of 0.05 over the entire volume according to Monte Carlo simulations conducted with AFNI. For the two-sample t-tests, the correlation maps from both groups were entered into the analysis and the z-scores at each voxel were averaged within each group and then compared between groups. For each seed, the two-sample t-test was restricted to voxels within the mask defined by a logic ‘or’ between the two group-maps of each seed resulting from the one-sample t-tests described above. Clusters with z-scores significantly differed between two groups were determined by combining individual voxel threshold of p < 0.005 with a spatial cluster size (cluster size from 5 to 12 voxels among seeds) which yielded a false-positive level of 0.05 for voxels in these masks according to Monte Carlo simulations conducted with AFNI.
