Group analysis 3dttest 3dMEMA

Good morning,

I am new to using AFNI (I’ve mostly used SPM). My background is mainly in resting-state fMRI doing ALFF and functional connectivity analyses of schizophrenia patients and healthy controls. To get familiar with AFNI, I am trying to do some simple analyses of seed-to-voxel connectivity in 100 schizophrenia patients and 111 controls. I preprocessed the data, and have performed the initial analyses at the single subject level to obtain seed-to-voxel correlation maps for each subject (1. computing and extracting average fMRI time series from the seed using 3dmaskave, 2. generated the correlation map from the seed time series using 3dTcorr1D and then 3. performed R–>Z transform using 3dcalc -expr atanh).

As a first analysis, I would like to explore patient vs. control differences in seed-to-voxel connectivity. In a second analysis, I would like to perform a regression on the seed-to-voxel correlation maps for the 100 schizophrenia subjects, exploring the impact of 3 behavioral factors: visual hallucination severity, auditory hallucination severity, and also model their interaction (visual severity*auditory severity).

Is 3dttest a good option for the group (patients vs. controls) analysis? Is 3dMEMA a good option for regression analysis in schizophrenia patients? I attended the AFNI workshop in October but we did not get a chance to do much interactive sessions with group analyses, so I thought I would verify that these are the best programs for the analyses I want to perform.

Thanks for your input,



Use 3dttest++ with option -covariates to include those two behavioral measures plus their interaction. Do the two group have substantial different average values for those two behavioral measures?