Dear everyone,
1) you can enter exact onsets, but SPM seems to take into account full TRs only (at least for the following settings). Say you have a TR of 2 seconds, a window length of 10 seconds and an order of 5. Then the onsets 15 s or 16 s lead to the same time domain regressor as far as I can see (first peak is at scan #9).
2) you won't get different regressors for different "onset types" (start of the TR, middle of the TR, ...)
But check out the time domain regressors yourself.
Best,
Helmut
|