Hi all,
I have been performing VBM analysis in FSL and then using AFNI to perform my non-parametric tests using 3dKruksalWallis and 3dMannWhitney. To perform correction for multiple comparisons I've been using AlphaSim, which was probably developed with functional data in mind.
I'm wondering if anyone can think of any reason why using AlphaSim for grey matter data might be invalid. Given Rick's response from yesterday: "suppose I have random data with a given data grid and inherent blur, then what sort of clustering can be expected from mere chance?" I think it should be ok. What do you think?
Thanks,
Colm.