Dear list,
I am trying to run randomise for a 2 paired samples T test in a particular mask (about 7000 voxels). I would like to have some feedback to know whether I'm doing things right.
-Input: I am using the filtered_func_data.nii file coming from the paired T test performed in FEAT. Is that correct? The subjects are in the following order: subj1condA, subj2condA, subj3condA, subj1condB,subj2condB, subj3condB.
-design.mat: let's say I have 3 subjects...
EV1 EV2 EV3 EV4
1 1 0 0
1 0 1 0
1 0 0 1
-1 1 0 0
-1 0 1 0
-1 0 0 1
Therefore, the first EV models the 2 groups, and EV2 to EV4, the subject's mean.
-design.con:
1 0 0 0
-1 0 0 0
-design.grp: is it really necessary? If so, how would it be?
Then, the command would be:
randomise -i filtered_func_data.nii -o output -d design.mat -t design.con -m ROI.nii -T
Thanks!
Javier
|