Hi Tom,
The design matrix I am using is:
Patients: 1 0
Controls: 0 -1
Patients > Controls 1 -1
Controls > Patients -1 1
Using FEAT w/this design matrix I get activation (t>2.3) in all but the patient > control contrast.
Using this design.mat file for randomise, I only get significant clusters in the control > patients contrast. I'm confused about where the really robust activity in the individual patient and control group maps went...
Thanks,
A-M
|