Hi Alfonso,

Thanks so much for your reply. It was very helpful. We went ahead and ran the associations separately, as you suggested. We believe we find a very strong positive correlation between performance and group B and a very strong negative association between performance and Group A (screenshots are attached). In this case would it be a correct interpretation to say that the negative contrast we find in the cluster is driven by both of these associations? Or would it be attributable more specifically to one than the other? We also included a screenshot of the main effect of the two.

Another question: we have read in the forum that the issue of multiple comparisons in still debated. In this case, we selected 4 seeds in a seed-to-voxel analysis, which we ran separately. We kept the default cluster threshold value of 0.05 and height threshold of 0.001 and considered only the FWE-corrected results. Would we need to correct again for multiple comparisons considering the 4 seeds? As in, FWE-corrected 0.05/4?

Thanks again for your help.

Kaylah