Hi Mona,

I'm sorry, but I don't see how that is actually possible.
There would be no way for 3dDeconvolve to figure out the
exact form of your waver command.

I suspect there is something wrong with your script,
happening before the 3dDeconvolve command. Note that the
actual waver command is in that file, which is a little
odd. The file should have only times in it.

If you would like, please post the script lines that
create those timing files (more than just your sample line
above - that looks fine, but there must be some mistake to
be found).

---

On a similar note, you really do not need to bother running
such a waver command yourself. It's better to provide those
stimulus times directly to 3dDeconvolve, replacing -stim_file
with -stim_times.

For example, you might change:

-stim_file 7 al_tp_CMV.1D

to:

-stim_times 7 al_tp_CMV.txt GAM


That would do essentially the same thing as running waver,
but you would not need to worry about concatenating runs,
keeping a peak of 1 when they overlap, etc.

---

One more suggestion is that you might consider trying out
afni_proc.py. That program will create a complete single
subject processing script (in tcsh). You can even just use
that to compare with the script you are currently using.

Consider "afni_proc.py -help".

- rick