I have analyzed some fMRI data using afni_proc.py, and the first level results look OK. However, when I try to use 3dttest++ to run a group analysis I get an error message that the stat volumes of the different subjects are of different size. Some stat volumes have 54 * 64 * 50 voxels (3 * 3 * 3 mm) and some volumes have 64 * 76 * 60 voxels (2.5 * 2.5 * 2.5 mm). I don't see how / why this can happen, as all the original fMRI data and the T1 volumes have the same number of voxels and the same voxel size. Using 3dinfo, I see that the only thing that differs between the subjects is the extent of the volumes.

What should I do to be able to do my group analysis?