Hi Bente, it's hard to say without seeing the command line (always a good thing to include when posting a question), but it looks like you need to include '--surface subject hemi' where subject is the name of the subject whose surface the analysis is being performed on and hemi is the name of the hemisphere (lh or rh). This information is needed to compute the spatial smoothness.

doug

bbruijn@bidmc.harvard.edu wrote:

Dear FS experts,

I came across an error-message that I can't figure out, hopefully you can help me. I want to do a paired T-test (first without regressors) and I am following the steps on the Freesurfer website. Preprocessing and smoothing all went OK and I've just created the second fsgd file;

GroupDescriptorFile 1 Class Main Input subject1pair Main Input subject2pair Main Input subject3pair Main Input subject4pair Main Input subject5pair Main Input subject6pair Main Input subject7pair Main Input subject8pair Main

Because I don't want to include any regressors I put in the flag --osgm (is this right?) and didn't include the contrasts.

This is the error message I receive after putting in the mri_glmfit command:

ERROR: you must use '--surface subject hemi' with surface data

Did I do everything correct or did I forget a step?

Thank you so much in advance,

Bente _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer