Hi,

I'm currently working on task-based gPPI analysis (two task conditions) for three-groups of patients with 11 selected ROI's as physiological regressor and contrast difference between task conditions as psychological regressor. 

From the available literature on the analysis methods, I can understand that for seed-to-voxel and voxel-to-voxel analyses, the displayed results are corrected based on the cluster-extent thresholding across all voxels. However, while reporting the results that includes multiple ROI's and task contrasts across three-groups, should perform multiple comparison correction (FDR or Bonferroni methods) and add final corrected-p values? 

I apologize for the basic question and would appreciate any help!

Thank you in advance!

NST