The stimulus duration is just the amount of time that optseq will allocate to the stimulus. You can fill it however you like, so 300 + 1700 null is ok.

As to whether the stimulus onset should be synchronous with your TR, there are some other considerations. If you are going to assume a shape to the hemodynamic response, then it is not important (and many people prefer to do it this way)-- you can set dPSD as small as you like (though it might take a while). If you are not going to assume a shape (ie, an FIR model), then you are going to lose statistical power by a factor of TR/dPSD

doug

Bosky Ravindranath wrote:

Hello List, We are designing a rapid event related paradigm, wherein a stimulus is presented for a duration of 300ms. We have been using optseq for generating sequences in the past, however , optseq requires that the event duration be a multiple of of TR or dPSD( sampling interval). In that case is it ok to create a design with a stimulus duration of 2 s ( which is our TR) and then split it into 300ms + 1700 ms, where the 1700 ms adds to the null period immediately following the stimulus? ( There is a null condition between each stimulus). Also, is it not important to start a new volume at the beginning of each stimulus? Any input is appreciated. Regards, Bosky Ravindranath

Laboratory for the Study of Emotion and Cognition, Stony Brook University, NY,11790

---

---

Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer