Hi rick

Yes. the overlap area locates in the medial of brain. so you mean there will always are overlap areas no matter what i try, because the left and right hemi do share voxels?

I was trying to to consensus map to find out how many subjects activated in one voxels. say i have 15 subjects, the max value of the consensus map should be 15. well now i have problems. after 3dcalc to combine the left and right hemi, there are always some voxels whose value bigger to 15, in the medial brain. if there do have some voxels shared by left and right hemi, it's not surprise. But how am i gonna do to avoid this embarrassed situation?

And I used to do significant test on the surface domain nodes, and then map it to volume domain by the same way as above , now i think it's a problem too!

Thanks a lot.