[Freesurfer] correcting mri_glmfit results

Dear FreeSurfer experts,

I'm trying to run a mri_glmfit on my first time point (out of 3). I ran the glm with this: mri_glmfit --table $SUBJECTS_DIR/aparcs_stats_sc1.txt --fsgd $SUBJECTS_DIRfsgd_sc1.txt --C $SUBJECTS_DIR/group.mtx --surf 87kids_template lh --cortex --glmdir $SUBJECTS_DIR/glm/lh.thick_group.glmdir

So far so good :)

Now when I try to apply the Monte Carlo correction with this: mri_glmfit-sim --glmdir $SUBJECTS_DIR/qdec/87kids/glm/lh.thick_group.glmdir --cache-dir $SUBJECTS_DIR/average/mult-comp-cor --cache 3.0 neg --cwp 0.05 --2spaces

I get the error "fwhm: Undefined variable."

So I tried to run the glmfit with the --y flag and use the lh.thickness_sm10.mgh file but I created that when registering and smoothing ALL my time points to the template, so I get this error: "ERROR: dimension mismatch between y and X. y has 224 inputs, X has 72 rows."

I have looked at mri_glmfit-sim --help but there seems to be no flag for fwhm only an overwriting one.

Should I register and smooth the data again for the separate time points or is there another way to deal with that? What could I do to correct my glm data?

Cheers, Clara