Thanks very much for the explanation. Is it fair to conclude from your answer that if I run
3drefit -deoblique -xorigin cen -yorigin cen -zorigin cen 8101-T1+orig
3drefit -deoblique -xorigin cen -yorigin cen -zorigin cen 8101-EPI-01+orig
immediately after reconstructing the scans (before running 3dTshift), and if that gets rid of my alignment problems, I'm all set? That may be the simplest solution for me. There's no danger in calling 3drefit in this way on the EPI scan before slice-time correcting?
I suppose if there is danger I could call 3drefit after 3dTshift, but it's easier in my processing stream to do it the other way around (and a lot of data has already been processed with 3drefit having been called first.)
If that doesn't work, I'll try the other options you recommended.
Thanks again,
Dan