No, more like running the voxelwise comparisons only on voxels that have positive BP between 2 group.  Is that possible?

On another note, I am running into a puzzling "Segmentation fault" error with the following command:

mri_glmfit-sim --glmdir  lh_ad_hc.glmdir  --cache 2 neg --cwp  0.05 --2spaces

================================================

Reading source surface /Users/MacPro/Documents/NIMROD_DTI//fsaverage/surf/lh.white

Done reading source surface

Reading annotation /Users/MacPro/Documents/NIMROD_DTI//fsaverage/label/lh.aparc.annot

reading colortable from annotation file...

colortable with 36 entries read (originally /autofs/space/terrier_001/users/nicks/freesurfer/average/colortable_desikan_killiany.txt)

Segmentation fault 

================================================

Strangely, I can't seem to reproduce this error consistently too. Sometimes, after a pause of few minutes, the exact same command works! This happens on smoothed and non-smooth PET data.

I have attached the text outputs resulting from both scenarios 1) seg_fault and 2) no_seg_fault (about 10 minutes between each execution). 

The mri_glmfit-sim version is v .53.2.2 2015/08/28 19:00:13
I am running this on a Mac Pro with 64 GB RAM.

Thanks again

Best Wishes,
Elijah