Mike, have you tried the script here for making ROIs? http://montilab.psych.ucla.edu/fmri-wiki/ Naively, might have to do either with the precision or the mask you are using to make the ROI from. cheers ________________________________ From: FSL - FMRIB's Software Library <[log in to unmask]> on behalf of Mike <[log in to unmask]> Sent: 21 May 2017 14:36 To: [log in to unmask] Subject: [FSL] spherical mask problem Hi, I tried to use the following command to create a spherical mask: fslmaths point_56_n28_52.nii.gz -kernel sphere 10 -fmean -bin sphere_r10 (point_56_n28_52.nii.gz is a single voxel located in the MNI coordinate 56/-28/52, center of the sphere) When the radius was 6, 8, or 10, it worked well. But when the radius was 12 or 15, it looked weird: http://imgur.com/a/O6PGs [http://i.imgur.com/adLioCj.jpg?fb]<http://imgur.com/a/O6PGs> Imgur: The most awesome images on the Internet<http://imgur.com/a/O6PGs> imgur.com Imgur: The most awesome images on the Internet. Does anyone know why? Mike