 



From nperentos at gmail.com  Tue Sep 28 08:10:14 2010
From: nperentos at gmail.com (Nicholas Perentos)
Date: Tue, 28 Sep 2010 16:10:14 +0100
Subject: [Eeglablist] Sleep scoring
Message-ID: <AANLkTikkjkHsxdxPd7BLtuE1ODUJTgvgLbCaYncLyY4F@mail.gmail.com>

Hi there,

Is sleep scoring at all possible with EEGLAB? If not, is there someone that
might be able to point towards a freely available software or matlab code
for sleep scoring?

Any information is appreciated
Thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100928/ece15a46/attachment.html>

From jordicostafa at gmail.com  Tue Sep 28 09:15:37 2010
From: jordicostafa at gmail.com (Jordi Costa Faidella)
Date: Tue, 28 Sep 2010 12:15:37 -0400
Subject: [Eeglablist] Using ICA with interpolated channels
In-Reply-To: <082446FB6B234D8D9C596B4AA78AC9C5@mine>
References: <326D1E69-8542-4C65-A2D8-6023B058DF17@gmail.com>
	<082446FB6B234D8D9C596B4AA78AC9C5@mine>
Message-ID: <5CFF34E6-6E5D-4DE1-B1D5-608EE2FF5D46@gmail.com>

thank you David and Philip for your clear comments!

However, I still have a question for Robert Whelan about the FASTER method: in the paper, the method appears to perform first an interpolation of bad channels and then the ICA, constrained with PCA. Is this PCA that constrains the possible ICs implemented by default in the method or it is something you had to perform in order to be able to use FASTER with your dataset?

thank you,

Jordi 




El 28/09/2010, a las 12:03, Philip Michael Zeman escribi?:

> Hello Jordi
> 
> there are a number of methods of interpolation and not all are equal. However, generally:
> 
> an interpolated channel is the linear combination of 2 or more other real EEG channels.
> 
> Hence, a new dimension of data is not being created by doing the interpolation.
> 
> Hence, the interpolated channel does not benefit the ICA process.
> 
> This said,
> 
> it is pemissible to interpolate channels of back-projected (scalp projected) components "after" applying ICA.
> 
> 
> Phil
> 
> 
> =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
> Philip Michael Zeman B.Eng, Ph.D.
> Applied Brain and Vision Sciences Inc.
> Brain Function Analysis for Novel Paradigms and Serious Games
> Analysis of Pharmaceutical Effects on Brain Function
> http://www.abvsciences.com
> Latest Brain Research Result:
> http://www.spatialbrain.com
> Email:   pzeman at alumni.uvic.ca
> Phone: +1-250-589-4234
> LinkedIn Profile: http://ca.linkedin.com/in/philipmichaelzeman
> =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
> 
> ----- Original Message ----- From: "Jordi Costa Faidella" <jordicostafa at gmail.com>
> To: <eeglablist at sccn.ucsd.edu>
> Sent: Wednesday, September 22, 2010 5:43 PM
> Subject: [Eeglablist] Using ICA with interpolated channels
> 
> 
>> Dear EEGlab users,
>> 
>> this question arose when reading the new FASTER method, but I think is of general importance for all of us. Is it correct to perform an ICA on a dataset in which some of the channels have been interpolated?
>> 
>> thanks,
>> 
>> Jordi
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu 
> 




From balazs at cogpsyphy.hu  Tue Sep 28 04:55:23 2010
From: balazs at cogpsyphy.hu (balazs)
Date: Tue, 28 Sep 2010 13:55:23 +0200
Subject: [Eeglablist] ICA in multi-session experiment
Message-ID: <4CA1D7AB.7090000@cogpsyphy.hu>

Dear all,
I wonder if it were possible to perform a single ICA analysis ondata in 
case the same protocol is repeated at a number of occasions. As far as I 
know it is not advised in general because the electrode locations could 
be different. I would like to hear your opinions on the feasibility of 
using some sort of alignment of the potential field provided the 
electrode positions are co-registered. I am also looking forward to 
comments from those who believe that one can not really gain much with 
handling multi-session data this way.
Kind regards,
Laszlo

-- 
      Laszlo Balazs, Ph.D.          /      dr. Bal?zs L?szl?
 Institute for Psychology HAS       /   MTA Pszichol?giai Kutat?int?zet
                 P O B 398, Budapest, Hungary, H-1394
                Tel:+36(1)354-2410 | Fax:+36(1)354-2416
                      http://www.cogpsyphy.hu/balazs



__________ Information from ESET NOD32 Antivirus, version of virus signature database 5485 (20100928) __________

The message was checked by ESET NOD32 Antivirus.

http://www.eset.com




From grega.repovs at psy.ff.uni-lj.si  Tue Sep 28 13:14:31 2010
From: grega.repovs at psy.ff.uni-lj.si (Grega Repovs)
Date: Tue, 28 Sep 2010 22:14:31 +0200
Subject: [Eeglablist] Re. Using ICA with interpolated channels
In-Reply-To: <AANLkTim0wGSCd44phn_WNXqJfNgo0LyOn5qYoboqOA1W@mail.gmail.com>
References: <AANLkTim0wGSCd44phn_WNXqJfNgo0LyOn5qYoboqOA1W@mail.gmail.com>
Message-ID: <8826BC61-DCFF-42C8-B357-81AF19EC8DAE@psy.ff.uni-lj.si>

Dear Rob & Hugh,

Since there seem to be arguments against using the problematic channels both before as well as after interpolation, why not run ICA without those channels. So the procedure would be:

1/ Identify and remove bad channels
2/ Perform ICA on good channels only
3/ Remove bad ICA components
4/ Reconstruct good channels
5/ Interpolate bad channels

This way neither noise nor non-linearities would affect the ICA solution and bad channels can still be interpolated based on cleaned data.

I also have one other question with regards to FASTER. In your paper you compared it to SCADS. I was wondering, why did you not compare it to ERP PCA Toolkit by Joseph Dien, which also performs fully automated data preprocessing and employs algorithms similar to FASTER. I myself would be quite interested in that comparison.

All the best,

Grega Repovs



On Sep 28, 2010, at 12:04 PM, Robert Whelan wrote:

> Jordi Costa Faidella wrote "Is it correct to perform an ICA on a dataset in which some of the channels have been interpolated?"
> This is an interesting question and we considered both orders (each order has some advantages and disadvantages) for the FASTER method. Ultimately, we decided to run interpolation first followed ICA. Here was our rationale:
> As the EEGLAB manual recommends ? ?ICA works best when given a large amount of basically similar and mostly clean data.? (see p.59). Therefore, an ICA on a dataset in which some channels are noisy (perhaps with a lot of non-stereotypic data due to a problem with the electrode) may decrease the quality of the ICA (i.e., dissimilar activations are mixed into the ICs).
> On the other hand, interpolating before ICA raises a couple of issues 1) it reduces the dimensionality of the data and 2) introduces some non-linearity into the data (if the interpolation method was not linear), which is detrimental to the ICA solution. We dealt with Issue 1in FASTER by restricting the maximum number of ICs to correspond with the reduced rank of the data after interpolation.
> The choice then was between reducing the quality of the ICA by introducing noisy channels or reducing the quality of the ICA by the non-linearity introduced due to spherical interpolation. Although ICA assumes linearity, there is almost certainly some non-linearity in the signals recorded at the scalp, and the non-linearity introduced by spherical interpolation is likely only a small contributer to the overall non-linearity. In any case, based on pilot testing we found that when the ICA was done with noisy channels included (i.e., not interpolated out) the resulting components were less useful than when the data were cleaner (i.e., the channels were interpolated). As an aside, testing algorithms on real data proved much more informative than testing on the simulated data, perhaps due to the inclusion of non-stereotypic artefacts in the real data. 
> That said, we are certainly open to persuasion on this issue and/or suggestions about how to quantify which order is better. Also, might there be situations in which one order is superior to the other, perhaps depending on the maximum number of ICs that can be generated? 
> If there is demand, we can also configure FASTER so that the user can select the order of the processing steps. Email me directly robert.whelan at tcd.ie or whelanrob at gmail.com if this is something that people might want or with any other suggestions.
> 
> Best Regards, 
> 
> Rob & Hugh
> 
> -- 
> Robert Whelan, PhD
> Senior Research Scientist
> 
> Trinity Centre for Bioengineering
> Trinity College Dublin
> 
> Department of Neurology
> St. Vincent's University Hospital
> Elm Park, Dublin 4
> 
> webpage: http://www.mee.tcd.ie/~neuraleng/People/Robert
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu




-- 
Asist. Prof. Grega Repov?, Ph.D.
Department of Psychology
University of Ljubljana
A?ker?eva 2
SI-1000 Ljubljana
tel: +386 1 241 1175
email: grega.repovs at psy.ff.uni-lj.si

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100928/7cf943c0/attachment.html>

From jdien07 at mac.com  Tue Sep 28 17:02:06 2010
From: jdien07 at mac.com (Joseph Dien)
Date: Tue, 28 Sep 2010 17:02:06 -0700
Subject: [Eeglablist] Re. Using ICA with interpolated channels
In-Reply-To: <8826BC61-DCFF-42C8-B357-81AF19EC8DAE@psy.ff.uni-lj.si>
References: <AANLkTim0wGSCd44phn_WNXqJfNgo0LyOn5qYoboqOA1W@mail.gmail.com>
	<8826BC61-DCFF-42C8-B357-81AF19EC8DAE@psy.ff.uni-lj.si>
Message-ID: <5804C6E0-DA75-4BB5-B9D3-2FD611E1DEC2@mac.com>

Thanks for the plug Grega!  I would be interested as well.

In case it would be of help:

Dien, J. (2010). The ERP PCA Toolkit: An Open Source Program For Advanced Statistical Analysis of Event Related Potential Data. Journal of Neuroscience Methods, 187(1), 138-145.

http://sourceforge.net/projects/erppcatoolkit/

Cheers!

Joe


On Sep 28, 2010, at 1:14 PM, Grega Repovs wrote:

> Dear Rob & Hugh,
> 
> Since there seem to be arguments against using the problematic channels both before as well as after interpolation, why not run ICA without those channels. So the procedure would be:
> 
> 1/ Identify and remove bad channels
> 2/ Perform ICA on good channels only
> 3/ Remove bad ICA components
> 4/ Reconstruct good channels
> 5/ Interpolate bad channels
> 
> This way neither noise nor non-linearities would affect the ICA solution and bad channels can still be interpolated based on cleaned data.
> 
> I also have one other question with regards to FASTER. In your paper you compared it to SCADS. I was wondering, why did you not compare it to ERP PCA Toolkit by Joseph Dien, which also performs fully automated data preprocessing and employs algorithms similar to FASTER. I myself would be quite interested in that comparison.
> 
> All the best,
> 
> Grega Repovs
> 
> 
> 
> On Sep 28, 2010, at 12:04 PM, Robert Whelan wrote:
> 
>> Jordi Costa Faidella wrote "Is it correct to perform an ICA on a dataset in which some of the channels have been interpolated?"
>> This is an interesting question and we considered both orders (each order has some advantages and disadvantages) for the FASTER method. Ultimately, we decided to run interpolation first followed ICA. Here was our rationale:
>> As the EEGLAB manual recommends ? ?ICA works best when given a large amount of basically similar and mostly clean data.? (see p.59). Therefore, an ICA on a dataset in which some channels are noisy (perhaps with a lot of non-stereotypic data due to a problem with the electrode) may decrease the quality of the ICA (i.e., dissimilar activations are mixed into the ICs).
>> On the other hand, interpolating before ICA raises a couple of issues 1) it reduces the dimensionality of the data and 2) introduces some non-linearity into the data (if the interpolation method was not linear), which is detrimental to the ICA solution. We dealt with Issue 1in FASTER by restricting the maximum number of ICs to correspond with the reduced rank of the data after interpolation.
>> The choice then was between reducing the quality of the ICA by introducing noisy channels or reducing the quality of the ICA by the non-linearity introduced due to spherical interpolation. Although ICA assumes linearity, there is almost certainly some non-linearity in the signals recorded at the scalp, and the non-linearity introduced by spherical interpolation is likely only a small contributer to the overall non-linearity. In any case, based on pilot testing we found that when the ICA was done with noisy channels included (i.e., not interpolated out) the resulting components were less useful than when the data were cleaner (i.e., the channels were interpolated). As an aside, testing algorithms on real data proved much more informative than testing on the simulated data, perhaps due to the inclusion of non-stereotypic artefacts in the real data. 
>> That said, we are certainly open to persuasion on this issue and/or suggestions about how to quantify which order is better. Also, might there be situations in which one order is superior to the other, perhaps depending on the maximum number of ICs that can be generated? 
>> If there is demand, we can also configure FASTER so that the user can select the order of the processing steps. Email me directly robert.whelan at tcd.ie or whelanrob at gmail.com if this is something that people might want or with any other suggestions.
>> 
>> Best Regards, 
>> 
>> Rob & Hugh
>> 
>> -- 
>> Robert Whelan, PhD
>> Senior Research Scientist
>> 
>> Trinity Centre for Bioengineering
>> Trinity College Dublin
>> 
>> Department of Neurology
>> St. Vincent's University Hospital
>> Elm Park, Dublin 4
>> 
>> webpage: http://www.mee.tcd.ie/~neuraleng/People/Robert
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
> 
> 
> 
> 
> -- 
> Asist. Prof. Grega Repov?, Ph.D.
> Department of Psychology
> University of Ljubljana
> A?ker?eva 2
> SI-1000 Ljubljana
> tel: +386 1 241 1175
> email: grega.repovs at psy.ff.uni-lj.si
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100928/6f8313de/attachment.html>

From laurence.casini at univ-provence.fr  Thu Sep 30 01:34:13 2010
From: laurence.casini at univ-provence.fr (Laurence Casini)
Date: Thu, 30 Sep 2010 10:34:13 +0200
Subject: [Eeglablist] post-doc announcement
Message-ID: <4CA44B85.3090506@univ-provence.fr>

Dear colleagues,
Please find a call for a postdoctoral position available in Marseille.
Please feel free to circulate.
Thank you very much for your help.

Laurence Casini

Postdoctoral position on Feedback error-related poentials


The postdoctoral research position is available now and tenable for up 
to two years in a project funded by the "Agence Nationale de la 
Recherche" (ANR). Applications will be considered until the position is 
assigned.

The position will be based at the "Laboratoire de Neurobiologie de la 
Cognition" (CNRS& Universit? de Provence, Marseille). Applications will 
be considered until the position is assigned.

Research topics :
The research project is part of a multidisciplinary funding including 
physicists, mathematicians and neuroscientists, focused on the 
co-adaptation between an user and a BCI system. Our work in this project 
concerns feedback Error-related potentials.

The "Laboratoire de Neurobiologie de la Cognition (LNC)" is embedded in 
a rich neuroscientific environment, with active collaborations, both 
national and international. The LNC owns several high resolution EEG 
systems (from 64 to 128 electrodes) and has also access to a 3T fMRI 
machine devoted to research, and to a MEG machine.

Requirements : Applicants should hold a PhD in Cognitive Sciences or 
Neurosciences. Applications of PhD in Computer Sciences showing interest 
for experimental approaches in Neurosciences will also be considered.

Successful candidates should have an expertise in EEG technique, a 
programming experience and some interest in the field of BCI. Excellent 
oral and written English skills are a requirement.
Signal processing skills will also be an asset.


To apply, please send a full CV, in electronic form, to :

Laurence CASINI (mailto:laurence.casini at univ-provence.fr)
Boris BURLE (boris.burle at univ-provence.fr 
<mailto:boris.burle at univ-provence.fr>)



-- 
__________________________________________________________
ATTENTION CHANGEMENT DE TELEPHONE PROFESSIONNEL !!!!

Laurence Casini
Ma?tre de Conf?rences
UMR 6155 - P?le 3C
Centre St-Charles - Case C
3 place Victor Hugo
13331 Marseille cedex 03

tel : 33 4 13 55 09 41
fax : 33 4 13 55 09 58
e-mail : laurence.casini at univ-provence.fr

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100930/04804a24/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: post-doc.pdf
Type: application/pdf
Size: 39225 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100930/04804a24/attachment.pdf>

From nolanhu at tcd.ie  Wed Sep 29 03:47:39 2010
From: nolanhu at tcd.ie (Hugh Nolan)
Date: Wed, 29 Sep 2010 11:47:39 +0100
Subject: [Eeglablist] Using ICA with interpolated channels
Message-ID: <AANLkTi=0SxyvWSDmwpCsEzkWMyacBSnZPWzO-ZCN2u64@mail.gmail.com>

Hi Jordi,
In response to your question "Is this PCA that constrains the possible ICs
implemented by default in the method or it is something you had to perform
in order to be able to use FASTER with your dataset?", the PCA reduction is
performed by default in FASTER when ICA is run, to constrain the number of
independent components computed based on a) the k-value, a measure of the
number of points per IC (as mentioned in the EEGLAB handbook, and
"Information-based
modeling of event-related brain dynamics." by Onton et al), and b) the
number of interpolated channels, as mentioned previously. Is this necessary
to be able to use FASTER? Well, during testing of the FASTER method we ran
tests of various k-values for a number of datasets, using 15, 25, 40 and
also 1, to see the effect on the output. Generally, the high variance
components, particularly the EOG component, was little affected by this, but
the lower variance components were - this included pop-offs and other
transient artifacts. These lower variance components were often split into
numerous components, which made detection of artifacts difficult. It's a
difficult issue, to determine the correct number of sources - we found that
the value of k = 25 gave good results, so we stuck with that. Of course,
experimentation could show a different value works better on your own
particular dataset - if so, let us know, we are always eager to hear.
Hugh and Rob

-- 
Hugh Nolan,
Trinity Centre for Bioengineering,
Printing House,
Trinity College Dublin.

Tel: +353861297722
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100929/8939e691/attachment.html>

From whelanrob at gmail.com  Wed Sep 29 04:36:36 2010
From: whelanrob at gmail.com (Robert Whelan)
Date: Wed, 29 Sep 2010 12:36:36 +0100
Subject: [Eeglablist] Re. Using ICA with interpolated channels
In-Reply-To: <8826BC61-DCFF-42C8-B357-81AF19EC8DAE@psy.ff.uni-lj.si>
References: <AANLkTim0wGSCd44phn_WNXqJfNgo0LyOn5qYoboqOA1W@mail.gmail.com>
	<8826BC61-DCFF-42C8-B357-81AF19EC8DAE@psy.ff.uni-lj.si>
Message-ID: <AANLkTinkzXd3Sh-0qoa2r34Xv_nY-jAsqk=0NoB6ux2A@mail.gmail.com>

Dear Grega,

That is a great suggestion -- thank you. Jordi Costa Faidella emailed me
directly with the same suggestion yesterday and we've already started coding
up the approach that you describe -- should be done and tested in a day or
two. We will also run our EEG data through the new approach and quantify the
difference between interpolating channels vs. removing channels before ICA.

Re. Joseph Dien's ERP PCA toolkit. At the time of writing our paper we
wanted to pick a method from the literature for comparison (although the ERP
PCA toolkit has been available for a while), and with the publication of
Dien (2010) we will definitely compare the two approaches. Although I
haven't used the Toolkit yet, I read the Dien (2010) paper recently and it
looks great.

Thanks again,

Rob & Hugh

On Tue, Sep 28, 2010 at 9:14 PM, Grega Repovs <grega.repovs at psy.ff.uni-lj.si
> wrote:

> Dear Rob & Hugh,
>
> Since there seem to be arguments against using the problematic channels
> both before as well as after interpolation, why not run ICA without those
> channels. So the procedure would be:
>
> 1/ Identify and remove bad channels
> 2/ Perform ICA on good channels only
> 3/ Remove bad ICA components
> 4/ Reconstruct good channels
> 5/ Interpolate bad channels
>
> This way neither noise nor non-linearities would affect the ICA solution
> and bad channels can still be interpolated based on cleaned data.
>
> I also have one other question with regards to FASTER. In your paper you
> compared it to SCADS. I was wondering, why did you not compare it to ERP PCA
> Toolkit by Joseph Dien, which also performs fully automated data
> preprocessing and employs algorithms similar to FASTER. I myself would be
> quite interested in that comparison.
>
> All the best,
>
> Grega Repovs
>
>
>
> On Sep 28, 2010, at 12:04 PM, Robert Whelan wrote:
>
> Jordi Costa Faidella wrote "Is it correct to perform an ICA on a dataset
> in which some of the channels have been interpolated?"
>
> This is an interesting question and we considered both orders (each order
> has some advantages and disadvantages) for the FASTER method. Ultimately, we
> decided to run interpolation first followed ICA. Here was our rationale:
>
> As the EEGLAB manual recommends ? ?ICA works best when given a large amount
> of basically similar and mostly clean data.? (see p.59). Therefore, an ICA
> on a dataset in which some channels are noisy (perhaps with a lot of
> non-stereotypic data due to a problem with the electrode) may decrease the
> quality of the ICA (i.e., dissimilar activations are mixed into the ICs).
>
> On the other hand, interpolating before ICA raises a couple of issues 1) it
> reduces the dimensionality of the data and 2) introduces some non-linearity
> into the data (if the interpolation method was not linear), which is
> detrimental to the ICA solution. We dealt with Issue 1in FASTER by
> restricting the maximum number of ICs to correspond with the reduced rank of
> the data after interpolation.
>
> The choice then was between reducing the quality of the ICA by introducing
> noisy channels or reducing the quality of the ICA by the non-linearity
> introduced due to spherical interpolation. Although ICA assumes linearity,
> there is almost certainly some non-linearity in the signals recorded at the
> scalp, and the non-linearity introduced by spherical interpolation is likely
> only a small contributer to the overall non-linearity. In any case, based
> on pilot testing we found that when the ICA was done with noisy channels
> included (i.e., not interpolated out) the resulting components were less
> useful than when the data were cleaner (i.e., the channels were
> interpolated). As an aside, testing algorithms on real data proved much more
> informative than testing on the simulated data, perhaps due to the inclusion
> of non-stereotypic artefacts in the real data.
>
> That said, we are certainly open to persuasion on this issue and/or
> suggestions about how to quantify which order is better. Also, might there
> be situations in which one order is superior to the other, perhaps depending
> on the maximum number of ICs that can be generated?
>
> If there is demand, we can also configure FASTER so that the user can
> select the order of the processing steps. Email me directly
> robert.whelan at tcd.ie or whelanrob at gmail.com if this is something that
> people might want or with any other suggestions.
>
> Best Regards,
>
> Rob & Hugh
>
> --
> Robert Whelan, PhD
> Senior Research Scientist
>
> Trinity Centre for Bioengineering
> Trinity College Dublin
>
> Department of Neurology
> St. Vincent's University Hospital
> Elm Park, Dublin 4
>
> webpage: http://www.mee.tcd.ie/~neuraleng/People/Robert<http://www.mee.tcd.ie/%7Eneuraleng/People/Robert>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
>
>
>
>
> --
> Asist. Prof. Grega Repov?, Ph.D.
> Department of Psychology
> University of Ljubljana
> A?ker?eva 2
> SI-1000 Ljubljana
> tel: +386 1 241 1175
> email: grega.repovs at psy.ff.uni-lj.si
>
>


-- 
Robert Whelan, PhD
Senior Research Scientist

Trinity Centre for Bioengineering
Trinity College Dublin

Department of Neurology
St. Vincent's University Hospital
Elm Park, Dublin 4

webpage: http://www.mee.tcd.ie/~neuraleng/People/Robert<http://www.mee.tcd.ie/%7Eneuraleng/People/Robert>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100929/31438fe7/attachment.html>

From arno at ucsd.edu  Thu Sep 30 11:02:17 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Thu, 30 Sep 2010 11:02:17 -0700
Subject: [Eeglablist] Re. Using ICA with interpolated channels
In-Reply-To: <AANLkTinkzXd3Sh-0qoa2r34Xv_nY-jAsqk=0NoB6ux2A@mail.gmail.com>
References: <AANLkTim0wGSCd44phn_WNXqJfNgo0LyOn5qYoboqOA1W@mail.gmail.com>
	<8826BC61-DCFF-42C8-B357-81AF19EC8DAE@psy.ff.uni-lj.si>
	<AANLkTinkzXd3Sh-0qoa2r34Xv_nY-jAsqk=0NoB6ux2A@mail.gmail.com>
Message-ID: <D564EE56-5E8A-4636-8DF2-5650463B628F@ucsd.edu>

Dear Grega and Robert,

concerning interpolating channels before running ICA. Robert's analysis of pros and cons of interpolating channels before running ICA makes perfect sense. We personally at the Swartz center never interpolate data channels and rarely reduce the dimensionality of the data matrix before running ICA (unless we are using 256 channels and then we reduce the dimensionality to 150 before running ICA). As Robert pointed out both reducing the dimensionality using PCA and interpolating channels introduce non linearities. Spherical interpolation introduces non-linearities because a non-linear algorithm is used to interpolate channels. Pre-processing with PCA introduces non-linearities because some of the PCA components - the ones with the lowest eigenvalues - are discarded. Since PCA does not model the structure of the data (i.e. the brain sources), this introduces non linearity. It is hard enough to run ICA and get a clean decomposition for the purpose of analyzing brain source that it is better not to apply any procedure that would potentially introduce non-linearity. When running ICA for the purpose of removing artifacts, this is probably less critical.

Just wanted to add my grain of salt,
Cheers,

Arno

On Sep 29, 2010, at 4:36 AM, Robert Whelan wrote:

> Dear Grega, 
> 
> That is a great suggestion -- thank you. Jordi Costa Faidella emailed me directly with the same suggestion yesterday and we've already started coding up the approach that you describe -- should be done and tested in a day or two. We will also run our EEG data through the new approach and quantify the difference between interpolating channels vs. removing channels before ICA.
> 
> Re. Joseph Dien's ERP PCA toolkit. At the time of writing our paper we wanted to pick a method from the literature for comparison (although the ERP PCA toolkit has been available for a while), and with the publication of Dien (2010) we will definitely compare the two approaches. Although I haven't used the Toolkit yet, I read the Dien (2010) paper recently and it looks great.
> 
> Thanks again, 
> 
> Rob & Hugh
> 
> On Tue, Sep 28, 2010 at 9:14 PM, Grega Repovs <grega.repovs at psy.ff.uni-lj.si> wrote:
> Dear Rob & Hugh,
> 
> Since there seem to be arguments against using the problematic channels both before as well as after interpolation, why not run ICA without those channels. So the procedure would be:
> 
> 1/ Identify and remove bad channels
> 2/ Perform ICA on good channels only
> 3/ Remove bad ICA components
> 4/ Reconstruct good channels
> 5/ Interpolate bad channels
> 
> This way neither noise nor non-linearities would affect the ICA solution and bad channels can still be interpolated based on cleaned data.
> 
> I also have one other question with regards to FASTER. In your paper you compared it to SCADS. I was wondering, why did you not compare it to ERP PCA Toolkit by Joseph Dien, which also performs fully automated data preprocessing and employs algorithms similar to FASTER. I myself would be quite interested in that comparison.
> 
> All the best,
> 
> Grega Repovs
> 
> 
> 
> On Sep 28, 2010, at 12:04 PM, Robert Whelan wrote:
> 
>> Jordi Costa Faidella wrote "Is it correct to perform an ICA on a dataset in which some of the channels have been interpolated?"
>> This is an interesting question and we considered both orders (each order has some advantages and disadvantages) for the FASTER method. Ultimately, we decided to run interpolation first followed ICA. Here was our rationale:
>> As the EEGLAB manual recommends ? ?ICA works best when given a large amount of basically similar and mostly clean data.? (see p.59). Therefore, an ICA on a dataset in which some channels are noisy (perhaps with a lot of non-stereotypic data due to a problem with the electrode) may decrease the quality of the ICA (i.e., dissimilar activations are mixed into the ICs).
>> On the other hand, interpolating before ICA raises a couple of issues 1) it reduces the dimensionality of the data and 2) introduces some non-linearity into the data (if the interpolation method was not linear), which is detrimental to the ICA solution. We dealt with Issue 1in FASTER by restricting the maximum number of ICs to correspond with the reduced rank of the data after interpolation.
>> The choice then was between reducing the quality of the ICA by introducing noisy channels or reducing the quality of the ICA by the non-linearity introduced due to spherical interpolation. Although ICA assumes linearity, there is almost certainly some non-linearity in the signals recorded at the scalp, and the non-linearity introduced by spherical interpolation is likely only a small contributer to the overall non-linearity. In any case, based on pilot testing we found that when the ICA was done with noisy channels included (i.e., not interpolated out) the resulting components were less useful than when the data were cleaner (i.e., the channels were interpolated). As an aside, testing algorithms on real data proved much more informative than testing on the simulated data, perhaps due to the inclusion of non-stereotypic artefacts in the real data. 
>> That said, we are certainly open to persuasion on this issue and/or suggestions about how to quantify which order is better. Also, might there be situations in which one order is superior to the other, perhaps depending on the maximum number of ICs that can be generated? 
>> If there is demand, we can also configure FASTER so that the user can select the order of the processing steps. Email me directly robert.whelan at tcd.ie or whelanrob at gmail.com if this is something that people might want or with any other suggestions.
>> 
>> Best Regards, 
>> 
>> Rob & Hugh
>> 
>> -- 
>> Robert Whelan, PhD
>> Senior Research Scientist
>> 
>> Trinity Centre for Bioengineering
>> Trinity College Dublin
>> 
>> Department of Neurology
>> St. Vincent's University Hospital
>> Elm Park, Dublin 4
>> 
>> webpage: http://www.mee.tcd.ie/~neuraleng/People/Robert
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
> 
> 
> 
> 
> -- 
> Asist. Prof. Grega Repov?, Ph.D.
> Department of Psychology
> University of Ljubljana
> A?ker?eva 2
> SI-1000 Ljubljana
> tel: +386 1 241 1175
> email: grega.repovs at psy.ff.uni-lj.si
> 
> 
> 
> 
> -- 
> Robert Whelan, PhD
> Senior Research Scientist
> 
> Trinity Centre for Bioengineering
> Trinity College Dublin
> 
> Department of Neurology
> St. Vincent's University Hospital
> Elm Park, Dublin 4
> 
> webpage: http://www.mee.tcd.ie/~neuraleng/People/Robert
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20100930/c285fa69/attachment.html>

From Ronald.Phlypo at ugent.be  Fri Oct  1 01:16:00 2010
From: Ronald.Phlypo at ugent.be (Ronald Phlypo)
Date: Fri, 01 Oct 2010 10:16:00 +0200
Subject: [Eeglablist] Using ICA with interpolated channels
In-Reply-To: <082446FB6B234D8D9C596B4AA78AC9C5@mine>
References: <326D1E69-8542-4C65-A2D8-6023B058DF17@gmail.com>
	<082446FB6B234D8D9C596B4AA78AC9C5@mine>
Message-ID: <4CA598C0.6010105@ugent.be>

  Dear Jordi and list members,


Personally, I do not fully agree that one does not benefit from spatial 
interpolation of the EEG before ICA analysis. Whilst the comment

<<Hence, a new dimension of data is not being created by doing the
interpolation.>>

does certainly hold true for linear interpolation schemes, this does no 
longer hold for nonlinear interpolation schemes (cubic splines, 
wavelets, etc.), since a linear unmixing process would see new data in 
there. At best, the dimension of the "cerebral data" would not increase, 
but it might well be that the signal to noise ratio augments by 
increasing the number of virtual channels (obtained through 
interpolation) included in your dataset, where I made the implicit 
assumption that the interpolation is a good approximation to the 
potential field, i.e., that the potential field has been well sampled 
from the start by choosing good electrode locations. The total dimension 
of your data thus increases by introducing new noise sources (which are 
the interpolation errors).

A question that arises to me is whether interpolation should be used or 
best least squares fitting (since the electrode measurements themselves 
include noise, which should not to be modelled by the "interpolated" 
potential field).

Hope this keeps the discussion alive and I would very much be interested 
if anyone would already have obtained some results on this !


Ronald




Le 28/09/2010 18:03, Philip Michael Zeman a ?crit :
> Hello Jordi
>
> there are a number of methods of interpolation and not all are equal.
> However, generally:
>
> an interpolated channel is the linear combination of 2 or more other real
> EEG channels.
>
> Hence, a new dimension of data is not being created by doing the
> interpolation.
>
> Hence, the interpolated channel does not benefit the ICA process.
>
> This said,
>
> it is pemissible to interpolate channels of back-projected (scalp projected)
> components "after" applying ICA.
>
>
> Phil
>
>
> =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
> Philip Michael Zeman B.Eng, Ph.D.
> Applied Brain and Vision Sciences Inc.
> Brain Function Analysis for Novel Paradigms and Serious Games
> Analysis of Pharmaceutical Effects on Brain Function
> http://www.abvsciences.com
> Latest Brain Research Result:
> http://www.spatialbrain.com
> Email:   pzeman at alumni.uvic.ca
> Phone: +1-250-589-4234
> LinkedIn Profile: http://ca.linkedin.com/in/philipmichaelzeman
> =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
>
> ----- Original Message -----
> From: "Jordi Costa Faidella"<jordicostafa at gmail.com>
> To:<eeglablist at sccn.ucsd.edu>
> Sent: Wednesday, September 22, 2010 5:43 PM
> Subject: [Eeglablist] Using ICA with interpolated channels
>
>
>> Dear EEGlab users,
>>
>> this question arose when reading the new FASTER method, but I think is of
>> general importance for all of us. Is it correct to perform an ICA on a
>> dataset in which some of the channels have been interpolated?
>>
>> thanks,
>>
>> Jordi
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu


From jdsitt at gmail.com  Fri Oct  1 01:36:48 2010
From: jdsitt at gmail.com (Jaco Sitt)
Date: Fri, 1 Oct 2010 10:36:48 +0200
Subject: [Eeglablist] Re. Using ICA with interpolated channels
In-Reply-To: <D564EE56-5E8A-4636-8DF2-5650463B628F@ucsd.edu>
References: <AANLkTim0wGSCd44phn_WNXqJfNgo0LyOn5qYoboqOA1W@mail.gmail.com>
	<8826BC61-DCFF-42C8-B357-81AF19EC8DAE@psy.ff.uni-lj.si>
	<AANLkTinkzXd3Sh-0qoa2r34Xv_nY-jAsqk=0NoB6ux2A@mail.gmail.com>
	<D564EE56-5E8A-4636-8DF2-5650463B628F@ucsd.edu>
Message-ID: <AANLkTikV2xF6got-dDaEpfBn0m28qGuFZLkLbcLv6PRD@mail.gmail.com>

Dear Arnaud,

Could you please give more details of the reduction of dimensionality
on the 256 channels recordings ?

I am running ICA on the whole 256 channels  to remove artifacts, but I
am concern regarding the quality of ICA in this case.

Kind regards,


-- 
Jacobo Diego Sitt, MD, PhD

Groupe Hospitalier Piti?-Salp?tri?re,
Service de Neurologie 1, & Centre du Cerveau et de la Mo?lle
47-83 Bd de l'H?pital, 75651 PARIS Cedex 13

INSERM U992CEA/Saclay, NeuroSpin B?t 145
CEA - Saclay, France
Point Courrier 156, 91191 Gif/Yvette Cedex



On Thu, Sep 30, 2010 at 8:02 PM, Arnaud Delorme <arno at ucsd.edu> wrote:
> Dear Grega and Robert,
> concerning interpolating channels before running ICA. Robert's analysis of
> pros and cons of interpolating channels before running ICA makes perfect
> sense. We personally at the Swartz center never interpolate data channels
> and rarely reduce the dimensionality of the data matrix before running ICA
> (unless we are using 256 channels and then we reduce the dimensionality to
> 150 before running ICA). As Robert pointed out both reducing the
> dimensionality using PCA and interpolating channels introduce no
> linearities. Spherical interpolation introduces non-linearities because a
> non-linear algorithm is used to interpolate channels. Pre-processing with
> PCA introduces non-linearities because some of the PCA components - the ones
> with the lowest eigenvalues - are discarded. Since PCA does not model the
> structure of the data (i.e. the brain sources), this introduces non
> linearity. It is hard enough to run ICA and get a clean decomposition for
> the purpose of analyzing brain source that it is better not to apply any
> procedure that would potentially introduce non-linearity. When running ICA
> for the purpose of removing artifacts, this is probably less critical.
> Just wanted to add my grain of salt,
> Cheers,
> Arno
> On Sep 29, 2010, at 4:36 AM, Robert Whelan wrote:
>
> Dear Grega,
>
> That is a great suggestion -- thank you. Jordi Costa Faidella emailed me
> directly with the same suggestion yesterday and we've already started coding
> up the approach that you describe -- should be done and tested in a day or
> two. We will also run our EEG data through the new approach and quantify the
> difference between interpolating channels vs. removing channels before ICA.
>
> Re. Joseph Dien's ERP PCA toolkit. At the time of writing our paper we
> wanted to pick a method from the literature for comparison (although the ERP
> PCA toolkit has been available for a while), and with the publication of
> Dien (2010) we will definitely compare the two approaches. Although I
> haven't used the Toolkit yet, I read the Dien (2010) paper recently and it
> looks great.
>
> Thanks again,
>
> Rob & Hugh
>
> On Tue, Sep 28, 2010 at 9:14 PM, Grega Repovs
> <grega.repovs at psy.ff.uni-lj.si> wrote:
>>
>> Dear Rob & Hugh,
>> Since there seem to be arguments against using the problematic channels
>> both before as well as after interpolation, why not run ICA without those
>> channels. So the procedure would be:
>> 1/ Identify and remove bad channels
>> 2/ Perform ICA on good channels only
>> 3/?Remove bad ICA components
>> 4/ Reconstruct good channels
>> 5/ Interpolate bad channels
>> This way neither noise nor non-linearities would affect the ICA solution
>> and bad channels can still be interpolated based on cleaned data.
>> I also have one other question with regards to FASTER. In your paper you
>> compared it to SCADS. I was wondering, why did you not compare it to ERP PCA
>> Toolkit by Joseph Dien, which also performs fully automated data
>> preprocessing and employs algorithms similar to FASTER. I myself would be
>> quite interested in that comparison.
>> All the best,
>> Grega Repovs
>>
>>
>> On Sep 28, 2010, at 12:04 PM, Robert Whelan wrote:
>>
>> Jordi Costa Faidella wrote "Is it correct to perform an ICA on a dataset
>> in which some of the channels have been interpolated?"
>>
>> This is an interesting question and we considered both orders (each order
>> has some advantages and disadvantages) for the FASTER method. Ultimately, we
>> decided to run interpolation first followed ICA. Here was our rationale:
>>
>> As the EEGLAB manual recommends ? ?ICA works best when given a large
>> amount of basically similar and mostly clean data.? (see p.59). Therefore,
>> an ICA on a dataset in which some channels are noisy (perhaps with a lot of
>> non-stereotypic data due to a problem with the electrode) may decrease the
>> quality of the ICA (i.e., dissimilar activations are mixed into the ICs).
>>
>> On the other hand, interpolating before ICA raises a couple of issues 1)
>> it reduces the dimensionality of the data and 2) introduces some
>> non-linearity into the data (if the interpolation method was not linear),
>> which is detrimental to the ICA solution. We dealt with Issue 1in FASTER by
>> restricting the maximum number of ICs to correspond with the reduced rank of
>> the data after interpolation.
>>
>> The choice then was between reducing the quality of the ICA by introducing
>> noisy channels or reducing the quality of the ICA by the non-linearity
>> introduced due to spherical interpolation. Although ICA assumes linearity,
>> there is almost certainly some non-linearity in the signals recorded at the
>> scalp, and the non-linearity introduced by spherical interpolation is likely
>> only a small contributer to the overall non-linearity. In any case, based on
>> pilot testing we found that when the ICA was done with noisy channels
>> included (i.e., not interpolated out) the resulting components were less
>> useful than when the data were cleaner (i.e., the channels were
>> interpolated). As an aside, testing algorithms on real data proved much more
>> informative than testing on the simulated data, perhaps due to the inclusion
>> of non-stereotypic artefacts in the real data.
>>
>> That said, we are certainly open to persuasion on this issue and/or
>> suggestions about how to quantify which order is better. Also, might there
>> be situations in which one order is superior to the other, perhaps depending
>> on the maximum number of ICs that can be generated?
>>
>> If there is demand, we can also configure FASTER so that the user can
>> select the order of the processing steps. Email me directly
>> robert.whelan at tcd.ie or whelanrob at gmail.com if this is something that people
>> might want or with any other suggestions.
>>
>> Best Regards,
>>
>> Rob & Hugh
>>
>> --
>> Robert Whelan, PhD
>> Senior Research Scientist
>>
>> Trinity Centre for Bioengineering
>> Trinity College Dublin
>>
>> Department of Neurology
>> St. Vincent's University Hospital
>> Elm Park, Dublin 4
>>
>> webpage: http://www.mee.tcd.ie/~neuraleng/People/Robert
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>>
>>
>>
>> --
>> Asist. Prof. Grega Repov?, Ph.D.
>> Department of Psychology
>> University of Ljubljana
>> A?ker?eva 2
>> SI-1000 Ljubljana
>> tel: +386 1 241 1175
>> email: grega.repovs at psy.ff.uni-lj.si
>
>
>
> --
> Robert Whelan, PhD
> Senior Research Scientist
>
> Trinity Centre for Bioengineering
> Trinity College Dublin
>
> Department of Neurology
> St. Vincent's University Hospital
> Elm Park, Dublin 4
>
> webpage: http://www.mee.tcd.ie/~neuraleng/People/Robert
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



From hubert.preissl at uni-tuebingen.de  Mon Oct  4 03:45:05 2010
From: hubert.preissl at uni-tuebingen.de (Hubert Preissl)
Date: Mon, 04 Oct 2010 12:45:05 +0200
Subject: [Eeglablist] =?iso-8859-15?q?Conference_and_Autumn_School_at_the_?=
 =?iso-8859-15?q?MEG_Center_in_T=FCbingen=2C_November_2010?=
Message-ID: <4CA9B031.5030305@uni-tuebingen.de>



*Lifelong Imaging Conference: 25^th -27^th November 2010*

*Autumn School "Analyze the brain": 24^th -25^th November 2010 *


Hello,
we are pleased to announce the upcoming "Autumn school" followed by the 
"Lifelong Imaging conference" in T?bingen organized by the MEG Center.

For application and further information on invited speakers and 
scientific program
please visit our website:

http://www.mp.uni-tuebingen.de/lli-konferenz/


We look forward to meet you in T?bingen!

Best regards ,

Hubert Preissl

ps: we encountered over the last week several problems with our web-server and

hope that it is running smoothly now. If you encounter any problems please contact:

  meg at med.uni-tuebingen.de  <mailto:meg at med.uni-tuebingen.de>.

-- 
Dr. Hubert Preissl
MEG-Center                   	phone: ++49-(0)7071-2987704
Otfried Mueller Str 47       	fax: ++49-(0)7071-295706
72076 T?bingen, Germany      	email: hubert.preissl at uni-tuebingen.de


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101004/01273d08/attachment.html>

From arno at ucsd.edu  Tue Oct  5 13:07:27 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Tue, 5 Oct 2010 13:07:27 -0700
Subject: [Eeglablist] GPUs and EEGLAB
Message-ID: <496C3318-1686-44A8-B820-6BB454A2594C@ucsd.edu>

This message is from Daniel Gardner concerning the following GPU wiki page:

http://sccn.ucsd.edu/wiki/GPU

Seconding and confirming Arno's recent analysis 
of GPU-enabled MATLAB for EEGLAB analysis, I 
agree that realizing the very great potential of 
GPUs for neurophysiology requires very careful 
analysis and programming that generic packages 
are unlikely to achieve.

Jonathan Victor and I have begun a project to 
extend our neuroanalysis.org 
information-theoretic and other spike train 
analytics to the GPU platform. Our experience, 
that of others in our project who have used GPUs 
for other areas of biomedicine, and our TC Dublin 
colleague Conor Houghton, all confirm that 
generic or library-based solutions show only 
modest performance gains.

Getting more than an order of magnitude 
improvement over standard twin-core processors 
requires very careful utilization of GPU 
architecture, including thread assignments and 
memory utilization, and this needs thoughtful 
analysis and manual recoding of parallel routines.

Of course this has to begin with a careful 
analysis that identifies bottlenecks, but 
bottlenecks that are good code targets for 
GPU-derived processing. In addition to being 
computationally intensive, these should be 
parsable into large numbers of threads, larger 
than the number of available cores. Each thread 
should have a good number of independent 
computations that utilize on-chip memory as much 
as possible. The trick is finding the algorithms 
for which the thread, core, and memory GPU 
architecture actually facilitates the needed 
computation. Data fetches, and bus transfers 
between the CPU-controlled part of a computer and 
the GPU slow things down, so the ideal case for 
speedup is where data gets transferred to the GPU 
and long calculations are done in parallel by 
hundreds of cores rather than serially (or in 
parallel by a small number of CPU cores). Also, 
the very new 448-core GPUs allow efficient use of 
many threads per core up to an efficiency limit 
of several thousand simultaneous threads.

Although we have just begun, we are willing to 
share our ongoing experience with the EEGLAB 
community over time; we begin with these initial 
suggestions. The analysis posted appears to have 
been carried out on an NVDIA C1060 (there is no 
2060), but we target these to the newer NVIDIA 
GPU, the C2050 and its native CUDA development 
environment:
- Efficient launch kernel processes to optimize 
instruction throughput, including straightforward 
execution paths for each warp,
- Design code so that flow control instructions 
(if?else, for, do, ?) control multi-thread warps, 
rather than individual threads,
- Use GPU card shared memory efficiently and 
appropriately, enabling block (multi-thread) 
multi-parameter-dependent and paired-trace 
calculations, as well as keeping values within 
registers to avoid GPU card 'local' (on-card but 
off-chip thread-specific) memory transfers, and
- Leverage the complex structured hierarchy of 
each of several components of the GPU-derived 
architecture so that memory transfers, bank 
utilization, thread-per-kernel and 
thread-per-block execution, are performed in the 
smallest number of clock cycles.
- Select the appropriate CUDA runtime math 
library (one is optimized for speed, the other 
for enhanced precision) for each routine.
- Be prepared for multiple cycles of optimization.

-- 
...Daniel Gardner
________________________________________________
| dan at med.cornell.edu
| dg458 at columbia.edu
|________________________________________________
| Dr. Daniel Gardner
| Professor of Physiology &amp; Biophysics
| Head, Laboratory of Neuroinformatics - D-404
| Weill Medical College of Cornell University
| 1300 York Avenue                              voice: (212) 746-6373
| New York, NY 10065 USA                    fax:   (212) 746-8355
| US cell:      +1 917 902-0654
| UK mobile: +44 (0) 7817 423 348
|________________________________________________




From franc_donkers at med.unc.edu  Wed Oct  6 11:55:40 2010
From: franc_donkers at med.unc.edu (Franc Donkers)
Date: Wed, 6 Oct 2010 14:55:40 -0400
Subject: [Eeglablist] Postdoctoral Fellowship at the University of North
 Carolina Chapel Hill
Message-ID: <011001cb6588$1226c740$367455c0$@unc.edu>

The Early Brain Development Research Program at the University of North
Carolina Chapel Hill seeks a postdoctoral fellow to lead a study of cortical
neuronal oscillatory activity and synchrony in this longitudinal study of
brain development.  In addition to the electrophysiological studies, all
children have structural, DTI, and resting state functional MRIs, as well as
developmental cognitive assessments, allowing integrative studies of
cortical brain development in normal and high risk children from birth to
age 6.  The program is supported by several NIH grants, including a Conte
Center for the Neuroscience of Mental Disorders.  The successful applicant
will join our multi-disciplinary team and work with a diverse group of
investigators, including Aysenil Belger, PhD and Franc Donkers Ph.D., who
direct the Electrophysiology Assessment Core at UNC.

 

Applicants must have a recent Ph.D. or M.D. and expertise in EEG acquisition
and analysis systems, such as  EEGLab/FieldTrip, Brain Vision Analyzer, or
EGI Net Station analysis software. Applicants who have experience in applied
programming (Matlab, C), and experience with infant/child EEG recording and
analysis are preferred. Good communication skills and an ability to work
with a dynamic team of investigators are necessary.

 

Please send a letter of interest, a CV, and 3 letters of reference to John
Gilmore MD jgilmore at med.unc.edu. 

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101006/38d5c62b/attachment.html>

From teresa.hawkes at gmail.com  Wed Oct  6 15:07:29 2010
From: teresa.hawkes at gmail.com (Teresa Hawkes)
Date: Wed, 6 Oct 2010 17:07:29 -0500
Subject: [Eeglablist] ADJAR Technique
Message-ID: <AANLkTik6c=2Nm9AwQUpK9vsHqdCJk8rxmRGdD+c4ouJA@mail.gmail.com>

Dear All,

We are using a task-switch paradigm with a very short ISI (10msec). We
know we are getting distortion of our ERPs due to overlap from
temporally adjacent ERPs. We are planning to use the ADJAR technique
(Woldorff, MG. (1993). Distortion of ERP averages due to overlap from
temporally adjacent ERPs: analysis and correction. Psychophysiology,
30, 98-119) to remove this distortion Has anyone used this technique,
and is there code already written for EEGLab?

Many thanks,

-- 
Teresa D. Hawkes, B.F.A.
Ph.D. Candidate
Department of Human Physiology
Institute of Neuroscience
University of Oregon
348 Gerlinger Hall,
1240 University of Oregon
Eugene, Oregon 97403-1240
Phone: 541-337-9443 (cell) or 541-346-0275 (laboratory)
thawkes at uoregon.edu


From Lars.Michels at kispi.uzh.ch  Thu Oct  7 06:58:41 2010
From: Lars.Michels at kispi.uzh.ch (Michels Lars)
Date: Thu, 7 Oct 2010 15:58:41 +0200
Subject: [Eeglablist] Two PhD positions in multimodality neuroimaging
Message-ID: <FD1E6887BC99C341800CBFD7ABC2F9C20259B6BC@exzh1.kispi.int>

Applications are invited for two doctoral research positions in a multimodality neuroimaging project funded by the Z?rich Institute of Human Physiology (ZIHP), to start on the 1st of January 2011. The project will investigate the major physiological and metabolic markers of typical and atypical brain development, using a combination of advanced brain imaging, e.g. simultaneous EEG and functional MRI, perfusion MRI, and MR-spectroscopy methods. Candidates should hold a Master's degree in neuroimaging, neuroscience, neurobiology, neuropsychology, neurophysiology, or physics, or a related field. Fluency in English, good oral and written German, and the ability to work within a multidisciplinary team are essential. Some experience of MRI, EEG or other brain imaging methods is an advantage.

 

Salaries are in accordance with the Swiss National Research Foundation:

1st year: 40'800.-

2nd year: 43'800.-

3rd year: 46'800.-

 

For further information about the project please contact:

 

Dr. Lars Michels (lars.michels at kispi.uzh.ch)

 

To apply please send your CV with a covering letter including a description of your research interests, a copy of your academic degree(s), and two academic referees to the address below. Reviews of applications will begin on the 25th of October and will continue until the positions are occupied.

 

Dr. Lars Michels

MR-Zentrum

University Children's Hospital

Steinwiesstrasse 75

8032 Z?rich

Switzerland

 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101007/a7cb2c18/attachment.html>

From lpxcr1 at nottingham.ac.uk  Thu Oct  7 06:48:17 2010
From: lpxcr1 at nottingham.ac.uk (Christopher Retzler)
Date: Thu, 7 Oct 2010 14:48:17 +0100
Subject: [Eeglablist] Analysis of ERSP data
Message-ID: <A5CE89FBB8146247B56B9D7A67A9EC680E5CEF@VUIEXCHC.ad.nottingham.ac.uk>

I am currently analysing ERSP data from a concurrent choice task and wondered if the following approach is OK?

 I pre-calculate the Frequency measures and then use the following code to create variables containing the ERSP data:

STUDY  = pop_erspparams(STUDY, 'freqrange', [0 35]);

[STUDY erspdata ersptimes erspfreqs] = std_erspplot(STUDY, ALLEEG, 'channels', {'E5', 'E6','E11', 'E12'}, 'plotsubjects', 'on');  % Frontal cluster of electrodes

 

I then extract the data for each subject in each condition for the frequency of interest (in this case beta) and the relevant time period:

Fz_choc = erspdata{1,2}(25:54,107:136, 1, 1);

Fz_cigs = erspdata{1,2}(25:54,107:136, 1, 1);

 

I then calculate the mean beta power for each subject and condition. This data can then be exported to SPSS for ANOVA analysis along with other behavioural variables etc. Is it OK to average the data in this way?

 

Thanks in advance.

 

Chris

This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system:
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101007/cb903afd/attachment.html>

From pdkieffaber at wm.edu  Thu Oct  7 13:18:04 2010
From: pdkieffaber at wm.edu (Paul Kieffaber)
Date: Thu, 7 Oct 2010 15:18:04 -0500
Subject: [Eeglablist] Inter-electrode distances
Message-ID: <AANLkTimtyXGcfhqMwR5kOTa-sE9wMS7Rf5W80AJpzBUZ@mail.gmail.com>

Hi all,
Is there an easy way to compute interelectrode distances?
I'm using the default 10-5 coordinates in EEGlab and am having trouble
making sense of the units returned by standard computations of
distance between polar coordinates.
Any tips, or an M-file you wouldn't mind sharing?

Many thanks,
Paul

-- 
-----------------------------------------------------
Paul Kieffaber, Ph.D.
Assistant Professor
Department of Psychology &
Program in Neuroscience
The College of William & Mary
(757) 221-1965
pdkieffaber at wm.edu
-----------------------------------------------------


From schwiedrzik at mpih-frankfurt.mpg.de  Thu Oct  7 07:33:32 2010
From: schwiedrzik at mpih-frankfurt.mpg.de (Caspar M. Schwiedrzik)
Date: Thu, 7 Oct 2010 16:33:32 +0200
Subject: [Eeglablist] Removal of distortion due to overlapping ERP
 responses - ADJAR filter
In-Reply-To: <201008021738.o72Hc0kb022339@post.webmailer.de>
References: <201008021738.o72Hc0kb022339@post.webmailer.de>
Message-ID: <AANLkTik1A0jwsxJFLZT5mu6Ebsiy3MG3mUQM4uZ=djTT@mail.gmail.com>

Hi everyone,
we would also be very interested in a Matlab implementation of ADJAR.
Thanks and kind regards,
Caspar M. Schwiedrzik



2010/8/2 Niko Busch <mail at nikobusch.net>:
> Hi everyone,
>
> I am looking for a way to remove the distortion from an ERP waveformthat is due to overlapping ERP responses to different events. I willemploy an experimental paradigm in which an experimental event A isfollowed by an event B after a variable SOA on each trial. My interestis in the ERP evoked by the Bs. However, the resulting ERP waveform -time-locked to event B - will be "contaminated" by the responses to theAs.
>
> It seems that people use two different approaches to remove the contribution of the As:
> 1. Include trials with A events only and then subtract the average of A from the average of A+B.
> 2. A convolution approach - the ADJAR method specified in this paper:
>
> Woldorff MG. Distortion of ERP averages due to overlap from temporallyadjacent ERPs: analysis and correction. Psychophysiology. 1993Jan;30(1):98-119.
>
> Has anyone implemented ADJAR or something similar in Eeglab/Matlab andis willing to share the code? Any help would be very much appreciated!
>
> Best,
> Niko
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
>



From arno at ucsd.edu  Thu Oct  7 20:31:30 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Thu, 7 Oct 2010 20:31:30 -0700
Subject: [Eeglablist] Inter-electrode distances
In-Reply-To: <AANLkTimtyXGcfhqMwR5kOTa-sE9wMS7Rf5W80AJpzBUZ@mail.gmail.com>
References: <AANLkTimtyXGcfhqMwR5kOTa-sE9wMS7Rf5W80AJpzBUZ@mail.gmail.com>
Message-ID: <C8BD06C6-9CF8-4232-A919-BD05DD6C5392@ucsd.edu>

Hi Paul,

what you probably want is the distance on the scalp of the electrodes. Assuming a spherical model, for this you have to get the angle between the two vectors pointing to each electrodes from the center of then head the multiply by the radius of the sphere to get the distance on the sphere. I would simply do the dot product of the 3-D Cartesian coordinates of each vector, then compute the angle using argcos (inverse cosine function) (divide by the dot product by the length of the two vector then do the argcos function), then compute angle (in radian) times radius to get the distance.

I have not tried it so I am not sure my fast reasoning is right (but it is a start). More at

http://en.wikipedia.org/wiki/Dot_product

Hope this helps,

Arno

On Oct 7, 2010, at 1:18 PM, Paul Kieffaber wrote:

> Hi all,
> Is there an easy way to compute interelectrode distances?
> I'm using the default 10-5 coordinates in EEGlab and am having trouble
> making sense of the units returned by standard computations of
> distance between polar coordinates.
> Any tips, or an M-file you wouldn't mind sharing?
> 
> Many thanks,
> Paul
> 
> -- 
> -----------------------------------------------------
> Paul Kieffaber, Ph.D.
> Assistant Professor
> Department of Psychology &
> Program in Neuroscience
> The College of William & Mary
> (757) 221-1965
> pdkieffaber at wm.edu
> -----------------------------------------------------
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu




From arno at ucsd.edu  Thu Oct  7 22:02:10 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Thu, 7 Oct 2010 22:02:10 -0700
Subject: [Eeglablist] Analysis of ERSP data
In-Reply-To: <A5CE89FBB8146247B56B9D7A67A9EC680E5CEF@VUIEXCHC.ad.nottingham.ac.uk>
References: <A5CE89FBB8146247B56B9D7A67A9EC680E5CEF@VUIEXCHC.ad.nottingham.ac.uk>
Message-ID: <1DF2E7F1-06F7-4F7F-84AA-4E2B09D81E5B@ucsd.edu>

Dear Chris,

yes, this is fine. There is a simpler and more elegant way. You only have 4 channels and EEGLAB will not plot power scalp maps if you have less than 5 channels. Add an extra channel and plot the scalp map in a given time-frequency range (18 22 Hz and 100 to 200 ms for instance). Then you can just collect the numbers and export them (no need to select specific frequencies or time range).

Hope this helps,

Arno

On Oct 7, 2010, at 6:48 AM, Christopher Retzler wrote:

> I am currently analysing ERSP data from a concurrent choice task and wondered if the following approach is OK?
> 
>  I pre-calculate the Frequency measures and then use the following code to create variables containing the ERSP data:
> 
> STUDY  = pop_erspparams(STUDY, ?freqrange?, [0 35]);
> [STUDY erspdata ersptimes erspfreqs] = std_erspplot(STUDY, ALLEEG, ?channels?, {?E5?, ?E6?,?E11?, ?E12?}, ?plotsubjects?, ?on?);  % Frontal cluster of electrodes
>  
> I then extract the data for each subject in each condition for the frequency of interest (in this case beta) and the relevant time period:
> 
> Fz_choc = erspdata{1,2}(25:54,107:136, 1, 1);
> Fz_cigs = erspdata{1,2}(25:54,107:136, 1, 1);
>  
> I then calculate the mean beta power for each subject and condition. This data can then be exported to SPSS for ANOVA analysis along with other behavioural variables etc. Is it OK to average the data in this way?
>  
> Thanks in advance.
>  
> Chris
> 
> This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it. Please do not use, copy or disclose the information contained in this message or in any attachment. Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
> 
> This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system: you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101007/9c62b458/attachment.html>

From simonshlomo.poil at cncr.vu.nl  Thu Oct  7 23:12:13 2010
From: simonshlomo.poil at cncr.vu.nl (Simon-Shlomo Poil)
Date: Fri, 8 Oct 2010 08:12:13 +0200
Subject: [Eeglablist] Inter-electrode distances
In-Reply-To: <C8BD06C6-9CF8-4232-A919-BD05DD6C5392@ucsd.edu>
References: <AANLkTimtyXGcfhqMwR5kOTa-sE9wMS7Rf5W80AJpzBUZ@mail.gmail.com>
	<C8BD06C6-9CF8-4232-A919-BD05DD6C5392@ucsd.edu>
Message-ID: <AANLkTi=4MP9RjMjAsAQMxTOqYiD3bD34h1Ut-mtMY5Lx@mail.gmail.com>

Hi Paul

There is a function in Joseph Dien's "ERP toolbox" call ep_closestchans that
does what you want.

Simon

On Oct 8, 2010 5:56 AM, "Arnaud Delorme" <arno at ucsd.edu> wrote:

Hi Paul,

what you probably want is the distance on the scalp of the electrodes.
Assuming a spherical model, for this you have to get the angle between the
two vectors pointing to each electrodes from the center of then head the
multiply by the radius of the sphere to get the distance on the sphere. I
would simply do the dot product of the 3-D Cartesian coordinates of each
vector, then compute the angle using argcos (inverse cosine function)
(divide by the dot product by the length of the two vector then do the
argcos function), then compute angle (in radian) times radius to get the
distance.

I have not tried it so I am not sure my fast reasoning is right (but it is a
start). More at

http://en.wikipedia.org/wiki/Dot_product

Hope this helps,

Arno

On Oct 7, 2010, at 1:18 PM, Paul Kieffaber wrote: > Hi all, > Is there an
easy way to compute inte...
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101008/d7b06b78/attachment.html>

From whelanrob at gmail.com  Fri Oct  8 07:26:32 2010
From: whelanrob at gmail.com (Robert Whelan)
Date: Fri, 8 Oct 2010 15:26:32 +0100
Subject: [Eeglablist] Re. Using ICA with interpolated channels
Message-ID: <AANLkTinV2+sqw3622+L905V1p65f=RqMn52v_hg_kUAR@mail.gmail.com>

Hi all,

Re. FASTER: the option to ignore bad channels when running ICA then
interpolate those channels after ICA is now in the latest version (FASTER
v1.1), along with a couple of other new options.

FASTER v1.1 is available on a new host at
http://sourceforge.net/projects/faster/files/  and there is a mailing list
at https://lists.sourceforge.net/lists/listinfo/faster-eeg-list for
FASTER-specific topics.

Best Regards,
Rob & Hugh

-- 
Robert Whelan, PhD
Senior Research Scientist

Trinity Centre for Bioengineering
Trinity College Dublin

Department of Neurology
St. Vincent's University Hospital
Elm Park, Dublin 4

webpage: http://www.mee.tcd.ie/~neuraleng/People/Robert<http://www.mee.tcd.ie/%7Eneuraleng/People/Robert>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101008/17a3f19d/attachment.html>

From jordicostafa at gmail.com  Fri Oct  8 14:25:56 2010
From: jordicostafa at gmail.com (Jordi Costa Faidella)
Date: Fri, 8 Oct 2010 17:25:56 -0400
Subject: [Eeglablist] Re. Using ICA with interpolated channels
In-Reply-To: <AANLkTinV2+sqw3622+L905V1p65f=RqMn52v_hg_kUAR@mail.gmail.com>
References: <AANLkTinV2+sqw3622+L905V1p65f=RqMn52v_hg_kUAR@mail.gmail.com>
Message-ID: <6A8EB8DF-CE3B-4F0C-98CA-93CE45F2F808@gmail.com>

Dear Rob & Hugh,

I'm very happy that you added that option after the open and nice discussion we all had in this forum. I'll definitely use FASTER in my new datasets and share the outcome with all of you.

thank you a lot,

Jordi

El 08/10/2010, a las 10:26, Robert Whelan escribi?:

> Hi all, 
> 
> Re. FASTER: the option to ignore bad channels when running ICA then interpolate those channels after ICA is now in the latest version (FASTER v1.1), along with a couple of other new options. 
> 
> FASTER v1.1 is available on a new host at http://sourceforge.net/projects/faster/files/  and there is a mailing list at https://lists.sourceforge.net/lists/listinfo/faster-eeg-list for FASTER-specific topics.
> 
> Best Regards, 
> Rob & Hugh
> 
> -- 
> Robert Whelan, PhD
> Senior Research Scientist
> 
> Trinity Centre for Bioengineering
> Trinity College Dublin
> 
> Department of Neurology
> St. Vincent's University Hospital
> Elm Park, Dublin 4
> 
> webpage: http://www.mee.tcd.ie/~neuraleng/People/Robert
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101008/0578a9e2/attachment.html>

From jdien07 at mac.com  Mon Oct 11 13:39:51 2010
From: jdien07 at mac.com (Joseph Dien)
Date: Mon, 11 Oct 2010 20:39:51 +0000 (UTC)
Subject: [Eeglablist] CNS Research Technologist position at the University
	of Maryland
Message-ID: <785c5556-c879-e170-8b4d-eb88126c81d0@me.com>

Hi,
    I am writing to let you know that we are currently searching for a CNS Research Technologist at the University of Maryland in College Park.  This would be a full-time position.  Candidates must hold U.S. citizenship.  An M.S. or higher with previous experience in CNS techniques (like EEG, MEG, fMRI, or fNIRS) would be preferred.  Compensation is quite competitive (minimum salary on par with typical post-doc salaries, scaling depending on qualifications).  See attached flyer for more information.  Let me know if you have any questions.
 
Joe
 
--------------------------------------------------------------------------------
 
Joseph Dien,
University of Maryland
 
E-mail: jdien07 at mac.com
Phone: 301-226-8848
Fax: 301-226-8811
http://homepage.mac.com/jdien07/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101011/bbe858e6/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: FRA- Research Technologist.pdf
Type: application/pdf
Size: 102176 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101011/bbe858e6/attachment.pdf>

From schalk at wadsworth.org  Tue Oct 12 07:58:11 2010
From: schalk at wadsworth.org (Gerwin Schalk)
Date: Tue, 12 Oct 2010 10:58:11 -0400
Subject: [Eeglablist] 2nd Intl. Workshop on Advances in Electrocorticography
Message-ID: <C0B25B4D-E80F-4D20-A6F9-5C1150CE6F51@wadsworth.org>

Dear colleague,

This is to announce the 2nd Intl. Workshop on Advances in Electrocorticography, which will be held prior to the upcoming annual meeting of the Society for Neuroscience in San Diego on Thu, Nov 11 - Fri, Nov. 12, 2010 at the Marriott Hotel and Marina, 333 West Harbor Drive, San Diego, California. 

World-class faculty will present the current state and recent advances in research and clinical application of electrocorticographic (ECoG) recordings.  The first day of this workshop focuses on current clinical use of electrocorticography and emerging clinical opportunities. The second day focuses on principles of ECoG signals and emerging research.

We hope that you will be able to join us in this exciting event with world-class speakers. The brochure with the complete program and registration information is available through the link below.

www.amc.edu/Academic/CME/documents/2010_Brain_Mapping_Brochure.pdf

Gerwin Schalk and Anthony Ritaccio

IMPORTANT NOTICE: This e-mail and any attachments may contain
confidential or sensitive information which is, or may be, legally
privileged or otherwise protected by law from further disclosure.  It
is intended only for the addressee.  If you received this in error or
from someone who was not authorized to send it to you, please do not
distribute, copy or use it or any attachments.  Please notify the
sender immediately by reply e-mail and delete this from your
system. Thank you for your cooperation.





From sjwebb at u.washington.edu  Tue Oct 12 09:51:59 2010
From: sjwebb at u.washington.edu (Sara Jane Webb)
Date: Tue, 12 Oct 2010 09:51:59 -0700
Subject: [Eeglablist] Postdoctoral Position - Clinical Electrophysiology
Message-ID: <A9DC30A4-294E-4E88-B0B3-774865FAF113@u.washington.edu>

Electrophysiological Endophenotypes in Autism

Postdoctoral Position available at the Autism Research Programs at the  
University of Washington and Seattle Children?s Research Institute  
(Seattle, Washington).

The University of Washington & Seattle Children?s Research Institute  
bring together scientists from several different disciplines who are  
working collaboratively to discover the genetic cause of autism and
the factors that impact outcome for children with this disorder. As a  
part of this effort, our lab utilizes electrical (endo)phenotypes with  
applications in clinical practice. For more information about the lab  
please see our website http://depts.washington.edu/pbslab/

The Fellow is expected to contribute to study design, data collection,  
data analysis, training and supervision of research assistants, co- 
authoring or leading publications, and the dissemination of findings.

Post-doctoral position (for up to 2 years) to examine broader  
phenotype characteristics of parents of children with autism and  
infants at-risk for autism.  This project uses EEG/ERP and behavioral  
methodology to study social attention in 1st degree relatives of  
individuals with autism as part of the UW Autism Center
of Excellence. The successful candidate will have a Ph.D. in  
developmental or cognitive psychology with experience in  
electrophysiology and an interest in clinical neuroscience.

Position available immediately.

Enquiries should be directed to: Dr. Sara Jane Webb sjwebb at u.washington.edu 
. Please provide cover letter, CV & contact information for 3  
references.

Sara Jane Webb, PhD
Research Assistant Professor of Psychiatry and Behavioral Sciences
and UW Autism Center, Research Program
http://depts.washington.edu/pbslab/
Box 357920; CHDD 314C; University of Washington
Seattle WA 98195
206.221.6461
sjwebb at u.washington.edu





From EJM9F at hscmail.mcc.virginia.edu  Tue Oct 12 10:59:06 2010
From: EJM9F at hscmail.mcc.virginia.edu (Modestino, Edward J *HS)
Date: Tue, 12 Oct 2010 13:59:06 -0400
Subject: [Eeglablist] ICA at Script Level
Message-ID: <CA7B140F2F20CA43AC79DF86B7357ABDD76551C52D@HSCSEMAIL32.hscs.virginia.edu>

Dear EEGLAB experts,

If you have a background in EEGLAB scripting and can help me with this, I would greatly appreciate this.



I have been referring to this, but not able to solve the problem: http://sccn.ucsd.edu/wiki/Chapter_02:_Writing_EEGLAB_Scripts



Recently, I have had great trouble running ICA on my data using the GUI due to memory constraints.  We have not been able to fix this otherwise.  So, I have moved to the script level using a low memory script level version of ICA.  I load my dataset into the GUI and then run the script through the GUI.  Here is the script I have been using.



[EEG.icaweights EEG.sphere] = runicalowmem(reshape(EEG.data, EEG.nbchan, EEG.trials*EEG.pnts));

EEG = eeg_checkset(EEG);

[ALLEEG EEG CURRENTSET] = eeg_store(ALLEEG, EEG, CURRENTSET);

eeglab redraw;



Here is the command window output:

"Input data size [62,1048575] = 62 channels, 1048575 frames

Finding 62 ICA components using logistic ICA.

Decomposing 272 frames per ICA weight ((3844)^2 = 1048575 weights, Initial learning rate will be 0.000157494, block size 70.

Learning rate will be multiplied by 0.9 whenever angledelta >= 60 deg.

More than 32 channels: default stopping weight change 1E-7

Training will end when wchange < 1e-007 or after 512 steps.

Online bias adjustment will be used.

Removing mean of each channel ...

Final training data range: -2048.65 to 1315.39

Computing the sphering matrix...

Starting weights are the identity matrix ...

Sphering the data ...

Beginning ICA training ...

Warning: Using 'state' to set RAND's internal state causes RAND, RANDI, and RANDN

to use legacy random number generators.

> In runicalowmem at 896

  In pop_runscript at 50

step 1 - lrate 0.000157, wchange 5985964.09709237, angledelta  0.0 deg

.

.

.

step 127 - lrate 0.000000, wchange 0.00000008, angledelta 50.9 deg

Sorting components in descending order of mean projected variance ..."





It appears from the command window that ICA ran on the data.  However, the current loaded dataset that ICA was run on says there are no ICA weights (GUI attached).  When I go to plot components, it says ICA needs to be run first.  So, clearly it ran but did not save the ICA to the dataset.



I thought that this following command would overwrite the variables in the workspace: [ALLEEG EEG CURRENTSET] = eeg_store(ALLEEG, EEG, CURRENTSET);



I thought this following command would update the display in the GUI: eeglab redraw;



However, this is not working.

Please help me if you can.
Thanks,
Ed Modestino

Edward Justin Modestino, Ph.D.
Postdoctoral Research Associate
Ray Westphal Neuroimaging Laboratory
Division of Perceptual Studies
Department of Psychiatry and Neurobehavioral Sciences
University of Virginia
Email: ejm9f at virginia.edu

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101012/af4eaf36/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: GUI.jpg
Type: image/jpeg
Size: 57485 bytes
Desc: GUI.jpg
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101012/af4eaf36/attachment.jpg>

From jdesjardins at brocku.ca  Wed Oct 13 13:45:01 2010
From: jdesjardins at brocku.ca (James Desjardins)
Date: Wed, 13 Oct 2010 16:45:01 -0400
Subject: [Eeglablist] ICA at Script Level
In-Reply-To: <CA7B140F2F20CA43AC79DF86B7357ABDD76551C52D@HSCSEMAIL32.hscs.virginia.edu>
References: <CA7B140F2F20CA43AC79DF86B7357ABDD76551C52D@HSCSEMAIL32.hscs.virginia.edu>
Message-ID: <20101013164501.v3sqwcshs0gk8o4g@webmail.brocku.ca>

Hi Ed,

I have not used this function before but I think that it may just be a  
matter of specifying the second output as EEG.icasphere (rather than  
EEG.sphere).

> [EEG.icaweights EEG.icasphere] = runicalowmem(reshape(EEG.data,   
> EEG.nbchan, EEG.trials*EEG.pnts));


... if you still have EEG.sphere in memory you could just try the  
following (rather than rerunning the ICA):

>> EEG.icasphere = EEG.sphere;
>> eeglab redraw;

I hope that this helps.

James Desjardins
Technician, MA Student
Department of Psychology, Behavioural Neuroscience
Cognitive and Affective Neuroscience Lab
Brock University
500 Glenridge Ave.
St. Catharines, ON, Canada
L2S 3A1
905-688-5550 x4676


Quoting "Modestino, Edward J *HS" <EJM9F at hscmail.mcc.virginia.edu>:

> Dear EEGLAB experts,
>
> If you have a background in EEGLAB scripting and can help me with   
> this, I would greatly appreciate this.
>
>
>
> I have been referring to this, but not able to solve the problem:   
> http://sccn.ucsd.edu/wiki/Chapter_02:_Writing_EEGLAB_Scripts
>
>
>
> Recently, I have had great trouble running ICA on my data using the   
> GUI due to memory constraints.  We have not been able to fix this   
> otherwise.  So, I have moved to the script level using a low memory   
> script level version of ICA.  I load my dataset into the GUI and   
> then run the script through the GUI.  Here is the script I have been  
>  using.
>
>
>
> [EEG.icaweights EEG.sphere] = runicalowmem(reshape(EEG.data,   
> EEG.nbchan, EEG.trials*EEG.pnts));
>
> EEG = eeg_checkset(EEG);
>
> [ALLEEG EEG CURRENTSET] = eeg_store(ALLEEG, EEG, CURRENTSET);
>
> eeglab redraw;
>
>
>
> Here is the command window output:
>
> "Input data size [62,1048575] = 62 channels, 1048575 frames
>
> Finding 62 ICA components using logistic ICA.
>
> Decomposing 272 frames per ICA weight ((3844)^2 = 1048575 weights,   
> Initial learning rate will be 0.000157494, block size 70.
>
> Learning rate will be multiplied by 0.9 whenever angledelta >= 60 deg.
>
> More than 32 channels: default stopping weight change 1E-7
>
> Training will end when wchange < 1e-007 or after 512 steps.
>
> Online bias adjustment will be used.
>
> Removing mean of each channel ...
>
> Final training data range: -2048.65 to 1315.39
>
> Computing the sphering matrix...
>
> Starting weights are the identity matrix ...
>
> Sphering the data ...
>
> Beginning ICA training ...
>
> Warning: Using 'state' to set RAND's internal state causes RAND,   
> RANDI, and RANDN
>
> to use legacy random number generators.
>
>> In runicalowmem at 896
>
>   In pop_runscript at 50
>
> step 1 - lrate 0.000157, wchange 5985964.09709237, angledelta  0.0 deg
>
> .
>
> .
>
> .
>
> step 127 - lrate 0.000000, wchange 0.00000008, angledelta 50.9 deg
>
> Sorting components in descending order of mean projected variance ..."
>
>
>
>
>
> It appears from the command window that ICA ran on the data.    
> However, the current loaded dataset that ICA was run on says there   
> are no ICA weights (GUI attached).  When I go to plot components, it  
>  says ICA needs to be run first.  So, clearly it ran but did not  
> save  the ICA to the dataset.
>
>
>
> I thought that this following command would overwrite the variables   
> in the workspace: [ALLEEG EEG CURRENTSET] = eeg_store(ALLEEG, EEG,   
> CURRENTSET);
>
>
>
> I thought this following command would update the display in the   
> GUI: eeglab redraw;
>
>
>
> However, this is not working.
>
> Please help me if you can.
> Thanks,
> Ed Modestino
>
> Edward Justin Modestino, Ph.D.
> Postdoctoral Research Associate
> Ray Westphal Neuroimaging Laboratory
> Division of Perceptual Studies
> Department of Psychiatry and Neurobehavioral Sciences
> University of Virginia
> Email: ejm9f at virginia.edu
>
>






From simonshlomo.poil at cncr.vu.nl  Wed Oct 13 15:15:05 2010
From: simonshlomo.poil at cncr.vu.nl (Simon-Shlomo Poil)
Date: Thu, 14 Oct 2010 00:15:05 +0200
Subject: [Eeglablist] ICA at Script Level
In-Reply-To: <CA7B140F2F20CA43AC79DF86B7357ABDD76551C52D@HSCSEMAIL32.hscs.virginia.edu>
References: <CA7B140F2F20CA43AC79DF86B7357ABDD76551C52D@HSCSEMAIL32.hscs.virginia.edu>
Message-ID: <AANLkTi=eg7xwTcuWM7kq5e10ZzH0oVheJNCd8Gy5A4gE@mail.gmail.com>

Dear Edward,

Did you try just to set  EEG.icaweights, EEG.sphere?
i.e. run the first line of your script. Then check if you can plot the
activations.

hope that works.
-Simon

-- 
Simon-Shlomo Poil

Neuronal Oscillations and Cognition Group (NOC)
Department of Integrative Neurophysiology (INF)
Center for Neurogenomics and Cognitive Research (CNCR)
Neuroscience Campus Amsterdam
VU University Amsterdam
De Boelelaan 1085, Room B-435
1081 HV Amsterdam, The Netherlands

E-mail: simonshlomo.poil at cncr.vu.nl
Phone: +31 20 5989408
Webpage: http://www.poil.dk and http://www.cncr.nl

2010/10/12 Modestino, Edward J *HS <EJM9F at hscmail.mcc.virginia.edu>:
> Dear EEGLAB experts,
>
> If you have a background in EEGLAB scripting and can help me with this, I
> would greatly appreciate this.
>
>
>
> I have been referring to this, but not able to solve the problem:
> http://sccn.ucsd.edu/wiki/Chapter_02:_Writing_EEGLAB_Scripts
>
>
>
> Recently, I have had great trouble running ICA on my data using the GUI due
> to memory constraints.? We have not been able to fix this otherwise.? So, I
> have moved to the script level using a low memory script level version of
> ICA.? I load my dataset into the GUI and then run the script through the
> GUI.? Here is the script I have been using.
>
>
>> [EEG.icaweights EEG.sphere] = runicalowmem(reshape(EEG.data, EEG.nbchan,
> EEG.trials*EEG.pnt
s));
>
> EEG = eeg_checkset(EEG);
>
> [ALLEEG EEG CURRENTSET] = eeg_store(ALLEEG, EEG, CURRENTSET);
>
> eeglab redraw;
>
>
>
> Here is the command window output:
>
> ?Input data size [62,1048575] = 62 channels, 1048575 frames
>
> Finding 62 ICA components using logistic ICA.
>
> Decomposing 272 frames per ICA weight ((3844)^2 = 1048575 weights, Initial
> learning rate will be 0.000157494, block size 70.
>
> Learning rate will be multiplied by 0.9 whenever angledelta >= 60 deg.
>
> More than 32 channels: default stopping weight change 1E-7
>
> Training will end when wchange < 1e-007 or after 512 steps.
>
> Online bias adjustment will be used.
>
> Removing mean of each channel ...
>
> Final training data range: -2048.65 to 1315.39
>
> Computing the sphering matrix...
>
> Starting weights are the identity matrix ...
>
> Sphering the data ...
>
> Beginning ICA training ...
>
> Warning: Using 'state' to set RAND's internal state causes RAND, RANDI, and
> RANDN
>
> to use legacy random number generators.
>
>> In runicalowmem at 896
>
> ? In pop_runscript at 50
>
> step 1 - lrate 0.000157, wchange 5985964.09709237, angledelta? 0.0 deg
>
> .
>
> .
>
> .
>
> step 127 - lrate 0.000000, wchange 0.00000008, angledelta 50.9 deg
>
> Sorting components in descending order of mean projected variance ...?
>
>
>
>
>
> It appears from the command window that ICA ran on the data.? However, the
> current loaded dataset that ICA was run on says there are no ICA weights
> (GUI attached).? When I go to plot components, it says ICA needs to be run
> first.? So, clearly it ran but did not save the ICA to the dataset.
>
>
>
> I thought that this following command would overwrite the variables in the
> workspace: [ALLEEG EEG CURRENTSET] = eeg_store(ALLEEG, EEG, CURRENTSET);
>
>
>
> I thought this following command would update the display in the GUI: eeglab
> redraw;
>
>
>
> However, this is not working.
>
>
>
> Please help me if you can.
>
> Thanks,
> Ed Modestino
>
>
>
> Edward Justin Modestino, Ph.D.
>
> Postdoctoral Research Associate
>
> Ray Westphal Neuroimaging Laboratory
>
> Division of Perceptual Studies
>
> Department of Psychiatry and Neurobehavioral Sciences
>
> University of Virginia
>
> Email: ejm9f at virginia.edu
>
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



From EJM9F at hscmail.mcc.virginia.edu  Thu Oct 14 07:11:07 2010
From: EJM9F at hscmail.mcc.virginia.edu (Modestino, Edward J *HS)
Date: Thu, 14 Oct 2010 10:11:07 -0400
Subject: [Eeglablist] ICA at Script Level
In-Reply-To: <20101013164501.v3sqwcshs0gk8o4g@webmail.brocku.ca>
References: <CA7B140F2F20CA43AC79DF86B7357ABDD76551C52D@HSCSEMAIL32.hscs.virginia.edu>
	<20101013164501.v3sqwcshs0gk8o4g@webmail.brocku.ca>
Message-ID: <CA7B140F2F20CA43AC79DF86B7357ABDD76551C6D3@HSCSEMAIL32.hscs.virginia.edu>

Thanks James and Simon, this code change worked!

-----Original Message-----
From: James Desjardins [mailto:jdesjardins at brocku.ca] 
Sent: Wednesday, October 13, 2010 4:45 PM
To: Modestino, Edward J *HS
Cc: eeglablist at sccn.ucsd.edu; Kelly, Edward *HS
Subject: Re: [Eeglablist] ICA at Script Level

Hi Ed,

I have not used this function before but I think that it may just be a  
matter of specifying the second output as EEG.icasphere (rather than  
EEG.sphere).

> [EEG.icaweights EEG.icasphere] = runicalowmem(reshape(EEG.data,   
> EEG.nbchan, EEG.trials*EEG.pnts));


... if you still have EEG.sphere in memory you could just try the  
following (rather than rerunning the ICA):

>> EEG.icasphere = EEG.sphere;
>> eeglab redraw;

I hope that this helps.

James Desjardins
Technician, MA Student
Department of Psychology, Behavioural Neuroscience
Cognitive and Affective Neuroscience Lab
Brock University
500 Glenridge Ave.
St. Catharines, ON, Canada
L2S 3A1
905-688-5550 x4676


Quoting "Modestino, Edward J *HS" <EJM9F at hscmail.mcc.virginia.edu>:

> Dear EEGLAB experts,
>
> If you have a background in EEGLAB scripting and can help me with   
> this, I would greatly appreciate this.
>
>
>
> I have been referring to this, but not able to solve the problem:   
> http://sccn.ucsd.edu/wiki/Chapter_02:_Writing_EEGLAB_Scripts
>
>
>
> Recently, I have had great trouble running ICA on my data using the   
> GUI due to memory constraints.  We have not been able to fix this   
> otherwise.  So, I have moved to the script level using a low memory   
> script level version of ICA.  I load my dataset into the GUI and   
> then run the script through the GUI.  Here is the script I have been  
>  using.
>
>
>
> [EEG.icaweights EEG.sphere] = runicalowmem(reshape(EEG.data,   
> EEG.nbchan, EEG.trials*EEG.pnts));
>
> EEG = eeg_checkset(EEG);
>
> [ALLEEG EEG CURRENTSET] = eeg_store(ALLEEG, EEG, CURRENTSET);
>
> eeglab redraw;
>
>
>
> Here is the command window output:
>
> "Input data size [62,1048575] = 62 channels, 1048575 frames
>
> Finding 62 ICA components using logistic ICA.
>
> Decomposing 272 frames per ICA weight ((3844)^2 = 1048575 weights,   
> Initial learning rate will be 0.000157494, block size 70.
>
> Learning rate will be multiplied by 0.9 whenever angledelta >= 60 deg.
>
> More than 32 channels: default stopping weight change 1E-7
>
> Training will end when wchange < 1e-007 or after 512 steps.
>
> Online bias adjustment will be used.
>
> Removing mean of each channel ...
>
> Final training data range: -2048.65 to 1315.39
>
> Computing the sphering matrix...
>
> Starting weights are the identity matrix ...
>
> Sphering the data ...
>
> Beginning ICA training ...
>
> Warning: Using 'state' to set RAND's internal state causes RAND,   
> RANDI, and RANDN
>
> to use legacy random number generators.
>
>> In runicalowmem at 896
>
>   In pop_runscript at 50
>
> step 1 - lrate 0.000157, wchange 5985964.09709237, angledelta  0.0 deg
>
> .
>
> .
>
> .
>
> step 127 - lrate 0.000000, wchange 0.00000008, angledelta 50.9 deg
>
> Sorting components in descending order of mean projected variance ..."
>
>
>
>
>
> It appears from the command window that ICA ran on the data.    
> However, the current loaded dataset that ICA was run on says there   
> are no ICA weights (GUI attached).  When I go to plot components, it  
>  says ICA needs to be run first.  So, clearly it ran but did not  
> save  the ICA to the dataset.
>
>
>
> I thought that this following command would overwrite the variables   
> in the workspace: [ALLEEG EEG CURRENTSET] = eeg_store(ALLEEG, EEG,   
> CURRENTSET);
>
>
>
> I thought this following command would update the display in the   
> GUI: eeglab redraw;
>
>
>
> However, this is not working.
>
> Please help me if you can.
> Thanks,
> Ed Modestino
>
> Edward Justin Modestino, Ph.D.
> Postdoctoral Research Associate
> Ray Westphal Neuroimaging Laboratory
> Division of Perceptual Studies
> Department of Psychiatry and Neurobehavioral Sciences
> University of Virginia
> Email: ejm9f at virginia.edu
>
>






From achim.andre at uqam.ca  Thu Oct 14 08:47:49 2010
From: achim.andre at uqam.ca (=?iso-8859-1?Q?Achim=2C_Andr=E9?=)
Date: Thu, 14 Oct 2010 11:47:49 -0400
Subject: [Eeglablist] Removal of distortion due to overlapping ERP
	responses - ADJAR filter
In-Reply-To: <AANLkTik1A0jwsxJFLZT5mu6Ebsiy3MG3mUQM4uZ=djTT@mail.gmail.com>
References: <201008021738.o72Hc0kb022339@post.webmailer.de>
	<AANLkTik1A0jwsxJFLZT5mu6Ebsiy3MG3mUQM4uZ=djTT@mail.gmail.com>
Message-ID: <B3C24EA955FF0C4EA14658997CD3E25E41DDC511@CAHIER.gst.uqam.ca>

The following MATLAB function extracts two 'clean' means that are thought to overlap additively at varying delays. I used it to separate components linked to stimulus from those linked to responses, even though it is highly optimistic, in this case, to assume that the processing of the stimulus is constant despite leading to varying reaction times. To simplify the process, the input are here the total of both raw conditions, instead of the means. In your case totS would be the total spatio-temporal ERP (or average multiplied by the number of trials in the average) time locked to event A and totR the ERP time locked to event B (see lated for number of data points). The third parameter is the list of all individual delays between A en B, expressed in sampling points (i.e. the amount of sliding of the clean responses relative to each other that is currently expressed in the input ERPs as cross contamination). The number of trials in the averages is obtained from the number of elements in argument TR (French abbreviation for Reaction Times, representing the delays between events in my application). On output, you get the respective clean versions, which are averages (not totals as for input).

On input, totS and totR are both (nchannels,npoints). The latter size must be such that, at your largest lag, both totals are aligned in time. For instance, if (a) you consider a minimum baseline of 50 sampling points, (b) your largest lag is 175 sampling points and (c) your second condition may produce a response up to 200 sampling points from stimulus B onset, npoints would be 425 (=50+175+200). Your total tots, time locked to your stimulus A, would cover -50 to +375 time points while your total totR, time locked to your stimulus B, would cover -225 to +200.

The procedures starts with poor estimates of the two 'clean' signals, MS and MR, as if they were known; these initial estimates are simply the respective contaminated averages obtained from the totals in input. 'Knowing' these, we can apply the delays to MS to see how it contaminates MR and subtract that contamination to better estimate MR. The estimate of MR is also smeared by the set of delays to estimate its contamination on MS. Subtracting it from MS yields a better estimate of MS. Iteration proceed until no point in MS or MR, at any channel, changed by more than 1e-5 (the value of crit set in the first line of code), subject to a maximum of 1000 iterations.

You may want to change the name of the function and of its arguments to something more meaningful in English (name and comments are in French), but is should do the job. If your stimuli A and B may take several values, you should consider effecting the separation independently for each combination of A and B, as the model implies that the responses to A and to B are each constant across trials.

Yours

Andr? Achim



function [MS,MR]=separation_S_R(totS,totR,TR)
% [MS,MR]=separation_S_R(totS,totR,TR);
% totS et totR sont des totaux (ncan,npts), tout comme sont les moyennes MS et MR en sortie et TR est la collection des d?lais
% de S ? R en nombre de points d'?chantillonnage (d'ou on inf?re le nb de cas pour faire les moyennes a partir des totaux)
% totS et totR sont consid?r?s align?s pour le plus grand TR (les points de m?me rang ?tant simultan?s dans les deux s?ries)
% by Andr? Achim, Universit? du Qu?bec ? Montr?al
crit=1e-5;
n=length(TR);
[ncan,npts]=size(totS);
MS=totS/n;
MR=totR/n;
TRm=max(TR);
deb=TRm-TR+1;    % d?but dans R (ou S) (la fin est npts) quand on soustrait R de S (ou S de R)
fin=npts-TRm+TR; % fin dans S (ou R) (le d?but est 1) quand on soustrait R de S (ou S de R)
for fois=1:1000  % maximum 1000 r??valuations de MS et MR
    tot=totS;
    for k=1:n,
        tot(:,1:fin(k))=tot(:,1:fin(k))-MR(:,deb(k):npts);
    end
    tot=tot/n;
    errS=max(max(abs(tot-MS)));
    MS=tot;
    tot=totR;
    for k=1:n,
        tot(:,deb(k):npts)=tot(:,deb(k):npts)-MS(:,1:fin(k));
    end
    tot=tot/n;
    errR=max(max(abs(tot-MR)));
    MR=tot;
    if errS<crit & errR<crit, break; end
end

-----Message d'origine-----
De?: eeglablist-bounces at sccn.ucsd.edu [mailto:eeglablist-bounces at sccn.ucsd.edu] De la part de Caspar M. Schwiedrzik
Envoy??: 7 octobre 2010 10:34
??: Niko Busch
Cc?: eeglablist at sccn.ucsd.edu
Objet?: Re: [Eeglablist] Removal of distortion due to overlapping ERP responses - ADJAR filter

Hi everyone,
we would also be very interested in a Matlab implementation of ADJAR.
Thanks and kind regards,
Caspar M. Schwiedrzik



2010/8/2 Niko Busch <mail at nikobusch.net>:
> Hi everyone,
>
> I am looking for a way to remove the distortion from an ERP waveformthat is due to overlapping ERP responses to different events. I willemploy an experimental paradigm in which an experimental event A isfollowed by an event B after a variable SOA on each trial. My interestis in the ERP evoked by the Bs. However, the resulting ERP waveform -time-locked to event B - will be "contaminated" by the responses to theAs.
>
> It seems that people use two different approaches to remove the contribution of the As:
> 1. Include trials with A events only and then subtract the average of A from the average of A+B.
> 2. A convolution approach - the ADJAR method specified in this paper:
>
> Woldorff MG. Distortion of ERP averages due to overlap from temporallyadjacent ERPs: analysis and correction. Psychophysiology. 1993Jan;30(1):98-119.
>
> Has anyone implemented ADJAR or something similar in Eeglab/Matlab andis willing to share the code? Any help would be very much appreciated!
>
> Best,
> Niko
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
>

_______________________________________________
Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu



From alenarto at ucla.edu  Wed Oct 13 08:45:57 2010
From: alenarto at ucla.edu (Agatha Lenartowicz)
Date: Wed, 13 Oct 2010 08:45:57 -0700
Subject: [Eeglablist] cutting up a dataset
Message-ID: <52C22D4B-8B4B-4BA6-A562-6AEAFE805C39@ucla.edu>

Hi all - What is the inverse of eeg_mergeset?

I have a continuous dataset - merged - that I'd like to unmerge. I haven't been able to find the function.
Agatha




From alenarto at ucla.edu  Thu Oct 14 16:26:16 2010
From: alenarto at ucla.edu (Agatha Lenartowicz)
Date: Thu, 14 Oct 2010 16:26:16 -0700
Subject: [Eeglablist] running fmrib pluging on sge cluster
Message-ID: <EB8942C4-C48A-4CA3-9572-0D215884AB5E@ucla.edu>

Dear experts who work on SGE clusters ~ 

I am running the fmrib plugin for removing BCG artefacts - on an SGE cluster. I launch an interactive session using:

qrsh -V -l i,h_rt=1:00:00,mem=4G -pe shared 8 -now n -N matlab cd /u/home/FMRI/mscohen/alenarto; matlab -desktop  


This allocates 4G*8node memory for my session. And yet I get out-of-memory errors. Any thoughts on what's going on? The fMRIB plugin can definitely handle my data in matlab - b/c it runs fine on my 4GB laptop.

Any help would be appreciated... 
Agatha






From alenarto at ucla.edu  Fri Oct 15 15:17:55 2010
From: alenarto at ucla.edu (Agatha Lenartowicz)
Date: Fri, 15 Oct 2010 15:17:55 -0700
Subject: [Eeglablist] fmrib plugin on linux SGE cluster
Message-ID: <107187BB-0F78-40F2-B634-72E873CD2EA5@ucla.edu>

Hi all - I am having lot's of problems running the fmrib plugin on a linux machine - SGE cluster. I first see errors about mex files (which I then recompiled) - that are then replaced by out of memory errors.

The out of memory errors are not right (I've tried assigning 32 G to my job with no luck). The analysis works just fine on my laptop (OSX 4G memory - same dataset and same EEGLAB version (in EEGLAB 9_0_2_2b).

If anyone has seen this before - do let me know! Doing this cleaning on my home machine works but is mighty slow.
Best
Agatha



~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Agatha Lenartowicz, Ph.D.
Post Doctoral Scholar
Laboratory of M.S. Cohen
http://www.brainmapping.org/MarkCohen/lab.html

UCLA Center for Cognitive Neuroscience
760 Westwood Plaza, Suite 17-369
Los Angeles, CA 90095

http://alenarto.bol.ucla.edu/




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101015/87309bac/attachment.html>

From artiabhishek82 at rediffmail.com  Mon Oct 18 01:28:59 2010
From: artiabhishek82 at rediffmail.com (Arti Abhishek)
Date: 18 Oct 2010 08:28:59 -0000
Subject: [Eeglablist] =?utf-8?q?Neuroscan_averaged_files?=
Message-ID: <20101018082859.33943.qmail@f4mail213.rediffmail.com>

Dear list,I am new to EEG lab. I have some Neuroscan average files (.avg). I &nbsp;want to import the files to EEG lab and get the topographical distribution. How can I do this?Many thanks,Arti
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101018/4746ec7b/attachment.html>

From j.martinovic at liverpool.ac.uk  Mon Oct 18 03:31:04 2010
From: j.martinovic at liverpool.ac.uk (Jasna Martinovic)
Date: Mon, 18 Oct 2010 11:31:04 +0100
Subject: [Eeglablist] postdoctoral position at the University of Aberdeen
Message-ID: <4CBC21E8.7090007@liverpool.ac.uk>

Research Fellow

University of Aberdeen - School of Psychology, College of Life Sciences 
& Medicine

Job Summary: The project will investigate the extent to which luminance 
and chromatic visual pathways function independently or interactively at 
different stages of the visual processing hierarchy. A series of 
experiments will explore the roles played by luminance and chromatic 
signals in low-level, mid-level and high-level vision. The project will 
be supervised by Dr Jasna Martinovic. The research will be conducted 
using a 128-electrode Biosemi ActiveTwo electrode montage and two Visage 
stimulus generators, one set in the EEG lab and the other one in the 
psychophysics lab.

The candidiate must have a PhD in psychology or a related discipline. 
Experience in perception or cognition research is essential, and 
expertise in electroencephalographic (EEG) research, visual 
psychophysics and use of Matlab for stimulus presentation, data 
collection and signal processing would be an advantage.

The position is full time for two years, at grade 6 of the university 
structure (?29,853). The project is intended to start in March 2011 or 
soon thereafter.

Should you require a visa to undertake paid employment in the UK you 
will be required to fulfil the minimum point criteria to be granted a 
Certificate of Sponsorship and Tier 2 visa. As appropriate, at the time 
an offer of appointment is made you will be asked to demonstrate that 
you fulfil the criteria in respect of financial maintenance and 
competency in English. Please do not hesitate to contact Mrs Sacha 
MacLennan (sacha.maclennan at abdn.ac.uk)  for further information on this.

Closing Date: 10/11/2010

Further information is available online at:
https://atsv7.wcn.co.uk/saf/set_session.cgi?owner=5042238&ownertype=fair&jcode=1160766&posting_code=267&season=0&language=203&url=/search_engine/jobs.cgi

Informal enquiries are welcome and should be made to Dr Jasna Martinovic 
(Tel: 01224 272240, e-mail: j.martinovic at abdn.ac.uk).

To apply click here 
https://atsv7.wcn.co.uk/saf/login.cgi?owner=5042238&ownertype=fair&jcode=1160766&external=52500 
 
Job Reference: YPS238R

----
Dr Jasna Martinovic
School of Psychology
University of Aberdeen
William Guild Building
Aberdeen
AB24 2UB

tel: 01224 272240
email: j.martinovic @ abdn.ac.uk
web: http://www.abdn.ac.uk/~psy527/dept/




From sjluck at ucdavis.edu  Mon Oct 18 11:13:27 2010
From: sjluck at ucdavis.edu (Steve Luck)
Date: Mon, 18 Oct 2010 11:13:27 -0700
Subject: [Eeglablist] Public release of ERPLAB Toolbox 1.0.0
Message-ID: <DA850C9E-67AF-4150-B166-714D99DA993F@ucdavis.edu>

We are pleased to announce the release of ERPLAB Toolbox 1.0.0, a free, open-source package of ERP data analysis routines.  ERPLAB plugs into EEGLAB, taking advantage of EEGLAB's excellent user interface and EEG processing abilities.  ERPLAB is designed to provide industrial strength sorting of trials for averaging, artifact detection, and ERP post processing.  Like EEGLAB, it can be run from a GUI for ease of use or from Matlab scripts for power and flexibility. The first version focuses on high-quality implementations of the most basic ERP processing functions, and advanced functions will be added in the future.

ERPLAB Toolbox can be downloaded for free at http://erpinfo.org/erplab.  This site also provides information about a mailing list and online bulletin boards for asking questions, requesting features, and reporting bugs.

We would like to thank the National Institute of Mental Health for financial support, and Scott Makeig and Arnauld Delorme for their support in integrating ERPLAB with EEGLAB.

Steve Luck & Javier Lopez-Calderon
UC-Davis Center for Mind & Brain

--------------------------------------------------------------------
Steven J. Luck, Ph.D.
Director, Center for Mind & Brain
Professor, Department of Psychology
University of California, Davis
Room 127
267 Cousteau Place
Davis, CA 95618
(530) 297-4424
sjluck at ucdavis.edu
Web: http://mindbrain.ucdavis.edu/people/sjluck
Calendar: http://www.google.com/calendar/embed?src=stevenjluck%40gmail.com&ctz=America/Los_Angeles
--------------------------------------------------------------------







-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101018/606ef316/attachment.html>

From jainsanket1 at gmail.com  Wed Oct 20 16:52:06 2010
From: jainsanket1 at gmail.com (Sanket Jain)
Date: Wed, 20 Oct 2010 18:52:06 -0500
Subject: [Eeglablist] Neuroscan averaged files
In-Reply-To: <20101018082859.33943.qmail@f4mail213.rediffmail.com>
References: <20101018082859.33943.qmail@f4mail213.rediffmail.com>
Message-ID: <AANLkTintPAZ9dNXF++NBaRaePefgSZmEQK5_ypCo=QPf@mail.gmail.com>

I believe there is a loadavg command in eeglab which you can use to read
neuroscan (avg) files. Kindly read the documentation for the function. If
that does not work, then you will have to use conventional matlab commands
to parse through .avg file. I assume .avg file is a text file. If not look
for some option within neuroscan to save your file as a text file.

Hope it helps,
Sanket

On Mon, Oct 18, 2010 at 3:28 AM, Arti Abhishek <
artiabhishek82 at rediffmail.com> wrote:

> Dear list,
>
> I am new to EEG lab. I have some Neuroscan average files (.avg). I  want to
> import the files to EEG lab and get the topographical distribution. How can
> I do this?
>
> Many thanks,
> Arti
>
> <http://sigads.rediff.com/RealMedia/ads/click_nx.ads/www.rediffmail.com/signatureline.htm at Middle?>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101020/5cd4664b/attachment.html>

From poem.ece88 at nctu.edu.tw  Wed Oct 20 18:22:50 2010
From: poem.ece88 at nctu.edu.tw (poem)
Date: Thu, 21 Oct 2010 09:22:50 +0800
Subject: [Eeglablist] cutting up a dataset
In-Reply-To: <mailman.28.1287612933.3066.eeglablist@sccn.ucsd.edu>
References: <mailman.28.1287612933.3066.eeglablist@sccn.ucsd.edu>
Message-ID: <4CBF95EA.7080706@nctu.edu.tw>

To Agatha:
How about trying pop_select to select the desired interval? I think you 
can get "unmerged" datasets by repeating using pop_select

poem

eeglablist-request at sccn.ucsd.edu ??:
> ??:
> [Eeglablist] cutting up a dataset
> ???:
> Agatha Lenartowicz <alenarto at ucla.edu>
> ??:
> Wed, 13 Oct 2010 08:45:57 -0700
> ??? (To):
> eeglablist at sccn.ucsd.edu
>
> ??? (To):
> eeglablist at sccn.ucsd.edu
>
>
> Hi all - What is the inverse of eeg_mergeset?
>
> I have a continuous dataset - merged - that I'd like to unmerge. I haven't been able to find the function.
> Agatha
>   


From Lars.Michels at kispi.uzh.ch  Fri Oct 22 01:02:03 2010
From: Lars.Michels at kispi.uzh.ch (Michels Lars)
Date: Fri, 22 Oct 2010 10:02:03 +0200
Subject: [Eeglablist] Postdoctoral Fellowship in multimodal developmental
	neuroimaging
Message-ID: <FD1E6887BC99C341800CBFD7ABC2F9C20259B877@exzh1.kispi.int>

Postdoctoral Fellowship in multimodal developmental neuroimaging 

 

Applications are invited for a 2-years postdoctoral fellowship in the human multimodal neuroimaging project "Linking the major system markers for typical and atypical brain development: a multimodal imaging and spectroscopy study" (http://www.zihp.uzh.ch/1610.php#45) funded by the Z?rich Institute of Human Physiology. 

This study will investigate the major physiological markers of brain development, using a combination of advanced magnetic resonance imaging (e.g., functional MRI) and MR-spectroscopy methods. The initial phase of the study will establish baseline neurotransmitter levels, cerebral blood flow (e.g., perfusion MRI) and EEG frequency and power at rest across a range of age groups. Examining the interactions between these markers and the changes they demonstrate with age and hormone levels will allow to better understanding the global and regional processes underlying brain maturation. The later phases will investigate changes in these physiological markers with (a) cognitive tasks and (b) attention deficit hyperactivity disorder (ADHD). The starting date of the position is January 2011.

 

The successful applicant will have a PhD research background in Cognitive Neuroscience, Neurophysiology, Psychology, Neuropsychology, or related fields. Fluency in English, good oral and written German, and the ability to work within a multidisciplinary team are essential. Applicants should be experienced at conducting fMRI and/or EEG studies and be familiar with analysis software, such as SPM/Matlab, BrainVoyager and/or FSL. Experience with stimulus presentation software (such as Presentation), UNIX, and programming languages a plus. 

 

Salaries are in accordance with the Swiss National Research Foundation (starting at around 80'000 CHF p.a.). 

 

APPLICATION INSTRUCTIONS: To apply by e-mail (max. 5 MB), please send a curriculum vitae, a personal statement describing research interests, 3 letters of recommendations, and up to 5 article reprints/preprints to:

 

Dr Lars Michels

lars.michels at kispi.uzh.ch

MR-Zentrum

University Children's Hospital

Steinwiesstrasse 75

Z?rich 8032

Switzerland

 

Reviews of applications will begin on the 7th of November and will continue until the positions are filled.

 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101022/00a3fce8/attachment.html>

From smakeig at ucsd.edu  Tue Oct 26 18:53:23 2010
From: smakeig at ucsd.edu (Scott Makeig)
Date: Tue, 26 Oct 2010 18:53:23 -0700
Subject: [Eeglablist] Swartz Center EEG lab manager
Message-ID: <AANLkTim-4BprgLdNXnL8G=5SG9foE-=LODAYEQeoP2TC@mail.gmail.com>

The Swartz Center (SCCN) at UCSD in La Jolla, California, is looking
to hire a laboratories manager to develop and run two innovative EEG
laboratories incorporating a new Mobile Brain/Body Imaging (MoBI)
paradigm. The description below is from the UCSD job website.
Applicants should apply through the link given on this web page:

http://jobs.ucsd.edu/bulletin/job.aspx?cat=search&sortby=rank&jobnum_in=55596&search=55596

DESCRIPTION:
Under the general supervision of the Center Director, the candidate
will manage all aspects of two human biobehavioral research
laboratories organizing and executing all techniques involved. The
candidate is expected to ensure smooth functioning of the
laboratories, as well as making significant innovative contributions
pertaining to the design and overall direction of their equipment and
research projects and propose possible changes or new directions when
appropriate. The incumbent will be responsible for 1) investigating
new and current technologies in the field (EEG, motion capture, eye
tracking, audiovisual systems, etc., 2) managing equipment purchase,
calibration, maintenance, repairs, and replacements; 3) supervising,
training, and evaluating research staff, including directing and
supervising the work of laboratory technicians; 4) providing work
direction and teaching laboratory procedures to undergraduate and
graduate student research assistants, and assisting postdoctoral
scholars in their experimental research; 5) managing and performing
general laboratory maintenance; 6) supervising human subjects
approval, recruitment, and payment processes; 7) managing and
supervising data archiving; 8) contributing to project reports and
assisting with grant and manuscript development including editing
technical sections of manuscripts; 9) monitoring and reconciling
experimental program funds and budgets; 10) working with experimenters
to perform or manage preliminary analysis, including statistical
analysis using existing computerized methods. The candidate will also
monitor lab safety issues and take any corrective or preventative
actions necessary.

-- 
Scott Makeig, Research Scientist and Director, Swartz Center for
Computational Neuroscience, Institute for Neural Computation,
University of California San Diego, La Jolla CA 92093-0961,
http://sccn.ucsd.edu/~scott


From Antje.Nuthmann at ed.ac.uk  Wed Oct 27 11:17:51 2010
From: Antje.Nuthmann at ed.ac.uk (Antje Nuthmann)
Date: Wed, 27 Oct 2010 20:17:51 +0200
Subject: [Eeglablist] readership (associate professorship) in psychology at
	University of Edinburgh, UK
Message-ID: <C4078E0B-CB7C-4BB0-9D9C-9D085A0E5CFB@ed.ac.uk>


Please forward this message to anyone who might be interested in this opportunity

---------------------------------------------------------------------------------
readership (associate professorship) in psychology at University of Edinburgh, UK
application deadline: 09-Nov-2010

Note: The department is required to fill the position within the realm of cognitive science. 

We have a new EEG lab:
http://www.ppls.ed.ac.uk/staff/resources/eeg_erp.php
We also have extensive eye-tracking facilities:
http://www.ppls.ed.ac.uk/staff/resources/eye_trackers.php

official advert:
The University of Edinburgh, home to one of the leading Psychology Departments in the UK, seeks to appoint a Reader in cognitive aspects of psychology, broadly defined. The successful applicant will have an international profile and will have a track record of high-impact publications since completing his/her Ph.D. Applications are invited from all qualified persons, but we are seeking to build on three existing international strengths of the Edinburgh Psychology Department: Differential Psychology, Psychology of Language, and Human Cognitive Neuroscience.
Tenable from March 2011.

further details and application pack:
http://www.jobs.ed.ac.uk/vacancies/index.cfm?fuseaction=vacancies.detail&vacancy_ref=3013493


--
Dr Antje Nuthmann
Psychology Department
University of Edinburgh, UK
E-Mail: Antje.Nuthmann at ed.ac.uk
web: www.nuthmann.de/antje
phone: +44 (0)131 650 3459

The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.






-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101027/ca22b93c/attachment.html>

From williamjjonesjr at gmail.com  Mon Nov  1 14:05:58 2010
From: williamjjonesjr at gmail.com (William Jones)
Date: Mon, 1 Nov 2010 17:05:58 -0400
Subject: [Eeglablist] fractional area latency
Message-ID: <AANLkTimK35+ty0NPp_QfEb7LT1jhdaRrUR4uR73RKCpm@mail.gmail.com>

Hi all,

Is is possible to do fractional area latency in EEGLAB? If so, does
anyone have a suggestion as to how to get started? If not, can someone
recommend a program with these capabilities. Unfortunately, it appears
I cannot carryout this analysis with my Neuroscan software.

Thanks in advance,
Billy


From isacor at us.es  Tue Nov  2 08:47:57 2010
From: isacor at us.es (isacor at us.es)
Date: Tue, 02 Nov 2010 16:47:57 +0100
Subject: [Eeglablist] newcrossf error
Message-ID: <f53bb33758c47.4cd040bd@us.es>

Hello everyone:
I tried to compute the cross-coherence between twochannels using eeglab 8.0.3.5b. My epoch time range is from -1800 ms to 1200 ms. The parameters used are:
cycles: [1.5 0.5]
padratio:4
baseline: [-1300 -1200]
alpha: 0.01

I returned the following error:

EEGLAB error in function newcrossf() at line 1272:
Undefined function or variable "baseln".

If I do not use the baseline I have no problems. But I'm interested to use the baseline.
With the current version of eeglab get the same error but at line 854.

Can somebody help me?

Thanks

Isabel Cordones
Fisiolog?a Animal Y Zoologia
Neurociencia y Comportamiento
Facultad de Biolog?a, Universidad de Sevilla
Avda. Reina Mercedes 6, 41012-Sevilla




From jordicostafa at gmail.com  Sat Oct 30 09:54:53 2010
From: jordicostafa at gmail.com (Jordi Costa Faidella)
Date: Sat, 30 Oct 2010 12:54:53 -0400
Subject: [Eeglablist] ICA on baseline corrected epochs
In-Reply-To: <AANLkTim-4BprgLdNXnL8G=5SG9foE-=LODAYEQeoP2TC@mail.gmail.com>
References: <AANLkTim-4BprgLdNXnL8G=5SG9foE-=LODAYEQeoP2TC@mail.gmail.com>
Message-ID: <73A80331-69F9-42EB-A611-55EA980E58EC@gmail.com>

Dear EEGlab users,

I have a question regarding the use of ICA on baseline corrected epochs. Playing around with a dataset, I realized that the ICA doesn't retrieve exactly the same results when I apply it on a continuous EEG file or in concatenated, baseline corrected epochs. At the beginning I thought it was only a matter of using different number of data points, or the information in the datapoints per se. Thus, I tried the following: I extracted and concatenated epochs from a single dataset in two different ways, one without and one with baseline correction. The reason I did that it was because, as far as I now, when data is baseline corrected we only can look at the differences in scalp topography between the baseline period and a given time period, but not the "real" topography that was recorded in the original continuous file. The results I obtained differed slightly in components with high weights, like artifacts etc., but differed quite a lot in smaller ICs. Thus, I'm wondering whether it would be more correct to apply ICA on a continuous dataset, or at least in concatenated epochs without baseline correction, than to apply it on a set of single baseline corrected epochs.
Does anyone know what is better? Or is it the same and my results were due to an error (unknown to me)?

thank you all,

Jordi


From brice.rebsamen at gmail.com  Wed Nov  3 21:31:56 2010
From: brice.rebsamen at gmail.com (brice rebsamen)
Date: Thu, 4 Nov 2010 12:31:56 +0800
Subject: [Eeglablist] Sleep scoring (PING)
Message-ID: <AANLkTi=WH5O_4-z8tpMGnkF6o-Sq5oqbDRXG4dgd2Pbi@mail.gmail.com>

I am also interested in that, so I am asking the same question again,
I hope this time there will be an answer...

Is sleep scoring at all possible with EEGLAB? If not, is there someone
that might be able to point towards a freely available software or
matlab code for sleep scoring?
Any information is appreciated.

Brice

On Tue, Sep 28, 2010 at 11:10 PM, Nicholas Perentos <nperentos at gmail.com> wrote:
> Hi there,
> Is sleep scoring at all possible with EEGLAB? If not, is there someone that
> might be able to point towards a freely available software or matlab code
> for sleep scoring?
> Any information is appreciated
> Thanks
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>


From brian.murphy at unitn.it  Thu Nov  4 04:25:46 2010
From: brian.murphy at unitn.it (Brian Murphy)
Date: Thu, 4 Nov 2010 12:25:46 +0100
Subject: [Eeglablist] Post-doc position in computational neuroscience of
	language
Message-ID: <4CD2983A.8020700@unitn.it>

[this posting concerns decoding from neuroimaging data, so I hope it is appropriate for the list]


The CLIC laboratory of the Center for Mind/Brain Sciences (CIMeC) of
the University of Trento announces the availability of:

* Post-doc position in computational neuroscience of language *


* Research environment *

The Language, Interaction and Computation lab (clic.cimec.unitn.it) is
a unit of the University of Trento's Center for Mind/Brain Sciences
(www.cimec.unitn.it) or CIMEC: an interdisciplinary center for the
research in brain and cognition including neuroscientists,
psychologists, (computational) linguists, computational
neuroscientists, and physicists. CLIC consists of researchers from the
Departments of Computer Science and Cognitive Science carrying out
research on a range of topics, including concept acquisition and
information extraction from very large multimodal corpora, combining
brain data and data from corpora to study cognition, and methods of
theoretical linguistics.


* Post-doc position in computational neuroscience of language *

A 2-year post-doctoral position in the computational neuroscience of
language with a focus on the organization of conceptual knowledge in
the brain will soon become available at CIMeC/CLIC.

The successful candidate will work as part of a larger project whose
objective is to combine empirical data of different types (corpus
co-occurrence patterns, elicitation experiments, neuroimaging data) to
arrive at a better understanding of the organization of conceptual
knowledge in the mind and brain. Your task will be to continue
on-going work which uses machine learning methods to extract
conceptual representations from recordings of neural activity (EEG,
MEG and fMRI).


The candidate should have strong technical knowledge of either computational linguistics or brain-decoding methods, and familiarity with theories of the lexicon. Programming skills are a must, and experience with neuroimaging techniques, experimental design (elicitation/behavioural) and signal processing would be a plus.


* How to apply *

For additional information please send an expression of interest (with
CV) to Brian Murphy (brian.murphy at unitn.it)



From drziaulhaq at gmail.com  Thu Nov  4 00:36:58 2010
From: drziaulhaq at gmail.com (zia haq)
Date: Thu, 4 Nov 2010 13:06:58 +0530
Subject: [Eeglablist] Sleep scoring (PING)
In-Reply-To: <AANLkTi=WH5O_4-z8tpMGnkF6o-Sq5oqbDRXG4dgd2Pbi@mail.gmail.com>
References: <AANLkTi=WH5O_4-z8tpMGnkF6o-Sq5oqbDRXG4dgd2Pbi@mail.gmail.com>
Message-ID: <AANLkTi=v+kh7rvZAO-ntSvciwA-jGTp1VL1dgwQ7n5xc@mail.gmail.com>

Hi,

It would really be interesting to look for sleep scoring with Matlab?EEGLab.

I have been using some programmes (paid) but these are inefficient and one
has to go for manual scoring invariably.

Any info/ideas would be appreciated.

Cheers

On Thu, Nov 4, 2010 at 10:01 AM, brice rebsamen <brice.rebsamen at gmail.com>wrote:

> I am also interested in that, so I am asking the same question again,
> I hope this time there will be an answer...
>
> Is sleep scoring at all possible with EEGLAB? If not, is there someone
> that might be able to point towards a freely available software or
> matlab code for sleep scoring?
> Any information is appreciated.
>
> Brice
>
> On Tue, Sep 28, 2010 at 11:10 PM, Nicholas Perentos <nperentos at gmail.com>
> wrote:
> > Hi there,
> > Is sleep scoring at all possible with EEGLAB? If not, is there someone
> that
> > might be able to point towards a freely available software or matlab code
> > for sleep scoring?
> > Any information is appreciated
> > Thanks
> > _______________________________________________
> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> > To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> > For digest mode, send an email with the subject "set digest mime" to
> > eeglablist-request at sccn.ucsd.edu
> >
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Dr Mohammad Zia Ul Haq
Senior Resident
Centre for Cognitive Neurosciences
Central Institute of Psychiatry
Ranchi, India - 834006
Mb: +919234687231
Fax: +916512233668
Alternate email: drziaulhaq at rediffmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101104/d0e8fe96/attachment.html>

From jdesjardins at brocku.ca  Thu Nov  4 11:25:34 2010
From: jdesjardins at brocku.ca (James Desjardins)
Date: Thu, 04 Nov 2010 14:25:34 -0400
Subject: [Eeglablist] Sleep scoring (PING)
In-Reply-To: <AANLkTi=WH5O_4-z8tpMGnkF6o-Sq5oqbDRXG4dgd2Pbi@mail.gmail.com>
References: <AANLkTi=WH5O_4-z8tpMGnkF6o-Sq5oqbDRXG4dgd2Pbi@mail.gmail.com>
Message-ID: <20101104142534.5v5ksmt64gw44wc0@webmail.brocku.ca>

Hi Brice and Nicholas,

If it is a matter of manually inserting custom event markers via  
display of the continuous EEG recording it can be done with EEGLAB.

Use the menu option: Edit > Visually edit events and identify bad channels

This will open the eegplot function with the option of manually  
inserting event markers. Simply CTRL-leftclick on the eegplot figure  
window where you would like to insert an event. This will bring up a  
UI for the insertion of a new event.

I hope that this helps.

James Desjardins
Technician, MA Student
Department of Psychology, Behavioural Neuroscience
Cognitive and Affective Neuroscience Lab
Brock University
500 Glenridge Ave.
St. Catharines, ON, Canada
L2S 3A1
905-688-5550 x4676


Quoting brice rebsamen <brice.rebsamen at gmail.com>:

> I am also interested in that, so I am asking the same question again,
> I hope this time there will be an answer...
>
> Is sleep scoring at all possible with EEGLAB? If not, is there someone
> that might be able to point towards a freely available software or
> matlab code for sleep scoring?
> Any information is appreciated.
>
> Brice
>
> On Tue, Sep 28, 2010 at 11:10 PM, Nicholas Perentos   
> <nperentos at gmail.com> wrote:
>> Hi there,
>> Is sleep scoring at all possible with EEGLAB? If not, is there someone that
>> might be able to point towards a freely available software or matlab code
>> for sleep scoring?
>> Any information is appreciated
>> Thanks
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to  
>  eeglablist-request at sccn.ucsd.edu
>






From simonshlomo.poil at cncr.vu.nl  Thu Nov  4 01:10:22 2010
From: simonshlomo.poil at cncr.vu.nl (Simon-Shlomo Poil)
Date: Thu, 4 Nov 2010 09:10:22 +0100
Subject: [Eeglablist] Sleep scoring (PING)
In-Reply-To: <AANLkTi=WH5O_4-z8tpMGnkF6o-Sq5oqbDRXG4dgd2Pbi@mail.gmail.com>
References: <AANLkTi=WH5O_4-z8tpMGnkF6o-Sq5oqbDRXG4dgd2Pbi@mail.gmail.com>
Message-ID: <AANLkTi=tuWgQnSwA8ar8TpDH7ArOfHmtgBmsDoxak-Lw@mail.gmail.com>

Dear Brice,

Yes, there is the fast toolbox for spm.
http://www.montefiore.ulg.ac.be/~phillips/FAST.html

I do not know if any sleep scoring functions for eeglab, but probably
the fast toolbox can be integrated with eeglab?

good luck
Simon

-- 
Simon-Shlomo Poil

Neuronal Oscillations and Cognition Group (NOC)
Department of Integrative Neurophysiology (INF)
Center for Neurogenomics and Cognitive Research (CNCR)
Neuroscience Campus Amsterdam
VU University Amsterdam
De Boelelaan 1085, Room B-435
1081 HV Amsterdam, The Netherlands

E-mail: simonshlomo.poil at cncr.vu.nl
Phone: +31 20 5989408
Webpage: http://www.poil.dk and http://www.cncr.nl


2010/11/4 brice rebsamen <brice.rebsamen at gmail.com>:
> I am also interested in that, so I am asking the same question again,
> I hope this time there will be an answer...
>
> Is sleep scoring at all possible with EEGLAB? If not, is there someone
> that might be able to point towards a freely available software or
> matlab code for sleep scoring?
> Any information is appreciated.
>
> Brice
>
> On Tue, Sep 28, 2010 at 11:10 PM, Nicholas Perentos <nperentos at gmail.com> wrote:
>> Hi there,
>> Is sleep scoring at all possible with EEGLAB? If not, is there someone that
>> might be able to point towards a freely available software or matlab code
>> for sleep scoring?
>> Any information is appreciated
>> Thanks
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
>



-- 
Simon-Shlomo Poil

Neuronal Oscillations and Cognition Group (NOC)
Department of Integrative Neurophysiology (INF)
Center for Neurogenomics and Cognitive Research (CNCR)
Neuroscience Campus Amsterdam
VU University Amsterdam
De Boelelaan 1085, Room B-435
1081 HV Amsterdam, The Netherlands

E-mail: simonshlomo.poil at cncr.vu.nl
Phone: +31 20 5989408
Webpage: http://www.poil.dk and http://www.cncr.nl


From arno at ucsd.edu  Thu Nov  4 14:37:08 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Thu, 4 Nov 2010 14:37:08 -0700
Subject: [Eeglablist] fractional area latency
In-Reply-To: <AANLkTimK35+ty0NPp_QfEb7LT1jhdaRrUR4uR73RKCpm@mail.gmail.com>
References: <AANLkTimK35+ty0NPp_QfEb7LT1jhdaRrUR4uR73RKCpm@mail.gmail.com>
Message-ID: <1DDEACEF-5345-4BE9-9E8E-EB7150ED6169@ucsd.edu>

Dear William,

there is a contributed function called "eeg_amplitudearea" that is only available from the command line.
Type "help eeg_amplitudearea" on the Matlab command line to see if it fits your needs.

Best regards,

A. Delorme

On Nov 1, 2010, at 2:05 PM, William Jones wrote:

> Hi all,
> 
> Is is possible to do fractional area latency in EEGLAB? If so, does
> anyone have a suggestion as to how to get started? If not, can someone
> recommend a program with these capabilities. Unfortunately, it appears
> I cannot carryout this analysis with my Neuroscan software.
> 
> Thanks in advance,
> Billy
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu




From arno at ucsd.edu  Thu Nov  4 14:39:23 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Thu, 4 Nov 2010 14:39:23 -0700
Subject: [Eeglablist] ICA on baseline corrected epochs
In-Reply-To: <73A80331-69F9-42EB-A611-55EA980E58EC@gmail.com>
References: <AANLkTim-4BprgLdNXnL8G=5SG9foE-=LODAYEQeoP2TC@mail.gmail.com>
	<73A80331-69F9-42EB-A611-55EA980E58EC@gmail.com>
Message-ID: <9C151526-E455-4891-B6DF-9C54D7C18A2E@ucsd.edu>

Dear Jordi,

according to our test and a paper by David Groppe, if the baseline is too short (~100 ms) ICA component reliability degrades critically. This is because you are introducing non-linearity by removing the baseline (if there are common generators that could be isolated by ICA and project linearly to each channel, you are making it harder for ICA by subtracting a constant from each channel activity - since this common source cannot project linearly to all channel any more).

Groppe, D.M., Makeig, S., & Kutas, M. (2009) Identifying reliable independent components via split-half comparisons. NeuroImage, 45 pp.1199-1211.

I would personally recommend a baseline period of 1 second.
Hope this helps,

Arno

On Oct 30, 2010, at 9:54 AM, Jordi Costa Faidella wrote:

> Dear EEGlab users,
> 
> I have a question regarding the use of ICA on baseline corrected epochs. Playing around with a dataset, I realized that the ICA doesn't retrieve exactly the same results when I apply it on a continuous EEG file or in concatenated, baseline corrected epochs. At the beginning I thought it was only a matter of using different number of data points, or the information in the datapoints per se. Thus, I tried the following: I extracted and concatenated epochs from a single dataset in two different ways, one without and one with baseline correction. The reason I did that it was because, as far as I now, when data is baseline corrected we only can look at the differences in scalp topography between the baseline period and a given time period, but not the "real" topography that was recorded in the original continuous file. The results I obtained differed slightly in components with high weights, like artifacts etc., but differed quite a lot in smaller ICs. Thus, I'm wondering whether!
>  it would be more correct to apply ICA on a continuous dataset, or at least in concatenated epochs without baseline correction, than to apply it on a set of single baseline corrected epochs.
> Does anyone know what is better? Or is it the same and my results were due to an error (unknown to me)?
> 
> thank you all,
> 
> Jordi
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu




From arno at ucsd.edu  Fri Nov  5 04:16:21 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Fri, 5 Nov 2010 04:16:21 -0700
Subject: [Eeglablist] newcrossf error
In-Reply-To: <f53bb33758c47.4cd040bd@us.es>
References: <f53bb33758c47.4cd040bd@us.es>
Message-ID: <23613928-E8FA-4F23-8ADB-91393559377B@ucsd.edu>

Dear Isabel,

this problem has been fixed in EEGLAB 9.x.
Alternatively, you can just download the patched version of newcrossf

http://sccn.ucsd.edu/svn/software/eeglab/functions/timefreqfunc/newcrossf.m

Best regards,

Arno


On Nov 2, 2010, at 8:47 AM, isacor at us.es wrote:

> Hello everyone:
> I tried to compute the cross-coherence between twochannels using eeglab 8.0.3.5b. My epoch time range is from -1800 ms to 1200 ms. The parameters used are:
> cycles: [1.5 0.5]
> padratio:4
> baseline: [-1300 -1200]
> alpha: 0.01
> 
> I returned the following error:
> 
> EEGLAB error in function newcrossf() at line 1272:
> Undefined function or variable "baseln".
> 
> If I do not use the baseline I have no problems. But I'm interested to use the baseline.
> With the current version of eeglab get the same error but at line 854.
> 
> Can somebody help me?
> 
> Thanks
> 
> Isabel Cordones
> Fisiolog?a Animal Y Zoologia
> Neurociencia y Comportamiento
> Facultad de Biolog?a, Universidad de Sevilla
> Avda. Reina Mercedes 6, 41012-Sevilla
> 
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu




From nickbedo at yahoo.com  Mon Nov  8 23:16:06 2010
From: nickbedo at yahoo.com (Nick Bedo)
Date: Mon, 8 Nov 2010 23:16:06 -0800 (PST)
Subject: [Eeglablist] STUDY Cross-Coherence
Message-ID: <3833.78389.qm@web62005.mail.re1.yahoo.com>

Hi Everyone,

I've been looking a channel cross-coherence (via newcrossf) for individual 
participants; but now that I have stepped up to the STUDY structure, I'm unsure 
as to how I can properly use the functions to find coherence values for 
electrode pairs across participants.

Any input is greatly appreciated.

Thanks in advance,
Nick


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101108/81a56b9a/attachment.html>

From mparvaz at gmail.com  Wed Nov 10 14:33:18 2010
From: mparvaz at gmail.com (MP)
Date: Wed, 10 Nov 2010 17:33:18 -0500
Subject: [Eeglablist] Electrode Location (.LOCS) file for Neuroscan 64
	channel electrode cap
Message-ID: <AANLkTin1fxVFDRUHqfsOuSwuCgr=DmP88AnHZJVZkSYi@mail.gmail.com>

Hello all,

In our experiments we don't digitize the channel location, however to do ICA
or LORETA, a channel location file (.locs) is needed. I was wondering if
there is an approximate .locs file for Neuroscan 64 electrode cap, and if
someone would kindly share it with me.

Thanks

- Muhammad
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101110/c74fd054/attachment.html>

From marco.buiatti at gmail.com  Tue Nov  9 02:32:00 2010
From: marco.buiatti at gmail.com (Marco Buiatti)
Date: Tue, 9 Nov 2010 11:32:00 +0100
Subject: [Eeglablist] ADJUST: automatic algorithm for EEG artifact removal
In-Reply-To: <AANLkTinCUXBxmyNbb9o4KMp6+2DjYXrkaP_sT8wccW1M@mail.gmail.com>
References: <AANLkTinCUXBxmyNbb9o4KMp6+2DjYXrkaP_sT8wccW1M@mail.gmail.com>
Message-ID: <AANLkTimKXHuecTszuGXmZ_QBx1Ka0hZ03qQOAqYVRaUb@mail.gmail.com>

Dear EEGLAB users,

I would like to present you ADJUST, an ICA-based completely automatic
algorithm for efficient removal of the most common physiological
artifacts in EEG data.

ADJUST identifies artifacted Independent Components (IC) of EEG data
by combining stereotyped artifact-specific spatial and temporal
features. Features are optimised to capture blinks, eye movements and
generic discontinuities. Once artifacted IC are identified, they can
be simply removed from the data while leaving the activity due to
neural sources almost unaffected.

The algorithm is described in detailed and validated on real data in
this recent publication:
Mognon, Jovicich, Bruzzone, Buiatti, "ADJUST: An automatic EEG
artifact detector based on the joint use of spatial and temporal
features", Psychophysiology, in press.

ADJUST has been implemented as a fully functional EEGLAB plugin, and
is downloadable here (together with the associated paper, a tutorial
and a sample dataset):

http://www.unicog.org/pm/pmwiki.php/MEG/RemovingArtifactsWithADJUST

This page is also linked in the EEGLAB plugin page:
http://sccn.ucsd.edu/wiki/EEGLAB_Plugins

Feel free to contact adjust.staff at gmail.com should you have any
questions on the algorithm and on its application on your data.

Best,

Marco


--
Marco Buiatti, PhD

CEA/DSV/I2BM / NeuroSpin
INSERM U992 - Cognitive Neuroimaging Unit
B?t 145 - Point Courrier 156
Gif sur Yvette F-91191  FRANCE
Ph:  +33(0)169.08.65.21
Fax: +33(0)169.08.79.73
E-mail: marco.buiatti at gmail.com
http://www.unicog.org/pm/pmwiki.php/Main/MarcoBuiatti

***********************************************



From gangadhar.garipelli at epfl.ch  Fri Nov  5 04:50:44 2010
From: gangadhar.garipelli at epfl.ch (Gangadhar Garipelli)
Date: Fri, 05 Nov 2010 12:50:44 +0100
Subject: [Eeglablist] How to deal with 1/f noise in the low frequency
 oscillations for on-line experiments?
Message-ID: <4CD3EF94.1020408@epfl.ch>

Dear all,

I work with low frequency oscillations of the brain while a human
subject is cognitively engaged in a task. From the off-line analysis
(using zero-phase band pass FIR filters on full-band EEG), I discovered
that task-related cognitive signals are located in the range of [0.2
0.3]Hz and in [0.6 0.8]Hz. The fluctuations/oscillations ( formally
called very low frequency oscillations VLFO, or infra slow oscillations
ISO) below 0.2Hz are REAL devil due to 1/f nature. The noise power is
>100 times higher than signal's power.

Now as per my experimental demands, I need to estimate on-line in
real-time the signals mentioned in the above range and manipulate
stimulus presentation. Ideally, this eventually means I need to have a
very sharp high pass filter with almost zero group/phase delay. Which
sounds impossible!

However, I should come up with a decent trade-off between SNR and
phase-delay. Do you have any suggestions? All suggestions ranging from
signal processing/machine-learning to hardware to solve this problem are
most welcome!

Thanks in advance!
Sincerely,
-- 
Gangadhar GARIPELLI,
Doctoral student,
EPFL, Switzerland



From demiral.007 at googlemail.com  Wed Nov 10 13:19:43 2010
From: demiral.007 at googlemail.com (Baris Demiral)
Date: Wed, 10 Nov 2010 21:19:43 +0000
Subject: [Eeglablist] Conducting ICA on correct or all epochs
Message-ID: <AANLkTin6+hTbFq2Y=kQQE=_ZUiPkihcbgeDegwjGeNGb@mail.gmail.com>

Hi everyone,

I am running a simple EEG experiment where I measure reaction times and
accuracies.

I want to use ICs for artifact removal, and I will report only the correct
trials (hits).

So would it be better to use the correct epochs for the ICA to correct for
the artifacts or is it OK to use all the epochs to detect the artifacts and
then run the artifact correction (pop_subcomp) followed by deleting the
incorrect epochs?

Thanks,
Baris


-- 
SB Demiral, PhD.
Department of Psychology
7 George Square
The University of Edinburgh
Edinburgh, EH8 9JZ
UK
Phone: +44 (0131) 6503063
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101110/9f28cddf/attachment.html>

From ross.avila at gmail.com  Mon Nov  8 19:00:34 2010
From: ross.avila at gmail.com (=?ISO-8859-1?Q?Ross_=C1vila?=)
Date: Mon, 8 Nov 2010 22:00:34 -0500
Subject: [Eeglablist] Alpha asymmetry spectral analysis
Message-ID: <AANLkTikyrpBe-01Wn9_mfbKDYNABp=uwYQ3mO8cda_x7@mail.gmail.com>

Hi everyone,

I am a graduate student working on an alpha asymmetry study and I am
struggling with the spectral analysis.

I have collected 8 minutes of resting EEG divided into 8 60-second
blocks.  Upon loading the data into EEGLAB, it recognizes a single
8-minute long epoch.  I'm just confused about what I need to do at
this point, and there is no one at the University of South Florida
that researches alpha asymmetry (they all do ERP studies here).  I
have little to no experience in MATLAB (part of what appealed to me
about the EEGLAB interface), so any help in the form of coding might
need to be "dumbed" down and pretty basic.

I am struggling specifically with the following questions:
1) Do I need to extract epochs? If so, how?
2) How do I actually get the alpha asymmetry?  I've read about using
FFT, but don't know how to actually do this.  Also read about
"windowing" but again do not understand exactly what this means.

Any help would be immensely appreciated.

Thank you,
Ross

-- 
Ross T. ?vila
Cognitive Neuropsychology Lab
Doctoral Program in Clinical Psychology
University of South Florida



From bkuhr at uni-osnabrueck.de  Tue Nov  9 06:18:09 2010
From: bkuhr at uni-osnabrueck.de (Benjamin Kuhr)
Date: Tue, 9 Nov 2010 15:18:09 +0100 (CET)
Subject: [Eeglablist] Out of memory
Message-ID: <52957.131.173.133.29.1289312289.squirrel@myuos.uni-osnabrueck.de>

Hi,

I tried to import a large *.bdf file, 614 MB. I get the "Out of memory" -
message, but not on all computers and for some reason it seems as it does
not depend on the actual memory. It works on a weaker system, but not on
the current one, which has 3.8 GiB of memory and a Intel Core 2 Duo CPU
R7500 @ 2.93 GHz.

Even on identical systems with the same hardware and same software I can
load the file only on one of them. I tired it with Windows XP and Ubuntu
10 on the system mentioned above, no difference. Any suggestions?

Thanks in advance
Benjamin Kuhr




From Beatrice.Jobst at ait.ac.at  Thu Nov  4 04:20:43 2010
From: Beatrice.Jobst at ait.ac.at (Jobst Beatrice)
Date: Thu, 4 Nov 2010 12:20:43 +0100
Subject: [Eeglablist] Activity Power Spectrum with funktion "pop_prop()"
Message-ID: <9F69795E29C890408AC2DAF646C89BB37998BF7452@MAILBOX.arc.local>

Dear All,

I have a question about plotting the activity power spectrum (frequency vs. power) with the function "pop_prop()":
I have a dataset, which has already been examined with another software, and now I want to compare the results with EEGLab. My problem is, that the calculations for the Power Spectrum in EEGLab produces completely different and also unrealistic results, the function values are completely different ones.
Where could be the problem? Is it possible, that the results are wrong because I don't have the Signal Processing Toolbox for Matlab? How can I face this problem without buying the toolbox? Can there be other reasons for the problem?

Thank you for your help!!

Best regards,
Beatrice Jobst
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101104/97758aed/attachment.html>

From bradley.voytek at gmail.com  Wed Nov 10 20:49:34 2010
From: bradley.voytek at gmail.com (Bradley Voytek)
Date: Wed, 10 Nov 2010 20:49:34 -0800
Subject: [Eeglablist] Alpha asymmetry spectral analysis
In-Reply-To: <AANLkTikyrpBe-01Wn9_mfbKDYNABp=uwYQ3mO8cda_x7@mail.gmail.com>
References: <AANLkTikyrpBe-01Wn9_mfbKDYNABp=uwYQ3mO8cda_x7@mail.gmail.com>
Message-ID: <318EE8D8-F9C4-47F7-864A-1104433E0EB7@gmail.com>

Ross:

I'll gladly skype chat with you. Much easier to explain that way than over email.

Or if you're going to San Diego for SfN we can chat there.

::brad



On Nov 8, 2010, at 19:00, Ross ?vila <ross.avila at gmail.com> wrote:

> Hi everyone,
> 
> I am a graduate student working on an alpha asymmetry study and I am
> struggling with the spectral analysis.
> 
> I have collected 8 minutes of resting EEG divided into 8 60-second
> blocks.  Upon loading the data into EEGLAB, it recognizes a single
> 8-minute long epoch.  I'm just confused about what I need to do at
> this point, and there is no one at the University of South Florida
> that researches alpha asymmetry (they all do ERP studies here).  I
> have little to no experience in MATLAB (part of what appealed to me
> about the EEGLAB interface), so any help in the form of coding might
> need to be "dumbed" down and pretty basic.
> 
> I am struggling specifically with the following questions:
> 1) Do I need to extract epochs? If so, how?
> 2) How do I actually get the alpha asymmetry?  I've read about using
> FFT, but don't know how to actually do this.  Also read about
> "windowing" but again do not understand exactly what this means.
> 
> Any help would be immensely appreciated.
> 
> Thank you,
> Ross
> 
> -- 
> Ross T. ?vila
> Cognitive Neuropsychology Lab
> Doctoral Program in Clinical Psychology
> University of South Florida
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu



From tarikbelbahar at gmail.com  Wed Nov 10 21:06:25 2010
From: tarikbelbahar at gmail.com (Tarik S Bel-Bahar)
Date: Wed, 10 Nov 2010 21:06:25 -0800
Subject: [Eeglablist] Alpha asymmetry spectral analysis
In-Reply-To: <AANLkTikyrpBe-01Wn9_mfbKDYNABp=uwYQ3mO8cda_x7@mail.gmail.com>
References: <AANLkTikyrpBe-01Wn9_mfbKDYNABp=uwYQ3mO8cda_x7@mail.gmail.com>
Message-ID: <AANLkTi=yxZhErdwdeDY9pR3081jmR0ezdbF_S89LSVJH@mail.gmail.com>

Greetings Ross,

I had similar questions a while back, and hopefully the below helps.
Good luck with articulating your data analysis path!
You may also want to consider other types of asymmetries, other
spectral bands, ratios between bands, and source
estimation, if your data is dense enough.

Cheers, Tarik


**********START****************************

First off, read through all of the eeglab documentation, with some
focus on importing data
and on having eeglab detect or import events.

Next, make sure that you are saving the events inside or with your raw
EEG file, so as to make
sure there is some way for that eeglab can detect them, or that you
can use the extra epochs tool on later.
If you have no "events", try to make sure that your file has segment
start events or markers saved into it.
There are other options depending on the system you are using. Eeglab
should be able to read any extra event information in your eeg files.

Next, search in the EEGLAB list archives for terms with alpha
asymmetry. You can download them all to one folder, and then search
for the text string of interest across the whole archives. I think
it's the archives are also searchable online. [there is some alpha
asymmetry information there, with code suggestions].
You may also want to specifically contact a researcher who has
published an alpha asymmetry article with EEGLAB.
Also, see the eeglab help documentation about running?spectopo and
similar time-frequency functions.
Within eeglab, why don't you take a look at some of the code below. Good luck!


[spec439 frq439] = spectopo (EEG.data, 250, 250, 'overlap', 125,
'winsize', 250,'plot', 'off', 'chanlocs', 'GSN129.sfp');
xlswrite('spec439.xls', spec439(:,1:256));
xlswrite('frq439.xls', frq439);

or

left= 24;
right = 25;
[spec_24 freqs ] = spectopo(EEG.data(right,:), 250, 250, 'plot', 'off')
[spec_25 freqs ] = spectopo(EEG.data(left,:), 250, 250, 'plot', 'off')
[tmp alpha_ind] = min( abs(freqs-8));
asymmetry1 = spec_24(alpha_ind) - spec_25(alpha_ind);
asymmetry2 = spec_25(alpha_ind) - spec_24(alpha_ind);
f439= fopen('439.xls', 'w+');
dlmwrite('439.xls', asymmetry1);


***I've also include a sample MATLAB non-EEGLAB function from John Allen below,
which he was kind enough to share when I had questions similar to yours.

function [fs,pwr] = geeg_block_fft(time_series,hz,win)
%GEEG_BLOCK_FFT - get power for multiple time series
%
%?? Syntax: [fs,pwr] = geeg_block_fft(time_series,hz,win)
%
%?? Input:
%?????? time_series - [nPoints x nSeries] time series
%?????? hz - sampling rate
%?????? win - window to apply (e.g. hamming(2048))
%?? Output:
%?????? fs - frequencies
%?????? pwr - [nFFTPoints x nSeries] power at each frequency

[nPoints, nSeries] = size(time_series);

%% Windowing
% Check window size against block size
if (size(win,2) < nSeries)
??? % repeat window to make matrix the same size as time_series
??? win = repmat(win,1,nSeries);
end
% Apply window
time_series = time_series.*win;

%% Compute FFT, power
% Take fft
FFTX = fft(time_series);
% Calculate the number of unique points
NumUniquePts = ceil((nPoints+1)/2);
% FFT is symmetric, throw away second half
FFTX = FFTX(1:NumUniquePts,:);
% Take the magnitude of fft of x
MX = abs(FFTX);
% Scale the fft so that it is not a function of the length of x
MX = MX./nPoints;
% Take the square of the magnitude of fft of x.
pwr = MX.^2;
% Multiply by 2 because you threw out second half of FFTX above
pwr = pwr.*2;
% DC Component should be unique, i.e. undo multiply by 2.
pwr(1,:) = pwr(1,:)./2;
% Nyquist component should also be unique.
if ~rem(nPoints,2)
?? % Here NFFT is even; therefore, Nyquist point is included.
?? pwr(end,:) = pwr(end,:)./2;
end

%% Compute frequency vector
% This is an evenly spaced frequency vector with NumUniquePts points.
fs = (0:NumUniquePts-1)* hz / nPoints;
fs=fs';



**********END****************************













On Mon, Nov 8, 2010 at 7:00 PM, Ross ?vila <ross.avila at gmail.com> wrote:
>
> Hi everyone,
>
> I am a graduate student working on an alpha asymmetry study and I am
> struggling with the spectral analysis.
>
> I have collected 8 minutes of resting EEG divided into 8 60-second
> blocks. ?Upon loading the data into EEGLAB, it recognizes a single
> 8-minute long epoch. ?I'm just confused about what I need to do at
> this point, and there is no one at the University of South Florida
> that researches alpha asymmetry (they all do ERP studies here). ?I
> have little to no experience in MATLAB (part of what appealed to me
> about the EEGLAB interface), so any help in the form of coding might
> need to be "dumbed" down and pretty basic.
>
> I am struggling specifically with the following questions:
> 1) Do I need to extract epochs? If so, how?
> 2) How do I actually get the alpha asymmetry? ?I've read about using
> FFT, but don't know how to actually do this. ?Also read about
> "windowing" but again do not understand exactly what this means.
>
> Any help would be immensely appreciated.
>
> Thank you,
> Ross
>
> --
> Ross T. ?vila
> Cognitive Neuropsychology Lab
> Doctoral Program in Clinical Psychology
> University of South Florida
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu



From andrewhill at ucla.edu  Wed Nov 10 22:36:56 2010
From: andrewhill at ucla.edu (Andrew Hill)
Date: Wed, 10 Nov 2010 22:36:56 -0800
Subject: [Eeglablist] Out of memory
In-Reply-To: <52957.131.173.133.29.1289312289.squirrel@myuos.uni-osnabrueck.de>
References: <52957.131.173.133.29.1289312289.squirrel@myuos.uni-osnabrueck.de>
Message-ID: <D41E8623-0898-4862-94F8-C9B248827B48@ucla.edu>

Hi Ben,

Only when running a 64-bit version of your hardware/operating system and 64-bit Matlab can you address enough RAM to import huge files - that might be a factor if you are experiencing different behavior on different machines.   I can only get Windows/Mac 32-bit machines to import about 350-400mb files max, either .bdf or .cnt.

Out of memory is a common problem when importing from other formats, b/c if i recall correctly Matlab makes more than one copy of the data on import.  Not sure if it's true with the bdf import you are using (there are a couple in EEGLab, and I don't know what either is doing)

Various workarounds exist, like first selecting a subset of channels, segmenting the recording (and then importing and appending .sets), and discarding the DC information to convert your .bdf to .edf - this is assuming your bdf is BioSemi.   If so, grab the Converter utility here http://www.biosemi.com/download.htm and "convert" your file.  this will discard the offset (default settings should be fine) and the resulting file will be a lot smaller and easier to import.  

You may want to detrend your data afterwards to ensure the the offset is cleanly out (using something like EEG.data = detrend(EEG.data); )  I've not seen it make much difference in my data, but have also seen it recommended with importing BioSemi data. 

There are also Reducer and Cropper tools available at that same link for channel/time selection, and a Decimator tool that can reduce the sampling rate if you recorded at some larger than necessary

Best,
Andrew


On Nov 9, 2010, at 6:18 AM, Benjamin Kuhr wrote:

> Hi,
> 
> I tried to import a large *.bdf file, 614 MB. I get the "Out of memory" -
> message, but not on all computers and for some reason it seems as it does
> not depend on the actual memory. It works on a weaker system, but not on
> the current one, which has 3.8 GiB of memory and a Intel Core 2 Duo CPU
> R7500 @ 2.93 GHz.
> 
> Even on identical systems with the same hardware and same software I can
> load the file only on one of them. I tired it with Windows XP and Ubuntu
> 10 on the system mentioned above, no difference. Any suggestions?
> 
> Thanks in advance
> Benjamin Kuhr
> 
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101110/ecd90069/attachment.html>

From tarikbelbahar at gmail.com  Wed Nov 10 22:53:10 2010
From: tarikbelbahar at gmail.com (Tarik S Bel-Bahar)
Date: Wed, 10 Nov 2010 22:53:10 -0800
Subject: [Eeglablist] Electrode Location (.LOCS) file for Neuroscan 64
 channel electrode cap
In-Reply-To: <AANLkTin1fxVFDRUHqfsOuSwuCgr=DmP88AnHZJVZkSYi@mail.gmail.com>
References: <AANLkTin1fxVFDRUHqfsOuSwuCgr=DmP88AnHZJVZkSYi@mail.gmail.com>
Message-ID: <AANLkTikfp8MEG6NGMymBrFJ0RUPCowXAR_Svu-xD0Mqv@mail.gmail.com>

Greetings Muhammad,

If you google search "neuroscan 64 locs file", it seems that one of
the first five links takes you right to
EEGLAB's page on electrode location files. Therein it seems that you
can login to the SCCN server
as guest and look through several folders of location files. In the
first instance when you have questions,
please try to use google and a have a close read and/or search of the
EEGLAB wiki and tutorials.

Respectfully,
Tarik




On Wed, Nov 10, 2010 at 2:33 PM, MP <mparvaz at gmail.com> wrote:
> Hello all,
> In our experiments we don't digitize the channel location, however to do ICA
> or LORETA, a channel location file (.locs) is needed. I was wondering if
> there is an approximate .locs file for Neuroscan 64 electrode cap, and if
> someone would kindly share it with me.
> Thanks
> - Muhammad
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>


From dgroppe at cogsci.ucsd.edu  Thu Nov 11 07:33:28 2010
From: dgroppe at cogsci.ucsd.edu (David Groppe)
Date: Thu, 11 Nov 2010 10:33:28 -0500
Subject: [Eeglablist] Conducting ICA on correct or all epochs
In-Reply-To: <AANLkTin6+hTbFq2Y=kQQE=_ZUiPkihcbgeDegwjGeNGb@mail.gmail.com>
References: <AANLkTin6+hTbFq2Y=kQQE=_ZUiPkihcbgeDegwjGeNGb@mail.gmail.com>
Message-ID: <AANLkTinarbscZ1s8YN=s_BNmFr0kdiCcxFzy76dgXQR8@mail.gmail.com>

Hi Baris,
   ICA's performance will generally degrade as the number of
electrical sources increases (see
http://www.cogsci.ucsd.edu/~dgroppe/PUBLICATIONS/GroppeCSO2008.pdf).
The incorrect trials probably have some EEG activity not present (or
at least less present) in the correct trials.  So if you have
sufficient data to run ICA on just the correct trials, it would
probably be better just to use the correct trials.  If you don't have
enough data using just the correct trials though, you'll probably be
fine using the all the trials, since surely a lot of the EEG activity
is common to both sets of trials.
         hope this helps,
             -David Groppe


On Wed, Nov 10, 2010 at 4:19 PM, Baris Demiral
<demiral.007 at googlemail.com> wrote:
> Hi everyone,
> I am running a simple EEG experiment where I measure reaction times and
> accuracies.
> I want to use ICs for artifact removal, and I will report only the correct
> trials (hits).
> So would it be better to use the correct epochs for the ICA to correct for
> the artifacts or is it OK to use all the epochs to detect the artifacts and
> then run the artifact correction (pop_subcomp) followed by?deleting the
> incorrect epochs?
> Thanks,
> Baris
>
> --
> SB Demiral, PhD.
> Department of Psychology
> 7 George Square
> The University of Edinburgh
> Edinburgh, EH8 9JZ
> UK
> Phone: +44 (0131) 6503063
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
David Groppe, Ph.D.
dgroppe at cogsci.ucsd.edu
http://www.cogsci.ucsd.edu/~dgroppe/



From gangadhar.garipelli at epfl.ch  Thu Nov 11 04:40:51 2010
From: gangadhar.garipelli at epfl.ch (Gangadhar Garipelli)
Date: Thu, 11 Nov 2010 13:40:51 +0100
Subject: [Eeglablist] How to deal with 1/f noise in the low frequency
 oscillations for on-line experiments?
In-Reply-To: <692C29CF-6415-4099-845E-899D71D32F83@columbia.edu>
References: <4CD3EF94.1020408@epfl.ch>
	<692C29CF-6415-4099-845E-899D71D32F83@columbia.edu>
Message-ID: <4CDBE453.4090704@epfl.ch>

Hello Philip & eeglablist,

Thank you for your prompt response. Yes, you are correct! I should apply
high pass filtering to remove a variety of neuronal and non-neuronal
artifacts in the very slow frequency range[I, II]. Eventually, a high
pass of 0.1Hz cut-off is good enough. But the problem is, I need to have
a sharp transition for the filter and should have "almost" no
group/phase delay for the on-line experiment.

"I quote again that I need to implement such a filter on-line and almost
in real-time wherein I can afford a time delay (resulting from filter's
group delay) of ~1s."

If I manage to have such a filter, I would not loose any signals in the
range [0.2 0.6] & [0.6 0.8]Hz that are relevant to my experiment.

I am expecting various suggestions, using either signal processing or
machine learning approaches or combination of both. I had suggestions
already, using matched filters, lock-in amplification, amplitude
modulation using choppers, exploiting spatial information etc. But none
of them are satisfactory as of now.

If one of you dealt with such a problem already, I would be very
grateful to learn from your experience! :-)

References :
[I] Vanhatalo S, Viopio J, Kaila K. Full-band EEG (FbEEG): An emerging
standard in electroencephalography. Clin. Neurophysiol., 116(1):1-8, 2005.

[II] Vanhatalo et al., 2004b S. Vanhatalo, J. Voipio and K. Kaila,
Infraslow EEG activity In: E. Niedermeyer and F. Lopes da Silva,
Editors, Electroencephalography: basic principles, clinical
applications, and related fields, Lippincott-Williams & Wilkins, Boston,
MA (2004).

Sincerly
Ganga

On 11/11/2010 05:16 AM, PHILIP GRIEVE wrote:
> not sure what noise source you are referring to - do you mean the noise from the DC coupled EEG amplifier?  i would think that the real problem in discerning these very slow cortical signals is the electrode artifact voltages generated at the contact of the skin and the electrode - these "half-cell" potentials drift around and can be very large as the potentials from the two EEG electrodes oppose each other and their difference cause artifact at the mm volt level - a high pass filter can remove this artifacts but also will remove your signal! 
> 
> 
> On Nov 5, 2010, at 7:50 AM, Gangadhar Garipelli wrote:
> 
>> Dear all,
>>
>> I work with low frequency oscillations of the brain while a human
>> subject is cognitively engaged in a task. From the off-line analysis
>> (using zero-phase band pass FIR filters on full-band EEG), I discovered
>> that task-related cognitive signals are located in the range of [0.2
>> 0.3]Hz and in [0.6 0.8]Hz. The fluctuations/oscillations ( formally
>> called very low frequency oscillations VLFO, or infra slow oscillations
>> ISO) below 0.2Hz are REAL devil due to 1/f nature. The noise power is
>>> 100 times higher than signal's power.
>>
>> Now as per my experimental demands, I need to estimate on-line in
>> real-time the signals mentioned in the above range and manipulate
>> stimulus presentation. Ideally, this eventually means I need to have a
>> very sharp high pass filter with almost zero group/phase delay. Which
>> sounds impossible!
>>
>> However, I should come up with a decent trade-off between SNR and
>> phase-delay. Do you have any suggestions? All suggestions ranging from
>> signal processing/machine-learning to hardware to solve this problem are
>> most welcome!
>>
>> Thanks in advance!
>> Sincerely,
>> -- 
>> Gangadhar GARIPELLI,
>> Doctoral student,
>> EPFL, Switzerland
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
>>
> 


From mklados at med.auth.gr  Wed Nov 10 23:21:23 2010
From: mklados at med.auth.gr (Klados Manousos)
Date: Thu, 11 Nov 2010 09:21:23 +0200
Subject: [Eeglablist] Out of memory
In-Reply-To: <52957.131.173.133.29.1289312289.squirrel@myuos.uni-osnabrueck.de>
References: <52957.131.173.133.29.1289312289.squirrel@myuos.uni-osnabrueck.de>
Message-ID: <AANLkTiknU1gxDsiccZ6m4j0GC9q_SZAr2Mtxg7nOnT_5@mail.gmail.com>

Hello to all,

Both systems you menioned are running the same programes? Because one system
may run in background applications that need more memory than the other....

Try to swap your physical memory from the hard drive... With that way i
achieved to load big files in EEGLAB

2010/11/9 Benjamin Kuhr <bkuhr at uni-osnabrueck.de>

> Hi,
>
> I tried to import a large *.bdf file, 614 MB. I get the "Out of memory" -
> message, but not on all computers and for some reason it seems as it does
> not depend on the actual memory. It works on a weaker system, but not on
> the current one, which has 3.8 GiB of memory and a Intel Core 2 Duo CPU
> R7500 @ 2.93 GHz.
>
> Even on identical systems with the same hardware and same software I can
> load the file only on one of them. I tired it with Windows XP and Ubuntu
> 10 on the system mentioned above, no difference. Any suggestions?
>
> Thanks in advance
> Benjamin Kuhr
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Manousos A. Klados
PhD Candidate -- Research Assistant
Group of Applied Neurosciences
Lab of Medical Informatics
School of Medicine
Aristotle University of Thessaloniki
P.O. Box 323 54124 Thessaloniki Greece
_________________________________________________
Tel: +30-2310-999332
Fax:+30-2310-999263
Website: http://lomiweb.med.auth.gr/gan/mklados

________________________________________________________________
??? ????? ???????: ??? ?????? ???? ?? mail ????? ?????????? ?? ??????????.
Acting by Reacting: By not printing this e-mail I help protect the
environment.
________________________________________________________________
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101111/eabbfe99/attachment.html>

From tarikbelbahar at gmail.com  Wed Nov 10 23:00:43 2010
From: tarikbelbahar at gmail.com (Tarik S Bel-Bahar)
Date: Wed, 10 Nov 2010 23:00:43 -0800
Subject: [Eeglablist] Conducting ICA on correct or all epochs
In-Reply-To: <AANLkTin6+hTbFq2Y=kQQE=_ZUiPkihcbgeDegwjGeNGb@mail.gmail.com>
References: <AANLkTin6+hTbFq2Y=kQQE=_ZUiPkihcbgeDegwjGeNGb@mail.gmail.com>
Message-ID: <AANLkTimws=fRs7EMZbNhK9anmwzdf=hsbDO4aJt1pWir@mail.gmail.com>

Hi Baris, I am not sure, but it might depend on what you want your
artifactual ICA components to reflect, artifacts during correct trials
only, or across the participant's whole performance. With some
individuals, incorrect trials may be more informative for the ICA
approach. You may want to compare the cleaned re-composed data after
artifactual IC removal via both all three possible methods (only
correct trials, only incorrect trials, and all trials), and make a
decision from there. Overall, it is hard to imagine a scenario where
the brain activity during incorrect trials (or perhaps those preceding
correct ones) are completely irrelevant, so in my humble opinion, I
would do the tripartite comparison above, seeing if my statistical
results change based on different cleaning paths. If no difference
amongst the cleaning techniques, then using the "cleaned from only
correct trial artifacts" would be defendable. Cheers, Tarik


On Wed, Nov 10, 2010 at 1:19 PM, Baris Demiral
<demiral.007 at googlemail.com> wrote:
> Hi everyone,
> I am running a simple EEG experiment where I measure reaction times and
> accuracies.
> I want to use ICs for artifact removal, and I will report only the correct
> trials (hits).
> So would it be better to use the correct epochs for the ICA to correct for
> the artifacts or is it OK to use all the epochs to detect the artifacts and
> then run the artifact correction (pop_subcomp) followed by?deleting the
> incorrect epochs?
> Thanks,
> Baris
>
> --
> SB Demiral, PhD.
> Department of Psychology
> 7 George Square
> The University of Edinburgh
> Edinburgh, EH8 9JZ
> UK
> Phone: +44 (0131) 6503063
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



From gangadhar.garipelli at epfl.ch  Fri Nov 12 06:36:32 2010
From: gangadhar.garipelli at epfl.ch (Gangadhar Garipelli)
Date: Fri, 12 Nov 2010 15:36:32 +0100
Subject: [Eeglablist] How to deal with 1/f noise in
In-Reply-To: <000001cb8272$b75eff50$261cfdf0$@unl.pt>
References: <4CD3EF94.1020408@epfl.ch> <000001cb8272$b75eff50$261cfdf0$@unl.pt>
Message-ID: <4CDD50F0.4060406@epfl.ch>

Hello Arnaldo,

Thanks for the attempt! :-)

Yes, it is a bit tricky to record such slow oscillations. Conventional
or classical-EEG is usually high pass at 0.5Hz. However, FbEEG is
becoming a standard. Please check Vanhatalo et al, 2005 [1] for an
excellent report on FbEEG and hardware (using DC coupled amplifiers)
other related requirements.

Reference :
[I] Vanhatalo S, Viopio J, Kaila K. Full-band EEG (FbEEG): An emerging
standard in electroencephalography. Clin. Neurophysiol., 116(1):1-8, 2005.


On 11/12/2010 03:04 PM, Arnaldo Batista wrote:
> Hi
> 
> Can?t help you, but thought being generally the EEG data high-pass filtered
> in the acquisition step, and how can you retain signal at such low
> frequencies?
> 
> Thanks
> 
> Arnaldo
> 
> 
> 
> -----Original Message-----
> From: eeglablist-bounces at sccn.ucsd.edu
> [mailto:eeglablist-bounces at sccn.ucsd.edu] On Behalf Of Gangadhar Garipelli
> Sent: 05 November 2010 11:51
> To: eeglablist at sccn.ucsd.edu
> Subject: [Eeglablist] How to deal with 1/f noise in the low frequency
> oscillations for on-line experiments?
> 
> Dear all,
> 
> I work with low frequency oscillations of the brain while a human
> subject is cognitively engaged in a task. From the off-line analysis
> (using zero-phase band pass FIR filters on full-band EEG), I discovered
> that task-related cognitive signals are located in the range of [0.2
> 0.3]Hz and in [0.6 0.8]Hz. The fluctuations/oscillations ( formally
> called very low frequency oscillations VLFO, or infra slow oscillations
> ISO) below 0.2Hz are REAL devil due to 1/f nature. The noise power is
>> 100 times higher than signal's power.
> 
> Now as per my experimental demands, I need to estimate on-line in
> real-time the signals mentioned in the above range and manipulate
> stimulus presentation. Ideally, this eventually means I need to have a
> very sharp high pass filter with almost zero group/phase delay. Which
> sounds impossible!
> 
> However, I should come up with a decent trade-off between SNR and
> phase-delay. Do you have any suggestions? All suggestions ranging from
> signal processing/machine-learning to hardware to solve this problem are
> most welcome!
> 
> Thanks in advance!
> Sincerely,

-- 
Gangadhar GARIPELLI,
Doctoral assistant,
EPFL-STI-CNBI,
ELB 141, Station-11,
CH-1015, Lausanne, Switzerland.



From alexandre.lehmann at gmail.com  Sat Nov 13 21:23:52 2010
From: alexandre.lehmann at gmail.com (Alexandre Lehmann)
Date: Sun, 14 Nov 2010 00:23:52 -0500
Subject: [Eeglablist] Out of memory
In-Reply-To: <AANLkTiknU1gxDsiccZ6m4j0GC9q_SZAr2Mtxg7nOnT_5@mail.gmail.com>
References: <52957.131.173.133.29.1289312289.squirrel@myuos.uni-osnabrueck.de>
	<AANLkTiknU1gxDsiccZ6m4j0GC9q_SZAr2Mtxg7nOnT_5@mail.gmail.com>
Message-ID: <AANLkTimTDq+f37GLfKFc5T0vRoCpLwPjmYAccawRNCu0@mail.gmail.com>

Hello All,

Klados, when you say "Try to swap your physical memory from the hard
drive.", you mean adding some pagfile in memory preferences in windows ? Or
are you refering to another procedure or another OS ?

I did try to increase my virtual memory by creating a pagefile of 4Gb, but
even 150Mb bdf files would still give an out of memory error.

Thanks,

Regards

Alexandre



On Thu, Nov 11, 2010 at 2:21 AM, Klados Manousos <mklados at med.auth.gr>wrote:

> Hello to all,
>
> Both systems you menioned are running the same programes? Because one
> system may run in background applications that need more memory than the
> other....
>
> Try to swap your physical memory from the hard drive... With that way i
> achieved to load big files in EEGLAB
>
> 2010/11/9 Benjamin Kuhr <bkuhr at uni-osnabrueck.de>
>
> Hi,
>>
>> I tried to import a large *.bdf file, 614 MB. I get the "Out of memory" -
>> message, but not on all computers and for some reason it seems as it does
>> not depend on the actual memory. It works on a weaker system, but not on
>> the current one, which has 3.8 GiB of memory and a Intel Core 2 Duo CPU
>> R7500 @ 2.93 GHz.
>>
>> Even on identical systems with the same hardware and same software I can
>> load the file only on one of them. I tired it with Windows XP and Ubuntu
>> 10 on the system mentioned above, no difference. Any suggestions?
>>
>> Thanks in advance
>> Benjamin Kuhr
>>
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>>
>
>
>
> --
> Manousos A. Klados
> PhD Candidate -- Research Assistant
> Group of Applied Neurosciences
> Lab of Medical Informatics
> School of Medicine
> Aristotle University of Thessaloniki
> P.O. Box 323 54124 Thessaloniki Greece
> _________________________________________________
> Tel: +30-2310-999332
> Fax:+30-2310-999263
> Website: http://lomiweb.med.auth.gr/gan/mklados
>
> ________________________________________________________________
> ??? ????? ???????: ??? ?????? ???? ?? mail ????? ?????????? ?? ??????????.
> Acting by Reacting: By not printing this e-mail I help protect the
> environment.
> ________________________________________________________________
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101114/eb985df4/attachment.html>

From agb at fct.unl.pt  Fri Nov 12 06:04:47 2010
From: agb at fct.unl.pt (Arnaldo Batista)
Date: Fri, 12 Nov 2010 14:04:47 -0000
Subject: [Eeglablist] How to deal with 1/f noise in
In-Reply-To: <4CD3EF94.1020408@epfl.ch>
References: <4CD3EF94.1020408@epfl.ch>
Message-ID: <000001cb8272$b75eff50$261cfdf0$@unl.pt>

Hi

Can?t help you, but thought being generally the EEG data high-pass filtered
in the acquisition step, and how can you retain signal at such low
frequencies?

Thanks

Arnaldo



-----Original Message-----
From: eeglablist-bounces at sccn.ucsd.edu
[mailto:eeglablist-bounces at sccn.ucsd.edu] On Behalf Of Gangadhar Garipelli
Sent: 05 November 2010 11:51
To: eeglablist at sccn.ucsd.edu
Subject: [Eeglablist] How to deal with 1/f noise in the low frequency
oscillations for on-line experiments?

Dear all,

I work with low frequency oscillations of the brain while a human
subject is cognitively engaged in a task. From the off-line analysis
(using zero-phase band pass FIR filters on full-band EEG), I discovered
that task-related cognitive signals are located in the range of [0.2
0.3]Hz and in [0.6 0.8]Hz. The fluctuations/oscillations ( formally
called very low frequency oscillations VLFO, or infra slow oscillations
ISO) below 0.2Hz are REAL devil due to 1/f nature. The noise power is
>100 times higher than signal's power.

Now as per my experimental demands, I need to estimate on-line in
real-time the signals mentioned in the above range and manipulate
stimulus presentation. Ideally, this eventually means I need to have a
very sharp high pass filter with almost zero group/phase delay. Which
sounds impossible!

However, I should come up with a decent trade-off between SNR and
phase-delay. Do you have any suggestions? All suggestions ranging from
signal processing/machine-learning to hardware to solve this problem are
most welcome!

Thanks in advance!
Sincerely,
-- 
Gangadhar GARIPELLI,
Doctoral student,
EPFL, Switzerland

_______________________________________________
Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
For digest mode, send an email with the subject "set digest mime" to
eeglablist-request at sccn.ucsd.edu




From pzeman at alumni.uvic.ca  Sun Nov 14 16:16:51 2010
From: pzeman at alumni.uvic.ca (Philip Michael Zeman)
Date: Sun, 14 Nov 2010 16:16:51 -0800
Subject: [Eeglablist] How to deal with 1/f noise in
In-Reply-To: <4CDD50F0.4060406@epfl.ch>
References: <4CD3EF94.1020408@epfl.ch> <000001cb8272$b75eff50$261cfdf0$@unl.pt>
	<4CDD50F0.4060406@epfl.ch>
Message-ID: <46AECCF169BF47A4B27CA4C6725CEF21@mine>

Hello Gangadhar

If you have infinite time (or grad students): something for you to try:

you might also try looking at higher frequencies for this signal you search 
for at the low end of the spectrum.  I'm finding that some of these low 
frequencies signals are related to high-frequency signals.  It makes some 
sense if you subscribe to the idea that alot of these signals are resulting 
from non-linear mixing.  For example, in a study a few years back, I was 
tracking theta-band activities.  In my search, I also check to see if I 
could find some theta-band modulated activities in the gamma frequency band. 
(Like a low frequency signal multiplied with a higher frequency carrie 
wave.)  I did find statistically significant differences between conditions 
this way but demodulating the signal (assuming a gamma band carrier 
frequency).

Phil

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Philip Michael Zeman B.Eng, Ph.D.
Applied Brain and Vision Sciences Inc.
Brain Function Analysis for Novel Paradigms and Serious Games
Analysis of Pharmaceutical Effects on Brain Function
http://www.abvsciences.com
Latest Brain Research Result:
http://www.spatialbrain.com
Email:   pzeman at alumni.uvic.ca
Phone: +1-250-589-4234
Skype: philip_michael_zeman
LinkedIn Profile: http://ca.linkedin.com/in/philipmichaelzeman
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

Gangadhar Garipelli <gangadhar.garipelli at epfl.ch>
----- Original Message ----- 
From: "Gangadhar Garipelli" <gangadhar.garipelli at epfl.ch>
To: "Arnaldo Batista" <agb at fct.unl.pt>
Cc: <eeglablist at sccn.ucsd.edu>
Sent: Friday, November 12, 2010 6:36 AM
Subject: Re: [Eeglablist] How to deal with 1/f noise in


> Hello Arnaldo,
>
> Thanks for the attempt! :-)
>
> Yes, it is a bit tricky to record such slow oscillations. Conventional
> or classical-EEG is usually high pass at 0.5Hz. However, FbEEG is
> becoming a standard. Please check Vanhatalo et al, 2005 [1] for an
> excellent report on FbEEG and hardware (using DC coupled amplifiers)
> other related requirements.
>
> Reference :
> [I] Vanhatalo S, Viopio J, Kaila K. Full-band EEG (FbEEG): An emerging
> standard in electroencephalography. Clin. Neurophysiol., 116(1):1-8, 2005.
>
>
> On 11/12/2010 03:04 PM, Arnaldo Batista wrote:
>> Hi
>>
>> Can?t help you, but thought being generally the EEG data high-pass 
>> filtered
>> in the acquisition step, and how can you retain signal at such low
>> frequencies?
>>
>> Thanks
>>
>> Arnaldo
>>
>>
>>
>> -----Original Message-----
>> From: eeglablist-bounces at sccn.ucsd.edu
>> [mailto:eeglablist-bounces at sccn.ucsd.edu] On Behalf Of Gangadhar 
>> Garipelli
>> Sent: 05 November 2010 11:51
>> To: eeglablist at sccn.ucsd.edu
>> Subject: [Eeglablist] How to deal with 1/f noise in the low frequency
>> oscillations for on-line experiments?
>>
>> Dear all,
>>
>> I work with low frequency oscillations of the brain while a human
>> subject is cognitively engaged in a task. From the off-line analysis
>> (using zero-phase band pass FIR filters on full-band EEG), I discovered
>> that task-related cognitive signals are located in the range of [0.2
>> 0.3]Hz and in [0.6 0.8]Hz. The fluctuations/oscillations ( formally
>> called very low frequency oscillations VLFO, or infra slow oscillations
>> ISO) below 0.2Hz are REAL devil due to 1/f nature. The noise power is
>>> 100 times higher than signal's power.
>>
>> Now as per my experimental demands, I need to estimate on-line in
>> real-time the signals mentioned in the above range and manipulate
>> stimulus presentation. Ideally, this eventually means I need to have a
>> very sharp high pass filter with almost zero group/phase delay. Which
>> sounds impossible!
>>
>> However, I should come up with a decent trade-off between SNR and
>> phase-delay. Do you have any suggestions? All suggestions ranging from
>> signal processing/machine-learning to hardware to solve this problem are
>> most welcome!
>>
>> Thanks in advance!
>> Sincerely,
>
> -- 
> Gangadhar GARIPELLI,
> Doctoral assistant,
> EPFL-STI-CNBI,
> ELB 141, Station-11,
> CH-1015, Lausanne, Switzerland.
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to 
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to 
> eeglablist-request at sccn.ucsd.edu 



From gangadhar.garipelli at epfl.ch  Mon Nov 15 01:08:49 2010
From: gangadhar.garipelli at epfl.ch (Gangadhar Garipelli)
Date: Mon, 15 Nov 2010 10:08:49 +0100
Subject: [Eeglablist] How to deal with 1/f noise in
In-Reply-To: <46AECCF169BF47A4B27CA4C6725CEF21@mine>
References: <4CD3EF94.1020408@epfl.ch> <000001cb8272$b75eff50$261cfdf0$@unl.pt>
	<4CDD50F0.4060406@epfl.ch> <46AECCF169BF47A4B27CA4C6725CEF21@mine>
Message-ID: <4CE0F8A1.9080000@epfl.ch>

Hello Phil,

I guess you are speaking about cross-frequency coupling [1]!
If that is so, thanks a lot for reinforcing me that I am moving in right 
direction! :-)
And I hope this not to consume infinite time!

However, I would like to figure out the complete limits of low-frequency 
features for real-time experimental manipulations.

[1] 
http://www.cell.com/trends/cognitive-sciences/abstract/S1364-6613(10)00206-8

Ganga

On 11/15/10 1:16 AM, Philip Michael Zeman wrote:
> Hello Gangadhar
>
> If you have infinite time (or grad students): something for you to try:
>
> you might also try looking at higher frequencies for this signal you 
> search for at the low end of the spectrum.  I'm finding that some of 
> these low frequencies signals are related to high-frequency signals.  
> It makes some sense if you subscribe to the idea that alot of these 
> signals are resulting from non-linear mixing.  For example, in a study 
> a few years back, I was tracking theta-band activities.  In my search, 
> I also check to see if I could find some theta-band modulated 
> activities in the gamma frequency band. (Like a low frequency signal 
> multiplied with a higher frequency carrie wave.)  I did find 
> statistically significant differences between conditions this way but 
> demodulating the signal (assuming a gamma band carrier frequency).
>
> Phil
>
> =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
> Philip Michael Zeman B.Eng, Ph.D.
> Applied Brain and Vision Sciences Inc.
> Brain Function Analysis for Novel Paradigms and Serious Games
> Analysis of Pharmaceutical Effects on Brain Function
> http://www.abvsciences.com
> Latest Brain Research Result:
> http://www.spatialbrain.com
> Email:   pzeman at alumni.uvic.ca
> Phone: +1-250-589-4234
> Skype: philip_michael_zeman
> LinkedIn Profile: http://ca.linkedin.com/in/philipmichaelzeman
> =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
>
> Gangadhar Garipelli <gangadhar.garipelli at epfl.ch>
> ----- Original Message ----- From: "Gangadhar Garipelli" 
> <gangadhar.garipelli at epfl.ch>
> To: "Arnaldo Batista" <agb at fct.unl.pt>
> Cc: <eeglablist at sccn.ucsd.edu>
> Sent: Friday, November 12, 2010 6:36 AM
> Subject: Re: [Eeglablist] How to deal with 1/f noise in
>
>
>> Hello Arnaldo,
>>
>> Thanks for the attempt! :-)
>>
>> Yes, it is a bit tricky to record such slow oscillations. Conventional
>> or classical-EEG is usually high pass at 0.5Hz. However, FbEEG is
>> becoming a standard. Please check Vanhatalo et al, 2005 [1] for an
>> excellent report on FbEEG and hardware (using DC coupled amplifiers)
>> other related requirements.
>>
>> Reference :
>> [I] Vanhatalo S, Viopio J, Kaila K. Full-band EEG (FbEEG): An emerging
>> standard in electroencephalography. Clin. Neurophysiol., 116(1):1-8, 
>> 2005.
>>
>>
>> On 11/12/2010 03:04 PM, Arnaldo Batista wrote:
>>> Hi
>>>
>>> Can?t help you, but thought being generally the EEG data high-pass 
>>> filtered
>>> in the acquisition step, and how can you retain signal at such low
>>> frequencies?
>>>
>>> Thanks
>>>
>>> Arnaldo
>>>
>>>
>>>
>>> -----Original Message-----
>>> From: eeglablist-bounces at sccn.ucsd.edu
>>> [mailto:eeglablist-bounces at sccn.ucsd.edu] On Behalf Of Gangadhar 
>>> Garipelli
>>> Sent: 05 November 2010 11:51
>>> To: eeglablist at sccn.ucsd.edu
>>> Subject: [Eeglablist] How to deal with 1/f noise in the low frequency
>>> oscillations for on-line experiments?
>>>
>>> Dear all,
>>>
>>> I work with low frequency oscillations of the brain while a human
>>> subject is cognitively engaged in a task. From the off-line analysis
>>> (using zero-phase band pass FIR filters on full-band EEG), I discovered
>>> that task-related cognitive signals are located in the range of [0.2
>>> 0.3]Hz and in [0.6 0.8]Hz. The fluctuations/oscillations ( formally
>>> called very low frequency oscillations VLFO, or infra slow oscillations
>>> ISO) below 0.2Hz are REAL devil due to 1/f nature. The noise power is
>>>> 100 times higher than signal's power.
>>>
>>> Now as per my experimental demands, I need to estimate on-line in
>>> real-time the signals mentioned in the above range and manipulate
>>> stimulus presentation. Ideally, this eventually means I need to have a
>>> very sharp high pass filter with almost zero group/phase delay. Which
>>> sounds impossible!
>>>
>>> However, I should come up with a decent trade-off between SNR and
>>> phase-delay. Do you have any suggestions? All suggestions ranging from
>>> signal processing/machine-learning to hardware to solve this problem 
>>> are
>>> most welcome!
>>>
>>> Thanks in advance!
>>> Sincerely,
>>
>> -- 
>> Gangadhar GARIPELLI,
>> Doctoral assistant,
>> EPFL-STI-CNBI,
>> ELB 141, Station-11,
>> CH-1015, Lausanne, Switzerland.
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to 
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to 
>> eeglablist-request at sccn.ucsd.edu 
>

-- 
Gangadhar GARIPELLI,
Doctoral assistant,
EPFL-STI-CNBI,
ELB 133, Station-11,
CH-1015, Lausanne, Switzerland.

Tel : +41-21-6936985
Fax : +41-21-6935305
gangadhar.garipelli at epfl.ch
http://people.epfl.ch/gangadhar.garipelli



From derek at stenljus.se  Wed Nov 17 02:01:52 2010
From: derek at stenljus.se (derek eder)
Date: Wed, 17 Nov 2010 11:01:52 +0100
Subject: [Eeglablist] How to deal with 1/f noise in
In-Reply-To: <4CDD50F0.4060406@epfl.ch>
References: <4CD3EF94.1020408@epfl.ch> <000001cb8272$b75eff50$261cfdf0$@unl.pt>
	<4CDD50F0.4060406@epfl.ch>
Message-ID: <4CE3A810.6010804@stenljus.se>

If amplifier 1/f noise is suspected, one possible sanity check is to 
record the same signal on multiple amplifier-channels and then sum* them.

The noise** contribution to the resulting average should be reduced by: 
   1/sqrt(n.channels)

Compare the resulting low frequency activity with those from individual 
channels.

Depending on the input-design / input-impedance of your amplifiers, 
there are practical limits to the number of channels that one can 
parallel from the same electrodes,  but 3 or 4 should not be a problem.


~Derek


*   perhaps de-mean and scale first if your channels have offsets and/or 
gain differences.
** assuming uncorrelated noise from individual channels


On 11/12/2010 03:36 PM, Gangadhar Garipelli wrote:
> Hello Arnaldo,
>
> Thanks for the attempt! :-)
>
> Yes, it is a bit tricky to record such slow oscillations. Conventional
> or classical-EEG is usually high pass at 0.5Hz. However, FbEEG is
> becoming a standard. Please check Vanhatalo et al, 2005 [1] for an
> excellent report on FbEEG and hardware (using DC coupled amplifiers)
> other related requirements.
>
> Reference :
> [I] Vanhatalo S, Viopio J, Kaila K. Full-band EEG (FbEEG): An emerging
> standard in electroencephalography. Clin. Neurophysiol., 116(1):1-8, 2005.
>
>
> On 11/12/2010 03:04 PM, Arnaldo Batista wrote:
>> Hi
>>
>> Can?t help you, but thought being generally the EEG data high-pass filtered
>> in the acquisition step, and how can you retain signal at such low
>> frequencies?
>>
>> Thanks
>>
>> Arnaldo
>>
>>
>>
>> -----Original Message-----
>> From: eeglablist-bounces at sccn.ucsd.edu
>> [mailto:eeglablist-bounces at sccn.ucsd.edu] On Behalf Of Gangadhar Garipelli
>> Sent: 05 November 2010 11:51
>> To: eeglablist at sccn.ucsd.edu
>> Subject: [Eeglablist] How to deal with 1/f noise in the low frequency
>> oscillations for on-line experiments?
>>
>> Dear all,
>>
>> I work with low frequency oscillations of the brain while a human
>> subject is cognitively engaged in a task. From the off-line analysis
>> (using zero-phase band pass FIR filters on full-band EEG), I discovered
>> that task-related cognitive signals are located in the range of [0.2
>> 0.3]Hz and in [0.6 0.8]Hz. The fluctuations/oscillations ( formally
>> called very low frequency oscillations VLFO, or infra slow oscillations
>> ISO) below 0.2Hz are REAL devil due to 1/f nature. The noise power is
>>> 100 times higher than signal's power.
>> Now as per my experimental demands, I need to estimate on-line in
>> real-time the signals mentioned in the above range and manipulate
>> stimulus presentation. Ideally, this eventually means I need to have a
>> very sharp high pass filter with almost zero group/phase delay. Which
>> sounds impossible!
>>
>> However, I should come up with a decent trade-off between SNR and
>> phase-delay. Do you have any suggestions? All suggestions ranging from
>> signal processing/machine-learning to hardware to solve this problem are
>> most welcome!
>>
>> Thanks in advance!
>> Sincerely,


-- 
Derek Eder
tlf 0704 915 715
www.stenljus.se



From eskappenman at ucdavis.edu  Wed Nov 17 13:40:39 2010
From: eskappenman at ucdavis.edu (Emily Kappenman)
Date: Wed, 17 Nov 2010 13:40:39 -0800
Subject: [Eeglablist] 2011 ERP Boot Camp
Message-ID: <AANLkTi=_We1vscQZgpakGvWonDYAmVEgVT2830cJt=te@mail.gmail.com>

The UC-Davis ERP Boot Camp, an NIH-funded summer workshop on the ERP
technique, will be held July 11-20 2011. Please forward this
announcement to students, postdocs, and faculty who might be
interested in attending.  (For additional information, see
erpinfo.org/the-erp-bootcamp).

The ERP Boot Camp is a 10-day introduction to the ERP technique held
each summer at UC Davis. It is intended for beginning and intermediate
ERP researchers, and for both basic scientists and
clinical/translational researchers.

The topics will include:

1) Where do ERPs come from? What do they mean?
2) ERP components
3) The design and interpretation of ERP experiments
4) EEG data acquisition
5) Filtering, artifact rejection, and artifact correction
6) Measuring and analyzing ERP components
7) ERP localization
8) Time-frequency analysis
9) Setting up and running an ERP lab

The Boot Camp consists of lectures on these topics, structured
discussions, individual consultations, and a substantial laboratory
component.  It is led by Steve Luck, and the faculty includes many
distinguished ERP researchers from UC Davis and other universities.

Participants at previous Boot Camps have come from around the world
and have ranged from beginning graduate students to full professors.
They have included psychologists, neuroscientists, psychiatrists,
neurologists, and speech pathologists. Typically, we expect that
students and postdocs should have had at least 6 months of significant
ERP (or related) experience before attending the Boot Camp.

We strongly encourage the participation of individuals from
underrepresented groups.

Funding is available from NIMH to defray some or all of the costs of
attending the Boot Camp, and scholarships will be provided to all
participants who are U.S. citizens or permanent residents.
Unfortunately, scholarships are not available for international
participants. We typically accept 25-28 U.S. citizens and permanent
residents, along with 2-5 international participants.

The application consists of a CV, a 1-2 page statement of background
and interests, and (for students and postdocs) a letter of
recommendation.

Applications for the 2011 session are now being accepted at
erpinfo.org/the-erp-bootcamp. Applications are due on March 1, 2011.
Questions should be directed to erpbootcamp at gmail.com.

-- 
--------------------------------------------------------------------
Emily S. Kappenman
UC Davis Center for Mind and Brain
267 Cousteau Place
Davis, CA 95618
eskappenman at ucdavis.edu
--------------------------------------------------------------------


From cambridge02238 at yahoo.com  Mon Nov 15 07:37:20 2010
From: cambridge02238 at yahoo.com (Tae Kim)
Date: Mon, 15 Nov 2010 07:37:20 -0800 (PST)
Subject: [Eeglablist] ERSP Image Question
Message-ID: <383987.74477.qm@web33603.mail.mud.yahoo.com>

Hello, 
?
I wonder what the green line and the blue line under the upper panel in the ERSP image notate.? The manual says it is the ERSP envelope.??Then is the green high mean dB values, relative to baseline at each time in the epoch, and the blue low mean dB values relative to baseline?
?
Sincerely,
?
tk




????? Tae Kim


??????









      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101115/cb80ad4a/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Timef.gif
Type: image/gif
Size: 37301 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101115/cb80ad4a/attachment.gif>

From gangadhar.garipelli at epfl.ch  Fri Nov 19 02:43:11 2010
From: gangadhar.garipelli at epfl.ch (Gangadhar Garipelli)
Date: Fri, 19 Nov 2010 11:43:11 +0100
Subject: [Eeglablist] How to deal with 1/f noise in
In-Reply-To: <4CE3A810.6010804@stenljus.se>
References: <4CD3EF94.1020408@epfl.ch>
	<000001cb8272$b75eff50$261cfdf0$@unl.pt>	<4CDD50F0.4060406@epfl.ch>
	<4CE3A810.6010804@stenljus.se>
Message-ID: <4CE654BF.3080908@epfl.ch>

Hello Derek,

Yes! indeed I am doing this already by applying a spatial smoothing
filter (SSF).

But this step works better if I band pass in the range (0.1 1)Hz to
remove any signals below 0.1Hz and common average reference the signals
before SSF! This is the reason, "I ran into the need for narrow bandpass"!

As a side note, I believe that 1/f noise is not only result of amplifier
but due to the inherent structure of neural oscillations [1] and could
also be due to slow variations in the gel-skin contact conductance.

[1] Gyorgy Buzsaki, Rhythms of the Brain, Oxford University Press, 1
edition,2006.

Thanks again! :-)
Ganga

On 11/17/2010 11:01 AM, derek eder wrote:
> If amplifier 1/f noise is suspected, one possible sanity check is to 
> record the same signal on multiple amplifier-channels and then sum* them.
> 
> The noise** contribution to the resulting average should be reduced by: 
>    1/sqrt(n.channels)
> 
> Compare the resulting low frequency activity with those from individual 
> channels.
> 
> Depending on the input-design / input-impedance of your amplifiers, 
> there are practical limits to the number of channels that one can 
> parallel from the same electrodes,  but 3 or 4 should not be a problem.
> 
> 
> ~Derek
> 
> 
> *   perhaps de-mean and scale first if your channels have offsets and/or 
> gain differences.
> ** assuming uncorrelated noise from individual channels
> 


From jordicostafa at gmail.com  Fri Nov 19 10:05:31 2010
From: jordicostafa at gmail.com (Jordi Costa Faidella)
Date: Fri, 19 Nov 2010 13:05:31 -0500
Subject: [Eeglablist] problems importing neuroscan .cnt with trigger
	information
In-Reply-To: <AANLkTi=_We1vscQZgpakGvWonDYAmVEgVT2830cJt=te@mail.gmail.com>
References: <AANLkTi=_We1vscQZgpakGvWonDYAmVEgVT2830cJt=te@mail.gmail.com>
Message-ID: <12AA00E5-CA9D-42DD-9E76-96037457774D@gmail.com>

dear eeglab users,

I'm having a problem when importing neuroscan .cnt files into eeglab format with the triggers. It is the first time I'm doing an experiment in which the participants need to give an answer by pressing a button, so never before I had to face this issue. The trigger for the answer appears in the .cnt file while recording it, while viewing it with neuroscan software and when extracting the .evt or .ev2 files. However, when I import the .cnt into eeglab, those triggers coding the participant's answers are missing. If I try to import the .cnt and then an .ev2 file eeglab gives me an error. However, I wouldn't like to depend on neuroscan software, and as long as the trigger information is in the .cnt file, there should be a way to import the .cnt into eeglab with the information of all triggers. Does anyone faced this problem before? I would be very pleased if someone could help me or post some written script to solve this issue.

thank you a lot,

Jordi Costa Faidella




From lutobu at gmail.com  Thu Nov 18 16:41:46 2010
From: lutobu at gmail.com (ludwing torres)
Date: Fri, 19 Nov 2010 01:41:46 +0100
Subject: [Eeglablist] How to import the leadfield matrix
Message-ID: <AANLkTikUkuodMXLn0gbbHC1FnFgCkn7nuiSkww3q0i7k@mail.gmail.com>

Hello.
Is there any way to export the leadfield matrix or the inverse model as an
matlab array?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101119/3a775882/attachment.html>

From maximilien.chaumon at gmail.com  Thu Nov 18 18:40:10 2010
From: maximilien.chaumon at gmail.com (Maximilien Chaumon)
Date: Thu, 18 Nov 2010 18:40:10 -0800
Subject: [Eeglablist] Neuromag sensor locations
Message-ID: <AANLkTim0GWpWxVorzZW5VmnqakGuy9X2fzNwM4sNAmpB@mail.gmail.com>

Hello EEGlabbers,
I'm trying to import sensor locations from my .fif files out of a neuromag
machine and can't get to stick it into eeglab structure.
Has anyone ever transformed sensor locations from neuromag to eeglab?
Any help is appreciated,
Thank you,
Max
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101118/4c2a65eb/attachment.html>

From zmb25 at cam.ac.uk  Fri Nov 19 08:50:27 2010
From: zmb25 at cam.ac.uk (=?ISO-8859-1?Q?Zara_Bergstr=F6m?=)
Date: 19 Nov 2010 16:50:27 +0000
Subject: [Eeglablist] Is ITC biased by trial numbers?
Message-ID: <Prayer.1.3.3.1011191650270.9238@hermes-2.csi.cam.ac.uk>

Dear EEG experts,

is the ITC measure as implemented by EEGLAB biased towards lower trial 
numbers (i.e. higher itc when fewer trials are used in the computation, as 
some measures of phase coherence supposedly are), and if so, how do you 
deal with that issue when comparing conditions with different trial 
numbers? Do you think it is appropriate to compute baseline corrected ITC, 
which might help?

I analysed epoched datasets (-1-2s using default pre-stimulus baseline for 
ERSP) with wavelets using newtimef to get complex ITC values (using the 
default 'phasecoher' option), converted the ITC to real numbers between 0-1 
(absitc=sqrt(real(itc).^2+imag(itc).^2)), and averaged these into 
participant x condition x time x frequency ITC matrices for use in group 
level statistics.

The attached line plot shows grand average (24 subjects) ITC averaged 
across the alpha band (8-12 hz) for four conditions. The condition with the 
fewest average trial numbers (red line) has significantly higher ITC than 
the other conditions throughout the trial, even before stimulus onset, 
which cannot be explained by psychological factors since these conditions 
were presented randomly intermixed. If I however were to subtract the 
average baseline period ITC from the post-stimulus data, it seems that the 
difference would disappear. Would that be an appropriate step to take here?

Do you think this pattern is caused by a trial number bias, or have I done 
something wrong in the analysis pipeline?

Any thoughts would be very much appreciated.

Thanks very much for you time,

Zara Bergstrom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: AVG_ITC_4conditions.bmp
Type: image/bmp
Size: 1802394 bytes
Desc: AVG_ITC_4conditions.bmp
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101119/17cfdbae/attachment.bmp>

From maximilien.chaumon at gmail.com  Sat Nov 20 18:24:23 2010
From: maximilien.chaumon at gmail.com (Maximilien Chaumon)
Date: Sat, 20 Nov 2010 18:24:23 -0800
Subject: [Eeglablist] Neuromag sensor locations
In-Reply-To: <32CC77C0C8A7AD4B9410934642608E1F01D425@exchccr1.neuro.gu.se>
References: <AANLkTim0GWpWxVorzZW5VmnqakGuy9X2fzNwM4sNAmpB@mail.gmail.com>
	<32CC77C0C8A7AD4B9410934642608E1F01D425@exchccr1.neuro.gu.se>
Message-ID: <AANLkTik=Lb4LdGNczhVJJQNnU78kw4v=rWroWsDsABxc@mail.gmail.com>

Thanks Elena,

I found a way, described here under "comments" to importing neuromag .fif
data.
http://sccn.ucsd.edu/wiki/A01:_Importing_Continuous_and_Epoched_Data#Supported_Data_Formats

Thanks,
Max

2010/11/20 Elena Orekhova <Elena.Orekhova at neuro.gu.se>

>  Hi Max,
>
> I am not sure how it is now, but until recently there was no EEGlab method
> for that.
>
> I use MNE matlab toolbox  for fif to matlab conversion:
>
> raw = fiff_setup_read_raw(infile.fif);
> [ data, times] = fiff_read_raw_segment(raw, raw.first_samp, raw.last_samp
> );
>
> You can download the MNE at
> http://www.nmr.mgh.harvard.edu/martinos/userInfo/data/sofMNE.php
>
> Hope it helped,
>
> Elena
>
>
>  ------------------------------
> *From:* eeglablist-bounces at sccn.ucsd.edu [eeglablist-bounces at sccn.ucsd.edu]
> on behalf of Maximilien Chaumon [maximilien.chaumon at gmail.com]
> *Sent:* Friday, November 19, 2010 3:40 AM
>
> *To:* eeglablist at sccn.ucsd.edu
> *Subject:* [Eeglablist] Neuromag sensor locations
>
>  Hello EEGlabbers,
> I'm trying to import sensor locations from my .fif files out of a neuromag
> machine and can't get to stick it into eeglab structure.
> Has anyone ever transformed sensor locations from neuromag to eeglab?
> Any help is appreciated,
> Thank you,
> Max
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101120/046bfa65/attachment.html>

From maximilien.chaumon at gmail.com  Sun Nov 21 13:49:39 2010
From: maximilien.chaumon at gmail.com (Maximilien Chaumon)
Date: Sun, 21 Nov 2010 13:49:39 -0800
Subject: [Eeglablist] Fieldtrip2EEGlab
Message-ID: <AANLkTinS+M-PX_aHbo8FXoitapCY6kQjf-m2R3J3VvMS@mail.gmail.com>

Hi EEGlisters,
I would like to import epoched fieldtrip data to EEGLab.
Has anyone ever made a fieldtrip2eeglab function?
Thanks,
Max
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101121/65b45324/attachment.html>

From jordicostafa at gmail.com  Fri Nov 19 15:12:38 2010
From: jordicostafa at gmail.com (Jordi Costa Faidella)
Date: Fri, 19 Nov 2010 18:12:38 -0500
Subject: [Eeglablist] problems importing neuroscan .cnt with
	trigger	information
In-Reply-To: <4CE6F66F.4060105@cogpsyphy.hu>
References: <AANLkTi=_We1vscQZgpakGvWonDYAmVEgVT2830cJt=te@mail.gmail.com>
	<12AA00E5-CA9D-42DD-9E76-96037457774D@gmail.com>
	<4CE6F66F.4060105@cogpsyphy.hu>
Message-ID: <A74B1384-966A-455B-98CF-13A3DAF70466@gmail.com>

Thanks a lot Gabor, that really helped! I didn't imagine it had been such an easy thing to solve.
Greetings from NYC, soon back in Barcelona...
Jordi

ps:Toomas, I've been using eeglab only with neuroscan data and never had such a problem you say. I've used it under Windows (32 and 64-bit), Mac and Linux. My Matlabs have been 2008b, but I don't think using 2010 will make any difference. Have you checked if the 16 or 32 bit option when importing the data fits with the data you acquired? Hope this helps... I get nice continuous data, ERPs, TFs, and whatever I do...




El 19/11/2010, a las 17:13, STEFANICS, Gabor escribi?:

> Hi Jordi,
> 
> you need to specify the import function to import keystrokes, adding 'keystroke', 'on', parameters, like this:
> 
> EEG = pop_loadcnt('H:\CONTROL\data\01\01_k_k_01.cnt' , 'dataformat', 'int32', 'keystroke', 'on');
> 
> Greeting from Budapest,
> Gabor
> 
> 
> On 2010. 11. 19. 19:05, Jordi Costa Faidella wrote:
>> dear eeglab users,
>> 
>> I'm having a problem when importing neuroscan .cnt files into eeglab format with the triggers. It is the first time I'm doing an experiment in which the participants need to give an answer by pressing a button, so never before I had to face this issue. The trigger for the answer appears in the .cnt file while recording it, while viewing it with neuroscan software and when extracting the .evt or .ev2 files. However, when I import the .cnt into eeglab, those triggers coding the participant's answers are missing. If I try to import the .cnt and then an .ev2 file eeglab gives me an error. However, I wouldn't like to depend on neuroscan software, and as long as the trigger information is in the .cnt file, there should be a way to import the .cnt into eeglab with the information of all triggers. Does anyone faced this problem before? I would be very pleased if someone could help me or post some written script to solve this issue.
>> 
>> thank you a lot,
>> 
>> Jordi Costa Faidella
>> 
>> 
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
> 
> -- 
> Consider the environment, please don't print this email unless you really need to.
> 
> <gstefan.vcf>




From daniel.cassel at atr.jp  Mon Nov 15 19:12:23 2010
From: daniel.cassel at atr.jp (Daniel Cassel)
Date: Tue, 16 Nov 2010 12:12:23 +0900
Subject: [Eeglablist] eegthresh Question
Message-ID: <4CE1F697.7020504@atr.jp>

Hello,
I want to reject components by extreme values (pop_eegthresh / 
eegthresh). In the documentation it says if you reject components 
threshold limit is in std dev. and that the components
are normalized to have std dev 1 before thresholding. When I try to do 
so, however, no components are rejected no matter how low the threshold 
is. When I examined the code I realized it was because
the components don't seem to be normalized before they are thresholded. 
Does anyone know precisely where the components should be normalized in 
the code?
Thank you.
Daniel


From brian.murphy at unitn.it  Sun Nov 21 03:24:26 2010
From: brian.murphy at unitn.it (Brian Murphy)
Date: Sun, 21 Nov 2010 12:24:26 +0100
Subject: [Eeglablist] Neuromag sensor locations
In-Reply-To: <mailman.0.1290196802.2293.eeglablist@sccn.ucsd.edu>
References: <mailman.0.1290196802.2293.eeglablist@sccn.ucsd.edu>
Message-ID: <4CE9016A.3010209@unitn.it>

Hi Max,

I have some routines that take MEG sensor locations, and EEG electrode 
locations (recorded with a Polhemus system) and transform them into a 
common EEGLAB space. The scripts are pretty preliminary and brittle, but 
I'd be happy to share them,

best,

Brian
>
> Hello EEGlabbers,
> I'm trying to import sensor locations from my .fif files out of a neuromag machine and can't get to stick it into eeglab structure.
> Has anyone ever transformed sensor locations from neuromag to eeglab?
>
>
> Any help is appreciated,
> Thank you,
> Max
>
>
>   


-- 
Brian Murphy
Post-Doctoral Researcher
Language, Interaction and Computation Lab
Centre for Mind/Brain Sciences
University of Trento
http://clic.cimec.unitn.it/brian/



From annemurphy7 at gmail.com  Sun Nov 21 18:19:39 2010
From: annemurphy7 at gmail.com (Anne Murphy)
Date: Mon, 22 Nov 2010 13:19:39 +1100
Subject: [Eeglablist] Using PCA on continuous EEG data
Message-ID: <AANLkTimM=8YP2q3EYb59zPFmr2CxC3V-hzG=vvOugWA2@mail.gmail.com>

Dear EEGLAB users,

I am a new user of EEGlab (self-taught) so please excuse me if this is an
unintelligent question.
I have 30 datasets that I have cleaned of ocular/muscle movement artifacts
and cacatenated together within a STUDY.
Each dataset is 2 minutes long of continuous EEG data that I downloaded from
EGI Net Station (not ERP data).
I also ran individual ICA's (runica) on each dataset prior to building the
study.

Within the Study, I have run the preclustering array and now have 1
parentcluster of 3841 ICs.
My next step is to run the Clustering components algorithm using Kmeans
(stat. toolbox).
When I do this, I get the error message  "X must have more rows than the
number of clusters".
It does not matter what number I enter into the "number of clusters to
compute" entry box (from 1-9999), I still get the same error message.

As I am keen to move forward to the next step of producing the PCA spectra
plots, I am a little at a loss as to what to do next?
Any help would be most appreciated!

kind regards,

Anne Murphy
B. Psych (Hons)
University of New South Wales
Sydney, Australia
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101122/fcc6e742/attachment.html>

From bradley.voytek at gmail.com  Mon Nov 22 11:20:38 2010
From: bradley.voytek at gmail.com (Bradley Voytek)
Date: Mon, 22 Nov 2010 11:20:38 -0800
Subject: [Eeglablist] Is ITC biased by trial numbers?
In-Reply-To: <Prayer.1.3.3.1011191650270.9238@hermes-2.csi.cam.ac.uk>
References: <Prayer.1.3.3.1011191650270.9238@hermes-2.csi.cam.ac.uk>
Message-ID: <AANLkTin+6sdTW_Gy-7meCJ_mVK9Y9kcU+A4tYQm8g_WE@mail.gmail.com>

Zara:

The best approach would be to do resampling statistics.

So the question you want to know is: does your stimulus affect ITC in
some task-dependent manner? So for each subject you have an ITC value
for each time point, for each condition.

Let's say each condition has 100 trials per subject, except for the
condition shown by the red line, which has 50. Let's say you want to
compare condition red to blue. What you do now is you take your 100
blue trials, calculate ITC, and do the same separately for your 50 red
trials. Your concern is that the number of trials biases the ITC.

To address this, calculate a difference ITC (red - blue) for each
subject, giving you a real ITC difference. What's important to you
here are the trial labels: red v. blue. If you want to see if your
difference between conditions is real effect of trial label (red v.
blue), or an artifact of number of samples, you can resample your
data.

Take your 100 blue trials, 50 red trials, and create a vector of 150
trials for subject 1. Now, randomly select 100 trials from that vector
of 150 total trials, and calculate ITC from that. This gives you a
"surrogate blue" ITC. Do the same for the remaining 50 trials in the
vector to get a "surrogate red". Subtract the two to get a surrogate
difference. Repeat at least 500 times (though I recommend 10000).

Now you have a distribution (at each time point) of 10000 surrogate
ITC differences *from your actual data*. That is, you now know, given
your real data, what is the probability of getting an ITC difference
between red and blue just based on the actual data? In other words,
does the number of samples bias the ITC estimate?

Statistically, you can calculate a z-score from this distribution at
each time point, for each subject, by taking the REAL ITC difference
at that time point, subtracting the mean ITC difference from the 10000
surrogate values at that same time point, and dividing by the standard
deviation of the distribution of the 10000 values from that same time
point. From this z-score you know the probability (or significance) of
your difference, and whether it is biased by the number of trials.

Of course, because ITC is a bounded value in the range of [0,1],
technically you might want to normalize all the ITC values, but
because you're calculating a difference, in the end you should be ok.

::brad

2010/11/19 Zara Bergstr?m <zmb25 at cam.ac.uk>:
> Dear EEG experts,
>
> is the ITC measure as implemented by EEGLAB biased towards lower trial
> numbers (i.e. higher itc when fewer trials are used in the computation, as
> some measures of phase coherence supposedly are), and if so, how do you deal
> with that issue when comparing conditions with different trial numbers? Do
> you think it is appropriate to compute baseline corrected ITC, which might
> help?
>
> I analysed epoched datasets (-1-2s using default pre-stimulus baseline for
> ERSP) with wavelets using newtimef to get complex ITC values (using the
> default 'phasecoher' option), converted the ITC to real numbers between 0-1
> (absitc=sqrt(real(itc).^2+imag(itc).^2)), and averaged these into
> participant x condition x time x frequency ITC matrices for use in group
> level statistics.
>
> The attached line plot shows grand average (24 subjects) ITC averaged across
> the alpha band (8-12 hz) for four conditions. The condition with the fewest
> average trial numbers (red line) has significantly higher ITC than the other
> conditions throughout the trial, even before stimulus onset, which cannot be
> explained by psychological factors since these conditions were presented
> randomly intermixed. If I however were to subtract the average baseline
> period ITC from the post-stimulus data, it seems that the difference would
> disappear. Would that be an appropriate step to take here?
>
> Do you think this pattern is caused by a trial number bias, or have I done
> something wrong in the analysis pipeline?
>
> Any thoughts would be very much appreciated.
>
> Thanks very much for you time,
>
> Zara Bergstrom
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



From kevin_spencer at hms.harvard.edu  Mon Nov 22 11:24:07 2010
From: kevin_spencer at hms.harvard.edu (Spencer, Kevin M.)
Date: Mon, 22 Nov 2010 14:24:07 -0500
Subject: [Eeglablist] Is ITC biased by trial numbers?
In-Reply-To: <Prayer.1.3.3.1011191650270.9238@hermes-2.csi.cam.ac.uk>
References: <Prayer.1.3.3.1011191650270.9238@hermes-2.csi.cam.ac.uk>
Message-ID: <C5D74C5905A21845B94A6D13332EDC4D419BBC5B80@ITCCRMAIL02.MED.HARVARD.EDU>

Dear Zara,

Yes, ITC (in general, not just in EEGLAB) is biased by the number of trials used to compute it. One recent paper that discusses this issue is Edwards et al., 2009, J Neurophysiology.

In my experience, the best way to overcome this problem is to select approximately equal numbers of trials for each condition. With respect to baseline correction, not just the overall ITC value but the range of effects can be biased by the number of trials, so baseline correction is not the solution. I've tried statistical mapping with the permutation test, but this still seems to report too many false positives.

Good luck,
Kevin
--------------------------------------------------------------------------------------------------
Kevin M. Spencer, Ph.D. 
Director, Neural Dynamics Laboratory (http://ndl.hms.harvard.edu)
Research Health Scientist, VA Boston Healthcare System
Assistant Professor of Psychiatry, Harvard Medical School
--------------------------------------------------------------------------------------------------

________________________________________
From: eeglablist-bounces at sccn.ucsd.edu [eeglablist-bounces at sccn.ucsd.edu] On Behalf Of Zara Bergstr?m [zmb25 at cam.ac.uk]
Sent: Friday, November 19, 2010 11:50 AM
To: eeglablist at sccn.ucsd.edu
Subject: [Eeglablist] Is ITC biased by trial numbers?

Dear EEG experts,

is the ITC measure as implemented by EEGLAB biased towards lower trial
numbers (i.e. higher itc when fewer trials are used in the computation, as
some measures of phase coherence supposedly are), and if so, how do you
deal with that issue when comparing conditions with different trial
numbers? Do you think it is appropriate to compute baseline corrected ITC,
which might help?

I analysed epoched datasets (-1-2s using default pre-stimulus baseline for
ERSP) with wavelets using newtimef to get complex ITC values (using the
default 'phasecoher' option), converted the ITC to real numbers between 0-1
(absitc=sqrt(real(itc).^2+imag(itc).^2)), and averaged these into
participant x condition x time x frequency ITC matrices for use in group
level statistics.

The attached line plot shows grand average (24 subjects) ITC averaged
across the alpha band (8-12 hz) for four conditions. The condition with the
fewest average trial numbers (red line) has significantly higher ITC than
the other conditions throughout the trial, even before stimulus onset,
which cannot be explained by psychological factors since these conditions
were presented randomly intermixed. If I however were to subtract the
average baseline period ITC from the post-stimulus data, it seems that the
difference would disappear. Would that be an appropriate step to take here?

Do you think this pattern is caused by a trial number bias, or have I done
something wrong in the analysis pipeline?

Any thoughts would be very much appreciated.

Thanks very much for you time,

Zara Bergstrom


From ralphj at rpi.edu  Mon Nov 22 19:55:24 2010
From: ralphj at rpi.edu (Jason Ralph)
Date: Mon, 22 Nov 2010 22:55:24 -0500
Subject: [Eeglablist] timing issues
Message-ID: <AANLkTi=q=yTbxqFcbzYvfQ0ks6CiE2WJGEbp9o5Jx9PS@mail.gmail.com>

Hi,

I am having some major timing issues in my ERP studies.  I'm finding P300
peak latencies around 500 ms in a simple flanker task and there is no P1/N1
wave to speak of.

I suspect that the problem is a result of our display setup.  Stimuli are
presented on an LCD monitor connected through a KVM switch.  Does anyone
know if the input lag from this could have caused these problems?  We use
EGI 32 channel nets for acquisition and send triggers via a custom software
from a Mac Mini (e.g. not using E-prime).


Jason

-- 
Jason Ralph
CogWorks Laboratory
Cognitive Science Department
Rensselaer Polytechnic Institute
110 8th Street, Troy, NY 12180
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101122/7afd7d61/attachment.html>

From jakob.scherer at gmail.com  Mon Nov 22 23:28:43 2010
From: jakob.scherer at gmail.com (Jakob Scherer)
Date: Tue, 23 Nov 2010 08:28:43 +0100
Subject: [Eeglablist] Bcilab, csp and dsp
Message-ID: <AANLkTimn-isC3AZ6o_Ewg9b8w63+i6jRN2pxqaE90bmY@mail.gmail.com>

Hello everyone,
i work on single-trial eeg classification, and i am looking for
implementations of transformations. several questions arose:

1. is there any implementation of the discriminative spatial patterns
(DSP) algorithm? does not necessarily have to be in written in matlab

2. for the common spatial patterns (CSP) algorithm i thought using the
biosig4octmat-2.51\biosig\t300_FeatureExtraction\csp.m

3. is the BCILAB toolbox already available and - if yes - where can i
download it?


thanks a lot in advance,
jakob


From Michiel.Spape at nottingham.ac.uk  Tue Nov 23 00:53:00 2010
From: Michiel.Spape at nottingham.ac.uk (Michiel Spape)
Date: Tue, 23 Nov 2010 08:53:00 +0000
Subject: [Eeglablist] Is ITC biased by trial numbers?
In-Reply-To: <Prayer.1.3.3.1011191650270.9238@hermes-2.csi.cam.ac.uk>
References: <Prayer.1.3.3.1011191650270.9238@hermes-2.csi.cam.ac.uk>
Message-ID: <09DAEA8BC192C94EB62C8E71FC35A5D92900819DD8@EXCHANGE3.ad.nottingham.ac.uk>

Hi Zara,
Having some experience with ITC, or at least cross-coherence, I would say yes - since the ITC of using only one trial is necessarily 1. I find that subtracting the baseline does help somewhat, but usually, one still find effects to be larger (and often more variable) if there are fewer trials. This, I guess, is more or less the same thing one would see in typical ERP type of studies - any component being larger depending on the number of trials one averages over. One idea I've heard before on this list is to correct for this is by using random subsets of the same number of trials as in the condition with fewest trials and average over these, so that they become comparable in magnitude. Does anyone else have suggested as to what might be possible pitfalls in this type of analysis? Any rules of thumb on the amount of subsets to use, for instance?

Best,
Mich

Michiel Spap?
Research Fellow
Perception & Action group
University of Nottingham
School of Psychology
www.cognitology.eu


-----Original Message-----
From: eeglablist-bounces at sccn.ucsd.edu [mailto:eeglablist-bounces at sccn.ucsd.edu] On Behalf Of Zara Bergstr?m
Sent: 19 November 2010 16:50
To: eeglablist at sccn.ucsd.edu
Subject: [Eeglablist] Is ITC biased by trial numbers?

Dear EEG experts,

is the ITC measure as implemented by EEGLAB biased towards lower trial numbers (i.e. higher itc when fewer trials are used in the computation, as some measures of phase coherence supposedly are), and if so, how do you deal with that issue when comparing conditions with different trial numbers? Do you think it is appropriate to compute baseline corrected ITC, which might help?

I analysed epoched datasets (-1-2s using default pre-stimulus baseline for
ERSP) with wavelets using newtimef to get complex ITC values (using the default 'phasecoher' option), converted the ITC to real numbers between 0-1 (absitc=sqrt(real(itc).^2+imag(itc).^2)), and averaged these into participant x condition x time x frequency ITC matrices for use in group level statistics.

The attached line plot shows grand average (24 subjects) ITC averaged across the alpha band (8-12 hz) for four conditions. The condition with the fewest average trial numbers (red line) has significantly higher ITC than the other conditions throughout the trial, even before stimulus onset, which cannot be explained by psychological factors since these conditions were presented randomly intermixed. If I however were to subtract the average baseline period ITC from the post-stimulus data, it seems that the difference would disappear. Would that be an appropriate step to take here?

Do you think this pattern is caused by a trial number bias, or have I done something wrong in the analysis pipeline?

Any thoughts would be very much appreciated.

Thanks very much for you time,

Zara Bergstrom
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system:
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.


From mparvaz at gmail.com  Tue Nov 23 09:10:30 2010
From: mparvaz at gmail.com (MP)
Date: Tue, 23 Nov 2010 12:10:30 -0500
Subject: [Eeglablist] .locs file for Neuroscan 64 channel cap
Message-ID: <AANLkTinAFnk0DNHJ8XA2GCq+Tqkm4v8=Z6-YbW--LqFu@mail.gmail.com>

Hello all,

I am sending this message again, because the last time I sent it, I was
directed  to the Eeglab downloads page for the .locs file. On the Eeglab FTP
server, there are only .map files for Neuroscan 64 channel cap, and it says
that Eeglab is currently unable to convert it into .locs file.

So, is there any other source where I can find this file? Or, is there
anyone who is using Neuroscan 64 channel system and would like to share
their .locs file with me?

Note that these 64 channels include HEO, VEO, M1, M2, CB1 and CB2 electrodes
as well.

Thanks
- Muhammad Parvaz
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101123/1dee737e/attachment.html>

From zahra.hirji at gmail.com  Tue Nov 23 13:25:23 2010
From: zahra.hirji at gmail.com (Zahra Hirji)
Date: Tue, 23 Nov 2010 16:25:23 -0500
Subject: [Eeglablist] MEG eye blink removal
Message-ID: <AANLkTim_hn4bWv_ntgi88B+6PBXxgirJCAHbaOXutSQP@mail.gmail.com>

Hello,

Is there a recommended ICA algorithm to use on MEG data for eye blink
artifact removal? We are using a 151-channel CTF system.

Zahra
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101123/761ced3f/attachment.html>

From arno at ucsd.edu  Wed Nov 24 13:31:33 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Wed, 24 Nov 2010 13:31:33 -0800
Subject: [Eeglablist] .locs file for Neuroscan 64 channel cap
In-Reply-To: <AANLkTinAFnk0DNHJ8XA2GCq+Tqkm4v8=Z6-YbW--LqFu@mail.gmail.com>
References: <AANLkTinAFnk0DNHJ8XA2GCq+Tqkm4v8=Z6-YbW--LqFu@mail.gmail.com>
Message-ID: <3B5C1200-23CE-4FD6-8F67-8C6F11A0C0F1@ucsd.edu>

Dear Muhammad,

simply press the button "Look up locs" in the channel editor to look up standard locations for your channels (choose the BESA option).
Prior to doing that, you may rename your channels

M1 -> TP9
M2 -> TP10

I am not 100% sure that these are totally equivalent but this is standard usage.

HEO -> HEOG
VEO -> VEOG

I do not think CB1 and CB2 are regular 10/5 channels or have any equivalent. You may use the location below which are also in the regular BESA space.

http://sccn.ucsd.edu/~arno/fam2data/delorme_locfile.loc

Best regards,

A. Delorme

ps: more references 

10/20, 10/10, and 10/5 systems revisited: Their validity as relative head-surface-based positioning systems
Valer Jurcak1, Daisuke Tsuzuki1 and Ippeita Dan




On Nov 23, 2010, at 9:10 AM, MP wrote:

> Hello all,
> 
> I am sending this message again, because the last time I sent it, I was directed  to the Eeglab downloads page for the .locs file. On the Eeglab FTP server, there are only .map files for Neuroscan 64 channel cap, and it says that Eeglab is currently unable to convert it into .locs file. 
> 
> So, is there any other source where I can find this file? Or, is there anyone who is using Neuroscan 64 channel system and would like to share their .locs file with me?
> 
> Note that these 64 channels include HEO, VEO, M1, M2, CB1 and CB2 electrodes as well.
> 
> Thanks
> - Muhammad Parvaz
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101124/f1d67e91/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: PastedGraphic-1.tiff
Type: image/tiff
Size: 26868 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101124/f1d67e91/attachment.tiff>

From dgroppe at cogsci.ucsd.edu  Wed Nov 24 10:17:34 2010
From: dgroppe at cogsci.ucsd.edu (David Groppe)
Date: Wed, 24 Nov 2010 10:17:34 -0800
Subject: [Eeglablist] Bcilab, csp and dsp
In-Reply-To: <AANLkTimn-isC3AZ6o_Ewg9b8w63+i6jRN2pxqaE90bmY@mail.gmail.com>
References: <AANLkTimn-isC3AZ6o_Ewg9b8w63+i6jRN2pxqaE90bmY@mail.gmail.com>
Message-ID: <AANLkTinazcpXxC0BXw=FhUNf9sk=DFXqNf2vL9+k9g5X@mail.gmail.com>

Hi Jakob,
  You can download BCILAB and a BCILAB tutorial from the program of
the recent Swartz Center for Computational Neuroscience EEGLAB
workshop:
   http://sccn.ucsd.edu/wiki/Twelfth_EEGLAB_Workshop#Workshop_Program

It's a great piece of software.  Christian Kothe's done a great job.
        -David

On Mon, Nov 22, 2010 at 11:28 PM, Jakob Scherer <jakob.scherer at gmail.com> wrote:
> Hello everyone,
> i work on single-trial eeg classification, and i am looking for
> implementations of transformations. several questions arose:
>
> 1. is there any implementation of the discriminative spatial patterns
> (DSP) algorithm? does not necessarily have to be in written in matlab
>
> 2. for the common spatial patterns (CSP) algorithm i thought using the
> biosig4octmat-2.51\biosig\t300_FeatureExtraction\csp.m
>
> 3. is the BCILAB toolbox already available and - if yes - where can i
> download it?
>
>
> thanks a lot in advance,
> jakob
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
>



-- 
David Groppe, Ph.D.
dgroppe at cogsci.ucsd.edu
http://www.cogsci.ucsd.edu/~dgroppe/


From izadewa at yahoo.com  Thu Nov 25 06:48:17 2010
From: izadewa at yahoo.com (Bagas Isadewa)
Date: Thu, 25 Nov 2010 06:48:17 -0800 (PST)
Subject: [Eeglablist] Importing data format
Message-ID: <929403.41306.qm@web58301.mail.re3.yahoo.com>

dear All

my  name is Bagas, I'm college student from Sepuluh Nopember Institute of  
Technology (Surabaya, Indonesia) majoring in Engineering Physics.
I'm  going to do my final project research using EEGLAB, but I have some  
trouble here because I record EEG data using Biologic Ceegraph and also using 
ProFusion (.slp) that  doesn't seem compatible with EEGLAB. I read the table 
from here ( 
http://sccn.ucsd.edu/wiki/A01:_Importing_Continuous_and_Epoched_Data). The table 
explain that Biologic and ProFusion was unverified.
Do you have any suggestion for this kind of problem ? 

Thank you


best regards,

Bagas Isadewa



      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101125/278c99ba/attachment.html>

From arno at ucsd.edu  Sat Nov 27 23:36:05 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Sat, 27 Nov 2010 23:36:05 -0800
Subject: [Eeglablist] ERSP Image Question
In-Reply-To: <383987.74477.qm@web33603.mail.mud.yahoo.com>
References: <383987.74477.qm@web33603.mail.mud.yahoo.com>
Message-ID: <D4098A2C-C126-4D9A-B20D-C7AE51D751A8@ucsd.edu>

Dear Kim,

these lines are the min and max accross time or frequencies (and the dotted lines represent significance). These curves are detailed further in the help message of the newtimef function.
Best regards,

A. Delorme

On Nov 15, 2010, at 7:37 AM, Tae Kim wrote:

> Hello,
>  
> I wonder what the green line and the blue line under the upper panel in the ERSP image notate.  The manual says it is the ERSP envelope.  Then is the green high mean dB values, relative to baseline at each time in the epoch, and the blue low mean dB values relative to baseline?
>  
> Sincerely,
>  
> tk
> 
> 
> 
>       Tae Kim
> 
>       
> 
> 
> 
> 
> 
> 
> 
> <Timef.gif>_______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101127/be9b0bac/attachment.html>

From arno at ucsd.edu  Sat Nov 27 21:36:57 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Sat, 27 Nov 2010 21:36:57 -0800
Subject: [Eeglablist] Activity Power Spectrum with funktion "pop_prop()"
In-Reply-To: <9F69795E29C890408AC2DAF646C89BB37998BF7452@MAILBOX.arc.local>
References: <9F69795E29C890408AC2DAF646C89BB37998BF7452@MAILBOX.arc.local>
Message-ID: <432752F3-5AE0-43B9-9FC7-2D74B6DEC701@ucsd.edu>

Dear Beatrice,

the EEGLAB function that computes spectrum produces by default a spectrum that only uses 15% of the data (to speed up computation). That is probably why you observe a different spectrum compared to your software. You may change the default to use 100% in the graphic interface or on the Matlab command line. Another reason could be that the parameters for computing the spectrum (window size, padding, and overlap between windows) are different.

Best regards,

Arno

ps: there was a bug a couple of years ago where the spectrum function was only using the first trial. This has been fixed in 2009 in all version including EEGLAB 6.03b.

On Nov 4, 2010, at 4:20 AM, Jobst Beatrice wrote:

> Dear All,
>  
> I have a question about plotting the activity power spectrum (frequency vs. power) with the function ?pop_prop()?:
> I have a dataset, which has already been examined with another software, and now I want to compare the results with EEGLab. My problem is, that the calculations for the Power Spectrum in EEGLab produces completely different and also unrealistic results, the function values are completely different ones.
> Where could be the problem? Is it possible, that the results are wrong because I don?t have the Signal Processing Toolbox for Matlab? How can I face this problem without buying the toolbox? Can there be other reasons for the problem?
>  
> Thank you for your help!!
>  
> Best regards,
> Beatrice Jobst
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" toeeglablist-request at sccn.ucsd.edu

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101127/b6548e5e/attachment.html>

From arno at ucsd.edu  Sat Nov 27 23:26:39 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Sat, 27 Nov 2010 23:26:39 -0800
Subject: [Eeglablist] Out of memory
In-Reply-To: <AANLkTimTDq+f37GLfKFc5T0vRoCpLwPjmYAccawRNCu0@mail.gmail.com>
References: <52957.131.173.133.29.1289312289.squirrel@myuos.uni-osnabrueck.de>
	<AANLkTiknU1gxDsiccZ6m4j0GC9q_SZAr2Mtxg7nOnT_5@mail.gmail.com>
	<AANLkTimTDq+f37GLfKFc5T0vRoCpLwPjmYAccawRNCu0@mail.gmail.com>
Message-ID: <09954569-A6C4-4FAD-925C-0C1020CE5FAF@ucsd.edu>

Dear Alexandre,

increasing virtual or swap memory will most likely not help because Matlab only seems to be able to allocate real physical non-paged memory.

The most affected operating system seems to be Windows 32-bit with recent versions of Matlab where even with more than 2Gb physical memory, it is sometimes impossible to open 100Mb files.

The solutions are
- buy even more RAM
- Close all programs, remove Windows services (Adobe etc?), reboot
- Change of OS (Windows 7 might seems to have less problems than Win XP)
- Try different memory manager ?start Matlab from the DOS command line with matlab ?memmgr fast option
- Use older versions of Matlab that behave better with 32-bit systems (how old?)
- Look at http://www.mathworks.com/support/tech-notes/1100/1107.html

Under OSx or linux, Matlab cannot allocated inactive memory. You may free it by tiping "du -sx /" (in OSx you will see the blue inactive memory decreasing).

Arno

ps: under Matlab 2010b Linux Fedora Core 64-bit, we have successfully allocated matrices of up to 74Gb.

On Nov 13, 2010, at 9:23 PM, Alexandre Lehmann wrote:

> Hello All,
> 
> Klados, when you say "Try to swap your physical memory from the hard drive.", you mean adding some pagfile in memory preferences in windows ? Or are you refering to another procedure or another OS ?
> 
> I did try to increase my virtual memory by creating a pagefile of 4Gb, but even 150Mb bdf files would still give an out of memory error.
> 
> Thanks,
> 
> Regards
> 
> Alexandre
> 
> 
> 
> On Thu, Nov 11, 2010 at 2:21 AM, Klados Manousos <mklados at med.auth.gr> wrote:
> Hello to all,
> 
> Both systems you menioned are running the same programes? Because one system may run in background applications that need more memory than the other....
> 
> Try to swap your physical memory from the hard drive... With that way i achieved to load big files in EEGLAB
> 
> 2010/11/9 Benjamin Kuhr <bkuhr at uni-osnabrueck.de>
> 
> Hi,
> 
> I tried to import a large *.bdf file, 614 MB. I get the "Out of memory" -
> message, but not on all computers and for some reason it seems as it does
> not depend on the actual memory. It works on a weaker system, but not on
> the current one, which has 3.8 GiB of memory and a Intel Core 2 Duo CPU
> R7500 @ 2.93 GHz.
> 
> Even on identical systems with the same hardware and same software I can
> load the file only on one of them. I tired it with Windows XP and Ubuntu
> 10 on the system mentioned above, no difference. Any suggestions?
> 
> Thanks in advance
> Benjamin Kuhr
> 
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
> 
> 
> 
> -- 
> Manousos A. Klados
> PhD Candidate -- Research Assistant
> Group of Applied Neurosciences
> Lab of Medical Informatics
> School of Medicine
> Aristotle University of Thessaloniki
> P.O. Box 323 54124 Thessaloniki Greece
> _________________________________________________
> Tel: +30-2310-999332
> Fax:+30-2310-999263
> Website: http://lomiweb.med.auth.gr/gan/mklados
> 
> ________________________________________________________________
> ??? ????? ???????: ??? ?????? ???? ?? mail ????? ?????????? ?? ??????????.
> Acting by Reacting: By not printing this e-mail I help protect the environment.
> ________________________________________________________________
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101127/3d17f17f/attachment.html>

From sjwebb at u.washington.edu  Tue Nov 30 10:07:11 2010
From: sjwebb at u.washington.edu (Sara Jane Webb)
Date: Tue, 30 Nov 2010 10:07:11 -0800
Subject: [Eeglablist] Special Interest Group Autism & EEG/MEG
Message-ID: <C45A8277-992D-4A58-ABE4-939AFE259B9B@u.washington.edu>

Hello all,

I am writing to inform you about the opportunity for an EEG/MEG in  
Autism Special Interest Group at the International Meeting for Autism  
in San Diego 2011. To facilitate this SIG, I am soliciting feedback  
from you about the following items:

1) Your ability/interest in attending the SIG at IMFAR in San Diego  
2011.

2) Suggestions for ways to interact at the 1 hour SIG at IMFAR.

3) Suggestions for ways to interact throughout the year.

Please send back you responses within the next two weeks to sjwebb at u.washington.edu

Best,

Sara
Sara Jane Webb, PhD
Associate Professor of Psychiatry and Behavioral Sciences
Autism Research Program
http://depts.washington.edu/pbslab/
Box 357920; CHDD 314C; University of Washington
Seattle WA 98195
206.221.6461
sjwebb at u.washington.edu

Confidentiality Notice:  Because email is not secure, please be aware  
that we cannot guarantee the confidentiality of information sent by  
email.  If you are not the intended recipient, please notify the  
sender by reply email, and then destroy all copies of the message and  
any attachments.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101130/98421112/attachment.html>

From mzivot at psych.umass.edu  Tue Nov 30 05:02:38 2010
From: mzivot at psych.umass.edu (Matthew Zivot)
Date: Tue, 30 Nov 2010 08:02:38 -0500
Subject: [Eeglablist] Out of memory
In-Reply-To: <09954569-A6C4-4FAD-925C-0C1020CE5FAF@ucsd.edu>
References: <52957.131.173.133.29.1289312289.squirrel@myuos.uni-osnabrueck.de>
	<AANLkTiknU1gxDsiccZ6m4j0GC9q_SZAr2Mtxg7nOnT_5@mail.gmail.com>
	<AANLkTimTDq+f37GLfKFc5T0vRoCpLwPjmYAccawRNCu0@mail.gmail.com>
	<09954569-A6C4-4FAD-925C-0C1020CE5FAF@ucsd.edu>
Message-ID: <20101130080238.171867hz58tyqrxq@umail.oit.umass.edu>

I've been having similar memory problems importing EGI raw files.  As  
a fix, I'm playing around with the pop_readsegegi function.  It calls  
the readegi function, which reads the eeg data in as doubles, in which  
each number require 8 bytes of storage.  I have converted these to  
int8 (for the event data) which requires 1 byte and int16 (for the eeg  
data) which requires 2 bytes.  Considering the amount of data points,  
this greatly decreases the memory usage.

int8 storage can only integers from -128 to 127 but I believe that the  
event channel is only zeros and ones.  int16 storage can only hold  
integers from -32,768 to 32,767 but I believe that the eeg data is  
exclusively integers and there is no value close to 32,767 in my data.

I am not well versed in EEG analysis.  Can anyone see problems with  
this approach?

Thank you,
Matthew Zivot

Quoting Arnaud Delorme <arno at ucsd.edu>:

> Dear Alexandre,
>
> increasing virtual or swap memory will most likely not help because  
> Matlab only seems to be able to allocate real physical non-paged  
> memory.
>
> The most affected operating system seems to be Windows 32-bit with  
> recent versions of Matlab where even with more than 2Gb physical  
> memory, it is sometimes impossible to open 100Mb files.
>
> The solutions are
> - buy even more RAM
> - Close all programs, remove Windows services (Adobe etc?), reboot
> - Change of OS (Windows 7 might seems to have less problems than Win XP)
> - Try different memory manager ?start Matlab from the DOS command  
> line with matlab ?memmgr fast option
> - Use older versions of Matlab that behave better with 32-bit  
> systems (how old?)
> - Look at http://www.mathworks.com/support/tech-notes/1100/1107.html
>
> Under OSx or linux, Matlab cannot allocated inactive memory. You may  
> free it by tiping "du -sx /" (in OSx you will see the blue inactive  
> memory decreasing).
>
> Arno
>
> ps: under Matlab 2010b Linux Fedora Core 64-bit, we have  
> successfully allocated matrices of up to 74Gb.
>
> On Nov 13, 2010, at 9:23 PM, Alexandre Lehmann wrote:
>
>> Hello All,
>>
>> Klados, when you say "Try to swap your physical memory from the  
>> hard drive.", you mean adding some pagfile in memory preferences in  
>> windows ? Or are you refering to another procedure or another OS ?
>>
>> I did try to increase my virtual memory by creating a pagefile of  
>> 4Gb, but even 150Mb bdf files would still give an out of memory  
>> error.
>>
>> Thanks,
>>
>> Regards
>>
>> Alexandre
>>
>>
>>
>> On Thu, Nov 11, 2010 at 2:21 AM, Klados Manousos  
>> <mklados at med.auth.gr> wrote:
>> Hello to all,
>>
>> Both systems you menioned are running the same programes? Because  
>> one system may run in background applications that need more memory  
>> than the other....
>>
>> Try to swap your physical memory from the hard drive... With that  
>> way i achieved to load big files in EEGLAB
>>
>> 2010/11/9 Benjamin Kuhr <bkuhr at uni-osnabrueck.de>
>>
>> Hi,
>>
>> I tried to import a large *.bdf file, 614 MB. I get the "Out of memory" -
>> message, but not on all computers and for some reason it seems as it does
>> not depend on the actual memory. It works on a weaker system, but not on
>> the current one, which has 3.8 GiB of memory and a Intel Core 2 Duo CPU
>> R7500 @ 2.93 GHz.
>>
>> Even on identical systems with the same hardware and same software I can
>> load the file only on one of them. I tired it with Windows XP and Ubuntu
>> 10 on the system mentioned above, no difference. Any suggestions?
>>
>> Thanks in advance
>> Benjamin Kuhr
>>
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime"  
>> to eeglablist-request at sccn.ucsd.edu
>>
>>
>>
>> --
>> Manousos A. Klados
>> PhD Candidate -- Research Assistant
>> Group of Applied Neurosciences
>> Lab of Medical Informatics
>> School of Medicine
>> Aristotle University of Thessaloniki
>> P.O. Box 323 54124 Thessaloniki Greece
>> _________________________________________________
>> Tel: +30-2310-999332
>> Fax:+30-2310-999263
>> Website: http://lomiweb.med.auth.gr/gan/mklados
>>
>> ________________________________________________________________
>> ??? ????? ???????: ??? ?????? ???? ?? mail ????? ?????????? ?? ??????????.
>> Acting by Reacting: By not printing this e-mail I help protect the  
>> environment.
>> ________________________________________________________________
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime"  
>> to eeglablist-request at sccn.ucsd.edu
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime"  
>> to eeglablist-request at sccn.ucsd.edu
>
>




From lutobu at gmail.com  Mon Nov 29 12:20:52 2010
From: lutobu at gmail.com (ludwing torres)
Date: Mon, 29 Nov 2010 21:20:52 +0100
Subject: [Eeglablist] Why imaginary parts in dipole fitting
Message-ID: <AANLkTimRpEXRvn2hfOaumN+zFJWYQLQhc-L0XSEFjFL_@mail.gmail.com>

Hello.
Im getting this warning when I use dipole fitting:

Warning: Imaginary parts of complex X, Y, and/or Z arguments ignored
> In dipplot at 624
  In pop_dipplot at 189
  In pop_multifit at 214
Scaling components to RMS microvolt
Done.

And when I go to EEG.dipfit.model.posxyz, I get some imaginary dipole
positions.
I'm using a 33 channel definition, when I try to make the warp I get the
next error:
Scaling components to RMS microvolt
Scaling components to RMS microvolt
readlocs(): 'elc' format assumed from file extension
Reading file (lines): 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160
170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350
360 370 380 390 400 410 420 430 440 450 460 470 480 490 500 510 520 530 540
550 560 570 580 590 600 610 620 630 640 650 660 670 680 690 698
converting units from 'mm' to 'dm'
??? Subscripted assignment between dissimilar structures.

Error in ==> ft_electroderealign at 166
    template(i) = ft_convert_units(template(i), elec.unit); % ensure that
the units are consistent with the electrodes

Error in ==> coregister>warp_chans at 650
    elec3 = ft_electroderealign(cfg);

Error in ==> coregister at 200
                    [ tmp dat.transform ] = warp_chans(dat.elec1, dat.elec2,
tmpelec2.label(clist2), 'traditional');

??? Error using ==> waitfor
Error while evaluating uicontrol Callback

Please if someone knows why, give me an answer please.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101129/a749839c/attachment.html>

From l.garcia.d at gmail.com  Tue Nov 30 05:44:52 2010
From: l.garcia.d at gmail.com (Luis Garcia Dominguez)
Date: Tue, 30 Nov 2010 08:44:52 -0500
Subject: [Eeglablist] Exporting matlab output files to NetStation
In-Reply-To: <D0CD6C29-E8CD-4F5F-BC33-CF4AD4D620B1@mac.com>
References: <Gophermail.2.0.0908071150150.10361@vs-w.tc.umn.edu>
	<D0CD6C29-E8CD-4F5F-BC33-CF4AD4D620B1@mac.com>
Message-ID: <AANLkTi=6+x+T_XjSiKXnESW1zNd+8ixSoQ0btFhpb5nU@mail.gmail.com>

Hello,
I am aware the problem of exporting from eeglab to egi raw files have been
posted before but i have found no definitely answer. Is there any new
function around that performs this translation?
Thanks all,
Luis

On Fri, Aug 7, 2009 at 11:35 PM, Joseph Dien <jdien07 at mac.com> wrote:

> I should be able to help with this.  I've been working a lot with
> Matlab and EGI file formats and wrote the FieldTrip I/O modules for
> them.  First I need Arno's help to track down a couple bugs I just
> found in FieldTrip's EEGlab I/O code (having just squashed a couple
> bugs in my own code - bugs are like cockroaches, they seem to breed
> constantly).  I'll let you know as soon as we've gotten everything
> worked out.
>
> Cheers!
>
> Joe
>
> On Aug 7, 2009, at 12:50 PM, Camelia Hostinar wrote:
>
> > Hi everyone,
> >
> > We successfully converted NetStation EGI files to RAW files and
> > opened them
> > in eeglab to run ICA, but now we would like to export the pruned
> > output
> > files back to NetStation so we can run some of our already created
> > scripts.
> > The applications and instructions found at
> > ftp://ftp.egi.com/pub/support/3rdPartySoftwareSupport/matlab do not
> > seem to
> > work...Does anyone know how to bring those files back to NetStation?
> >
> > Thank you,
> > Camelia
> >
> >
> > --
> > Camelia Hostinar
> > University of Minnesota
> > Graduate Student, Institute of Child Development
> > 51 E. River Road, Minneapolis, MN 55455
> >
> >
> > _______________________________________________
> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> > To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> > For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
>
>
> --------------------------------------------------------------------------------
>
> Joseph Dien,
> Senior Research Scientist
> Center for Advanced Study of Language
> University of Maryland , Box 25
> College Park , MD 20742-0025
>
> E-mail: jdien07 at mac.com
> Phone: 301-226-8800
> Fax: 301-226-8811
> http://homepage.mac.com/jdien07/
>
>
>
>
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101130/a1aab4e1/attachment.html>

From jdien07 at mac.com  Tue Nov 30 20:14:13 2010
From: jdien07 at mac.com (Joseph Dien)
Date: Tue, 30 Nov 2010 23:14:13 -0500
Subject: [Eeglablist] Exporting matlab output files to NetStation
In-Reply-To: <AANLkTi=6+x+T_XjSiKXnESW1zNd+8ixSoQ0btFhpb5nU@mail.gmail.com>
References: <Gophermail.2.0.0908071150150.10361@vs-w.tc.umn.edu>
	<D0CD6C29-E8CD-4F5F-BC33-CF4AD4D620B1@mac.com>
	<AANLkTi=6+x+T_XjSiKXnESW1zNd+8ixSoQ0btFhpb5nU@mail.gmail.com>
Message-ID: <8C3E4516-CE75-4CFB-B6B3-FD19EF6441CB@mac.com>

Sure!  Just use my ERP PCA Toolkit.

http://sourceforge.net/projects/erppcatoolkit/

Dien, J. (2010). The ERP PCA Toolkit: An Open Source Program For Advanced Statistical Analysis of Event Related Potential Data. Journal of Neuroscience Methods, 187(1), 138-145.

Cheers!

Joe


On Nov 30, 2010, at 8:44 AM, Luis Garcia Dominguez wrote:

> Hello, 
> I am aware the problem of exporting from eeglab to egi raw files have been posted before but i have found no definitely answer. Is there any new function around that performs this translation?
> Thanks all,
> Luis
> 
> On Fri, Aug 7, 2009 at 11:35 PM, Joseph Dien <jdien07 at mac.com> wrote:
> I should be able to help with this.  I've been working a lot with
> Matlab and EGI file formats and wrote the FieldTrip I/O modules for
> them.  First I need Arno's help to track down a couple bugs I just
> found in FieldTrip's EEGlab I/O code (having just squashed a couple
> bugs in my own code - bugs are like cockroaches, they seem to breed
> constantly).  I'll let you know as soon as we've gotten everything
> worked out.
> 
> Cheers!
> 
> Joe
> 
> On Aug 7, 2009, at 12:50 PM, Camelia Hostinar wrote:
> 
> > Hi everyone,
> >
> > We successfully converted NetStation EGI files to RAW files and
> > opened them
> > in eeglab to run ICA, but now we would like to export the pruned
> > output
> > files back to NetStation so we can run some of our already created
> > scripts.
> > The applications and instructions found at
> > ftp://ftp.egi.com/pub/support/3rdPartySoftwareSupport/matlab do not
> > seem to
> > work...Does anyone know how to bring those files back to NetStation?
> >
> > Thank you,
> > Camelia
> >
> >
> > --
> > Camelia Hostinar
> > University of Minnesota
> > Graduate Student, Institute of Child Development
> > 51 E. River Road, Minneapolis, MN 55455
> >
> >
> > _______________________________________________
> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> > To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> > For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
> 
> 
> 
> 





-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101130/23d84116/attachment.html>

From jdien07 at mac.com  Tue Nov 30 21:37:19 2010
From: jdien07 at mac.com (Joseph Dien)
Date: Wed, 01 Dec 2010 00:37:19 -0500
Subject: [Eeglablist] CNS Research Assistant position at the University of
	Maryland
Message-ID: <B8326861-0D79-429B-97D3-C4F0BC3D0FC8@mac.com>

Hi,
   I am writing to let you know that we are currently searching for two CNS Research Assistants at the Center for Advanced Study of Language, here at the University of Maryland in College Park.  These would be full-time positions.  A B.S. or higher with previous experience in CNS techniques (like EEG, MEG, fMRI, or fNIRS) would be preferred.  Compensation is quite competitive.  See attached flyer for more information.  Let me know if you have any questions.

Joe

--------------------------------------------------------------------------------

Joseph Dien,
University of Maryland

E-mail: jdien07 at mac.com
Phone: 301-226-8848
Fax: 301-226-8811
http://homepage.mac.com/jdien07/

-------------- next part --------------
A non-text attachment was scrubbed...
Name: FRA-CogNS Nov 2010.pdf
Type: application/pdf
Size: 60199 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101201/b7506366/attachment.pdf>

From Beatrice.Jobst at ait.ac.at  Fri Nov 26 00:59:28 2010
From: Beatrice.Jobst at ait.ac.at (Jobst Beatrice)
Date: Fri, 26 Nov 2010 09:59:28 +0100
Subject: [Eeglablist] Export EEG-Data
Message-ID: <9F69795E29C890408AC2DAF646C89BB37998D8BFA1@MAILBOX.arc.local>

Dear all,

I want to export EEG-Data into an ASCII-text-file. I've already tried it with the function pop_export(), but I'm not happy with the outcome.
The problem is that the time column doesn't make any sense and there are way more columns than electrode-columns.
How can it be displayed differently?

For better illustration I attached the above mentioned text-file.

Yours sincerely,
Beatrice Jobst



-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: EEG-File.txt
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101126/dab47218/attachment.txt>

From bargad at hotmail.com  Wed Dec  1 03:03:37 2010
From: bargad at hotmail.com (Gadi Bartur)
Date: Wed, 1 Dec 2010 13:03:37 +0200
Subject: [Eeglablist] wrong chanel location
Message-ID: <DUB102-ds61F05DBA60413C3428CFDDC260@phx.gbl>

  Hi,
  I am preprocessing data collected with 64 EEG and 14 emg channels. the EEG C?s (C1,C2......C6) and the EMG channels connected to the C slot (of the AD-box) get confused, and while trying to edit the channels in the EDIT ?> channel locations The emg appear at the end of the list with the location of the EEG C?s while the EEG C?s C1/C3/C5/C2/C4/C6/ are empty .
  Any ideas ? How can I edit/change the emg channels before the preprocessing stage?
  thanks
  Gadi 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101201/de367085/attachment.html>

From snkartik at gmail.com  Thu Dec  2 07:47:54 2010
From: snkartik at gmail.com (Kartik Samala Naga)
Date: Thu, 2 Dec 2010 21:17:54 +0530
Subject: [Eeglablist] cognitive levels
Message-ID: <AANLkTimX6ifDpPud-52nCz=Pf-+xzK4q7y37dhnO0HeC@mail.gmail.com>

hi all,
         how much minimum data should one record to classify a eeg wave
dominance(alpha,beta).how to classify a cognitive state.is there a
possibility to predict exactly by fft?


kartik
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101202/6f448a0b/attachment.html>

From demiral.007 at googlemail.com  Wed Dec  1 12:38:27 2010
From: demiral.007 at googlemail.com (Baris Demiral)
Date: Wed, 1 Dec 2010 20:38:27 +0000
Subject: [Eeglablist] colorbar in topoplot?
Message-ID: <AANLkTik2=bxHoiJSjJ04-WTJ44iYhaea2336KPptHo=d@mail.gmail.com>

Dear all,
I use topoplot function to generate some difference maps, but the 'colorbar'
option is not available.
colorbar is actually available in pop_topoplot function though.

Any suggestions? I need to see the max and min.

Thanks,
Baris

-- 
SB Demiral, PhD.
Department of Psychology
7 George Square
The University of Edinburgh
Edinburgh, EH8 9JZ
UK
Phone: +44 (0131) 6503063
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101201/9fcf2cca/attachment.html>

From demiral.007 at googlemail.com  Wed Dec  1 13:23:08 2010
From: demiral.007 at googlemail.com (Baris Demiral)
Date: Wed, 1 Dec 2010 21:23:08 +0000
Subject: [Eeglablist] Conducting ICA on correct or all epochs
In-Reply-To: <AANLkTinarbscZ1s8YN=s_BNmFr0kdiCcxFzy76dgXQR8@mail.gmail.com>
References: <AANLkTin6+hTbFq2Y=kQQE=_ZUiPkihcbgeDegwjGeNGb@mail.gmail.com>
	<AANLkTinarbscZ1s8YN=s_BNmFr0kdiCcxFzy76dgXQR8@mail.gmail.com>
Message-ID: <AANLkTinEK3VoGVUEffSQwH2H=X9R4+nbg6daFtvkQh8b@mail.gmail.com>

Dear David,

Thank you for your e-mail. I should mention that I am enthusiastically
following your work on the reliability of the ICA and other EEG related
issues. I also plan to use Mass Univariate ERP toolbox sometime in future.

It took me a while to test and play with the data. Well, here is what I have
done, and my evaluation:

I had max 40 trials from each condition. Participants made mean of 6
incorrect decisions per condition, so the odds of answering a question
correctly was 34/6. I analyzed the data in this order:

1-Filtered, re-referenced (bi-mastoid), epoched data goes into ICA (epochs
with very gross artifacts are removed as suggested by ADJUST algorithm
leading to around 1 or at most 2 epochs to be rejected)
2- Baseline to -200-0ms
3-If correct trials should be used, then select the correct epochs and go to
5, else go to 4
4-Use all trials
5- Use ADJUST algorithm to detect and remove problematic ICs automatically
6-Take out the incorrect trials if there are any
7-Export data for statistics

I also ran the classical method of rejecting the epochs with over 100
microvolts observed on the EOG electrodes (H1, H2 and VA2) before
baselining, leading to the elimination of mean of 6-7 epochs per participant
most probably due to the eye blinks. But, note that even though this method
is quiet common, it somehow ignores the individual EOG potential difference
strengths (some subjects might have eye blinks less than 100 microvolts
sometimes).

When I used all the epochs for feeding ICA, the statistical output (ran
ANOVA) was almost mirroring the statistical output I obtained from the
classical method, and the components which I was suspecting of being
artifacts disappeared. When I used the correct trials only, some earlier and
later components (mainly centro-temporal components which I believe not
artifacts) disappear/attenuated and in one case one new components appeared
mainly frontally.

My overall experience suggest that using 40 epochs per condition including
the incorrect epochs might be better IF you are only concerned about
'artifact correction' via a toolbox like ADJUST which is mainly depending on
the ICAs. Ignoring ICA reliability issues and assuming ADJUST treats the
data similarly every time, I think we need to use more trials.

best,
Baris

On Thu, Nov 11, 2010 at 3:33 PM, David Groppe <dgroppe at cogsci.ucsd.edu>wrote:

> Hi Baris,
>   ICA's performance will generally degrade as the number of
> electrical sources increases (see
> http://www.cogsci.ucsd.edu/~dgroppe/PUBLICATIONS/GroppeCSO2008.pdf).
> The incorrect trials probably have some EEG activity not present (or
> at least less present) in the correct trials.  So if you have
> sufficient data to run ICA on just the correct trials, it would
> probably be better just to use the correct trials.  If you don't have
> enough data using just the correct trials though, you'll probably be
> fine using the all the trials, since surely a lot of the EEG activity
> is common to both sets of trials.
>         hope this helps,
>             -David Groppe
>
>
> On Wed, Nov 10, 2010 at 4:19 PM, Baris Demiral
> <demiral.007 at googlemail.com> wrote:
> > Hi everyone,
> > I am running a simple EEG experiment where I measure reaction times and
> > accuracies.
> > I want to use ICs for artifact removal, and I will report only the
> correct
> > trials (hits).
> > So would it be better to use the correct epochs for the ICA to correct
> for
> > the artifacts or is it OK to use all the epochs to detect the artifacts
> and
> > then run the artifact correction (pop_subcomp) followed by deleting the
> > incorrect epochs?
> > Thanks,
> > Baris
> >
> > --
> > SB Demiral, PhD.
> > Department of Psychology
> > 7 George Square
> > The University of Edinburgh
> > Edinburgh, EH8 9JZ
> > UK
> > Phone: +44 (0131) 6503063
> >
> > _______________________________________________
> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> > To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> > For digest mode, send an email with the subject "set digest mime" to
> > eeglablist-request at sccn.ucsd.edu
> >
>
>
>
> --
> David Groppe, Ph.D.
> dgroppe at cogsci.ucsd.edu
> http://www.cogsci.ucsd.edu/~dgroppe/
>



-- 
SB Demiral, PhD.
Department of Psychology
7 George Square
The University of Edinburgh
Edinburgh, EH8 9JZ
UK
Phone: +44 (0131) 6503063
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101201/a2719beb/attachment.html>

From arno at ucsd.edu  Thu Dec  2 19:49:03 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Thu, 2 Dec 2010 19:49:03 -0800
Subject: [Eeglablist] colorbar in topoplot?
In-Reply-To: <AANLkTik2=bxHoiJSjJ04-WTJ44iYhaea2336KPptHo=d@mail.gmail.com>
References: <AANLkTik2=bxHoiJSjJ04-WTJ44iYhaea2336KPptHo=d@mail.gmail.com>
Message-ID: <037E3BFA-BED7-47C0-9478-CBC0ADAA43A1@ucsd.edu>

Dear Baris,

after calling topoplot simply type "cbar" or "colorbar".

Best,

Arno

On Dec 1, 2010, at 12:38 PM, Baris Demiral wrote:

> Dear all, 
> I use topoplot function to generate some difference maps, but the 'colorbar' option is not available. 
> colorbar is actually available in pop_topoplot function though.
> 
> Any suggestions? I need to see the max and min.
> 
> Thanks,
> Baris
> 
> -- 
> SB Demiral, PhD.
> Department of Psychology 
> 7 George Square
> The University of Edinburgh
> Edinburgh, EH8 9JZ
> UK
> Phone: +44 (0131) 6503063
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu




From smakeig at gmail.com  Sat Dec  4 17:42:36 2010
From: smakeig at gmail.com (Scott Makeig)
Date: Sat, 4 Dec 2010 17:42:36 -0800
Subject: [Eeglablist] Conducting ICA on correct or all epochs
In-Reply-To: <AANLkTinEK3VoGVUEffSQwH2H=X9R4+nbg6daFtvkQh8b@mail.gmail.com>
References: <AANLkTin6+hTbFq2Y=kQQE=_ZUiPkihcbgeDegwjGeNGb@mail.gmail.com>
	<AANLkTinarbscZ1s8YN=s_BNmFr0kdiCcxFzy76dgXQR8@mail.gmail.com>
	<AANLkTinEK3VoGVUEffSQwH2H=X9R4+nbg6daFtvkQh8b@mail.gmail.com>
Message-ID: <AANLkTi=+syfgBDX423XAhrnOuM7g2BnGj+COT6HBNNe-@mail.gmail.com>

Baris -

Below did you leave out a step '6.5 Back-project the remaining ICs to the
scalp channel space, then select a channel of interest.' ?

If so, you may be ignoring the strongest part of ICA-based filtering, namely
looking for significant effects at the single IC source level. Of course
here one wants then to compare ICs (and their effects) across subjects, for
which a great deal of machinery exists in EEGLAB.

Best,

Scott Makeig


On Wed, Dec 1, 2010 at 1:23 PM, Baris Demiral <demiral.007 at googlemail.com>wrote:

> Dear David,
>
> Thank you for your e-mail. I should mention that I am enthusiastically
> following your work on the reliability of the ICA and other EEG related
> issues. I also plan to use Mass Univariate ERP toolbox sometime in future.
>
> It took me a while to test and play with the data. Well, here is what I
> have done, and my evaluation:
>
> I had max 40 trials from each condition. Participants made mean of 6
> incorrect decisions per condition, so the odds of answering a question
> correctly was 34/6. I analyzed the data in this order:
>
> 1-Filtered, re-referenced (bi-mastoid), epoched data goes into ICA (epochs
> with very gross artifacts are removed as suggested by ADJUST algorithm
> leading to around 1 or at most 2 epochs to be rejected)
> 2- Baseline to -200-0ms
> 3-If correct trials should be used, then select the correct epochs and go
> to 5, else go to 4
> 4-Use all trials
> 5- Use ADJUST algorithm to detect and remove problematic ICs automatically
> 6-Take out the incorrect trials if there are any
> 7-Export data for statistics
>
> I also ran the classical method of rejecting the epochs with over 100
> microvolts observed on the EOG electrodes (H1, H2 and VA2) before
> baselining, leading to the elimination of mean of 6-7 epochs per participant
> most probably due to the eye blinks. But, note that even though this method
> is quiet common, it somehow ignores the individual EOG potential difference
> strengths (some subjects might have eye blinks less than 100 microvolts
> sometimes).
>
> When I used all the epochs for feeding ICA, the statistical output (ran
> ANOVA) was almost mirroring the statistical output I obtained from the
> classical method, and the components which I was suspecting of being
> artifacts disappeared. When I used the correct trials only, some earlier and
> later components (mainly centro-temporal components which I believe not
> artifacts) disappear/attenuated and in one case one new components appeared
> mainly frontally.
>
> My overall experience suggest that using 40 epochs per condition including
> the incorrect epochs might be better IF you are only concerned about
> 'artifact correction' via a toolbox like ADJUST which is mainly depending on
> the ICAs. Ignoring ICA reliability issues and assuming ADJUST treats the
> data similarly every time, I think we need to use more trials.
>
> best,
> Baris
>
> On Thu, Nov 11, 2010 at 3:33 PM, David Groppe <dgroppe at cogsci.ucsd.edu>wrote:
>
>> Hi Baris,
>>   ICA's performance will generally degrade as the number of
>> electrical sources increases (see
>> http://www.cogsci.ucsd.edu/~dgroppe/PUBLICATIONS/GroppeCSO2008.pdf).
>> The incorrect trials probably have some EEG activity not present (or
>> at least less present) in the correct trials.  So if you have
>> sufficient data to run ICA on just the correct trials, it would
>> probably be better just to use the correct trials.  If you don't have
>> enough data using just the correct trials though, you'll probably be
>> fine using the all the trials, since surely a lot of the EEG activity
>> is common to both sets of trials.
>>         hope this helps,
>>             -David Groppe
>>
>>
>> On Wed, Nov 10, 2010 at 4:19 PM, Baris Demiral
>> <demiral.007 at googlemail.com> wrote:
>> > Hi everyone,
>> > I am running a simple EEG experiment where I measure reaction times and
>> > accuracies.
>> > I want to use ICs for artifact removal, and I will report only the
>> correct
>> > trials (hits).
>> > So would it be better to use the correct epochs for the ICA to correct
>> for
>> > the artifacts or is it OK to use all the epochs to detect the artifacts
>> and
>> > then run the artifact correction (pop_subcomp) followed by deleting the
>> > incorrect epochs?
>> > Thanks,
>> > Baris
>> >
>> > --
>> > SB Demiral, PhD.
>> > Department of Psychology
>> > 7 George Square
>> > The University of Edinburgh
>> > Edinburgh, EH8 9JZ
>> > UK
>> > Phone: +44 (0131) 6503063
>> >
>> > _______________________________________________
>> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> > To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> > For digest mode, send an email with the subject "set digest mime" to
>> > eeglablist-request at sccn.ucsd.edu
>> >
>>
>>
>>
>> --
>> David Groppe, Ph.D.
>> dgroppe at cogsci.ucsd.edu
>> http://www.cogsci.ucsd.edu/~dgroppe/
>>
>
>
>
> --
> SB Demiral, PhD.
> Department of Psychology
> 7 George Square
> The University of Edinburgh
> Edinburgh, EH8 9JZ
> UK
> Phone: +44 (0131) 6503063
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Scott Makeig, Research Scientist and Director, Swartz Center for
Computational Neuroscience, Institute for Neural Computation, University of
California San Diego, La Jolla CA 92093-0559, http://sccn.ucsd.edu/~scott
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101204/bcb36a6d/attachment.html>

From dgroppe at cogsci.ucsd.edu  Thu Dec  2 18:57:28 2010
From: dgroppe at cogsci.ucsd.edu (David Groppe)
Date: Thu, 2 Dec 2010 21:57:28 -0500
Subject: [Eeglablist] colorbar in topoplot?
In-Reply-To: <AANLkTik2=bxHoiJSjJ04-WTJ44iYhaea2336KPptHo=d@mail.gmail.com>
References: <AANLkTik2=bxHoiJSjJ04-WTJ44iYhaea2336KPptHo=d@mail.gmail.com>
Message-ID: <AANLkTimyyEYnA6Jpkke0XZOupHYpK7E7rW0cfsQPqZXs@mail.gmail.com>

On Wed, Dec 1, 2010 at 3:38 PM, Baris Demiral
<demiral.007 at googlemail.com> wrote:
> Dear all,
> I use topoplot function to generate some difference maps, but the 'colorbar'
> option is not available.
> colorbar is actually available in pop_topoplot function though.
> Any suggestions? I need to see the max and min.

Hi Baris,
  You can use the EEGLAB function cbar.m.  For example:

dat=EEG.data(:,100,1);
figure; topoplot(dat,EEG.chanlocs); cbar('vert',0,[-1 1]*max(abs(dat)));

My attached function sig_topo.m shows how to do things like add labels
to the color bar and to mark significant electrodes on the topography.
 You might be able to modify it to suit your needs.
     cheers,
       -David

> Thanks,
> Baris
>
> --
> SB Demiral, PhD.
> Department of Psychology
> 7 George Square
> The University of Edinburgh
> Edinburgh, EH8 9JZ
> UK
> Phone: +44 (0131) 6503063
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
David Groppe, Ph.D.
dgroppe at cogsci.ucsd.edu
http://www.cogsci.ucsd.edu/~dgroppe/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: sig_topo.m
Type: application/octet-stream
Size: 11317 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101202/c229fac8/attachment.obj>

From dgroppe at cogsci.ucsd.edu  Mon Dec  6 10:12:56 2010
From: dgroppe at cogsci.ucsd.edu (David Groppe)
Date: Mon, 6 Dec 2010 13:12:56 -0500
Subject: [Eeglablist] Release of Mass Univariate ERP Toolbox
Message-ID: <AANLkTin=yuDoAGVke-86V458We2SzBxvWHW0ZGniC+bg@mail.gmail.com>

Dear Colleagues,

We are pleased to announce the release of the Mass Univariate ERP
Toolbox, a freely available set of MATLAB functions for performing
mass univariate analyses of event-related brain potentials. A mass
univariate analysis is the analysis of a massive number of
simultaneously measured dependent variables via the performance of
univariate hypothesis tests (e.g., t-tests).  Savvy corrections for
multiple comparisons are applied to make spurious findings unlikely
while still retaining a useful degree of statistical power. This
approach is popular in the neuroimaging community but has not been
commonly used by ERP researchers.

The advantages of mass univariate analyses include:
  -They reduce the need for a priori defined time windows/regions of interest
  -They can reveal unexpected effects even when a priori time
windows/regions of interest are available
  -They take full advantage of the spatial and temporal resolution of
EEG and are good for providing lower bounds on the temporal onsets of
effects (e.g., the earliest time point at which some variable or
manipulation affects stimulus processing)

The disadvantages of mass univariate analyses include:
  -Some loss of statistical power due to correction for multiple
comparisons (though much less power loss than Bonferroni correction)
  -Some popular corrections for multiple comparisons are not
guaranteed to work and may not provide the degree of certainty
provided by selective analyses of a priori time windows/regions of
interest

The current version of the toolbox features:
   1) Within and between-subject t-tests with false discovery rate
controls and control of the family-wise error
rate via permutation tests.
   2) Compatibility with EEGLAB and ERPLAB data structures
   3) Several novel methods for interactive data visualization that
should facilitate exploring your data and communicating your results
(see the attached images).

This toolbox was produced by members of Prof. Marta Kutas?s lab of the
Department of Cognitive Science at the University of California, San
Diego.  Instructions for downloading the toolbox, a tutorial, and
supportive information are all available here:
   http://openwetware.org/wiki/Mass_Univariate_ERP_Toolbox

Please contact me with any questions and we hope the software can be
of use to you,
      -David Groppe

-- 
David Groppe, Ph.D.
dgroppe at cogsci.ucsd.edu
http://www.cogsci.ucsd.edu/~dgroppe/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Gui_erp_tmax_xo.jpg
Type: image/jpeg
Size: 605722 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101206/a1bd5aad/attachment.jpg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: raster_xo.jpg
Type: image/jpeg
Size: 392032 bytes
Desc: not available
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101206/a1bd5aad/attachment-0001.jpg>

From wambua.kazi at gmail.com  Mon Dec  6 10:31:34 2010
From: wambua.kazi at gmail.com (Wambua Kazi)
Date: Mon, 6 Dec 2010 13:31:34 -0500
Subject: [Eeglablist] code for converting EEGLAB to fieldtrip data formats
	(and vice versa)
Message-ID: <AANLkTinbvNZU2M6K7-57roBZ6EqOZY9iOTHsfrFEKYL6@mail.gmail.com>

Hello,
   Does anyone have code for converting from EEGLAB EEG data structures to a
fieldtrip structure (and vice versa) that she or he would like to share?
thank you,
     Wambua
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101206/3c40752e/attachment.html>

From saim_rasheed at hotmail.com  Thu Dec  9 09:45:53 2010
From: saim_rasheed at hotmail.com (Saim Rasheed)
Date: Thu, 9 Dec 2010 22:45:53 +0500
Subject: [Eeglablist] epoch length varied during computation
Message-ID: <SNT119-W383A2BAD24AB54EEE525ADEA2E0@phx.gbl>


Hi,
 
I am using EEGLAB v9.0.3.4b
Created a STUDY with 9 different conditions and 5 subjects. Sampling rate is 256 Hz.
preComputed channel measures with default parameters using GUI.
epoch length is -500 to 1000 ms.
I have a problem while precomputing channel measures as per following message displayed.
 
"Each trial contains samples from -500 ms before to
  996 ms after the timelocking event.
  Image frequency direction: normal
Using 3 cycles at lowest frequency to 64 at highest.
Value of 'timesout' must be <= frame-winsize, 'timesout' adjusted to 99
Generating 99 time points (56.6 to 439.5 ms)
Distribution of data point for time/freq decomposition is perfectly uniform
The window size used is 285 samples (1113.28 ms) wide.
Estimating 100 log-spaced frequencies from 3.0 Hz to 128.0 Hz.
Processing time point (of 99): 10 20 30 40 50 60 70 80 90
Computing the mean baseline spectrum"
 
ERSP plots are not displayed according to my epoch lenght. It only displayes ERSP for the time points computed above (red text in bold). Indeed it should compute for the range -500 to 1000 ms. Where can I fix this parameter?
Moreover, my data is already bandpass filtered from 1 to 30 Hz but it computes frequencies from 3 Hz to 128 Hz. How to confine computation from 1 to 30 Hz. Although I can display from 1 to 30 Hz.
 
Probably I am missing some parameter fixing.
 
Any one please help. 
thanking you in advance.
 

Saim
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101209/77e91d75/attachment.html>

From Marcos.Osorno at jhuapl.edu  Tue Dec  7 06:09:09 2010
From: Marcos.Osorno at jhuapl.edu (Osorno, Marcos)
Date: Tue, 7 Dec 2010 09:09:09 -0500
Subject: [Eeglablist] EEGLab Related Careers
Message-ID: <AC78983C72177B4D9D1C14F7F4AEBA2141C453B2D1@aplesstripe.dom1.jhuapl.edu>

Good morning,

I hope this is within the scope of this list as it is only tangentially related to EEGLab. We are looking for individuals with EEG and neuroimaging experience at the Johns Hopkins University Applied Physics Laboratory.

Brain-Computer Interface (BCI) Engineer
https://owa.jhuapl.edu/psc/cg89prod_cg/EMPLOYEE/HRMS/c/HRS_HRAM.HRS_CE.GBL?Page=HRS_CE_JOB_DTL&Action=A&JobOpeningId=101679&SiteId=1&PostingSeq=1

Non-Invasive Neuroimaging Post Doctoral Fellow
https://owa.jhuapl.edu/psc/cg89prod_cg/EMPLOYEE/HRMS/c/HRS_HRAM.HRS_CE.GBL?Page=HRS_CE_JOB_DTL&Action=A&JobOpeningId=101680&SiteId=1&PostingSeq=1

VR

Marcos Osorno
=-=-=-=-=-=-=-=-=
Knowledge Operations Research Engineer
JHU Applied Physics Laboratory
marcos.osorno at jhuapl.edu
(240) 228-9187 office
(240) 393-8322 cell

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101207/0d02ee6f/attachment.html>

From jm733 at georgetown.edu  Thu Dec  9 08:20:07 2010
From: jm733 at georgetown.edu (Jacob Martin)
Date: Thu, 9 Dec 2010 11:20:07 -0500
Subject: [Eeglablist] Wireless keyboards for EEG?
Message-ID: <AANLkTi=aNvkdggqOV9ZJsxu9AnKmkNFoCAeO8y76MtKb@mail.gmail.com>

Hello,

We are considering purchasing a wireless keyboard to obtain subject
response entry during EEG recordings.

Does anyone have suggestions as to the technology that would interfere
the least with the signal and still work ok?  (e.g. Bluetooth, RF, or
IR).

Or, am I better sticking with a wired PS/2 or USB keyboard?

Thanks for any pointers!

Sincerely,
Jacob

--
Jacob G. Martin, Ph.D.
Postdoctoral Research Fellow
Georgetown University Medical Center
Department of Neuroscience
3970 Reservoir Rd NW
Research Building WP-01
Washington, DC 20007
202-687-6983


From ijaganjac at yahoo.com  Tue Dec  7 08:47:11 2010
From: ijaganjac at yahoo.com (Indir Jaganjac)
Date: Tue, 7 Dec 2010 08:47:11 -0800 (PST)
Subject: [Eeglablist]  cognitive levels
Message-ID: <525028.56114.qm@web39701.mail.mud.yahoo.com>

Hi Kartik,
?
?
?
?
?
?
?

I would suggest reading?the presentation from the 12th EEGLAB workshop "Time-Frequency analysis of biophysical time series", by Dr.Arnaud Delorme. The frequency ranges are:
?
30-60 Hz? gamma
18-21 Hz? beta
9-11 Hz? alpha
4-7 Hz? theta
0.5-2 Hz? delta
?

Perhaps it's good to record at least 20 minutes for basic three cognitive states; sleep, alert, high-alert. Then import EEG data>filter the data>basic FIR filter. In pop_eegfilt() just enter values in lower edge of the frequency pass band (Hz) and higher edge of the
frequency pass band (Hz). Then Tools>Run ICA.? In this way you can generate datasets for training and use function classify from statistics toolbox: 
class = classify(sample,training,group,type,prior).

FFT can help in the sense that training and classification are done more accurately?in time-frequency domain.
?
?
?
?
?
?
?

regards,

I. Jaganjac?? 
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?

From: Kartik Samala Naga <snkartik at gmail.com>
Precedence: list
MIME-Version: 1.0
To: eeglablist at sccn.ucsd.edu
Date: Thu, 2 Dec 2010 21:17:54 +0530
Message-ID: <AANLkTimX6ifDpPud-52nCz=Pf-+xzK4q7y37dhnO0HeC at mail.gmail.com>
Content-Type: multipart/alternative; boundary=000325550e5a52c76c04966f5899
Subject: [Eeglablist] cognitive levels
Message: 1
--000325550e5a52c76c04966f5899
Content-Type: text/plain; charset=ISO-8859-1
hi all,
???????? how much minimum data should one record to classify a eeg wave
dominance(alpha,beta).how to classify a cognitive state.is there a
possibility to predict exactly by fft?

kartik
--000325550e5a52c76c04966f5899
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
<br clear=3D"all"><br>hi all,<br>=A0 =A0 =A0 =A0=A0 how much minimum data s=
hould one record to classify a eeg wave dominance(alpha,beta).how to classi=
fy a cognitive <a href=3D"


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101207/b2233328/attachment.html>

From andre.schmidt at bli.uzh.ch  Tue Dec  7 08:08:33 2010
From: andre.schmidt at bli.uzh.ch (=?iso-8859-1?Q?Andr=E9_Schmidt?=)
Date: Tue, 7 Dec 2010 17:08:33 +0100
Subject: [Eeglablist] ICA to correct for eye movements
Message-ID: <005701cb9628$ff668c40$fe33a4c0$@schmidt@bli.uzh.ch>

Dear Colleagues,

 

I try to apply ica (runica) to correct for eye movements. EEG recordings
were made from 64 scalp electrodes using the ActiveTwo system (Biosemi, The
Netherlands).

 

Firstly, I like to know whether I have to read the data, as bdf or edf file?

 

Secondly, should I import continous data or not?

 

Thanks in advance

Regards

andr?

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101207/f2c6ad41/attachment.html>

From alexandre.lehmann at gmail.com  Mon Dec  6 17:52:30 2010
From: alexandre.lehmann at gmail.com (Alexandre Lehmann)
Date: Mon, 6 Dec 2010 19:52:30 -0600
Subject: [Eeglablist] Out of memory
In-Reply-To: <09954569-A6C4-4FAD-925C-0C1020CE5FAF@ucsd.edu>
References: <52957.131.173.133.29.1289312289.squirrel@myuos.uni-osnabrueck.de>
	<AANLkTiknU1gxDsiccZ6m4j0GC9q_SZAr2Mtxg7nOnT_5@mail.gmail.com>
	<AANLkTimTDq+f37GLfKFc5T0vRoCpLwPjmYAccawRNCu0@mail.gmail.com>
	<09954569-A6C4-4FAD-925C-0C1020CE5FAF@ucsd.edu>
Message-ID: <AANLkTikk_BcF1GSm8aqgd9ko5NgyhWLwEWEhR-12qrC3@mail.gmail.com>

Dear list

For your information a switch to Linux appears to be easing memory problems.
When importing bdf and cnt files of around 150 Mo, the previous system (32
bits Windows 7 with 4Go RAM) would return an out of memory error whereas the
current test system (32 bits Ubuntu 9 with 1.5 Go RAM) can import them
without problem.

Regards

Alexandre

On Sun, Nov 28, 2010 at 1:26 AM, Arnaud Delorme <arno at ucsd.edu> wrote:

> Dear Alexandre,
>
> increasing virtual or swap memory will most likely not help because Matlab
> only seems to be able to allocate real physical non-paged memory.
>
> The most affected operating system seems to be Windows 32-bit with recent
> versions of Matlab where even with more than 2Gb physical memory, it is
> sometimes impossible to open 100Mb files.
>
> The solutions are
> - buy even more RAM
> - Close all programs, remove Windows services (Adobe etc...), reboot
> - Change of OS (Windows 7 might seems to have less problems than Win XP)
> - Try different memory manager "start Matlab from the DOS command line
> with matlab -memmgr fast option
> - Use older versions of Matlab that behave better with 32-bit systems (how
> old?)
> - Look at http://www.mathworks.com/support/tech-notes/1100/1107.html
>
> Under OSx or linux, Matlab cannot allocated inactive memory. You may free
> it by tiping "du -sx /" (in OSx you will see the blue inactive memory
> decreasing).
>
> Arno
>
> ps: under Matlab 2010b Linux Fedora Core 64-bit, we have successfully
> allocated matrices of up to 74Gb.
>
> On Nov 13, 2010, at 9:23 PM, Alexandre Lehmann wrote:
>
> Hello All,
>
> Klados, when you say "Try to swap your physical memory from the hard
> drive.", you mean adding some pagfile in memory preferences in windows ? Or
> are you refering to another procedure or another OS ?
>
> I did try to increase my virtual memory by creating a pagefile of 4Gb, but
> even 150Mb bdf files would still give an out of memory error.
>
> Thanks,
>
> Regards
>
> Alexandre
>
>
>
> On Thu, Nov 11, 2010 at 2:21 AM, Klados Manousos <mklados at med.auth.gr>wrote:
>
>> Hello to all,
>>
>> Both systems you menioned are running the same programes? Because one
>> system may run in background applications that need more memory than the
>> other....
>>
>> Try to swap your physical memory from the hard drive... With that way i
>> achieved to load big files in EEGLAB
>>
>> 2010/11/9 Benjamin Kuhr <bkuhr at uni-osnabrueck.de>
>>
>> Hi,
>>>
>>> I tried to import a large *.bdf file, 614 MB. I get the "Out of memory" -
>>> message, but not on all computers and for some reason it seems as it does
>>> not depend on the actual memory. It works on a weaker system, but not on
>>> the current one, which has 3.8 GiB of memory and a Intel Core 2 Duo CPU
>>> R7500 @ 2.93 GHz.
>>>
>>> Even on identical systems with the same hardware and same software I can
>>> load the file only on one of them. I tired it with Windows XP and Ubuntu
>>> 10 on the system mentioned above, no difference. Any suggestions?
>>>
>>> Thanks in advance
>>> Benjamin Kuhr
>>>
>>>
>>> _______________________________________________
>>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>>> To unsubscribe, send an empty email to
>>> eeglablist-unsubscribe at sccn.ucsd.edu
>>> For digest mode, send an email with the subject "set digest mime" to
>>> eeglablist-request at sccn.ucsd.edu
>>>
>>
>>
>>
>> --
>> Manousos A. Klados
>> PhD Candidate -- Research Assistant
>> Group of Applied Neurosciences
>> Lab of Medical Informatics
>> School of Medicine
>> Aristotle University of Thessaloniki
>> P.O. Box 323 54124 Thessaloniki Greece
>> _________________________________________________
>> Tel: +30-2310-999332
>> Fax:+30-2310-999263
>> Website: http://lomiweb.med.auth.gr/gan/mklados
>>
>> ________________________________________________________________
>> ??? ????? ???????: ??? ?????? ???? ?? mail ????? ?????????? ?? ??????????.
>> Acting by Reacting: By not printing this e-mail I help protect the
>> environment.
>> ________________________________________________________________
>>
>> _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101206/ddcefd80/attachment.html>

From matt.mollison at gmail.com  Sat Dec 11 12:26:42 2010
From: matt.mollison at gmail.com (Matt Mollison)
Date: Sat, 11 Dec 2010 13:26:42 -0700
Subject: [Eeglablist] code for converting EEGLAB to fieldtrip data
 formats (and vice versa)
In-Reply-To: <AANLkTinbvNZU2M6K7-57roBZ6EqOZY9iOTHsfrFEKYL6@mail.gmail.com>
References: <AANLkTinbvNZU2M6K7-57roBZ6EqOZY9iOTHsfrFEKYL6@mail.gmail.com>
Message-ID: <AANLkTi=q9FKXAkhp097qyLuvr+2atc1g8zOHP142rgO_@mail.gmail.com>

Wambua, something like this should convert segmented events in EEGLAB .set
files to FieldTrip structures. Not sure how to go back to EEGLAB. Hope it
helps.

Matt

% ========================================
eventVales = {'target','lure'};

% the amount of time before and after the segmented events, in seconds
prepost = [0.5 1.5];

% importing eeglab .set files
ftype = 'eeglab_set';

% initialize the data structure
data = struct;

% add each event value as a field in the data struct
for evVal = 1:length(eventValues)

  % path to the eeglab .set file
  infile = fullfile('path/to/files',[eventValues{evVal},'.set']);

  cfg = [];
  cfg.dataset = infile;
  cfg.headerfile = infile;
  cfg.dataformat = ftype;
  cfg.headerformat = ftype;
  cfg.continuous = 'no';

  cfg.trialdef.prestim = prepost(1);
  cfg.trialdef.poststim = prepost(2);
  cfg.trialfun = 'trialfun_general';
  cfg.trialdef.eventtype = 'trigger';
  cfg.trialdef.eventvalue = eventValues{evVal};
  cfg = ft_definetrial(cfg);

  data.(eventValues{evVal}) = ft_preprocessing(cfg);
end
% ========================================

--
Univ. of Colorado at Boulder
Dept. of Psychology and Neuroscience
matthew.mollison at colorado.edu
http://psych.colorado.edu/~mollison/


On Mon, Dec 6, 2010 at 11:31 AM, Wambua Kazi <wambua.kazi at gmail.com> wrote:

> Hello,
>    Does anyone have code for converting from EEGLAB EEG data structures to
> a fieldtrip structure (and vice versa) that she or he would like to share?
> thank you,
>      Wambua
>
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101211/e9029e6e/attachment.html>

From mferroukhi at gmail.com  Sat Dec 11 12:45:35 2010
From: mferroukhi at gmail.com (ferroukhi merzak)
Date: Sat, 11 Dec 2010 21:45:35 +0100
Subject: [Eeglablist] Wireless keyboards for EEG?
In-Reply-To: <AANLkTi=aNvkdggqOV9ZJsxu9AnKmkNFoCAeO8y76MtKb@mail.gmail.com>
References: <AANLkTi=aNvkdggqOV9ZJsxu9AnKmkNFoCAeO8y76MtKb@mail.gmail.com>
Message-ID: <AANLkTinG7c069DLyHUe7iUkcmkvAS31c-Xz5ifcDHx8x@mail.gmail.com>

HELLO
TRY FOR WIRELESS EEG WITH MSP430

2010/12/9 Jacob Martin <jm733 at georgetown.edu>

> Hello,
>
> We are considering purchasing a wireless keyboard to obtain subject
> response entry during EEG recordings.
>
> Does anyone have suggestions as to the technology that would interfere
> the least with the signal and still work ok?  (e.g. Bluetooth, RF, or
> IR).
>
> Or, am I better sticking with a wired PS/2 or USB keyboard?
>
> Thanks for any pointers!
>
> Sincerely,
> Jacob
>
> --
> Jacob G. Martin, Ph.D.
> Postdoctoral Research Fellow
> Georgetown University Medical Center
> Department of Neuroscience
> 3970 Reservoir Rd NW
> Research Building WP-01
> Washington, DC 20007
> 202-687-6983
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Merzak FERROUKHI
Enseignant Chercheur
Laboratoire d'Instrumentation
Facult? d'Electronique et d'Informatique
USTHB
Mobile +213 770 61 44 93
 Fax + 213 21 24 71 87
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101211/282e7314/attachment.html>

From arno at ucsd.edu  Sat Dec 11 15:57:43 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Sat, 11 Dec 2010 15:57:43 -0800
Subject: [Eeglablist] code for converting EEGLAB to fieldtrip data
	formats (and vice versa)
In-Reply-To: <AANLkTi=q9FKXAkhp097qyLuvr+2atc1g8zOHP142rgO_@mail.gmail.com>
References: <AANLkTinbvNZU2M6K7-57roBZ6EqOZY9iOTHsfrFEKYL6@mail.gmail.com>
	<AANLkTi=q9FKXAkhp097qyLuvr+2atc1g8zOHP142rgO_@mail.gmail.com>
Message-ID: <064E93A2-A759-425A-8EB4-258F3A718851@ucsd.edu>

Dear Wambua,

there is the "eeglab2fieldtrip" structure to convert EEGLAB to FieldTrip (in EEGLAB code). Also we contributed files to the FILE-IO project of Fieldtrip to import EEGLAB datasets into Fieldtrip and any other project using FILE-IO. These functions also convert the EEGLAB event structure to the Fiedltrip event format.

There is no know sets of function to convert Fieldtrip back to EEGLAB although I had heard rumors a couple of years ago.

Best,

Arno

On Dec 11, 2010, at 12:26 PM, Matt Mollison wrote:

> Wambua, something like this should convert segmented events in EEGLAB .set files to FieldTrip structures. Not sure how to go back to EEGLAB. Hope it helps.
> 
> Matt
> 
> % ========================================
> eventVales = {'target','lure'};
> 
> % the amount of time before and after the segmented events, in seconds
> prepost = [0.5 1.5];
> 
> % importing eeglab .set files
> ftype = 'eeglab_set';
> 
> % initialize the data structure
> data = struct;
> 
> % add each event value as a field in the data struct
> for evVal = 1:length(eventValues)
> 
>   % path to the eeglab .set file
>   infile = fullfile('path/to/files',[eventValues{evVal},'.set']);
>   
>   cfg = [];
>   cfg.dataset = infile;
>   cfg.headerfile = infile;
>   cfg.dataformat = ftype;
>   cfg.headerformat = ftype;
>   cfg.continuous = 'no';
>   
>   cfg.trialdef.prestim = prepost(1);
>   cfg.trialdef.poststim = prepost(2);
>   cfg.trialfun = 'trialfun_general';
>   cfg.trialdef.eventtype = 'trigger';
>   cfg.trialdef.eventvalue = eventValues{evVal};
>   cfg = ft_definetrial(cfg);
>   
>   data.(eventValues{evVal}) = ft_preprocessing(cfg);
> end
> % ========================================
> 
> --
> Univ. of Colorado at Boulder
> Dept. of Psychology and Neuroscience
> matthew.mollison at colorado.edu
> http://psych.colorado.edu/~mollison/
> 
> 
> On Mon, Dec 6, 2010 at 11:31 AM, Wambua Kazi <wambua.kazi at gmail.com> wrote:
> Hello,
>    Does anyone have code for converting from EEGLAB EEG data structures to a fieldtrip structure (and vice versa) that she or he would like to share?
> thank you,
>      Wambua
> 
> 
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101211/2361dba6/attachment.html>

From arno at ucsd.edu  Sat Dec 11 16:07:35 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Sat, 11 Dec 2010 16:07:35 -0800
Subject: [Eeglablist] epoch length varied during computation
In-Reply-To: <SNT119-W383A2BAD24AB54EEE525ADEA2E0@phx.gbl>
References: <SNT119-W383A2BAD24AB54EEE525ADEA2E0@phx.gbl>
Message-ID: <CBCB678D-B09D-4606-BDEE-BE25633C722B@ucsd.edu>

Dear Sain,

> ERSP plots are not displayed according to my epoch lenght. It only displayes ERSP for the time points computed above (red text in bold). Indeed it should compute for the range -500 to 1000 ms. Where can I fix this parameter?


You cannot. To compute time-frequency decompositions, it is necessary to extract time windows. The latency of each time-frequency estimate corresponds to the center of each of these windows. Therefore it is not possible to obtain a time-frequency estimate at -500 ms (it would mean that your window size is 1 sample and it does not make sense to perform FFT or wavelet decomposition on a single sample). For instance, assuming data limits of -500 ms to +1000 millisecond, a sampling frequency of 1000 Hz (1000 samples per second), and a window size of 500 samples, the earliest time-frequency estimate may be obtained at -250 ms  latency (-500 ms is -500 samples compared to the time-locking event and the first window spam from latency -500 ms to 0 ms and has its center at -250 ms). Hope it makes more sense.


> Moreover, my data is already bandpass filtered from 1 to 30 Hz but it computes frequencies from 3 Hz to 128 Hz. How to confine computation from 1 to 30 Hz.


Use the 'freqs' option of the newtimef function (or enter [1 30] for the frequency limit in the graphic interface).

Best regards,

Arno

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101211/f10072be/attachment.html>

From andrewhill at ucla.edu  Sat Dec 11 16:10:32 2010
From: andrewhill at ucla.edu (Andrew Hill)
Date: Sat, 11 Dec 2010 16:10:32 -0800
Subject: [Eeglablist] Wireless keyboards for EEG?
In-Reply-To: <AANLkTi=aNvkdggqOV9ZJsxu9AnKmkNFoCAeO8y76MtKb@mail.gmail.com>
References: <AANLkTi=aNvkdggqOV9ZJsxu9AnKmkNFoCAeO8y76MtKb@mail.gmail.com>
Message-ID: <BB44D7B8-3BD0-4CAE-AE99-21CA93B3A1D8@ucla.edu>

Don't do it.  :)  Stay away from both wireless and USB, unless it's a special zero-latency USB protocol-based keyboard (e.g. a special keyboard designed for response collection that has an additional circuit board to monitor and adjust for the issues with USB).

Otherwise a USB protocol will have variable polling latency, and will introduce a (non-fixed) latency into each button-press that will range from about 7ms to over 30 ms, depending on various factors.  Most wireless keyboards are bluetooth, which has a much less predictable lag and jitter than that of USB.

So.. use a PS/2 keyboard or other serial or parallel response box, unless time resolution of 20-50ms is ok.

Best,
Andrew

On Dec 9, 2010, at 8:20 AM, Jacob Martin wrote:

> Hello,
> 
> We are considering purchasing a wireless keyboard to obtain subject
> response entry during EEG recordings.
> 
> Does anyone have suggestions as to the technology that would interfere
> the least with the signal and still work ok?  (e.g. Bluetooth, RF, or
> IR).
> 
> Or, am I better sticking with a wired PS/2 or USB keyboard?
> 
> Thanks for any pointers!
> 
> Sincerely,
> Jacob
> 
> --
> Jacob G. Martin, Ph.D.
> Postdoctoral Research Fellow
> Georgetown University Medical Center
> Department of Neuroscience
> 3970 Reservoir Rd NW
> Research Building WP-01
> Washington, DC 20007
> 202-687-6983
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu




From sklein at berkeley.edu  Sat Dec 11 16:57:09 2010
From: sklein at berkeley.edu (Stanley Klein)
Date: Sat, 11 Dec 2010 16:57:09 -0800
Subject: [Eeglablist] ICA to correct for eye movements
In-Reply-To: <963391316762182335@unknownmsgid>
References: <963391316762182335@unknownmsgid>
Message-ID: <AANLkTikvpVUZw=0yq+LEifvdi+MYdhfVe5L3oG45RB=p@mail.gmail.com>

Dear Andre,
   Keren, Yuval-Greenberg & Deouell (Neuroimage 49, 2010 p 2248-2263)  has
an excellent article on using ICA to remove the eye movement artifacts from
EEG.
Stan

On Tue, Dec 7, 2010 at 8:08 AM, Andr? Schmidt <andre.schmidt at bli.uzh.ch>wrote:

>  Dear Colleagues,
>
>
>
> I try to apply ica (runica) to correct for eye movements. EEG recordings
> were made from 64 scalp electrodes using the ActiveTwo system (Biosemi, The
> Netherlands).
>
>
>
> Firstly, I like to know whether I have to read the data, as bdf or edf
> file?
>
>
>
> Secondly, should I import continous data or not?
>
>
>
> Thanks in advance
>
> Regards
>
> andr?
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101211/63935e34/attachment.html>

From bradley.voytek at gmail.com  Sat Dec 11 17:52:01 2010
From: bradley.voytek at gmail.com (Bradley Voytek)
Date: Sat, 11 Dec 2010 17:52:01 -0800
Subject: [Eeglablist] epoch length varied during computation
In-Reply-To: <SNT119-W383A2BAD24AB54EEE525ADEA2E0@phx.gbl>
References: <SNT119-W383A2BAD24AB54EEE525ADEA2E0@phx.gbl>
Message-ID: <AANLkTikbr86rWoyQi5g6w5HOr1xJ4V_u23xPdKzXOPkF@mail.gmail.com>

Zach:

Because EEGLAB uses sliding-window wavelets on epoched data to
decompose the ERSP, your parameters will have a large effect on your
decomposition window. This line is key:

The window size used is 285 samples (1113.28 ms) wide.

Because your window size if 285 samples (determined by the number of
cycles used, the lowest frequency of interest, etc.), and your
sampling rate is probably 265 samples/second, then your decomposition
has no resolution at any time points before 56.6 ms or after 439.5 ms.

The math is simple:

* 1113.28/2 = 556.64.
* -500 + 556.64 = 56.64
* 996 - 556.64 = 439.36

You need to increase your epoch length or--if I remember newtimef
correctly--you might be able to increase your padratio to change the
size of these windows.

::brad

On Thu, Dec 9, 2010 at 09:45, Saim Rasheed <saim_rasheed at hotmail.com> wrote:
> Hi,
>
> I am using EEGLAB v9.0.3.4b
> Created a STUDY with 9 different conditions and 5 subjects. Sampling rate is
> 256 Hz.
> preComputed channel measures with default parameters using GUI.
> epoch length is -500 to 1000 ms.
> I have a problem while precomputing channel measures as per following
> message displayed.
>
> "Each trial contains samples from -500 ms before to
> ? 996 ms after the timelocking event.
> ? Image frequency direction: normal
> Using 3 cycles at lowest frequency to 64 at highest.
> Value of 'timesout' must be <= frame-winsize, 'timesout' adjusted to 99
> Generating 99 time points (56.6 to 439.5 ms)
> Distribution of data point for time/freq decomposition is perfectly uniform
> The window size used is 285 samples (1113.28 ms) wide.
> Estimating 100 log-spaced frequencies from 3.0 Hz to 128.0 Hz.
> Processing time point (of 99): 10 20 30 40 50 60 70 80 90
> Computing the mean baseline spectrum"
>
> ERSP plots are not displayed according to my epoch lenght. It only displayes
> ERSP for the time points computed above (red text in bold). Indeed it should
> compute for the range -500 to 1000 ms. Where can I fix this parameter?
> Moreover, my data is already bandpass filtered from 1 to 30 Hz but it
> computes frequencies?from 3 Hz to 128 Hz. How to confine computation from 1
> to 30 Hz. Although I can display from 1 to 30 Hz.
>
> Probably I am missing some parameter fixing.
>
> Any one please help.
> thanking you in advance.
>
>
> Saim
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



From mahesh.casiraghi at gmail.com  Sat Dec 11 18:34:16 2010
From: mahesh.casiraghi at gmail.com (Mahesh Casiraghi)
Date: Sat, 11 Dec 2010 21:34:16 -0500
Subject: [Eeglablist] How to correctly break down AR runica() in case of
	huge sets.
Message-ID: <AANLkTimG5BKhx41EqakjQVhgPDtcRRDRtHcJD3L82tB_@mail.gmail.com>

Dear more experienced EEGLabbers and ICA experts,


supposing one has to work with quite large datsets (several channels, very
high sample rate, long record lengths) and would therefore be unable to load
in memory several gigs of data altogether:

A) Is it methodologically problematic to run independent ICAs on subgroups
of trials and then separately perform AR (blinks and scalp detected ECG
components rejection) on each of them?

B) Assuming it would not be, as I tend indeed to think, a so recommendable
way, is there a methodologically proof way to combine all the obtained - and
presumably heterogeneous - sphere, weights and weights(-1) matrices in 3
single Sph, W, and W(-1) matrices and then use these new to backproject
after component rejection?

C) More precisely, let's suppose we have 700 trials and we run 7 independent
ICAs each time on 100 of them.

a) I would proceed in picking-up separately (subjective criteria, adjust,
faster or whatever one may prefer) the to-be-rejected components,
independently from each subgroup of trials.
b) I would then remove subgroup by subgroup the respective w(-1) columns and
EEG.icaact rows according to the discarded components.
c) I would merge the obtained 7 EEG.icasphere, the 7 EEG.icaweights, and the
7 EEG.icawinv, in 3 single matrices of equal dimensions, averaging through
nanmean (given the fact we are likely to pick up a different amount of
components from each of the trial subgroups and we would need consistent
matrix dimensions).
d) I would finally independently backproject subgroup by subgroup using the
same averaged EEG.icawinv and EEG.icasphere and each time the EEG.icaact of
the current subgroup of trials.

According to my first speculations, following a->b->c->d we should come up
with something analogous to the output of a big global ICA.

Am I wrong?

D) Did someone among you already try to run something like that and is
perhaps willing to provide some feedbacks-impressions?


Cheers,

Mahesh




Mahesh M. Casiraghi
PhD candidate - Cognitive Sciences
Roberto Dell'Acqua Lab, University of Padova
Pierre Jolicoeur Lab, Univesit? de Montr?al
mahesh.casiraghi at umontreal.ca

I have the conviction that when Physiology will be far enough advanced, the
poet, the philosopher, and the physiologist will all understand each other.
Claude Bernard
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101211/1309866e/attachment.html>

From grighi at gmail.com  Mon Dec 13 13:10:43 2010
From: grighi at gmail.com (Giulia Righi)
Date: Mon, 13 Dec 2010 16:10:43 -0500
Subject: [Eeglablist] significance of negative power in an ERSP
In-Reply-To: <mailman.0.1292270402.3063.eeglablist@sccn.ucsd.edu>
Message-ID: <C92BF603.1C910%grighi@gmail.com>

HI all

I am sorry to bother the list with such a theoretical question.
However I was wondering if anyone has a good reference to think about the
significance of negative power as measured in an ERSP.
What neural processes can it be related to?

thank you
giulia


<><><><><><><><><><><><><><><>
Giulia Righi, PhD
Postdoctoral Research Fellow
Laboratories of Cognitive Neuroscience
Division of Developmental Medicine
Children?s Hospital Boston/Harvard Medical School
1 Autumn St.
Boston, MA 02215-5365
Ph: (857) 218-5211 | Fax: (617) 730-0518
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101213/0bed6a80/attachment.html>

From japalmer29 at gmail.com  Mon Dec 13 13:36:40 2010
From: japalmer29 at gmail.com (Jason Palmer)
Date: Mon, 13 Dec 2010 13:36:40 -0800
Subject: [Eeglablist] How to correctly break down AR runica() in case
	of	huge sets.
In-Reply-To: <AANLkTimG5BKhx41EqakjQVhgPDtcRRDRtHcJD3L82tB_@mail.gmail.com>
References: <AANLkTimG5BKhx41EqakjQVhgPDtcRRDRtHcJD3L82tB_@mail.gmail.com>
Message-ID: <015f01cb9b0d$d582cb70$80886250$@gmail.com>

Hi Mahesh,

 

Merging the results by simple averaging probably won?t work since the
components are returned in random order (even after the variance sorting,
components won?t necessarily have the same index.) Using matcorr() or a
similar component matching algorithm before averaging is one possibility.

 

But it seems to me that averaging will not improve anything in your
situation. As long as you have enough data in each data block that ICA runs
on, then the components you get should be well determined, allowing you to
remove the artifacts separately, and use the separate unmixing matrices to
decompose the different subsets.

 

I?m not sure what kind of analysis you?re doing, but for many purposes, you
want to identify brain components of interest and then analyze the
activations and possibly localize them. In this case you only need to match
up the components of interest in the separate decompositions, e.g. a frontal
midline ERN component, and collect all the trials with the activations
produced by the respective ICA unmixing matrices.

 

Again, as long as you use as much data as you can load (possibly overlapping
data blocks), the decompositions should be good by themselves. Comparing the
components of interest across decompositions will give you an idea of how
stable the components you?re looking at really are in your dataset. You
might also look into characterizing the variance of the component maps in a
bootstrapping sense, using a large number of resampled blocks.

 

It would also be possible to modify the ICA algorithm to swap out data from
the disk, but as I said, I doubt using all the data would improve the
results over using as much data as you can load into memory. To me it makes
more sense to verify the stability of the components you?re interested in,
and use the separate ICA unmixing/sphere matrices on their corresponding
data blocks, and separately back-project the components of interest, and
then collect all the trials for the final analysis.

 

Hope this is useful.

 

Best,

Jason

 

 

From: eeglablist-bounces at sccn.ucsd.edu
[mailto:eeglablist-bounces at sccn.ucsd.edu] On Behalf Of Mahesh Casiraghi
Sent: Saturday, December 11, 2010 6:34 PM
To: eeglablist at sccn.ucsd.edu
Subject: [Eeglablist] How to correctly break down AR runica() in case of
huge sets.

 

Dear more experienced EEGLabbers and ICA experts,

 

 

supposing one has to work with quite large datsets (several channels, very
high sample rate, long record lengths) and would therefore be unable to load
in memory several gigs of data altogether:

 

A) Is it methodologically problematic to run independent ICAs on subgroups
of trials and then separately perform AR (blinks and scalp detected ECG
components rejection) on each of them?

 

B) Assuming it would not be, as I tend indeed to think, a so recommendable
way, is there a methodologically proof way to combine all the obtained - and
presumably heterogeneous - sphere, weights and weights(-1) matrices in 3
single Sph, W, and W(-1) matrices and then use these new to backproject
after component rejection?

 

C) More precisely, let's suppose we have 700 trials and we run 7 independent
ICAs each time on 100 of them. 

 

a) I would proceed in picking-up separately (subjective criteria, adjust,
faster or whatever one may prefer) the to-be-rejected components,
independently from each subgroup of trials.  

b) I would then remove subgroup by subgroup the respective w(-1) columns and
EEG.icaact rows according to the discarded components.

c) I would merge the obtained 7 EEG.icasphere, the 7 EEG.icaweights, and the
7 EEG.icawinv, in 3 single matrices of equal dimensions, averaging through
nanmean (given the fact we are likely to pick up a different amount of
components from each of the trial subgroups and we would need consistent
matrix dimensions).  

d) I would finally independently backproject subgroup by subgroup using the
same averaged EEG.icawinv and EEG.icasphere and each time the EEG.icaact of
the current subgroup of trials.

 

According to my first speculations, following a->b->c->d we should come up
with something analogous to the output of a big global ICA.

 

Am I wrong?

 

D) Did someone among you already try to run something like that and is
perhaps willing to provide some feedbacks-impressions?

 

 

Cheers,

 

Mahesh

 

 

  

 

Mahesh M. Casiraghi

PhD candidate - Cognitive Sciences

Roberto Dell'Acqua Lab, University of Padova

Pierre Jolicoeur Lab, Univesit? de Montr?al

mahesh.casiraghi at umontreal.ca

 

I have the conviction that when Physiology will be far enough advanced, the
poet, the philosopher, and the physiologist will all understand each other.

Claude Bernard

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101213/96321b84/attachment.html>

From naturalgump at gmail.com  Mon Dec 13 04:51:13 2010
From: naturalgump at gmail.com (gump forrest)
Date: Mon, 13 Dec 2010 20:51:13 +0800
Subject: [Eeglablist] asking help for 64 channel GSN-Hydrocel's complete
	10-20 equivalents
Message-ID: <AANLkTikuYknNYPfmBQqR=NFSzB=TL4AHckC8VMRWqCwy@mail.gmail.com>

Hi,
  I am looking for a list of  the complete 10-20 correspondence for EGI's
64-channel net (GSN-HydroCel-65 1.0). I've read the technote of the sensor
layouts, but it is a bit fuzzy. Does anyone would help me?
Thanks in advance.

Best,
gump
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101213/5396edb9/attachment.html>

From brian.murphy at unitn.it  Wed Dec 15 02:26:54 2010
From: brian.murphy at unitn.it (Brian Murphy)
Date: Wed, 15 Dec 2010 11:26:54 +0100
Subject: [Eeglablist] significance of negative power in an ERSP
In-Reply-To: <mailman.0.1292356802.20148.eeglablist@sccn.ucsd.edu>
References: <mailman.0.1292356802.20148.eeglablist@sccn.ucsd.edu>
Message-ID: <4D0897EE.1090401@unitn.it>

Hi Giulia,

negative power in ERSP is always relative to the baseline that you used. 
So it does not mean negative power per se - just that the power is lower 
in that band than it was before your event of interest.

Basic references for understanding the general meaning of spectral power 
are:
Event-related EEG/MEG synchronization and desynchronization: basic 
principles
da Pfurtscheller, FH Lopes da Silva - Clinical Neurophysiology, 1999 - 
Elsevier
Mining event-related brain dynamics
da S Makeig, S Debener, J Onton? - Trends in Cognitive Sciences, 2004 - 
Elsevier
The cognitive correlates of human brain oscillations
da MJ Kahana - Journal of Neuroscience, 2006 - neuro.cjb.net
Human gamma-frequency oscillations associated with attention and memory
da O Jensen, J Kaiser? - TRENDS in Neurosciences, 2007 - Elsevier

best,

Brian

eeglablist-request at sccn.ucsd.edu wrote:
> Send eeglablist mailing list submissions to
> 	eeglablist at sccn.ucsd.edu
>
> To subscribe or unsubscribe via the World Wide Web, visit
> 	http://sccn.ucsd.edu/mailman/listinfo/eeglablist
> or, via email, send a message with subject or body 'help' to
> 	eeglablist-request at sccn.ucsd.edu
>
> You can reach the person managing the list at
> 	eeglablist-owner at sccn.ucsd.edu
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of eeglablist digest..."
>   
> ------------------------------------------------------------------------
>
> Subject:
> [Eeglablist] significance of negative power in an ERSP
> From:
> Giulia Righi <grighi at gmail.com>
> Date:
> Mon, 13 Dec 2010 22:10:43 +0100
> To:
> "eeglablist at sccn.ucsd.edu" <eeglablist at sccn.ucsd.edu>
>
> To:
> "eeglablist at sccn.ucsd.edu" <eeglablist at sccn.ucsd.edu>
>
>
>
>
>
>
> HI all
>
>
>
> I am sorry to bother the list with such a theoretical question.
>
> However I was wondering if anyone has a good reference to think about the significance of negative power as measured in an ERSP.
>
> What neural processes can it be related to?
>
>
>
> thank you
>
> giulia
>
>
>
>
>
> <><><><><><><><><><><><><><><> 
>
> Giulia Righi, PhD
>
> Postdoctoral Research Fellow
>
> Laboratories of Cognitive Neuroscience
>
> Division of Developmental Medicine
>
> Children?s Hospital Boston/Harvard Medical School
>
> 1 Autumn St.
>
> Boston, MA 02215-5365
>
> Ph: (857) 218-5211 | Fax: (617) 730-0518
>
> ------------------------------------------------------------------------
>
> Subject:
> Re: [Eeglablist] How to correctly break down AR runica() in case of 
> huge sets.
> From:
> Jason Palmer <japalmer29 at gmail.com>
> Date:
> Mon, 13 Dec 2010 22:36:40 +0100
> To:
> 'Mahesh Casiraghi' <mahesh.casiraghi at gmail.com>, 
> "eeglablist at sccn.ucsd.edu" <eeglablist at sccn.ucsd.edu>
>
> To:
> 'Mahesh Casiraghi' <mahesh.casiraghi at gmail.com>, 
> "eeglablist at sccn.ucsd.edu" <eeglablist at sccn.ucsd.edu>
>
>
>
>
> Hi Mahesh,
>
> Merging the results by simple averaging probably won?t work since the 
> components are returned in random order (even after the variance 
> sorting, components won?t necessarily have the same index.) Using 
> matcorr() or a similar component matching algorithm before averaging 
> is one possibility.
>
> But it seems to me that averaging will not improve anything in your 
> situation. As long as you have enough data in each data block that ICA 
> runs on, then the components you get should be well determined, 
> allowing you to remove the artifacts separately, and use the separate 
> unmixing matrices to decompose the different subsets.
>
> I?m not sure what kind of analysis you?re doing, but for many 
> purposes, you want to identify brain components of interest and then 
> analyze the activations and possibly localize them. In this case you 
> only need to match up the components of interest in the separate 
> decompositions, e.g. a frontal midline ERN component, and collect all 
> the trials with the activations produced by the respective ICA 
> unmixing matrices.
>
> Again, as long as you use as much data as you can load (possibly 
> overlapping data blocks), the decompositions should be good by 
> themselves. Comparing the components of interest across decompositions 
> will give you an idea of how stable the components you?re looking at 
> really are in your dataset. You might also look into characterizing 
> the variance of the component maps in a bootstrapping sense, using a 
> large number of resampled blocks.
>
> It would also be possible to modify the ICA algorithm to swap out data 
> from the disk, but as I said, I doubt using all the data would improve 
> the results over using as much data as you can load into memory. To me 
> it makes more sense to verify the stability of the components you?re 
> interested in, and use the separate ICA unmixing/sphere matrices on 
> their corresponding data blocks, and separately back-project the 
> components of interest, and then collect all the trials for the final 
> analysis.
>
> Hope this is useful.
>
> Best,
>
> Jason
>
> *From:* eeglablist-bounces at sccn.ucsd.edu 
> [mailto:eeglablist-bounces at sccn.ucsd.edu] *On Behalf Of *Mahesh 
> Casiraghi *Sent:* Saturday, December 11, 2010 6:34 PM *To:* 
> eeglablist at sccn.ucsd.edu *Subject:* [Eeglablist] How to correctly 
> break down AR runica() in case of huge sets.
>
> Dear more experienced EEGLabbers and ICA experts,
>
> supposing one has to work with quite large datsets (several channels, 
> very high sample rate, long record lengths) and would therefore be 
> unable to load in memory several gigs of data altogether:
>
> A) Is it methodologically problematic to run independent ICAs on 
> subgroups of trials and then separately perform AR (blinks and scalp 
> detected ECG components rejection) on each of them?
>
> B) Assuming it would not be, as I tend indeed to think, a so 
> recommendable way, is there a methodologically proof way to combine 
> all the obtained - and presumably heterogeneous - sphere, weights and 
> weights(-1) matrices in 3 single Sph, W, and W(-1) matrices and then 
> use these new to backproject after component rejection?
>
> C) More precisely, let's suppose we have 700 trials and we run 7 
> independent ICAs each time on 100 of them.
>
> a) I would proceed in picking-up separately (subjective criteria, 
> adjust, faster or whatever one may prefer) the to-be-rejected 
> components, independently from each subgroup of trials.
>
> b) I would then remove subgroup by subgroup the respective w(-1) 
> columns and EEG.icaact rows according to the discarded components.
>
> c) I would merge the obtained 7 EEG.icasphere, the 7 EEG.icaweights, 
> and the 7 EEG.icawinv, in 3 single matrices of equal dimensions, 
> averaging through nanmean (given the fact we are likely to pick up a 
> different amount of components from each of the trial subgroups and we 
> would need consistent matrix dimensions).
>
> d) I would finally independently backproject subgroup by subgroup 
> using the same averaged EEG.icawinv and EEG.icasphere and each time 
> the EEG.icaact of the current subgroup of trials.
>
> According to my first speculations, following a->b->c->d we should 
> come up with something analogous to the output of a big global ICA.
>
> Am I wrong?
>
> D) Did someone among you already try to run something like that and is 
> perhaps willing to provide some feedbacks-impressions?
>
> Cheers,
>
> Mahesh
>
> Mahesh M. Casiraghi
>
> PhD candidate - Cognitive Sciences
>
> Roberto Dell'Acqua Lab, University of Padova
>
> Pierre Jolicoeur Lab, Univesit? de Montr?al
>
> mahesh.casiraghi at umontreal.ca <mailto:mahesh.casiraghi at umontreal.ca>
>
> I have the conviction that when Physiology will be far enough 
> advanced, the poet, the philosopher, and the physiologist will all 
> understand each other.
>
> Claude Bernard
>
>
> ------------------------------------------------------------------------
>
> Subject:
> [Eeglablist] asking help for 64 channel GSN-Hydrocel's complete 10-20 
> equivalents
> From:
> gump forrest <naturalgump at gmail.com>
> Date:
> Mon, 13 Dec 2010 13:51:13 +0100
> To:
> "eeglablist at sccn.ucsd.edu" <eeglablist at sccn.ucsd.edu>
>
> To:
> "eeglablist at sccn.ucsd.edu" <eeglablist at sccn.ucsd.edu>
>
>
>
> Hi, I am looking for a list of the complete 10-20 correspondence for 
> EGI's 64-channel net (GSN-HydroCel-65 1.0). I've read the technote of 
> the sensor layouts, but it is a bit fuzzy. Does anyone would help me? 
> Thanks in advance. Best, gump
>
>
>
>   


-- 
Brian Murphy
Post-Doctoral Researcher
Language, Interaction and Computation Lab
Centre for Mind/Brain Sciences
University of Trento
http://clic.cimec.unitn.it/brian/



From mahesh.casiraghi at gmail.com  Mon Dec 13 14:43:38 2010
From: mahesh.casiraghi at gmail.com (Mahesh Casiraghi)
Date: Mon, 13 Dec 2010 17:43:38 -0500
Subject: [Eeglablist] How to correctly break down AR runica() in case of
 huge sets.
In-Reply-To: <015f01cb9b0d$d582cb70$80886250$@gmail.com>
References: <AANLkTimG5BKhx41EqakjQVhgPDtcRRDRtHcJD3L82tB_@mail.gmail.com>
	<015f01cb9b0d$d582cb70$80886250$@gmail.com>
Message-ID: <AANLkTimPS7hyiM4kHkmLoaDvEnyptu1+4j3BHrMwQFyf@mail.gmail.com>

Dear Jason,

thank you for all the useful hints provided along with your response.

I see your points, but I am still not sure on how performing ICA on
arbitrarily sampled subgroups of trials can be methodologically proof. I try
with a simple example: let's suppose we have to run an ICA for artifact
rejection in one experiment with 5 conditions with 100 trials each, and for
some reasons the physical features of the stimuli in condition 3
dramatically increase the probability of eyeblinks in those trials where
those stimuli are present. Let's consider the extreme situation in which all
the blinks would be concentrated only in 100% of trials belonging to the
experimental condition 3. Then we run, according to what reported above, 5
independent ICAs, each one on 100 trials.

Now, if we do not care about the proportion of trials per condition in each
of the 5 ICAs, we can end up with two prototypical situations:

- one where we will have the 100 trials of the critical condition 3 all in
one ICA (and thus we will be likely to observe a big "blink component"
accounting for a lot of variance),

- and one where we will have 20 trials of condition 3 (and so with blinks)
in each one of the 5 ICAs, and we will therefore likely to observe 5 blink
components explaining approximately 1/5 of the big blink component above,
but this time in each one 5 ICAs.

The point is: as far as the sum of the accounted variances by each of the 5
components is not identical to the one accounted by the single big
component, we know that we introduce a bias in performing solution 1 rather
than solution 2. Perhaps that does not imply that this difference reflects
the amount of neural activity we erroneously removed from one condition
rather than from another, but as far as our subgroups are not balanced
condition-wise, it means that we will introduce some artifactual
condition-related variability in our data.

In line, the question is: should we concern about the proportion of trials
per experimental condition introduced in each of the "subgroup ICAs",
whenever we would need to decompose the ICA as a consequence of processing
constraints?

And, if we do, to what extent our final backprojected data will eventually
be equal to the output of a big global ICA?

Third and last question:
I see you write:

*It would also be possible to modify the ICA algorithm to swap out data from
the disk, but as I said, I doubt using all the data would improve the
results over using as much data as you can load into memory. *

Is there a function (or in alternative a relatively easy way) to run runica
like that, or to run it in parallel on multiple machines - cores in a
Cluster-GPU like manner, maybe making use of the parallel processing
toolbox?

Hope I did not abuse of your helpfulness with all those kind of issues.

Cheers,

Mahesh



Mahesh M. Casiraghi
PhD candidate - Cognitive Sciences
Roberto Dell'Acqua Lab, University of Padova
Pierre Jolicoeur Lab, Univesit? de Montr?al
mahesh.casiraghi at umontreal.ca

I have the conviction that when Physiology will be far enough advanced, the
poet, the philosopher, and the physiologist will all understand each other.
Claude Bernard



On Mon, Dec 13, 2010 at 4:36 PM, Jason Palmer <japalmer29 at gmail.com> wrote:

> Hi Mahesh,
>
>
>
> Merging the results by simple averaging probably won?t work since the
> components are returned in random order (even after the variance sorting,
> components won?t necessarily have the same index.) Using matcorr() or a
> similar component matching algorithm before averaging is one possibility.
>
>
>
> But it seems to me that averaging will not improve anything in your
> situation. As long as you have enough data in each data block that ICA runs
> on, then the components you get should be well determined, allowing you to
> remove the artifacts separately, and use the separate unmixing matrices to
> decompose the different subsets.
>
>
>
> I?m not sure what kind of analysis you?re doing, but for many purposes, you
> want to identify brain components of interest and then analyze the
> activations and possibly localize them. In this case you only need to match
> up the components of interest in the separate decompositions, e.g. a frontal
> midline ERN component, and collect all the trials with the activations
> produced by the respective ICA unmixing matrices.
>
>
>
> Again, as long as you use as much data as you can load (possibly
> overlapping data blocks), the decompositions should be good by themselves.
> Comparing the components of interest across decompositions will give you an
> idea of how stable the components you?re looking at really are in your
> dataset. You might also look into characterizing the variance of the
> component maps in a bootstrapping sense, using a large number of resampled
> blocks.
>
>
>
> It would also be possible to modify the ICA algorithm to swap out data from
> the disk, but as I said, I doubt using all the data would improve the
> results over using as much data as you can load into memory. To me it makes
> more sense to verify the stability of the components you?re interested in,
> and use the separate ICA unmixing/sphere matrices on their corresponding
> data blocks, and separately back-project the components of interest, and
> then collect all the trials for the final analysis.
>
>
>
> Hope this is useful.
>
>
>
> Best,
>
> Jason
>
>
>
>
>
> *From:* eeglablist-bounces at sccn.ucsd.edu [mailto:
> eeglablist-bounces at sccn.ucsd.edu] *On Behalf Of *Mahesh Casiraghi
> *Sent:* Saturday, December 11, 2010 6:34 PM
> *To:* eeglablist at sccn.ucsd.edu
> *Subject:* [Eeglablist] How to correctly break down AR runica() in case of
> huge sets.
>
>
>
> Dear more experienced EEGLabbers and ICA experts,
>
>
>
>
>
> supposing one has to work with quite large datsets (several channels, very
> high sample rate, long record lengths) and would therefore be unable to load
> in memory several gigs of data altogether:
>
>
>
> A) Is it methodologically problematic to run independent ICAs on subgroups
> of trials and then separately perform AR (blinks and scalp detected ECG
> components rejection) on each of them?
>
>
>
> B) Assuming it would not be, as I tend indeed to think, a so recommendable
> way, is there a methodologically proof way to combine all the obtained - and
> presumably heterogeneous - sphere, weights and weights(-1) matrices in 3
> single Sph, W, and W(-1) matrices and then use these new to backproject
> after component rejection?
>
>
>
> C) More precisely, let's suppose we have 700 trials and we run 7
> independent ICAs each time on 100 of them.
>
>
>
> a) I would proceed in picking-up separately (subjective criteria, adjust,
> faster or whatever one may prefer) the to-be-rejected components,
> independently from each subgroup of trials.
>
> b) I would then remove subgroup by subgroup the respective w(-1) columns
> and EEG.icaact rows according to the discarded components.
>
> c) I would merge the obtained 7 EEG.icasphere, the 7 EEG.icaweights, and
> the 7 EEG.icawinv, in 3 single matrices of equal dimensions, averaging
> through nanmean (given the fact we are likely to pick up a different amount
> of components from each of the trial subgroups and we would need consistent
> matrix dimensions).
>
> d) I would finally independently backproject subgroup by subgroup using the
> same averaged EEG.icawinv and EEG.icasphere and each time the EEG.icaact of
> the current subgroup of trials.
>
>
>
> According to my first speculations, following a->b->c->d we should come up
> with something analogous to the output of a big global ICA.
>
>
>
> Am I wrong?
>
>
>
> D) Did someone among you already try to run something like that and is
> perhaps willing to provide some feedbacks-impressions?
>
>
>
>
>
> Cheers,
>
>
>
> Mahesh
>
>
>
>
>
>
>
>
>
> Mahesh M. Casiraghi
>
> PhD candidate - Cognitive Sciences
>
> Roberto Dell'Acqua Lab, University of Padova
>
> Pierre Jolicoeur Lab, Univesit? de Montr?al
>
> mahesh.casiraghi at umontreal.ca
>
>
>
> I have the conviction that when Physiology will be far enough advanced, the
> poet, the philosopher, and the physiologist will all understand each other.
>
> Claude Bernard
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101213/c10085de/attachment.html>

From smakeig at ucsd.edu  Wed Dec 15 08:22:01 2010
From: smakeig at ucsd.edu (Scott Makeig)
Date: Wed, 15 Dec 2010 08:22:01 -0800
Subject: [Eeglablist] Technical positions available at the Swartz Center,
	UCSD
Message-ID: <AANLkTinmHOp7C0QGArNUu6NkTftz+zuH1mo1=_BT1FJa@mail.gmail.com>

Several staff positions at the Swartz Center for Computational Neuroscience
(SCCN) are currently open on the jobs website of the University of
California San Diego (see jobs.UCSD.edu, search: SCCN):

- Laboratory manager helping design and managing data collection and data
archiving at an innovative laboratory combining EEG, motion capture, eye
tracking, and other measures.

- EEGLAB and HeadIT project programmer using Matlab, Ruby on Rails, and
other scripting languages to advance EEGLAB and HeadIT (forthcoming), an
NIH-supported human electrophysiology, anatomic data, and integrated tools
resource.

- Binary programmer (C++, Pascal)  for DataRiver, a multi-platform,
interactive multimodal data acquisition environment (
sccn.ucsd.edu/wiki/DataSuite).

 - Video game programmer to produce a complex, interactive multi-person
experiment game environment combining a standard video gaming tool
environment with multimodal (DataRiver), multi-subject data acquisition.

To apply for these jobs, please use http://jobs.ucsd.edu - search for 'SCCN'

In addition, we are beginning to search for an experimental research
coordinator to work with Center faculty to secure, coordinate, and report
multiple research projects involving innovative human cognitive EEG and
multimodal imaging. A full-time or half-time academic or staff position is
possible for a suitable candidate who should have considerable research
experience. Interested parties please send an inquiry to Scott Makeig (
smakeig at ucsd.edu).

-- 
Scott Makeig, Research Scientist and Director, Swartz Center for
Computational Neuroscience, Institute for Neural Computation, University of
California San Diego, La Jolla CA 92093-0961,
http://sccn.ucsd.edu/~scott<http://sccn.ucsd.edu/%7Escott>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101215/50a31dc7/attachment.html>

From arno at ucsd.edu  Thu Dec 16 08:42:52 2010
From: arno at ucsd.edu (Arnaud Delorme)
Date: Thu, 16 Dec 2010 08:42:52 -0800
Subject: [Eeglablist] Research engineer position in neuroimaging - Paris,
	France
Message-ID: <2BC3E89A-8951-4245-B0F8-15EDBB6EE010@ucsd.edu>

(Message posted on behalf of Dr Sid Kouider)

Research engineer position in neuroimaging - Paris, France
 
The Department of Cognitive Science at the Ecole Normale Sup?rieure (Paris) invites applications for a research engineer position to work on Neuroimaging data in the framework of the DynaMind ERC project held by Dr Sid Kouider. The successful applicant will mainly be in charge of managing and analyzing data obtained on human subjects with fMRI in adults and high-density EEG in infants.  The successful applicant will also be involved in the design and elaboration of research protocols.
 
Candidates should have at least a master degree in a relevant domain and a solid training in neuroimaging data analysis, with robust experience in signal processing and programming. Experience with Matlab and the SPM toolbox is desirable but not a prerequisite.
 
The successful applicant will be affiliated to the Perception, Attention and Consciousness team of Dr Sid Kouider. The Department of Cognitive Science offers a rich research environment composed of a multidisciplinary set of research units on philosophy, cognitive psychology, cognitive neurosciences and computational neurosciences. Our research team, at the interface of these disciplines, is working on the cognitive and neural determinants of consciousness, with an emphasis on the distinction between unconscious perception and perceptual awareness. Our lab is  located in central Paris, within the historical Quartier Latin.
 
Consideration of applicants will commence December 15th, 2010 and will continue until the position is filled.
The net salary will range between 2 100 to 2 800 Euros per month depending on qualifications.
 
The appointment is for two years initially but can be extended later on.
 
Please send a CV (including the names and contact information of two references) to sid.kouider at ens.fr
Do not hesitate to contact us for further questions.
 
 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Dr Sid Kouider, CNRS research scientist
Ecole Normale Sup?rieure
LSCP, Pavillon Jardin,
29 rue d'Ulm, 75005 Paris, France
www.lscp.net/persons/sidk/
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 


From ijaganjac at yahoo.com  Thu Dec 16 08:52:28 2010
From: ijaganjac at yahoo.com (Indir Jaganjac)
Date: Thu, 16 Dec 2010 08:52:28 -0800 (PST)
Subject: [Eeglablist] eeglablist Digest, Vol 74, Issue 15
In-Reply-To: <mailman.0.1292443201.29517.eeglablist@sccn.ucsd.edu>
Message-ID: <800367.79135.qm@web39705.mail.mud.yahoo.com>

Hello,
?
?
?
?
I was wondering whether there are any available data on EEG during fully-immersive high-resolution VR.? I read reports from Human Interface Technology Lab (HITlab)?at the University of Washington, ?on fully-immersive VR and EEG/fMRI simultaneous measurements.? 
?
?
?
?
regards,
?
?
I. Jaganjac?

?
?

?
?
?
?
?
?
?
?

--- On Wed, 12/15/10, eeglablist-request at sccn.ucsd.edu <eeglablist-request at sccn.ucsd.edu> wrote:


From: eeglablist-request at sccn.ucsd.edu <eeglablist-request at sccn.ucsd.edu>
Subject: eeglablist Digest, Vol 74, Issue 15
To: eeglablist at sccn.ucsd.edu
Date: Wednesday, December 15, 2010, 12:00 PM


Send eeglablist mailing list submissions to
??? eeglablist at sccn.ucsd.edu

To subscribe or unsubscribe via the World Wide Web, visit
??? http://sccn.ucsd.edu/mailman/listinfo/eeglablist
or, via email, send a message with subject or body 'help' to
??? eeglablist-request at sccn.ucsd.edu

You can reach the person managing the list at
??? eeglablist-owner at sccn.ucsd.edu

When replying, please edit your Subject line so it is more specific
than "Re: Contents of eeglablist digest..."

Today's Topics:

???1. Technical positions available at the Swartz Center,??? UCSD
? ? ? (Scott Makeig)

_______________________________________________
eeglablist mailing list eeglablist at sccn.ucsd.edu
Eeglablist page: http://www.sccn.ucsd.edu/eeglab/eeglabmail.html
To unsubscribe, send an empty email to eeglablist-unsub at sccn.ucsd.edu
To switch to non-digest mode, send an empty email to eeglablist-nodigest at sccn.ucsd.edu


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101216/bd5ab2b4/attachment.html>

From japalmer29 at gmail.com  Thu Dec 16 22:04:07 2010
From: japalmer29 at gmail.com (Jason Palmer)
Date: Thu, 16 Dec 2010 22:04:07 -0800
Subject: [Eeglablist] How to correctly break down AR runica() in case of
	huge sets.
In-Reply-To: <AANLkTimPS7hyiM4kHkmLoaDvEnyptu1+4j3BHrMwQFyf@mail.gmail.com>
References: <AANLkTimG5BKhx41EqakjQVhgPDtcRRDRtHcJD3L82tB_@mail.gmail.com>	<015f01cb9b0d$d582cb70$80886250$@gmail.com>
	<AANLkTimPS7hyiM4kHkmLoaDvEnyptu1+4j3BHrMwQFyf@mail.gmail.com>
Message-ID: <009501cb9db0$3c343350$b49c99f0$@gmail.com>

Dear Mahesh,

 

I would suggest that you run ICA on contiguous blocks of data, as in your
prototypical situation (1), or e.g. 3 ICAs on conditions 1+2, 3, and 4+5.
The ICA model expects the data to be stationary, and EEG data should be more
stationary over contiguous data. The tradeoff is between having enough data
to get good components with ICA, and having too much data so that the data
cannot be modeled as being generated by a single complete basis set. So I
suggest you divide the data into segments with e.g. (nchan*5000) samples in
each segment, and run ICA on each segment, then try to match the components
you are interested in in the different decompositions, and back-project
those components using the separately determined activations.

 

Regardless of how you divide up the data, as long as there is enough data in
each segment to determine the components present (e.g. eye-blink), ICA
should be able in principle to remove all of the variance due to the
artifact. In your second example, if 20 trials of condition 3 contain enough
data to determine the eye-blink component, then you should be able to
separately remove the eye-blink from the blink trials in each segment using
that segment?s decomposition to remove and back-project. The main issue is
having enough data to determine the artifact or non-interesting components
and remove them, and to determine interesting components, and at the same
time have the data be stationary. High-pass filtering (before epoching) is
usually necessary to remove drift and achieve basic mean stationarity.

 

So given nonstationary data, a ?big global ICA? is not necessarily the best
thing, but having enough data to get good components is important. So as
long as you have enough data in your decompositions, you might even prefer
to do ICA on shorter segments, or single conditions, so that you can more
effectively remove non-stationary artifacts. Ideally you would find the type
of components you are interested in in the separate decompositions and be
able to back-project them. The power in the separately back-projected trials
should be comparable if you are back-projecting the same component(s) (e.g.
frontal midline, or mu).

 

There currently no simple way to do the swapping without modifying the code.
I will try to add such a facility in future development. We have a cluster
version of ICA at SCCN which you might be able to compile if you have a
Linux cluster. Parallelizing over multiple cores probably won?t help if your
issue is RAM.

 

Best,

Jason

 

From: Mahesh Casiraghi [mailto:mahesh.casiraghi at gmail.com] 
Sent: Monday, December 13, 2010 2:44 PM
To: japalmer29 at gmail.com; eeglablist at sccn.ucsd.edu
Subject: Re: [Eeglablist] How to correctly break down AR runica() in case of
huge sets.

 

Dear Jason,

 

thank you for all the useful hints provided along with your response.

 

I see your points, but I am still not sure on how performing ICA on
arbitrarily sampled subgroups of trials can be methodologically proof. I try
with a simple example: let's suppose we have to run an ICA for artifact
rejection in one experiment with 5 conditions with 100 trials each, and for
some reasons the physical features of the stimuli in condition 3
dramatically increase the probability of eyeblinks in those trials where
those stimuli are present. Let's consider the extreme situation in which all
the blinks would be concentrated only in 100% of trials belonging to the
experimental condition 3. Then we run, according to what reported above, 5
independent ICAs, each one on 100 trials. 

 

Now, if we do not care about the proportion of trials per condition in each
of the 5 ICAs, we can end up with two prototypical situations:

 

- one where we will have the 100 trials of the critical condition 3 all in
one ICA (and thus we will be likely to observe a big "blink component"
accounting for a lot of variance),

 

- and one where we will have 20 trials of condition 3 (and so with blinks)
in each one of the 5 ICAs, and we will therefore likely to observe 5 blink
components explaining approximately 1/5 of the big blink component above,
but this time in each one 5 ICAs.

 

The point is: as far as the sum of the accounted variances by each of the 5
components is not identical to the one accounted by the single big
component, we know that we introduce a bias in performing solution 1 rather
than solution 2. Perhaps that does not imply that this difference reflects
the amount of neural activity we erroneously removed from one condition
rather than from another, but as far as our subgroups are not balanced
condition-wise, it means that we will introduce some artifactual
condition-related variability in our data.

 

In line, the question is: should we concern about the proportion of trials
per experimental condition introduced in each of the "subgroup ICAs",
whenever we would need to decompose the ICA as a consequence of processing
constraints? 

 

And, if we do, to what extent our final backprojected data will eventually
be equal to the output of a big global ICA?

 

Third and last question: 

I see you write:

 

It would also be possible to modify the ICA algorithm to swap out data from
the disk, but as I said, I doubt using all the data would improve the
results over using as much data as you can load into memory. 

 

Is there a function (or in alternative a relatively easy way) to run runica
like that, or to run it in parallel on multiple machines - cores in a
Cluster-GPU like manner, maybe making use of the parallel processing
toolbox?

 

Hope I did not abuse of your helpfulness with all those kind of issues.

 

Cheers,

 

Mahesh

 

       

 

Mahesh M. Casiraghi

PhD candidate - Cognitive Sciences

Roberto Dell'Acqua Lab, University of Padova

Pierre Jolicoeur Lab, Univesit? de Montr?al

mahesh.casiraghi at umontreal.ca

 

I have the conviction that when Physiology will be far enough advanced, the
poet, the philosopher, and the physiologist will all understand each other.

Claude Bernard





On Mon, Dec 13, 2010 at 4:36 PM, Jason Palmer <japalmer29 at gmail.com> wrote:

Hi Mahesh,

 

Merging the results by simple averaging probably won?t work since the
components are returned in random order (even after the variance sorting,
components won?t necessarily have the same index.) Using matcorr() or a
similar component matching algorithm before averaging is one possibility.

 

But it seems to me that averaging will not improve anything in your
situation. As long as you have enough data in each data block that ICA runs
on, then the components you get should be well determined, allowing you to
remove the artifacts separately, and use the separate unmixing matrices to
decompose the different subsets.

 

I?m not sure what kind of analysis you?re doing, but for many purposes, you
want to identify brain components of interest and then analyze the
activations and possibly localize them. In this case you only need to match
up the components of interest in the separate decompositions, e.g. a frontal
midline ERN component, and collect all the trials with the activations
produced by the respective ICA unmixing matrices.

 

Again, as long as you use as much data as you can load (possibly overlapping
data blocks), the decompositions should be good by themselves. Comparing the
components of interest across decompositions will give you an idea of how
stable the components you?re looking at really are in your dataset. You
might also look into characterizing the variance of the component maps in a
bootstrapping sense, using a large number of resampled blocks.

 

It would also be possible to modify the ICA algorithm to swap out data from
the disk, but as I said, I doubt using all the data would improve the
results over using as much data as you can load into memory. To me it makes
more sense to verify the stability of the components you?re interested in,
and use the separate ICA unmixing/sphere matrices on their corresponding
data blocks, and separately back-project the components of interest, and
then collect all the trials for the final analysis.

 

Hope this is useful.

 

Best,

Jason

 

 

From: eeglablist-bounces at sccn.ucsd.edu
[mailto:eeglablist-bounces at sccn.ucsd.edu] On Behalf Of Mahesh Casiraghi
Sent: Saturday, December 11, 2010 6:34 PM
To: eeglablist at sccn.ucsd.edu
Subject: [Eeglablist] How to correctly break down AR runica() in case of
huge sets.

 

Dear more experienced EEGLabbers and ICA experts,

 

 

supposing one has to work with quite large datsets (several channels, very
high sample rate, long record lengths) and would therefore be unable to load
in memory several gigs of data altogether:

 

A) Is it methodologically problematic to run independent ICAs on subgroups
of trials and then separately perform AR (blinks and scalp detected ECG
components rejection) on each of them?

 

B) Assuming it would not be, as I tend indeed to think, a so recommendable
way, is there a methodologically proof way to combine all the obtained - and
presumably heterogeneous - sphere, weights and weights(-1) matrices in 3
single Sph, W, and W(-1) matrices and then use these new to backproject
after component rejection?

 

C) More precisely, let's suppose we have 700 trials and we run 7 independent
ICAs each time on 100 of them. 

 

a) I would proceed in picking-up separately (subjective criteria, adjust,
faster or whatever one may prefer) the to-be-rejected components,
independently from each subgroup of trials.  

b) I would then remove subgroup by subgroup the respective w(-1) columns and
EEG.icaact rows according to the discarded components.

c) I would merge the obtained 7 EEG.icasphere, the 7 EEG.icaweights, and the
7 EEG.icawinv, in 3 single matrices of equal dimensions, averaging through
nanmean (given the fact we are likely to pick up a different amount of
components from each of the trial subgroups and we would need consistent
matrix dimensions).  

d) I would finally independently backproject subgroup by subgroup using the
same averaged EEG.icawinv and EEG.icasphere and each time the EEG.icaact of
the current subgroup of trials.

 

According to my first speculations, following a->b->c->d we should come up
with something analogous to the output of a big global ICA.

 

Am I wrong?

 

D) Did someone among you already try to run something like that and is
perhaps willing to provide some feedbacks-impressions?

 

 

Cheers,

 

Mahesh

 

 

  

 

Mahesh M. Casiraghi

PhD candidate - Cognitive Sciences

Roberto Dell'Acqua Lab, University of Padova

Pierre Jolicoeur Lab, Univesit? de Montr?al

mahesh.casiraghi at umontreal.ca

 

I have the conviction that when Physiology will be far enough advanced, the
poet, the philosopher, and the physiologist will all understand each other.

Claude Bernard

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101216/e24c3a88/attachment.html>

From andre.schmidt at bli.uzh.ch  Wed Dec 22 02:15:03 2010
From: andre.schmidt at bli.uzh.ch (=?iso-8859-1?Q?Andr=E9_Schmidt?=)
Date: Wed, 22 Dec 2010 11:15:03 +0100
Subject: [Eeglablist] Remove components from data
Message-ID: <002f01cba1c1$19715950$4c540bf0$@schmidt@bli.uzh.ch>

Dear all

 

I will try to remove certain components from my EEG data after ICA to
correct for eye movements. After detecting the eye  components by maps more
or less distinctly , I removed those. 

Now I have following problems: Firstly, I cannot see this correction in the
frontal electrodes, what means the eye movement is still there. And
secondly, I will import this file into SPM to estimate effective
connectivity, but SPM was not able to convert the file. 

 

In analyzer software, I have to back-transform after ica. Is this analogous
in eeglab and when yes how is this tool called? Or what do you recommend me?

 

Thanks in advance

 

Kindly regards

Andr? Schmidt

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101222/e6bbf83c/attachment.html>

From andrewhill at ucla.edu  Mon Dec 20 14:21:59 2010
From: andrewhill at ucla.edu (Andrew Hill)
Date: Mon, 20 Dec 2010 14:21:59 -0800
Subject: [Eeglablist] ICA and Studys?
Message-ID: <821DA362-A563-42E7-95BD-38E3BD5D83C8@ucla.edu>

Just wondering if I'm doing something methodologically wrong:

I have several .bdf files with multiple behavioral events embedded in the event list - I wrote a script to import each one, filter, extract epochs of one type, run ICA on the epochs, and then save the resulting .set file.

When I create studys (via a script) with the epoched, ICA'd .sets, I get warning messages after I add each subject's files (6 per subject, using 2 different epoched files each from 3 different sessions):

Warning: ***** change STUDY design as it combines datasets with different ICA decompositions


The Studys I create this way appear to be ok: status says "ready to precluster", but have I done something "wrong"? 

Thanks,
Andrew





From bradley.voytek at gmail.com  Tue Dec 21 21:09:02 2010
From: bradley.voytek at gmail.com (Bradley Voytek)
Date: Tue, 21 Dec 2010 21:09:02 -0800
Subject: [Eeglablist] Online visual brain atlas: brainSCANr
In-Reply-To: <AANLkTimMyLwkmfqC6wwGY1bscmLB+PFW9WTrfEMEjinD@mail.gmail.com>
References: <AANLkTimMyLwkmfqC6wwGY1bscmLB+PFW9WTrfEMEjinD@mail.gmail.com>
Message-ID: <AANLkTikq5V9PBg1-pJXNz6O-bnJwZ1pUMhkPf1rqjo06@mail.gmail.com>

Dear friends and colleagues:

For years now I've been wanting a repository where I can easily check
to see what functions a brain region is associated with and what other
brain regions it's connected to.

Well, no one ever got around to making that happen, so my wife and I
just made our own.

Introducing brainSCANr!

http://www.brainscanr.com

brainSCANr (The Brain Systems, Associations, Connections, and Network
Relationships engine) works by searching PubMed for the co-occurence
of brain region name, cognitive/behavioral functions, and diseases to
build a connection matrix that you can graphically explore.

Want to know what the amygdala does and what it connects with? What
about Parkinson's disease? Attentional capacity? Just search! You can
also see all the papers used to populate the database.

All of the raw data is available for you to play with, too. Check out
all the details here:
http://www.brainscanr.com/Paper

And if you like it and think it's useful, please pass it along to
neuroscience friends and colleagues! And if you have any comments or
suggestions, let me know!

::brad


From tarikbelbahar at gmail.com  Wed Dec 22 11:33:48 2010
From: tarikbelbahar at gmail.com (Tarik S Bel-Bahar)
Date: Wed, 22 Dec 2010 11:33:48 -0800
Subject: [Eeglablist] Remove components from data
In-Reply-To: <-2997119363868756688@unknownmsgid>
References: <-2997119363868756688@unknownmsgid>
Message-ID: <AANLkTikg6BTxW_sMnT8zaxHb=E7CsVEduOHu3oYZqzjk@mail.gmail.com>

Hi Andre,

A few quick thoughts that might help.

1. First try to make sure you have removed all the components you believe to
be artifactual.
2. Then using the new file, save out as a text file and using only EEG as
output, which then should be importable into the application of your choice.
3. If you are examining your back projected single trial EEG data and cannot
see your eye movement corrections, it's possible that significant
artifactual ICs remain in the data.
4. You may want to consider ADJUST, FASTER, or Dien's PCA toolbox, which all
use EEGLAB to automatize removal of artifactual ICs.

Good luck!



On Wed, Dec 22, 2010 at 2:15 AM, Andr? Schmidt <andre.schmidt at bli.uzh.ch>wrote:

> Dear all
>
>
>
> I will try to remove certain components from my EEG data after ICA to
> correct for eye movements. After detecting the eye  components by maps more
> or less distinctly , I removed those.
>
> Now I have following problems: Firstly, I cannot see this correction in the
> frontal electrodes, what means the eye movement is still there. And
> secondly, I will import this file into SPM to estimate effective
> connectivity, but SPM was not able to convert the file.
>
>
>
> In analyzer software, I have to back-transform after ica. Is this analogous
> in eeglab and when yes how is this tool called? Or what do you recommend me?
>
>
>
> Thanks in advance
>
>
>
> Kindly regards
>
> Andr? Schmidt
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101222/64c9e98f/attachment.html>

From izadewa at yahoo.com  Thu Dec 23 01:09:13 2010
From: izadewa at yahoo.com (Bagas Isadewa)
Date: Thu, 23 Dec 2010 01:09:13 -0800 (PST)
Subject: [Eeglablist] (ask) about averaging and ICA
Message-ID: <618915.98563.qm@web58304.mail.re3.yahoo.com>

hello,

i have several questions, 

1. if I wanted to average EEG data to yields good ERP, how many minimum trials 
do I needed ? is there any paper or journals that explained about it ?
2. then, if I wanted to run ICA to extract ERP, do I need many trials or only 
single-trials in one epoch will be enough ??
3. is it necessary to run ICA first then do averaging or it is enough to use 
only one technique for extracting ERP ?

if you don't mind please give me an explanation
thank you very much


regards,
bagas isadewa


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101223/82e2579d/attachment.html>

From tarikbelbahar at gmail.com  Sun Dec 26 17:45:01 2010
From: tarikbelbahar at gmail.com (Tarik S Bel-Bahar)
Date: Sun, 26 Dec 2010 17:45:01 -0800
Subject: [Eeglablist] (ask) about averaging and ICA
In-Reply-To: <618915.98563.qm@web58304.mail.re3.yahoo.com>
References: <618915.98563.qm@web58304.mail.re3.yahoo.com>
Message-ID: <AANLkTikCs55efRtaM-gMg+E43vwVrYV0i0igjfvHovtu@mail.gmail.com>

Greetings Bagas,

1. Yes, there are multiple papers that refer to this issue. I would suggest
you start
with the HANDBOOKS by Tom Handy or Steve Luck, and these will answer your
basic questions about the relations
between trial counts, averaged ERPs, and the resultant signal-to-noise
ratios.
You can find these books by doing a Google search, and you can order them
from Amazon or another bookseller.
Here's the link for the Handy handbook:
http://mitpress.mit.edu/catalog/item/default.asp?tid=10253&ttype=2

2. Your second question is not clear. It's not clear what you mean by "many
trials" or only "single-trials" in one epoch.
For information about best practice using EEGLAB and ICA, please read
through the EEGLAB tutorial online.
A general rule of thumb is that more single trials are better than fewer
single trials, for an accurate ICA decomposition of the data.

3. Again, you should read and learn basic information about ERP techniques
from handbooks mentioned in Point 1 above,
or in articles that use ERPs (please search on Google Scholar for examples
that relate to your research questions).
It also not clear what you mean by using "only one technique".
You can generate average ERPs without using ICA, and you can also generate
ERPs after using ICA.
Please be aware that the emphasis in EEGLAB is on single-trial analysis,
although it can certainly be used for single subject average ERPs.

4. In the future please try to be more specific about what you are trying to
do, what kind of ERPs you are trying to generate.
Also it seems that you are unaware of the the many potential uses of ICA
(for artifact detection, for detection of independent modulators, etc..)
Please read through the articles describing EEGLAB (which you can find at
the EEGLAB site, or via Google Scholar).
Please also read through and do the EEGLAB tutorial, which will answer some
of your questions.
Then feel free to send some more questions after that.

Good luck Bagas!





On Thu, Dec 23, 2010 at 1:09 AM, Bagas Isadewa <izadewa at yahoo.com> wrote:

>  hello,
>
> i have several questions,
>
> 1. if I wanted to average EEG data to yields good ERP, how many minimum
> trials do I needed ? is there any paper or journals that explained about it
> ?
> 2. then, if I wanted to run ICA to extract ERP, do I need many trials or
> only single-trials in one epoch will be enough ??
> 3. is it necessary to run ICA first then do averaging or it is enough to
> use only one technique for extracting ERP ?
>
> if you don't mind please give me an explanation
> thank you very much
>
>
> regards,
> bagas isadewa
>
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101226/d2ba977f/attachment.html>

From andrewhill at ucla.edu  Sun Dec 26 17:14:24 2010
From: andrewhill at ucla.edu (Andrew Hill)
Date: Sun, 26 Dec 2010 20:14:24 -0500
Subject: [Eeglablist] (ask) about averaging and ICA
In-Reply-To: <618915.98563.qm@web58304.mail.re3.yahoo.com>
References: <618915.98563.qm@web58304.mail.re3.yahoo.com>
Message-ID: <3E723D09-3922-4D92-923E-72F91CA2F2E2@ucla.edu>


On Dec 23, 2010, at 4:09 AM, Bagas Isadewa wrote:

> hello,
> 
> i have several questions, 
> 
> 1. if I wanted to average EEG data to yields good ERP, how many minimum trials do I needed ? is there any paper or journals that explained about it ?

I suggest Steve Luck's book (An Introduction to the ERP Technique), and the documentation/tutorial that is part of ERPLab (an EEGLab plugin).

> 2. then, if I wanted to run ICA to extract ERP, do I need many trials or only single-trials in one epoch will be enough ??

Not really sure what you are asking... by definition an epoch will only have one trial.

> 3. is it necessary to run ICA first then do averaging or it is enough to use only one technique for extracting ERP ?

ICA first - averaging will transform the data in a way that will make ICA results not make sense.  I'm fairly certain this is true, but perhaps someone else will have a more well-reasoned answer.

Best,
Andrew 



> 
> if you don't mind please give me an explanation
> thank you very much
> 
> 
> regards,
> bagas isadewa
> 
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101226/8a6e879b/attachment.html>

From lzx72 at hotmail.com  Mon Dec 27 11:36:47 2010
From: lzx72 at hotmail.com (Zhongxu Liu)
Date: Mon, 27 Dec 2010 14:36:47 -0500
Subject: [Eeglablist] Non-parametric statistics regarding pooling trials
	across subject
Message-ID: <AANLkTindas-Z0MESUpSXG3aGcz-nKznRn1dJiwaO4HKE@mail.gmail.com>

Dear All,

In a study I am trying to test developmental (i.e., age) effect on ERSPs and
ERPs. Although I have more than 100 children tested, the number of
artifact-free trails for each subject is on average only about 20 and with
large variations (ranging from several to several dozens). I can regress
these ERSP/ERP measures on age while control for trial number, however, with
many subjects having very few trials, the regression analysis may be
influenced by those low-trial (consequently, high noise) data points.
Covariating the trial number may not solve the problem (if not adding new
problems).

After reading a previous discussion among Drs. Delorme, Scott, Rousselete,
and Robert (see below), I am thinking whether I can also use a nonparametric
statistical analysis: One option is to pool all the trials from all the
subjects, then, randomly split the trials into psudo Young and Old trials,
and calculate differences on ERSP and ERP measures. Repeat this e.g., 500
times to get null distribution. Then using true age information to split the
trials again into true Young and Old trials, and calculate the ERSP/ERP
differences again. Finally, I can check whether the real age differences
fall within or outside the 95% confidence interval of the null distribution.


The questions are: 1) Because there are 100 subjects, should the results be
generalized to the whole population from which I sampled? 2). Now with each
ERP/ERSP calculation using about 1000 trials (although the trials are from
different subjects), can I believe that the stability of the data can be
improved and I may have better chance to detect significant age effect? 3).
Will there be a difference (theoretically or in terms of statistical power)
if I randomly choose with replacement the psudo Young and Old trials (i.e.,
bootstrapping the pooled trials) instead of using permutation method?

Based on my limited reading, I did not find relevant literature on this kind
use of non-parametric methods. So I would really appreciate if someone can
give me some suggestions.

Have a good holiday!

Zhongxu





On Wed, Nov 18, 2009 at 6:20 PM, <eeglablist-request at sccn.ucsd.edu> wrote:

> Send eeglablist mailing list submissions to
>        eeglablist at sccn.ucsd.edu
>
> To subscribe or unsubscribe via the World Wide Web, visit
>        http://sccn.ucsd.edu/mailman/listinfo/eeglablist
> or, via email, send a message with subject or body 'help' to
>        eeglablist-request at sccn.ucsd.edu
>
> You can reach the person managing the list at
>        eeglablist-owner at sccn.ucsd.edu
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of eeglablist digest..."
>
> Today's Topics:
>
>   1. Re: statistics in EEGLAB (Robert Brown)
>
>
> ---------- Forwarded message ----------
> From: Robert Brown <bobrobbrown at googlemail.com>
> To: Guillaume Rousselet <g.rousselet at psy.gla.ac.uk>, arno delorme <
> arno at ucsd.edu>
> Date: Wed, 18 Nov 2009 12:39:16 +0200
> Subject: Re: [Eeglablist] statistics in EEGLAB
> thank you very much for your responses,
>
> I'll try to proceed with the strategy proposed by myself and supported by
> Scott and Guillaume (1. do the single trial analysis in a single subject. 2.
> look, if and where there is overlap in the effects).
>
> one question remains (just to be sure to be doing the right thing): is the
> eeglab newtimef function in case of condition comparisons providing me the
> required "single subject single trial" analysis? (that is, if the
> alpha-level output is showing me regions where in case of this subject there
> are significant differences between the conditions as evaluated on basis of
> single trials. I could not be sure of this, because the alpha level is also
> computed for the conditions). if not, then can you please lead me to the
> more appropriate way of doing this in eeglab (I know this is possible in
> fieldtrip, but I've already done all the other analysis and preprocessing in
> eeglab).
>
> thank you very much for your time and for your ideas.
>
> best,
> Bob
>
> 2009/11/17 Guillaume Rousselet <g.rousselet at psy.gla.ac.uk>
>
>> Robert,
>>
>> there is no theoretical reason to limit your statistical analyses to group
>> effects. It can be easily argued that you will have actually more power when
>> you do a comparison across 200 trials, rather than across 10 subjects.
>> Chances that a robust effect will occur by chance in 4 subjects are almost
>> null if you have enough trials and a clean signal. So you could do your
>> stats across trials in each subject, show the data for each subject, and
>> then report something like the number of subjects showing a significant
>> effect at any given time point, electrode, or for a given cluster, ICA...
>>
>> I've been exploring the possibilities of taking into account the variance
>> within observers in the following papers:
>>
>> Rousselet, G. A., Husk, J. S., Bennett, P. J., & Sekuler, A. B. (2008). Time
>> course and robustness of ERP object and face differences. *Journal of
>> Vision, 8*(12), 3, 1-18, http://journalofvision.org/18/12/13/,
>> doi:10.1167/1168.1112.1163.
>> Rousselet, G. A., Pernet, C. R., Bennett, P. J., & Sekuler, A. B. (2008).
>> Parametric study of EEG sensitivity to phase noise during face processing.
>> *BMC Neuroscience, 9:98*,
>> http://www.biomedcentral.com/1471-2202/1479/1498/.
>>
>> Best,
>>
>> GAR
>>
>>
>>
>>
>> On 13 Nov 2009, at 18:48, Scott Makeig wrote:
>>
>> I agree. For example, if there are 3 subjects, then simple binomial
>> probability can give no better a result than p <= 12.5%.  However, in the
>> case that each single-subject effect, across single trials, is significant
>> (e.g., at the p < .001% level), a much stronger inference can be derived
>> using reasonable subject distribution assumptions.
>>
>> Scott
>>
>> On Thu, Nov 12, 2009 at 4:03 AM, Robert Brown <bobrobbrown at googlemail.com
>> > wrote:
>>
>>> Dear Arno and All,
>>>
>>> thank you very much for your enlightening response.
>>>
>>> maybe one idea: let's say that I only have 4 subjects. the statistics
>>> based on "subject means" would be unreliable and I would not get any
>>> results. however, it could be that in case of each single subject there is a
>>> significant difference based on trials in the same time window, which would
>>> actually be a strong evidence for differences between the conditions and
>>> which could be written as "in case of each single subject p < .05
>>> (corrected)". I am sorry if this is not right, but I assume that there could
>>> be instances where the group statistics with 3-4 subjects would not show
>>> anything but the single trial statistics would. (good examples of important
>>> studies with so few subjects would be Tong & Engel, 2001 in Nature with 4
>>> subjects fMRI and Resulaj et al., 2009 in Nature with 3 subjects behavior.).
>>>
>>>
>>> to conclude: maybe the single trial statistics would work, if it a) would
>>> be calculated individually for each subject based on only this subjects
>>> single trials and then b) the (time-frequency) regions would be plotted,
>>> where all the subjects have significant differences based on their single
>>> trial analysis.
>>>
>>> thank you for your attention and good luck,
>>>
>>> Bob
>>>
>>> 2009/11/11 Arnaud Delorme <arno at ucsd.edu>
>>>
>>>> Dear Bob,
>>>>
>>>> thanks for the comments. I think you are using the statmode option
>>>> "trial" from the command line. This option is quite experimental. It was
>>>> implemented a while ago and is probably not forward compatible with more
>>>> recent changes. Also, the "statmode", "trials" option (assuming it was
>>>> working) should only be used to plot a single subjects. The reason is based
>>>> on the type of null hypothesis.
>>>>
>>>> When testing with 'statmode', 'subject' for two conditions, the NULL
>>>> hypothesis is: given the subjects I have recorded and given that these
>>>> subjects are a good representation of the general population of all possible
>>>> subjects, there is no difference between the ERP/spectrum/ERSP/ITC between
>>>> the two experimental conditions in the general subject population. Using
>>>> parametric, permutation, or bootstrap statistics (and assumptions) you may
>>>> either accept or reject this hypothesis at a given confidence level.
>>>>
>>>> When testing with 'statmode', 'trial' on a single subject (still two
>>>> conditions), the NULL hypothesis is : given the trials I have recorded and
>>>> given that these trials are a good representation of all the population of
>>>> trials for this subject, there is no difference between the
>>>> ERP/spectrum/ERSP/ITC between the two experimental conditions for this
>>>> subject. Again, using parametric, permutation, or bootstrap statistics (and
>>>> assumptions) you may either accept or reject this hypothesis at a given
>>>> confidence level.
>>>>
>>>> As you can see the two hypothesis are quite different. One makes an
>>>> inference about the population of subjects and the other one about the
>>>> population of trials.
>>>>
>>>> Now if you pool the trials from different subjects and attempt to
>>>> perform statistics, this is going to be more complex. The new hypothesis
>>>> would then be: given the trials I have recorded from my subjects and given
>>>> that these trials are a good representation of all the population of trials
>>>> from the general population of subjects, there is no difference between the
>>>> ERP/spectrum/ERSP/ITC between the two experimental conditions in the general
>>>> population of subjects. But the hypothesis is relatively biased because I
>>>> personally think that all the trials are *not* a good representation of
>>>> all the population of trials from the general population of subjects. The
>>>> trials are a good representation of all the trials from all the subjects
>>>> being presently recorded but not necessarily of the general subject
>>>> population. Therefore the real NULL hypothesis would be : given the trials I
>>>> have recorded from all of my subjects and given that these trials are a good
>>>> representation of all the population of trials from these subjects, there is
>>>> no difference between the ERP/spectrum/ERSP/ITC between the two experimental
>>>> conditions in the recorded subjects. As you see, rejecting the NULL this is
>>>> relatively limited as we care about the general population of subjects and
>>>> not the recorded subjects.
>>>>
>>>> If anybody has some better ideas (or Matlab function) of how to handle
>>>> the subject/trial problem (because it would be nice to include trials in
>>>> statistical analysis in order to make them more powerful), we will take
>>>> them.
>>>>
>>>> Best,
>>>>
>>>> Arno
>>>>
>>>> ps: we will remove the 'statmode', 'trial' option for now.
>>>> pps: for basic inferential statistics, you may also refer to this book
>>>> chapter http://sccn.ucsd.edu/~arno/mypapers/statistics.pdf<http://sccn.ucsd.edu/%7Earno/mypapers/statistics.pdf>
>>>>
>>>> On Nov 11, 2009, at 12:29 AM, Robert Brown wrote:
>>>>
>>>> Dear Arno & others,
>>>>
>>>> this does not seem to be as simple as Arno suggested (but thanks),
>>>>
>>>> 1. I have precomputed the values of these channels (with "savetrials",
>>>> "on")
>>>> 2. these channels all have data
>>>> 3. I can plot the data of the same channels when I use "statmode",
>>>> "subjects"
>>>> 4. I'm using EEGLAB v7.1.3.13b
>>>> 5. I now tried it with v7.1.7.18b and I still get the log of zero error
>>>> (you guys might be interested that in addition I now get, in case of
>>>> permutations and bootstrap, "??? Error using ==> reshape" in
>>>> statcond>surrogate at 438 and statcond at 301 and with this latest version
>>>> the reshape error even happens with the "statmode", "subjects")
>>>>
>>>> thus any other suggestions of what could be happening with my single
>>>> trial analysis in study would be very much appreciated.
>>>>
>>>> thank you very much and take care,
>>>> Bob
>>>>
>>>> 2009/11/11 Arnaud Delorme <arno at ucsd.edu>
>>>>
>>>>> Dear Bob,
>>>>>
>>>>> I think this might be because you are trying to plot ERSP of a channel
>>>>> that contains only 0. This error was also arising in old versions of EEGLAB
>>>>> when masking for significance.
>>>>>
>>>>> Hope this helps,
>>>>>
>>>>> Arno
>>>>>
>>>>>
>>>>> On Nov 7, 2009, at 11:38 AM, Robert Brown wrote:
>>>>>
>>>>>  Hi guys,
>>>>>>
>>>>>> I've been trying to get the study ersp analysis working on single
>>>>>> trials but I've not succeeded.
>>>>>>
>>>>>> in the function "std_readdata" I get the "Warning: Log of zero."
>>>>>> error, which is on the line ersp{c,g} = 20*log10(abs(ersp{c,g})); meaning
>>>>>> that the absolute value at some point is 0.
>>>>>> (This leads to) further errors:
>>>>>>
>>>>>> ??? Error using ==> set
>>>>>> Bad value for axes property: 'CLim'
>>>>>> Values must be increasing and non-NaN.
>>>>>>
>>>>>> Error in ==> caxis at 80
>>>>>>            set(ax,'CLim',arg);
>>>>>>
>>>>>> Error in ==> tftopo at 714
>>>>>> caxis([g.limits(5:6)]);
>>>>>>
>>>>>> I've tried to fix it but I'm not clever enough. any help would be
>>>>>> appreciated.
>>>>>>
>>>>>> thank you so much,,
>>>>>> Bob <ATT00001.txt>
>>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>
>>> _______________________________________________
>>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>>> To unsubscribe, send an empty email to
>>> eeglablist-unsubscribe at sccn.ucsd.edu
>>> For digest mode, send an email with the subject "set digest mime" to
>>> eeglablist-request at sccn.ucsd.edu
>>>
>>
>>
>>
>> --
>> Scott Makeig, Research Scientist and Director, Swartz Center for
>> Computational Neuroscience, Institute for Neural Computation, University of
>> California San Diego, La Jolla CA 92093-0961, http://sccn.ucsd.edu/~scott<http://sccn.ucsd.edu/%7Escott>
>>  _______________________________________________
>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>> To unsubscribe, send an empty email to
>> eeglablist-unsubscribe at sccn.ucsd.edu
>> For digest mode, send an email with the subject "set digest mime" to
>> eeglablist-request at sccn.ucsd.edu
>>
>>
>>
>>
>>
>> ************************************************************************************
>> *Guillaume A. Rousselet, Ph.D.*
>> *
>> *
>> Lecturer
>>
>> Centre for Cognitive Neuroimaging (CCNi)
>> Department of Psychology
>> Faculty of Information & Mathematical Sciences (FIMS)
>> University of Glasgow
>> 58 Hillhead Street
>> Glasgow, UK
>> G12 8QB
>>
>> The University of Glasgow, charity number SC004401
>>
>> http://web.me.com/rousseg/GARs_website/
>>
>> Email: g.rousselet at psy.gla.ac.uk
>> Fax. +44 (0)141 330 4606
>> Tel. +44 (0)141 330 6652
>> Cell +44 (0)791 779 7833
>>
>> *
>> *
>> *"For reasons I wish I understood, the spectacle of sync strikes a chord
>> in us, somewhere deep in our souls. It's a wonderful and terrifying thing.
>> Unlike many other phenomena, the witnessing of it touches people at a primal
>> level. Maybe we instinctively realize that if we ever find the source of
>> spontaneous order, we will have discovered the secret of the universe.*"
>>
>>
>> Steven Strogatz - *Sync* - 2003
>>
>> ************************************************************************************
>>
>>
>>
>>
>>
>
> _______________________________________________
> eeglablist mailing list eeglablist at sccn.ucsd.edu
> Eeglablist page: http://www.sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsub at sccn.ucsd.edu
> To switch to non-digest mode, send an empty email to
> eeglablist-nodigest at sccn.ucsd.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101227/576ad966/attachment.html>

From nancy.chen.psy at gmail.com  Mon Dec 27 18:33:50 2010
From: nancy.chen.psy at gmail.com (Xiaoxi Chen)
Date: Tue, 28 Dec 2010 10:33:50 +0800
Subject: [Eeglablist] channel location problem: ipsilateral and
	contralateral question
Message-ID: <AANLkTik8GCc-O1ifjn_GAwJyCE2RN5p_CUodZwWasRxB@mail.gmail.com>

Dear all,
My visual stimuli was presented on either *left* visual field or
*right*visual field, so I had to label channel locations on the
*ipsilateral *and *contralateral *sides, and then average the contralateral
side of both left presented stimuli and right presented stimuli (the same
with ipsilateral side of both left presented stimuli and right presented
stimuli ). Here I got a question:

How can this be done in EEGlab?

I  tried to isolate left and right stimuli into *two single datasets* and
load different channle location files separately, but I had to
*append*these two datasets to draw Sum/Compare ERP waves. While how
does EEGlab
append several datasets- based on the *number* of channel location, or the *
name* of channel location?  If these two datasets had different channel
location file, then the appended dataset would not have an unnified channel
location. How to solve this problem.
And how does  "Sum/Compare ERP waves" draw waves - based on number of
channel location, or the name of channel location?

Can you give some suggestions?

Thanks for your time.


regards,
-- 
Xiaoxi Chen,
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101228/1c875156/attachment.html>

From sjluck at ucdavis.edu  Tue Dec 28 13:29:49 2010
From: sjluck at ucdavis.edu (Steve Luck)
Date: Tue, 28 Dec 2010 13:29:49 -0800
Subject: [Eeglablist] eeglablist Digest, Vol 74, Issue 21
In-Reply-To: <mailman.0.1293566401.32005.eeglablist@sccn.ucsd.edu>
References: <mailman.0.1293566401.32005.eeglablist@sccn.ucsd.edu>
Message-ID: <B5974E76-866F-4AA6-8516-6555C6ADDE9E@ucdavis.edu>

Dear Xiaoxi,

This can be achieved very easily using the Channel Operations tool in ERPLAB Toolbox (erpinfo.org/erplab), which plugs into EEGLAB and can operate either on EEG or on averaged ERPs.  Specific instructions can be found at the end of the following manual page: http://erpinfo.org/erplab/erplab-documentation/erplab-manual/Bin_Operations.htm

Steve Luck


> 
> From: Xiaoxi Chen <nancy.chen.psy at gmail.com>
> Date: December 27, 2010 6:33:50 PM PST
> To: eeglablist at sccn.ucsd.edu
> Subject: [Eeglablist] channel location problem: ipsilateral and contralateral question
> 
> 
> Dear all,
> 
> My visual stimuli was presented on either left visual field or right visual field, so I had to label channel locations on the ipsilateral and contralateral sides, and then average the contralateral side of both left presented stimuli and right presented stimuli (the same with ipsilateral side of both left presented stimuli and right presented stimuli ). Here I got a question:
>  
> How can this be done in EEGlab? 
>  
> I  tried to isolate left and right stimuli into two single datasets and load different channle location files separately, but I had to append these two datasets to draw Sum/Compare ERP waves. While how does EEGlab append several datasets- based on the number of channel location, or the name of channel location?  If these two datasets had different channel location file, then the appended dataset would not have an unnified channel location. How to solve this problem.
> And how does  "Sum/Compare ERP waves" draw waves - based on number of channel location, or the name of channel location?
>  
> Can you give some suggestions?
>  
> Thanks for your time.
> 
>  
> regards,
> -- 
> Xiaoxi Chen,
> 
> 
> _______________________________________________
> eeglablist mailing list eeglablist at sccn.ucsd.edu
> Eeglablist page: http://www.sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsub at sccn.ucsd.edu
> To switch to non-digest mode, send an empty email to eeglablist-nodigest at sccn.ucsd.edu

--------------------------------------------------------------------
Steven J. Luck, Ph.D.
Director, Center for Mind & Brain
Professor, Department of Psychology
University of California, Davis
Room 127
267 Cousteau Place
Davis, CA 95618
(530) 297-4424
sjluck at ucdavis.edu
Web: http://mindbrain.ucdavis.edu/people/sjluck
Calendar: http://www.google.com/calendar/embed?src=stevenjluck%40gmail.com&ctz=America/Los_Angeles
--------------------------------------------------------------------







-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101228/8ac930eb/attachment.html>

From bradley.voytek at gmail.com  Tue Dec 28 22:06:38 2010
From: bradley.voytek at gmail.com (Bradley Voytek)
Date: Tue, 28 Dec 2010 22:06:38 -0800
Subject: [Eeglablist] channel location problem: ipsilateral and
 contralateral question
In-Reply-To: <AANLkTik8GCc-O1ifjn_GAwJyCE2RN5p_CUodZwWasRxB@mail.gmail.com>
References: <AANLkTik8GCc-O1ifjn_GAwJyCE2RN5p_CUodZwWasRxB@mail.gmail.com>
Message-ID: <AANLkTincHsdAi3LCM38i+CwfDxVJDBUeNJj6a-cfdZPH@mail.gmail.com>

Swap the data, not the channels. Leave the channels alone.

Just move the actual data from EEG.data from one hemisphere to the other.

::brad

On Mon, Dec 27, 2010 at 18:33, Xiaoxi Chen <nancy.chen.psy at gmail.com> wrote:
> Dear all,
>
> My visual stimuli was presented on either left visual field or right visual
> field, so I had to?label channel locations on the ipsilateral and
> contralateral sides, and then average the contralateral side of both left
> presented stimuli and right presented stimuli (the same with ipsilateral
> side of both left presented stimuli and right presented stimuli ). Here I
> got a question:
>
> How can this be done?in?EEGlab?
>
> I? tried to isolate left and right stimuli into two single datasets and load
> different channle location files separately, but I had to append these two
> datasets to draw Sum/Compare ERP waves. While?how does EEGlab append?several
> datasets- based on?the number?of channel location, or?the name of channel
> location?? If these two datasets had different channel location file, then
> the appended dataset would not have an unnified channel location. How to
> solve this problem.
> And how does??"Sum/Compare ERP waves" draw waves - based on number?of
> channel location, or?the name of channel location?
>
> Can you give some suggestions?
>
> Thanks for your time.
>
> regards,
> --
> Xiaoxi Chen,
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



From alenarto at ucla.edu  Wed Dec 29 08:31:35 2010
From: alenarto at ucla.edu (Agatha Lenartowicz)
Date: Wed, 29 Dec 2010 08:31:35 -0800
Subject: [Eeglablist] Online visual brain atlas: brainSCANr
In-Reply-To: <AANLkTikq5V9PBg1-pJXNz6O-bnJwZ1pUMhkPf1rqjo06@mail.gmail.com>
References: <AANLkTimMyLwkmfqC6wwGY1bscmLB+PFW9WTrfEMEjinD@mail.gmail.com>
	<AANLkTikq5V9PBg1-pJXNz6O-bnJwZ1pUMhkPf1rqjo06@mail.gmail.com>
Message-ID: <06D5B116-806E-40C8-8D82-8162CA059C98@ucla.edu>

Check out... http://www.cognitiveatlas.org/

Agatha

On Dec 21, 2010, at 9:09 PM, Bradley Voytek wrote:

> Dear friends and colleagues:
> 
> For years now I've been wanting a repository where I can easily check
> to see what functions a brain region is associated with and what other
> brain regions it's connected to.
> 
> Well, no one ever got around to making that happen, so my wife and I
> just made our own.
> 
> Introducing brainSCANr!
> 
> http://www.brainscanr.com
> 
> brainSCANr (The Brain Systems, Associations, Connections, and Network
> Relationships engine) works by searching PubMed for the co-occurence
> of brain region name, cognitive/behavioral functions, and diseases to
> build a connection matrix that you can graphically explore.
> 
> Want to know what the amygdala does and what it connects with? What
> about Parkinson's disease? Attentional capacity? Just search! You can
> also see all the papers used to populate the database.
> 
> All of the raw data is available for you to play with, too. Check out
> all the details here:
> http://www.brainscanr.com/Paper
> 
> And if you like it and think it's useful, please pass it along to
> neuroscience friends and colleagues! And if you have any comments or
> suggestions, let me know!
> 
> ::brad
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu




From dorothy.bishop at psy.ox.ac.uk  Tue Dec 28 05:05:56 2010
From: dorothy.bishop at psy.ox.ac.uk (Dorothy Bishop)
Date: Tue, 28 Dec 2010 13:05:56 +0000
Subject: [Eeglablist] EEG history fail, but solved
Message-ID: <01B4B1C924D5FC44B57B269548B562763C5909DF79@EXMBX05.ad.oak.ox.ac.uk>

Just in case others hit this problem:
I'm using version 9.0.4.4b

I was trying the automatic artefact rejection, which worked well.
However, I wanted to put this in a script.

When I typed EEG.history, I got
 EEG = pop_autorej(EEG, )

But if I put that command in a script, I get
Error: Unbalanced or unexpected parenthesis or bracket.

After reading help, I realised you need to add some of the 'optional' parameters, e.g.

EEG = pop_autorej(EEG,'electrodes',[1:36],'maxrej',5,'nogui','on');




Dorothy Bishop, Professor of Developmental Neuropsychology,
Dept of Experimental Psychology, University of Oxford, OX1 3UD.
tel +44 (0)1865 271369; fax +44 (0)1865 281255;
WEB: http://www.psy.ox.ac.uk/oscci/
Blog: http://deevybee.blogspot.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101228/94f7b39f/attachment.html>

From dorothy.bishop at psy.ox.ac.uk  Thu Dec 30 00:01:46 2010
From: dorothy.bishop at psy.ox.ac.uk (Dorothy Bishop)
Date: Thu, 30 Dec 2010 08:01:46 +0000
Subject: [Eeglablist] modifying head model for children in dipfit2
Message-ID: <01B4B1C924D5FC44B57B269548B562763C5909DF8D@EXMBX05.ad.oak.ox.ac.uk>

I am learning to use Dipfit2 with eeglab9_0_4_5s.

All works well if I accept default head model.

I would like, however, to modify the head model for children.



I followed the instructions for doing this that I found in:

http://sccn.ucsd.edu/pipermail/eeglablist/2007/001690.html



I kept everything the same as for spherical 4 shell model, except that I subsituted a new head model file with altered vol.r values to correspond to 8 yr old child. I also, on advice from a colleague, made

vol.c(3)=.0084



Output coordinates default to MNI when you select custom model file, and I kept this.



I then went to Edit Channel locations to set head radius to be same as last vol.r value, i.e. 70.3



I then ran autofit dipoles , specifying 'fit bilateral dipoles'.

But I end up with single dipoles in the middle of the head.



I am not sure if the problem is with my model specification, or whether there is a bug, and would be grateful for advice.



thanks




Dorothy Bishop, Professor of Developmental Neuropsychology,
Dept of Experimental Psychology, University of Oxford, OX1 3UD.
tel +44 (0)1865 271369; fax +44 (0)1865 281255;
WEB: http://www.psy.ox.ac.uk/oscci/
Blog: http://deevybee.blogspot.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101230/9bfdf48f/attachment.html>

From adimitri at uci.edu  Fri Dec 31 06:42:12 2010
From: adimitri at uci.edu (Andrew Dimitrijevic)
Date: Fri, 31 Dec 2010 09:42:12 -0500
Subject: [Eeglablist] modifying head model for children in dipfit2
In-Reply-To: <01B4B1C924D5FC44B57B269548B562763C5909DF8D@EXMBX05.ad.oak.ox.ac.uk>
References: <01B4B1C924D5FC44B57B269548B562763C5909DF8D@EXMBX05.ad.oak.ox.ac.uk>
Message-ID: <4D1DEBC4.9010008@uci.edu>

Hi Dorothy,

This is a common problem (at least with auditory evoked potentials). 
Some suggestions that worked for me in the past:

1) use "initial seed points" that are approximately in the auditory 
cortex (you can set the initial x,y,z points) before you do the dipole fit
2) make sure you have a good SNR (general comment for source analysis, 
but seems to be extra important for DipFit)

the reason that you're getting dipoles in the centre of the head is that 
it "truely" is the best explanation of the variance, albeit it doesn't 
make sense physiologically.

cheers

andrew


On 12/30/2010 3:01 AM, Dorothy Bishop wrote:
>
> I am learning to use Dipfit2 with eeglab9_0_4_5s.
>
> All works well if I accept default head model.
>
> I would like, however, to modify the head model for children.
>
> I followed the instructions for doing this that I found in:
>
> http://sccn.ucsd.edu/pipermail/eeglablist/2007/001690.html
>
> I kept everything the same as for spherical 4 shell model, except that 
> I subsituted a new head model file with altered vol.r values to 
> correspond to 8 yr old child. I also, on advice from a colleague, made
>
> vol.c(3)=.0084
>
> Output coordinates default to MNI when you select custom model file, 
> and I kept this.
>
> I then went to Edit Channel locations to set head radius to be same as 
> last vol.r value, i.e. 70.3
>
> I then ran autofit dipoles , specifying 'fit bilateral dipoles'.
>
> But I end up with single dipoles in the middle of the head.
>
> I am not sure if the problem is with my model specification, or 
> whether there is a bug, and would be grateful for advice.
>
> thanks
>
> Dorothy Bishop, Professor of Developmental Neuropsychology,
> Dept of Experimental Psychology, University of Oxford, OX1 3UD.
> tel +44 (0)1865 271369; fax +44 (0)1865 281255;
> WEB: http://www.psy.ox.ac.uk/oscci/
> Blog: http://deevybee.blogspot.com/
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101231/71316f11/attachment.html>

From izadewa at yahoo.com  Thu Dec 30 11:32:09 2010
From: izadewa at yahoo.com (Bagas Isadewa)
Date: Thu, 30 Dec 2010 11:32:09 -0800 (PST)
Subject: [Eeglablist] (ask) about averaging and ICA
In-Reply-To: <AANLkTikCs55efRtaM-gMg+E43vwVrYV0i0igjfvHovtu@mail.gmail.com>
References: <618915.98563.qm@web58304.mail.re3.yahoo.com>
	<AANLkTikCs55efRtaM-gMg+E43vwVrYV0i0igjfvHovtu@mail.gmail.com>
Message-ID: <2733.11950.qm@web161808.mail.bf1.yahoo.com>




hello,

thank you for the answers, especially for Mr. Andrew Hill and Mr. Tarik S 
Bel-Bahar
I apologize for my ambiguous and unclear question. It caused by my English is 
bad, i'm sorry.

I  will explain about my experiment, now i'm doing my final project to get  
bachelor degree's. I try to generate ERP with subject read an  alphabet/letter 
in a slideshow ( I chosed a/i/u/e/o letter for stimuli).  My goal is to get a 
ERP waveform related to a certain alphabet which  subject has read. Note that 
subject only thinking reading a words not  speak it and not produce any sound. 
The subject just say (or think) the  letter in mind.

for my question number :

1.thank you for  the suggestion, I take just 20 trials for each letter, actually 
I wanted  to do experiment (record many trials) again but i'm  facing a bit of 
problem here. To do an experiment, I must rent an EEG  in medical clinic which I 
must paid some money (unfortunately it quite  expensive). because of that I 
can't do an experiment as many as I  wanted.

2. I mean many N trials in many N epoch (is that right ?) and single-trial in 1 
epoch.


3.  Let me try to explain what I understand. To extract ERP there are two  ways 
: a. Do the averaging many trials and b. run ICA to the data. And  what I'm 
going to do is run ICA first, then I will average the result of  ICA 
decomposition


I admit I do not fully understand  ICA, but I still try reading the literature 
(it's little hard in  English, hehehe). There is a question in my mind, how do 
we prove the  result of ICA decomposition is right ? I mean, do the result will 
one  hundred percent represent the true neural signal ? 


thank you very much

regards,
bagas isadewa


________________________________
From: Tarik S Bel-Bahar <tarikbelbahar at gmail.com>
To: Bagas Isadewa <izadewa at yahoo.com>
Cc: EEGLAB <eeglablist at sccn.ucsd.edu>
Sent: Mon, December 27, 2010 8:45:01 AM
Subject: Re: [Eeglablist] (ask) about averaging and ICA


Greetings Bagas,
 
1. Yes, there are multiple papers that refer to this issue. I would suggest you 
start
with the HANDBOOKS by Tom Handy or Steve Luck, and these will answer your basic 
questions about the relations
between trial counts, averaged ERPs, and the resultant signal-to-noise ratios. 
You can find these books by doing a Google search, and you can order them from 
Amazon or another bookseller.
Here's the link for the Handy handbook: 
http://mitpress.mit.edu/catalog/item/default.asp?tid=10253&ttype=2
 
2. Your second question is not clear. It's not clear what you mean by "many 
trials" or only "single-trials" in one epoch.
For information about best practice using EEGLAB and ICA, please read through 
the EEGLAB tutorial online.
A general rule of thumb is that more single trials are better than fewer single 
trials, for an accurate ICA decomposition of the data.
 
3. Again, you should read and learn basic information about ERP techniques from 
handbooks mentioned in Point 1 above,
or in articles that use ERPs (please search on Google Scholar for examples that 
relate to your research questions). 

It also not clear what you mean by using "only one technique".
You can generate average ERPs without using ICA, and you can also generate ERPs 
after using ICA.
Please be aware that the emphasis in EEGLAB is on single-trial analysis, 
although it can certainly be used for single subject average ERPs.
 
4. In the future please try to be more specific about what you are trying to do, 
what kind of ERPs you are trying to generate.
Also it seems that you are unaware of the the many potential uses of ICA (for 
artifact detection, for detection of independent modulators, etc..)
Please read through the articles describing EEGLAB (which you can find at the 
EEGLAB site, or via Google Scholar).
Please also read through and do the EEGLAB tutorial, which will answer some of 
your questions.
Then feel free to send some more questions after that.
 
Good luck Bagas!
 
 
 

 
On Thu, Dec 23, 2010 at 1:09 AM, Bagas Isadewa <izadewa at yahoo.com> wrote:

hello,
>
>i have several questions, 
>
>1. if I wanted to average EEG data to yields good ERP, how many minimum trials 
>do I needed ? is there any paper or journals that explained about it ?
>2. then, if I wanted to run ICA to extract ERP, do I need many trials or only 
>single-trials in one epoch will be enough ??
>3. is it necessary to run ICA first then do averaging or it is enough to use 
>only one technique for extracting ERP ?
>
>if you don't mind please give me an explanation
>thank you very much
>
>
>regards,
>bagas isadewa
>
>
>
>_______________________________________________
>Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
>For digest mode, send an email with the subject "set digest mime" to 
>eeglablist-request at sccn.ucsd.edu
>



      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101230/16419e07/attachment.html>

From dorothy.bishop at psy.ox.ac.uk  Fri Dec 31 07:03:45 2010
From: dorothy.bishop at psy.ox.ac.uk (Dorothy Bishop)
Date: Fri, 31 Dec 2010 15:03:45 +0000
Subject: [Eeglablist] modifying head model for children in dipfit2
In-Reply-To: <4D1DEBC4.9010008@uci.edu>
References: <01B4B1C924D5FC44B57B269548B562763C5909DF8D@EXMBX05.ad.oak.ox.ac.uk>,
	<4D1DEBC4.9010008@uci.edu>
Message-ID: <01B4B1C924D5FC44B57B269548B562763C5909DF97@EXMBX05.ad.oak.ox.ac.uk>

I'm not sure about what you say because

a) Relatively minor changes to the head model have led to single dipoles in middle of head rather than sensibly placed dipoles. The residual variance was below 2% per dipole with the adult head model and is now much higher.
b) I have explicitly asked for 2 dipoles per component, yet can only see one.  I think I should be able to spot 2 superimposed dipoles in the output, even if the estimate is for a central location.

Dorothy Bishop, Professor of Developmental Neuropsychology,
Dept of Experimental Psychology, University of Oxford, OX1 3UD.
tel +44 (0)1865 271369; fax +44 (0)1865 281255;
WEB: http://www.psy.ox.ac.uk/oscci/
Blog: http://deevybee.blogspot.com/

________________________________
From: Andrew Dimitrijevic [adimitri at uci.edu]
Sent: 31 December 2010 14:42
To: eeglablist at sccn.ucsd.edu; Dorothy Bishop
Subject: Re: [Eeglablist] modifying head model for children in dipfit2

Hi Dorothy,

This is a common problem (at least with auditory evoked potentials). Some suggestions that worked for me in the past:

1) use "initial seed points" that are approximately in the auditory cortex (you can set the initial x,y,z points) before you do the dipole fit
2) make sure you have a good SNR (general comment for source analysis, but seems to be extra important for DipFit)

the reason that you're getting dipoles in the centre of the head is that it "truely" is the best explanation of the variance, albeit it doesn't make sense physiologically.

cheers

andrew


On 12/30/2010 3:01 AM, Dorothy Bishop wrote:

I am learning to use Dipfit2 with eeglab9_0_4_5s.

All works well if I accept default head model.

I would like, however, to modify the head model for children.



I followed the instructions for doing this that I found in:

http://sccn.ucsd.edu/pipermail/eeglablist/2007/001690.html



I kept everything the same as for spherical 4 shell model, except that I subsituted a new head model file with altered vol.r values to correspond to 8 yr old child. I also, on advice from a colleague, made

vol.c(3)=.0084



Output coordinates default to MNI when you select custom model file, and I kept this.



I then went to Edit Channel locations to set head radius to be same as last vol.r value, i.e. 70.3



I then ran autofit dipoles , specifying 'fit bilateral dipoles'.

But I end up with single dipoles in the middle of the head.



I am not sure if the problem is with my model specification, or whether there is a bug, and would be grateful for advice.



thanks




Dorothy Bishop, Professor of Developmental Neuropsychology,
Dept of Experimental Psychology, University of Oxford, OX1 3UD.
tel +44 (0)1865 271369; fax +44 (0)1865 281255;
WEB: http://www.psy.ox.ac.uk/oscci/
Blog: http://deevybee.blogspot.com/


_______________________________________________
Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu<mailto:eeglablist-unsubscribe at sccn.ucsd.edu>
For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu<mailto:eeglablist-request at sccn.ucsd.edu>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101231/fc1b1a96/attachment.html>

From smakeig at gmail.com  Thu Dec 30 16:11:30 2010
From: smakeig at gmail.com (Scott Makeig)
Date: Thu, 30 Dec 2010 16:11:30 -0800
Subject: [Eeglablist] modifying head model for children in dipfit2
In-Reply-To: <01B4B1C924D5FC44B57B269548B562763C5909DF8D@EXMBX05.ad.oak.ox.ac.uk>
References: <01B4B1C924D5FC44B57B269548B562763C5909DF8D@EXMBX05.ad.oak.ox.ac.uk>
Message-ID: <AANLkTi=A1xeQZdCeYVx0mScQJgN1sdEsqZDJs5_a3qtr@mail.gmail.com>

Dorothy -

The concept of using customized child head models is a fine one - I hope
Arno or someone on the list can point you to answers to your difficulties
with the current Dipfit2. Zeynep Akalin Acar 's NFT now includes inverse
dipole modeling, I believe, and is aimed first of all at working with
individual subject MRs, so perhaps NFT may also be of use to you here.

I am now working with Zeynep and with Ying Wu here to plan new, more
sophisticated developmental EEG studies. In particular, with first
collaborators we now plan to begin a project to build software returning a
standard developmental head model for a child of any specified age and
gender and hope to execute a pilot phase of this project this year. For this
and later phases, we will be looking for data and potentially for more
collaborators. If you or anyone on this list has an interest in
contributing, let me know privately.

Scott Makeig
smakeig at ucsd.edu


On Thu, Dec 30, 2010 at 12:01 AM, Dorothy Bishop <
dorothy.bishop at psy.ox.ac.uk> wrote:

>   I am learning to use Dipfit2 with eeglab9_0_4_5s.
>
> All works well if I accept default head model.
>
> I would like, however, to modify the head model for children.
>
>
>
> I followed the instructions for doing this that I found in:
>
> http://sccn.ucsd.edu/pipermail/eeglablist/2007/001690.html
>
>
>
> I kept everything the same as for spherical 4 shell model, except that I
> subsituted a new head model file with altered vol.r values to correspond to
> 8 yr old child. I also, on advice from a colleague, made
>
> vol.c(3)=.0084
>
>
>
> Output coordinates default to MNI when you select custom model file, and I
> kept this.
>
>
>
> I then went to Edit Channel locations to set head radius to be same as last
> vol.r value, i.e. 70.3
>
>
>
> I then ran autofit dipoles , specifying 'fit bilateral dipoles'.
>
> But I end up with single dipoles in the middle of the head.
>
>
>
> I am not sure if the problem is with my model specification, or whether
> there is a bug, and would be grateful for advice.
>
>
>
> thanks
>
>
>
>  Dorothy Bishop, Professor of Developmental Neuropsychology,
> Dept of Experimental Psychology, University of Oxford, OX1 3UD.
> tel +44 (0)1865 271369; fax +44 (0)1865 281255;
> WEB: http://www.psy.ox.ac.uk/oscci/
> Blog: http://deevybee.blogspot.com/
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>



-- 
Scott Makeig, Research Scientist and Director, Swartz Center for
Computational Neuroscience, Institute for Neural Computation, University of
California San Diego, La Jolla CA 92093-0559,
http://sccn.ucsd.edu/~scott<http://sccn.ucsd.edu/%7Escott>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20101230/bb4e5db8/attachment.html>