[Eeglablist] PSD for single subjects vs. STUDY function

Makoto Miyakoshi mmiyakoshi at ucsd.edu
Wed Oct 7 14:59:24 PDT 2015

```Dear Roxanne,

I've also seen another post related to the same issue; potential scale
confusion in single-trial analysis. Could you kindly post the question to
EEGLAB bugzilla?
https://sccn.ucsd.edu/bugzilla/enter_bug.cgi
Sorry for inconvenience and thank you for your patience and cooperation.

Makoto

On Tue, Oct 6, 2015 at 4:00 AM, Roxanne Sopp <roxannesopp at gmail.com> wrote:

> Dear all,
>
> my question is somewhat related to this older post below:
>
> I have precomputed power density in the STUDY function using the parameters 'specmode',[psd] in order to use the spectopo function.
>
> Now i am facing two problems:
>
> 1. The plot does not seem to be scaled in dB (as stated below), from the script i gather that these must be the absolute power values in µv²/Hz.. is that the case?
>
> 2. When I additionally use 'logtrials','on' as paramters the power values seem to be scaled in dB however they still do not correspond to the values that i calculated in the single subject mode using the spectotopo function.
>
> What could be the difference? And which one is the valid calculation?
>
> Thanks a lot in advance for any hints and suggestions!
>
> Best regards,
>
> Roxanne
>
> ______________________________________________________________________________
>
> Dear Marco,
>
> if you use no option to compute spectrum, then the units are power density (10*log(microvolt^2)/Hz) because the STUDY function uses the spectopo function. If you use 'specmode', 'fft', when you computed the spectrum then the unit is power values (10*log(microvolt^2)). These two are proportional but I agree that they should be made consistent. We have entered a bug report.
>
> Best regards,
>
> Arno
>
> On Sep 19, 2011, at 6:33 AM, Marco Montalto wrote:
>
> >* Dear all,
> *> >* I have created a STUDY and pre-computed measures for power spectra using the default spectopo parameters ("fft"). My question is the following: the numerical values I extract using this code (example is with channel Fpz):
> *> >* [ STUDY specdata specfreqs] = std_specplot(STUDY, ALLEEG, 'channels', { 'Fpz' });
> *> >* are power values or voltage or amplitude etc.?
> *> >* If they are not power values can someone explain to me how I can extract power values as I need these to calculate ERD/ERS?
> *> >* Thanks and regards,
> *>* Marco*
>
>
> _______________________________________________
> 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
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20151007/a9e5ebf9/attachment.html>
```