When the GLM for a group of subjects is computed using mri_glmfit, a residual FWHM value is computed and stored in the file 'FWHM.dat'.
When running a subsequent mri_glmfit-sim, using a pre-cached Monte-Carlo Z (mc-z) simulation for clusterwise correction for multiple comparisons, there appear to be separate simulations for a range of FWHM values stored (for each hemisphere, lh and rh) in /freesurfer/average/mult-comp-cor/fsaverage/?h/cortex/ from 01 to 30 (1 to 30 mm? - fwhm01, fwhm02, etc).
My question is, if the value in FWHM.dat for the original mri_glmfit for a hemisphere is say, 15.294 mm, does mri_glmfit-sim (mc-z) use interpolated data from fwhm15/ and fwhm16/ in its calculations, or does it choose the nearest value (in this case, fwhm15)?
Thanks, Chris Hyatt
Olin Neuropsychiatry Research Center Institute of Living
It uses the nearest value that is greater than the actual value.
doug
Chris Hyatt wrote:
When the GLM for a group of subjects is computed using mri_glmfit, a residual FWHM value is computed and stored in the file 'FWHM.dat'.
When running a subsequent mri_glmfit-sim, using a pre-cached Monte-Carlo Z (mc-z) simulation for clusterwise correction for multiple comparisons, there appear to be separate simulations for a range of FWHM values stored (for each hemisphere, lh and rh) in /freesurfer/average/mult-comp-cor/fsaverage/?h/cortex/ from 01 to 30 (1 to 30 mm? - fwhm01, fwhm02, etc).
My question is, if the value in FWHM.dat for the original mri_glmfit for a hemisphere is say, 15.294 mm, does mri_glmfit-sim (mc-z) use interpolated data from fwhm15/ and fwhm16/ in its calculations, or does it choose the nearest value (in this case, fwhm15)?
Thanks, Chris Hyatt
Olin Neuropsychiatry Research Center Institute of Living
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
freesurfer@nmr.mgh.harvard.edu
-
Chris Hyatt -
Douglas N Greve