I think the way I would do it would be to run mris_preproc for each group and time point to yield a stack of thicknesses for each. Then for each I would run

mri_segstats --i lh.stack.group1.tp1.mgh --excludeid 0 --sannot fsaverage lh /path/to/cluster.annot --avgwf lh.stack.group1.tp1.cluster.dat

lh.stack.group1.tp1.cluster.dat will have number of rows equal to the number of subjects in the stack and a number of columns equal to the number clusters. The value will be the mean thickness for that subject in that cluster. Note that your results may be a little different than your voxel-based results because this does not include smoothing. If you want them to match exactly, you can smooth the stack by the amount you used in your voxel-wise results

On 09/07/2016 07:33 AM, Clara Kühn wrote:

Dear Freesurfer Experts,

I did a longitudinal 2 stage model and now I would like to extract the thickness and area values from significant clusters that survived the Monte Carlo correction.

As I understood from the various e-mails in the archives expressing the same request there is a file "something.y.ocn.dat" where those thickness values are supposedly stored for each cluster and each subject. However, this file included "the average value of each subject in each cluster" (as it states in the group comparison tutorial), meaning the -log10(pvalue).

I would like to show that the significant difference between two groups is driven by a certain group and to do that I want to display the thickness and area values of each cluster in bar graphs.

What could I do to receive these values?

Cheers, Clara