I'm trying to design a task with the following 3 components per trial: a) subjects view a video clip for 5 seconds, b) subjects rate aspects of the video clip for 4 seconds and c) subjects look at a cross hair fixation

I'm most interested in the fMRI signal during component A - when subjects are viewing the video clip.

My question is how long should I make the crosshair fixation - is it best to a jitter the length (say 1-3 seconds)?  or should I try to made the trial length similar to the duration of the hemodynamic response (say 15 sec) and thus make the fixation about 6 seconds?

Thanks!
Fred