Hengyi,
> Is the corrected P value from your code meaning the P is for the
> cluster after the whole brain correction?
Yes, it corrects for the whole-brain, as saved in the R vector.
> How can i calculate a corrected P value of a voxel after the
> correction of a hypothesis region (e.g., suppose this region
> includes 1000 voxels)?
Random field theory needs to know both the search volume *and* the
topology of the search region, so it isn't as simple as specifying
number of voxels.
My best advice is to study spm2/spm_VOI.m, and see how search regions
are defined and then submitted to spm_resels() to calculate the R
vector.
Hope this helps.
-Tom
-- Thomas Nichols -------------------- Department of Biostatistics
http://www.sph.umich.edu/~nichols University of Michigan
[log in to unmask] 1420 Washington Heights
-------------------------------------- Ann Arbor, MI 48109-2029
|