The simple GLM should be the same. When you convert to nifti, you can either leave it as a surface overlay or you can run mri_surf2vol to create an actual volume of thicknesses. The latter may change the values and so change the results. Cluster-based correction for multiple comparisons will be different/wrong without taking into account the surface geometry.
doug
On 12/11/2013 10:58 AM, Govindarajan, Koushik Athreya wrote:
Dear Freesurfers,
I have a question regarding how Freesurfer’s paired glmfit and monte-carlo analysis compares with SPM8 GLM design analysis on cortical thickness. If I were to export thickness maps as nifti files and do a paired comparison on SPM8, would that be a fair comparison with Freesurfer or is Freesurfer’s analysis more appropriate for surface maps such as cortical thickness?
Thanks
Koushik
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer