External Email - Use Caution

Hello FreeSurfer Developers,

I'm attempting to build my own Z Monte Carlo on fsaverage5, as described on the BuildYourOwnMonteCarlo tutorial (https://secure-web.cisco.com/1uikaYhiWSF4KzsQTXjySqLmwrS4IRC0PJ5sIbxRfHQXCed...), but I get a segmentation fault.

As a reproducible example, the following command tries to compute 10 replications at FWHM=15 and save it in a subdirectory fsaverage5 of my current directory.

$ mri_mcsim --o ./fsaverage5 --base mc-z --surface fsaverage5 lh --nreps 10 --fwhm 15

Issuing this command leads to a segmentation fault. The full output is in the attachment (mri_mcsim_error.txt), and here are the last lines before the error happens:

Loading /Applications/freesurfer/7.1.1/subjects/fsaverage5/surf/lh.white group_avg_surface_area 84969.3 group_avg_vtxarea_loaded 1 Loading label file cortex Loading /Applications/freesurfer/7.1.1/subjects/fsaverage5/label/lh.cortex.label Segmentation fault: 11

Adding the «--debug» option gives no additional information. If I add the “—no-label” option, it works. However, I need those labels for later use in mri_glmfit-sim so that does not solve the issue.

Using fsaverage6 instead works fine: “$ mri_mcsim --o ./fsaverage6 --base mc-z --surface fsaverage6 lh --nreps 10 --fwhm 15” gives no error messages.

I've searched the list and no similar errors have been reported. Does anyone have any thoughts on how to trouble-shoot this one?

1) FreeSurfer version: freesurfer-darwin-macOS-7.1.1 2) Platform: macOS Big Sur 11.4

PS! I’m also aware that permutation testing is in general recommended rather than Z Monte Carlo simulations, but in this particular application the dataset is so large that permutation testing is likely not a viable option, due to computing constraints.

Best regards,

Øystein Sørensen, PhD Associate Professor in Statistics Center for Lifespan Changes in Brain and Cognition Department of Psychology University of Oslo http://secure-web.cisco.com/1WEu63aU9x0YgcndrjS1ZoT7n-mRQdI0c2tfDzloh2yF1pU0...