Hi all!
I would appreciate it if anyone in the know could verify that I'm not
making any grievous errors in setting up a conjunction for RFX.
I'm using bias-corrected .con images using the method proposed by
Calhoun et al. (2004), so I'm forced to use 2-tailed T-tests to produce
the .con images at the first level.
What I'm trying to do is to identify voxels that are activated in
both task 1 and task 2, with the intention of using these as the seed
area in a PPI. Extrapolating from previous posts on the list I'm
thinking that the correct way to do this is:
1: Perform two-tailed T tests [Task 1 > Baseline] and [Task 2>
Baseline] for each subject.
2: Then I take the resulting .con files into a flexible factorial with 1
factor and 2 levels, one for each task. The resulting RFX- model should
give me the voxels activated in both tasks.
Does this sound right?
Best wishes,
h.
|