Dear FS experts,

 

I would like to do a group analysis, not over a whole hemisphere, but only whithin a ROI, which I define by myself.

Following a description which I found in the list (https://mail.nmr.mgh.harvard.edu/pipermail//freesurfer/2010-January/013220.html), I painted a ROI, and ran mri_glmfit with this ROI in the --label option.

 

When I try to display the results in freewiev, the significance is still displayed for the whole left hemisphere. 

 

What am I doing wrong?

 

Thank you very much for your help,

Ines