Dear SPMers,
I'm trying to perform ROI analysis, however I don't know if I did it well.
Actually, I would like to examine if some regions which aren't activated
in a whole brain analysis presented intensity differences between
experimental conditions.
I've got 20 participants and 2 experimental conditions.
My first idea was to use REX toolbox in order to extract 20
eigenvariates from spm.mat 2nd level and use a statistic analysis to
compare the means.
However, I was told that percent of signal change - extracted with
marsbar toolbox - could be more appropriate.
My understanding is that percent of signal change is more time course
dependent than eigenvariates, is that right ?
Is there anyone to help me please ?
Best,
O.A.
|