Hi Zhuang,

Many of these questions are a bit general, making them a little
difficult to answer quickly. I suggest reading over some of the
class handouts.

Note that this web site only accepts jpeg images.

----

There is no default model in 3dDeconvolve. The default model
in afni_proc.py is GAM, but you should pick the model based on
what you expect.

How do you run 3dDeconvolve without -stim_times, do you mean
using -stim_file? If so, then you are still chosing the model.
If you mean just using -min/max_lag, then you are letting the
shape come from the data. I suggest reading these handouts:

[afni.nimh.nih.gov]
[afni.nimh.nih.gov]

Note that -TR_times is optional, and is used to specify the timing
of iresp (and sresp) output.

I suggest not using -stim_min/maxlag, but TENT functions instead.
Though your same question applies. The duration should be how long
you expect the response to take.

- rick