Hi Andrew,
Your description seems fine.
Correct me if I am wrong, but as far as I can tell
you are not asking anything about band passing, but
just whether censoring by time point removal in the
GLM is different from censoring by adding stick
regressors for those TRs.
There is no difference in the results with those 2
methods.
But given that the results are the same, what is
the point in calling data-removal preferred? For
what it's worth, that is how censoring has been
done in 3dDeconvolve since it was written in 1998.
What is on your mind about this?
- rick