Hi Heather,
You probably have 2 runs of data, are using a linear
polynomial for baseline, and have a significant shift
in the white matter regressor. What does the white
matter regressor look like? Assuming it does have a
shift, the "why?" would depend on the pre-processing
steps, which we do not know.
I take it that you are not using afni_proc,py for the
analysis. It might be a good idea to use it, at least
for comparison.
- rick