Dear all,
I want to ask a few questions about making a specific contrast. I have two
conditions in 1 experiment : CA CB (one rest R condition, modeled in 1
model for 17 subjects). I have another experiment with same conditions
(same participants) but passive viewing - NA and NB (one rest R condition,
in a single different model). I would like to get the result CA
inclusively masked by NA > CB inclusively masked by NB. Is there any valid
way to do this?
One option that I thought of can be to create masked images for each
subject (con-(CA>R) masked by NA>R for that subject using imcalc) and for
both conditions giving me 34(17x2) images - CA inclusively masked by NA
and CB inclusively masked by NB for each subject. Then I can use paired
t-test to get the result. I don't know if that is a valid statistical
procedure (and if yes then what threshold to use for NA mask).
Any help would be highly appreciated.
Best,
Atesh Koul
|