Hi Ozcelik,

Thanks for the message bump, life has been busy here...

It sounds like you just need to add a couple of -amean or
-bmean options (depending on whethere pre/post is your A
or B factor pair).

Note that it should be equivalent to running 3dMean on
sub-brick 0 of all of your subject 'betas' datasets, as
well as running it on sub-brick 4.

This assumes that your betas are already in %-change. Are
they? Did you scale your data before 3dDeconvolve?

---

If this is not clear enough, please post the top and bottom
of your 3dANOVA command (we don't need to see all of the
-dset options, but enough to understand the pattern).

---

A separate point. I see you are using 'all_runs" as input
to 3dDeconvolve, but nowhere do you specify a list of runs
via the '-concat' option. How many runs are in 'all_runs'?
If it is more than 1, you need to let 3dDeconvolve know.

- rick