[Eeglablist] How to interpret the ICA component in EEGLAB

[Makoto Miyakoshi]

Dear Naveen,

> Q: What is residual variance and how do i interpret independent
components based on  RV % ?  What RV% indicates ?

It's the error measure i.e. difference from the scalp map that would be
created by the estimated dipole. If it's large, it means your dipole
explains the scalp map only poorly. 15% r.v. thresholding was empirically
selected and no dedicated study performed upon it.

> Q: What different colors of the dipole indicates in the same independent
components and their directions?

The color does not matter. It's just for visualization.

> Q: Is that possible to explain physiological significance of multiple
subjects at one go using DIPFIT. cumulative study through DIPFIT

Probably yes. Our recommended group-level analysis solution STUDY uses the
similar idea.

Mokato

On Mon, Sep 7, 2015 at 5:10 AM, Naveen Bhardwaj <naveenno122k at gmail.com>
wrote:

> Thanks Mokato,
>
> I gone through DIPFIT tool ( which is good to plot source localization)
> but after plotting dipole  of the components i found difficulty in
> interpretation of components.
> So i have some questions,
>
> Q: What is residual variance and how do i interpret independent components
> based on  RV % ?  What RV% indicates ?
> Q: What different colors of the dipole indicates in the same independent
> components and their directions?
> Q: Is that possible to explain physiological significance of multiple
> subjects at one go using DIPFIT. cumulative study through DIPFIT
>
>
> Thanks for your help
>
> On Sat, Aug 22, 2015 at 4:18 AM, Makoto Miyakoshi <mmiyakoshi at ucsd.edu>
> wrote:
>
>> Dear Naveen,
>>
>> > Q: How to interpret the ICA component in EEGLAB, what are the steps for
>> analyses of ICA data?
>>
>> You have to run dipfit (i.e. source localization using equivalent current
>> dipole model) as well to fully identify their physiological significance.
>>
>> > Q: What color indicates in 2d, 3d topographic maps shown in Independent
>> components?
>>
>> Scalp topos represents normalized RMS amplitude of the IC. Note that
>> these topos are stationary across time (it's like orchestra members are
>> seated during performance).
>>
>> > Q: Can i analyze specific frequency band  and its power spectrum after
>> ICA ?
>>
>> Sure. If you create STUDY you can analyze them with GUI. If you code in
>> Matlab you can use output from functions (spectopo).
>>
>> > Q: Is that any tool in EEGLAB which help me to correlate the specific
>> band power v/s cognitive change during tasks.
>>
>> No I don't think so.
>> Cyril Pernet's LIMO may does it. I haven't used it yet so I'm not 100%
>> sure, but conceptually it must be able to take care of the need.
>> http://www.hindawi.com/journals/cin/2011/831409/
>>
>> Makoto
>>
>> On Mon, Aug 17, 2015 at 9:21 PM, Naveen Bhardwaj <naveenno122k at gmail.com>
>> wrote:
>>
>>> Dear EEGLAB list,
>>>
>>> I have 14 channel EEG tool through which i acquired continuous signals.
>>> Now i followed all the pre processing steps mention in EEGLAB tutorial,
>>> like import EDF file>>Select channel location>> Remove baseline>> basic
>>> band pass filter (0.1Hz to 40HZ)>> re-referencing channel data>> eye
>>> artifact removal manually>> RUNICA>>
>>> *i got 14 component from all 14 channels.*
>>> Q: How to interpret the ICA component in EEGLAB, what are the steps for
>>> analyses of ICA data?
>>>
>>> Q: What color indicates in 2d, 3d topographic maps shown in Independent
>>> components?
>>>
>>> Q: Can i analyze specific frequency band  and its power spectrum after
>>> ICA ?
>>>
>>> Q: Is that any tool in EEGLAB which help me to correlate the specific
>>> band power v/s cognitive change during tasks.
>>>
>>> Any suggestions or any kind of help regarding these questions will be
>>> highly appreciated.
>>>
>>>
>>> Many many thanks
>>>
>>> Naveen
>>> --
>>> *With Warm Regards *
>>>
>>>
>>> *Naveen Bhardwaj*
>>> *Research Scholar*
>>>
>>> *Skype Id- naveenb13*
>>> *Mobile: +91-9717173826 <%2B91-9717173826>*
>>>
>>>
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> 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
>>>
>>
>>
>>
>> --
>> Makoto Miyakoshi
>> Swartz Center for Computational Neuroscience
>> Institute for Neural Computation, University of California San Diego
>>
>
>
>
> --
> *With Warm Regards *
>
>
> *Naveen Bhardwaj*
> *Research Scholar*
> PhD (Instrument Technology - Human Machine Interaction)
> Instrument Design & Development Centre
> Indian Institute of Technology, Delhi
>
> Skype Id- naveenb13
> Mobile: +91-9717173826
>
>
>
>
>
>



-- 
Makoto Miyakoshi
Swartz Center for Computational Neuroscience
Institute for Neural Computation, University of California San Diego
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20150917/4d617ec5/attachment.html>