On 03/31/2014 10:05 AM, Mariam Sood wrote:
Hi Doug,
I have a few queries regarding FWHM used in multiple comparisons correction program mri_glmfit-sim
My data at present is analysed in FSL & Freesurfer following the procedure in the link http://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/FslFeatFreeSurfer. No volume based smoothing at single subject level (in FSL) is done at the moment, and a 4 mm surface level smoothing was applied during mris_preproc stage.
- I have noticed that mri_glmfit computed a FWHM of 8.578114 using residuals. How exactly is this value computed?
It is based on the correlation coefficient of the residuals of adjacent vertices.
- I understand that the cached simulation used in mri_glmfit-sim is selected based on the above computed smoothness value. Is this fine, or should I override? If I also smooth the data at the single subject level on the volume (in FSL), can I still follow the above process?
Do you mean over-ride because you smoothed at 4 but get 8.5? No, you would use the 8.5. If you smoothed in the volume, depending on how much, it is probably ok to continue the process (though technically not correct) doug
Many thanks, Mariam.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer