ok, thanks.
One last question.....When it tells me to save my path matrices at the end, it gives me 3 options (with a lag of 1, 2, and 3). I tried to read your paper but its way beyond my understanding in math (equations and stuff), which path matrix would give me the best model. What I did for my data was I ran 3dDeconvolve to remove any drift effects, motion, linear trends, etc without modeling my stimuli and used the errts or residuals as my signal. Then, I extracted residual signal from the blocks for the two tasks we have. We have a TR of 1s.
Thanks again