This is a question for the 1st level fMRI specification step.
I'm trying to include some regressors into my design matrix to look for
brain areas that modulate with certain ongoing measures. At the start
of each session, there is 30-60s where subjects are doing a different
task and so we have no ongoing measures for the primary task (i.e., they
haven't yet started the primary task). I don't want to trim away the
first 30-60s of each session because there are important trial types in
that period that I am analyzing.
So, is it possible to include regressors where SPM ignores the first
30-60s of those regressors? I thought that perhaps having NaN values
for the regressors during the first 30-60s would lead SPM to ignore that
time window but I don't know the inner workings of SPM enough to be
certain it would work. Does anyone know if it would indeed work, or if
there are any other ways to get SPM to ignore a regressor during certain
time windows?
Thanks in advance
Cheers,
Jason
|