[Eeglablist] per subject or pooled?

[Arnaud Delorme]

>I am new to single trial analysis.
>What is the best strategy to process data?
>
>(1) To process individual data and make a kind of averaging accross 
>subjects at the end or 
>(2) to pool trials from all subjects in a single analysis?
>  
>
We prefer to use the first solution.
The problem with the second solution is that some components might be 
specific of some subjects (since the same functional component has never 
exactly the same topographical map in all subjects). It might work 
though if you have more than 50 subjects?

>If I pool data, is it a valid and useful strategy to normalize the 
>amplitude in all trials (according to subject means and sds)?
>  
>
Yes, I think it is the minimum you can do to avoid having subjects 
dominating ICA components. When your decomposition is done, you should 
check if some subjects do not dominate some components (and since it 
will probably be the case, you will have to use solution (1)).
Hope this help

Arno
-- 

*Arnaud Delorme, Ph.D.*
Computational Neurobiology Lab, Salk Institute
10010 North Torrey Pines Road
La Jolla, CA 92037 USA

*Tel* : /(+1)-858-458-1927 ext 15/
*Fax* : /(+1)-858-458-1847/
*Web page *: www.sccn.ucsd.edu/~arno <http://www.sccn.ucsd.edu/%7Earno>
*To think upon*:

    Let your mind function freely without abiding anywhere or in anything.

        /PRINPRSU - from the Diamond Sutra/