Hi Gaurav,

> would you recommend that I average the errts residuals
> from all subjects and feed that averaged residual to FWHMx
> program to find out the smoothness of noise at group level
> and feed that smoothness to 3dClustSim to figure out the
> cluster size threshold at group level?

The current practice is to estimate the FWHM using 3dFWHMx with the residuals from each subject separately, and then average the FWHM values in the end (usually the values do not vary much across subjects).

> Is there a way to tell afni_proc.py to use 3dBlurToFWHM
> instead of 3dmerge to blur the data? I am dependent on afni_proc
> quite a bit. If afni_proc cannot do 3dBlurToFWHM then I will
> have to edit the proc.$subj scripts to replace 3dmerge with
> 3dBlurToFWHM.

Yes, you can do that by using the option -blur_to_fwhm in afni_proc.py

> I would like to fill up a volume with white noise and then run
> FWHMx on it, just for kicks. Can you suggest a way to
> generate a noise filled volume?

You may try this: 1) generate a 3D dataset of white noise using function rnorm in R; 2) write the data to an AFNI file with the following command in R:

write.AFNI('myData+tlrc', subBRKarray(brk=myData))

Gang