Hi,
I'd like to revive this discussion of how to divide massive clusters. I am
having a similar problem as this posting awhile ago -- 3dmerge to cluster
order results in a huge first cluster covering many different regions of
interests. Using 1erode doesn't make a difference in breaking up the
cluster, and I've increased the p threshold to 1e-6 with minimal
improvement (and also at the risk of losing parts of the cluster). If
possible, I would like to avoid breaking the cluster up according to
Talairach regions.....Any suggestions appreciated!
Thanks in advance,
g