Hi AFNI Experts,
I have a question about the generalized approach to the context-dependent correlation analysis (aka PPI). I noticed on Gang’s page, it says that steps 1 & 3 are identical to the standard PPI, but then says “follow steps 2, 4, and 5 except that at step 4 there are no more -1s.” Looking at step 4, it says “if you only consider one condition A, code the condition with 1’s and baseline time points with -1s.”
I am guessing that since there are multiple conditions in a gPPI, that I would instead have all 1’s for Condition A, and zeros (not -1s as the instruction for gPPI says) for everything else? Then do the same thing for each of my other conditions?
Then I could run waver on the interaction of the Seed region and each condition (& run) separately, to create distinct interaction terms for each condition?
Thanks,
Danny
