[Eeglablist] Study Plot channel ERSP GUI and script output

[Makoto Miyakoshi]

Dear Katharina,

To be honest, the STUDY design is also a mystery to me, particularly kind
kind of things. If what you reported is correct, it is certainly confusing,
at least to me. I'm happy for you that you found it out anyways!

You can give comment on this to developers. Please copy and paste your
message to EEGLAB bugzilla
https://sccn.ucsd.edu/bugzilla/enter_bug.cgi
I appreciate your cooperation and patience.

Makoto

On Mon, Oct 24, 2016 at 5:40 AM, Katharina Limbach <
katharina.limbach at uni-jena.de> wrote:

> Dear EEGLABers,
>
> I have a question concerning the data structure underlying the plots made
> with the GUI /Plot channel measures) and the output of the std_erspplot
> function.
>
> In short it seems to me, that the graphical output I get with the GUI and
> the std_erspplot function is organized differently to the output cell
> created by the std_ersp function, which is very confusing to me!
>
> Here is what I am doing:
> I have a study with a 2 x 2 design (novelty (Hits vs Correct Rejections
> (CR)) and voice age (young vs. old) and 24 participants. I have 4 .set
> files per participant and created the STUDY by using the condition input
> for novelty (Hits and CR) and session for voice age (1 = young, 2 = old). I
> set it up that way, so that I could later on built my designs by selecting
> Condition as my first independent variable and Session as my second
> independent variable.
>
> I then precompute the channel measures to get the ERSP. When I afterwards
> plot the results for one channel, I get a figure with 4 time-frequency
> graphs.
> The graphs are titled like this:
> "CP4, 1, CR"    "CP4,1,Hits"
> "CP4, 2, CR"    "CP4,2,Hits"
> So the first row seems to show both conditions for the first session (i.e.
> young voices) and the second row both conditions for the seond session
> (i.e. old voices).
>
> I then wanted to extract the data to do some some stats in a different
> program.
> I used the following code:
> [STUDY erspdata ersptimes erspfreqs] = std_erspplot(STUDY, ALLEEG, ...
>     'channels', {STUDY.changrp(55).name}); % 55 = channel CP4
>
> (The graphs look the same as when created from the GUI.) The erspdata
> output is a 2 x 2 cell array, and each element holds a 4-D matrix (22
> (frequencies) 250 (timepoints)     1 (channel)   24 (participants)). So far
> so good.
> However, I then assumed that this cell would be organized like the
> graphical output of the std_erspplot and the gui function.
> So that
> (1,1) 1,CR, (1,2) 1,Hits
> (2,1) 2,CR  (2,2) 2,Hits.
> However, that does not seem to be the case! It seems that the rows and
> columns are switched and that the cell array is actually organized like
> this:
> (1,1) 1,CR, (1,2) 2,CR
> (2,1) 1,Hits (2,2) 2,Hits.
>
> One way I tested this was to average the erspdata across participants (per
> cell) and then plot it using the Matlab imagesc function. The erspdata
> output seems to be organized such that the rows represent the Condition and
> the columns the Session (opposite to what it shown in the graphs).
>
> Why is this happening? Is this a bug, or where I am doing something wrong?
>
> I really appreciate your help,
> Katharina
>
>
>
>
> _______________________________________________
> 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/20161025/5227be47/attachment.html>