Doron:
I will defer to Rik based on his experience with models and contrasts.
I encountered the same mental hurdle in doing a repeated measures ANOVA.
Lukas Pezawas ([log in to unmask]) suggested using the
multisubject, conditions and covariates PET model. The only
disadvantage is that it excludes voxels < image mean/8, a holdover from
the PET days.
Enter 2 groups, 2 condtions (1 and 2, corresponding to time 1 and time
2). Then try the following F-contrasts:
for the effect of group -1 -1 1 1, 0 0 0 0....
for the effect of time -1 1 -1 1, 0 0 0 0.....
for the interaction of group and time 1 -1 -1 1, 0 0 0 0.....
Hope this helps,
Rob McClure
Robert K. McClure MD, Assistant Professor
Department of Psychiatry, School of Medicine
University of North Carolina at Chapel Hill
Address: Campus Box #7160
Chapel Hill, NC, 27510-7160, USA
E-mail: [log in to unmask]
Office phone: 919-843-6629
|