Hi,
My paradigm is this:

30s rest, 30s task1 (TR=2.5s), 12s rest, 30s task2(TR=2.5s),30s rest......

Now, in my proc.py, I am modeling it in 2 ways:

1. Provide stim files for task1 and task2 and using BLOCK(30,1) only

2. Provide stim files for task1, 12s rest (naming junk), task2 and using
regress_basis_multi option with
BLOCK(30,1), BLOCK(12,1), BLOCK(30,1) options

Which one seems right? To me, 2nd one make more sense as 12s rest between taks doesn't seem enough and if I don't model it, it will be part of baseline signal (I suppose).

Similarly, I carried the model to my PSC calculation.
There I have 3 TENT functions
TENT(-10,30,17), TENT(-4,12,7), TENT(-10,30,17)

But now my iresp file is giving some crazy numbers (like 30000, even though I am using scale option in AFNI)..

What seems to be the problem. Any help would be appreciated.

Thanks
Priyank