Dear Lukas,

One way to do that (e.g. if you are planning to use this mask for all of the analyses, not just this gPPI analysis) would be to enter that mask in the Setup.Options tab in the 'analysis mask' field. That will have CONN compute voxel-level BOLD timeseries only at voxels within the mask, and it will restrict all subsequent denoising and first-level analyses to only those voxels.

Another way to do that (e.g. if you want to only use this mask for a specific analysis) is to use masking at the level of your second-level analysis. After running gPPI analyses and group-level analyses normally, you can specify an additional mask that restricts second-level analyses to only voxels within that mask. You can specify this mask for example in the 'SPM display' option of the results explorer window (or alternatively, if using scripts to run your analyses, you can also specify this mask directly when running your second-level analyses using the conn_module('GLM',...'mask',...) option; see "help conn_module" for details)

Hope this helps

Alfonso

Originally posted by Lukas Van Oudenhove:

Dear Alfonso & team,

I would like to restrict the voxels in which I will do my gPPI analysis to a (single) mask consisting of voxels in a number of prehypothesized regions rather than the entire brain/grey matter.

Is there an option to do this in CONN, for gPPI or more generally for any voxel-based analysis?

Thanks very much in advance!

Best wishes,

Lukas