[Eeglablist] Time- frequency analysis

[Makoto Miyakoshi]

Dear Leonardo,

I haven't tested the time-frequency baseline parameters myself so I'm
afraid I'm not so much of your help.

> 1- For me, in order to select the baseline, I try first (with code)
applying several values (a few ms before or after 0), and best on the
worthy plot, I determine the suitable baseline. Is considered a good way to
select the baseline?

No I don't think so. It seems as if you are cherry picking. Use the most
reasonable and a priori definition of baseline and stick to it unless you
have clear reason not to do so.

> 2-Now, without code, when I use the eeglab, 0 is the value that is
selected by default in the time-frequency dialog (pop_newtimef()). In the
case of my continuous data, when I change select values for my baseline,
e.g., [0 300] or [-400 -200], I get the folllwing error message:
There are no sample points found in the default baseline. This may happen
even though data time limits overlap with the baseline period (because of
the time-freq. window width). Either disable the baseline, change the
baseline limits.

Sorry but I have no idea about this. If you like please post it to EEGLAB
Bugzilla
https://sccn.ucsd.edu/bugzilla/

> a- My major aim is to know when to select any of the options of drop-down
list (Use divisive baseline (DIV) as selecting them keep change the dB into
e.g., STD, Use standard deviation (STD), Use single trial DIV baseline, and
single trial baseline) because when I change the default option that
produces dB unit, it will be change into STD.

If you don't know which one to use because you don't know what they do, use
the default. If you know the differences but still not sure which one to
use, make a decision. If you can't, use the default.

> Is the STD (or any of the left 3 choice in drop-down list) produces dB
unit? In the 2 attached plots, I highlighted both default option and STD
option results where the dB changed to STD.

No, if the unit is STD it probably means X times deviant from the baseline
STD.

> b- When to select each of Use divisive baseline (DIV),Use standard
deviation (STD), Use single trial DIV baseline, and Use single trial STD
baseline?

The developer's implicit recommendation would be to use the default... I
would take so.

Makoto



On Fri, Dec 4, 2015 at 7:39 PM, Leonardo Lion <llionname at gmail.com> wrote:

> Dear Makoto and the list,
>
> Thanks a lot for your explanation and help. In fact, I still have some
> issues about this particular topic.
>
> 1- For me, in order to select the baseline, I try first (with code)
> applying several values (a few ms before or after 0), and best on the
> worthy plot, I determine the suitable baseline. Is considered a good way to
> select the baseline?
>
> 2-Now, without code, when I use the eeglab, 0 is the value that is
> selected by default in the time-frequency dialog (pop_newtimef()). In the
> case of my continuous data, when I change select values for my baseline,
> e.g., [0 300] or [-400 -200], I get the folllwing error message:
>
> There are no sample points found in the default baseline. This may happen
> even though data time limits overlap with the baseline period (because of
> the time-freq. window width). Either disable the baseline, change the
> baseline limits.
>
>
>
> How to solve this matter?
>
> 3-
>
> For baseline, when to select the option "Use standard deviation (STD)",
>> "Single trial DIV baseline", and "Single trial STD baseline"?  When we
>> change from one to another, does that change the dB unit or not, knowing
>> that the default option "Use divisive baseline" produces bar column in dB.
>> However, this bar color will be changed (e.g., std.) when we change the
>> option into "Use Single trial STD baseline".
>>
>> I believe the difference is how to take the baseline. It's my guess but
>> sounds like the default is the std of the baseline period after averaging,
>> while the 'single trial' option calculates it at single trials (I don't
>> know exactly how). The dB unit will be the same, but the baseline values
>> would be different since they are differently calculated.
>>
>>
> a- My major aim is to know when to select any of the options of drop-down
> list (Use divisive baseline (DIV) as selecting them keep change the dB into
> e.g., STD, Use standard deviation (STD), Use single trial DIV baseline, and
> single trial baseline) because when I change the default option that
> produces dB unit, it will be change into STD. Is the STD (or any of the
> left 3 choice in drop-down list) produces dB unit? In the 2 attached plots,
> I highlighted both default option and STD option results where the dB
> changed to STD.
>
> b- When to select each of Use divisive baseline (DIV),Use standard
> deviation (STD), Use single trial DIV baseline, and Use single trial STD
> baseline?
>
>
> Kind regards,
> Leonardo
>
>
>
> Makoto
>>
>> On Thu, Dec 3, 2015 at 9:42 AM, Leonardo Lion <llionname at gmail.com>
>> wrote:
>>
>>> Hi everyone,
>>>
>>> My question is about applying ht e Time- frequency analysis on my
>>> continuous dataset (I don't have any epoch in the data).
>>>
>>> In the EEGLAB, under the "Time-frequency transforms" option, there is a
>>> parameter called "Sub-epoch time limits". In the help of this parameter was
>>> written that "sub epochs may be extracted (note that this function aims at
>>> plotting data epochs not continuous data). You may select the new epoch
>>> limits in this edit box".
>>>
>>>
>>> 1- Does that mean, I can't do time-frequency analysis using
>>> "Time-frequency transforms" of EEGLAB because I don't have epochs in my
>>> data)? Is this could be a limitation of Time-frequency transforms (only
>>> dealing withe epoch data rather than continuous one)?
>>>
>>> 2- In same "Time-frequency transforms" option, what is the meaning of
>>> the word "log" that we find it in both "Log spaced" and "log power"
>>> parameters?
>>>
>>> 3- For baseline, when to select the option "Use standard deviation
>>> (STD)", "Single trial DIV baseline", and "Single trial STD baseline"?  When
>>> we change from one to another, does that change the dB unit or not, knowing
>>> that the default option "Use divisive baseline" produces bar column in dB.
>>> However, this bar color will be changed (e.g., std.) when we change the
>>> option into "Use Single trial STD baseline".
>>>
>>>
>>> Kind regards,
>>> Leonardo
>>>
>>> _______________________________________________
>>> 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
>>
>
>


-- 
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/20151207/c6d3a473/attachment.html>