Hi everybody,
we have run an experiment with six conditions described by two factors, one with 3 and one with 2 possible values, and a baseline condition. We would now like to determine whether the response in each voxel is influenced by each of the factors.
So far I've been using a set of GLT's in 3dDeconvolve to do this, by comparing pairwise across the rows and columns of the stimulus matrix.
Now I just tried to use 3dANOVA2 to the same purpose, and am told that my sample size is too small for computing the F-interaction. What exactly does that error message refer to, and where am I going wrong?
Any input would be greatly appreciated.
Cheers,
Kai