Thanks for that full output.
I am now running this command with "-child_epi .." and "-tshift_opts ..", and still not getting an error:
align_epi_anat.py -verb 5 \
-anat FT_anat+orig.HEAD \
-epi pb00.FT.r01.tcat+orig.HEAD \
-anat2epi \
-giant_move \
-cost lpc+ \
-epi_base mean \
-child_epi pb00.FT.r02.tcat+orig pb00.FT.r03.tcat+orig \
-tshift_opts '-ignore 4' \
-output_dir ./odir \
-save_vr \
-suffix _al \
-overwrite \
|& tee out_my_aea.txt
However, I think the different must be that you actually have slice shifts to apply to your data, and the EPIs that I have do not have those. I will ping you to get your datasets, if that is OK, to use those directly, but I suspect that is where the error is popping up; I think the 'mean' label is being misapplied there.
--pt