[Eeglablist] common baseline in std_ersp with 2 independent variables

[Arnaud Delorme]

Dear Anna,

to have the same baseline in all conditions, you need to pair the condition in the EEGLAB design interface. It is that simple. If the data is not paired, then the baseline is computed independently for all ERSP. 
If you have a design say 3x2 with 3 conditions and 2 groups where the conditions are paired and the groups are unpaired, a common baseline will be subtracted for all conditions in each group. If the group are paired as well, the same common baseline will be subtracted for all groups and conditions.
I will document this further in the wiki.
Hope this helps,

Arno

On Aug 6, 2012, at 7:46 PM, Makoto Miyakoshi wrote:

> Dear Arno,
> 
> These are questions about details, and I don't know exactly how they work.
> Could you answer to these questions?
> 
> Makoto
> 
> 2012/8/2 Anna-Antonia Pape <aa.pape at iwm-kmrc.de>:
>> Hello everyone.
>> 
>> I want to compute the ERSP on the component level in a study with 2
>> independent variables with 2 levels each.
>> So it's a 2x2 design, say    L vs R   x   + vs -  , i.e. 4
>> condition-pairs (L+, L-, R+, R-)
>> To save space on my disk, i have all conditions in one dataset, not
>> separate datasets.
>> I would like the baseline to be the same for all 4 conditions, e.g. the
>> mean of the 4 conditions. (I don't know whether this is sensible).
>> 
>> Anyway, I came up with two ideas to get there:
>> 
>> 1)  I tried using 'commonbase' as a value added in the pop_precomp GUI.
>> But if i understand it correctly by looking at the code of pop_precomp,
>> the way pop_precomp calls std_ersp, std_ersp calculates the ERSPs for
>> each condition-pair individually, i.e. in this fashion:
>> 
>> for subject
>>    for condition-pair
>>       for component
>>                          std_ersp()
>>       end
>>    end
>> end
>> 
>> So it seems to me as if i cannot use 'commonbase', if I want to have the
>> same baseline for all four conditions. Is this correct?
>> 
>> 
>> 2) I tried entering the precalculated mean of all the conditions via the
>> pop_precomp GUI window with the 'powbase' key.
>> It seems to me that the 'powbase' key somehow gets lost on the way and
>> is not handed over to std_ersp. Is this is correct?
>> 
>> I highly appreciate any comments on this.
>> 
>> Cheers from Tübingen,
>> Anna
>> 
>> --
>> Anna-Antonia Pape, M.Sc. of Neural  & Behavioral Science
>> 
>> email: aa.pape at iwm-kmrc.de
>> phone: +49 7071 979 301
>> 
>> Knowledge Media Research Center
>> Schleichstr.6
>> Room 536
>> 72076 Tübingen
>> Germany
>> 
>> www.iwm-kmrc.de
>> 
>> _______________________________________________
>> 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
> JSPS Postdoctral Fellow for Research Abroad
> Swartz Center for Computational Neuroscience
> Institute for Neural Computation, University of California San Diego