I just figured it out. It appears that this subject was heavily censored and the first several volumes in the errts were all zeros, so I thought that something was wrong. I also ran 3dhistog on it and it complained that all the voxels were zero, but it was not processing more than one subbrik. Anyway problem solved with the errts, sorry for the noise.
I've looked through the output of gen_ss_review_scripts.py -show_valid_opts for a option to set the volreg dataset, but nothing seemed appropriate. What command line argument should I use to provide that to the script?