There are 2 reasons were time points per stim could vary.
If one goes off the end of a run (which I suspect is what is
happening with the 6) it will be short. That is one reason
why I tend to suggest some rest period at the end of a run.
Also, the responses are applied by 3dDeconvolve for only
a limited duration (until they are very small). So depending
on how the onset (and offset) times align with the TRs, there
can be a difference of one time point between the response
curves of 2 events, as the curves are sampled at different
offsets.
- rick