I always use the 1st option. I've played around with it a bit and have not found that it makes that much of a difference for fMRI data with 3x3x5 voxel size. The default is projfrac=0 for historical reasons. It is just the way that I programmed it when I first wrote it (years ago now), and I'm always afraid to change a default.

doug

On 01/15/2014 01:45 PM, Josef Ling wrote:

I'd like to project functional data onto the surface for group analysis. I'd like to limit partial voluming when selecting values with mri_vol2surf.

I noticed that the default projfrac = 0 for mri_vol2surf which seems to me it would select a number of values contaminated by WM. Why is the default = 0?

...

From the examples below, which is the soundest method of selecting

representative GM only values? Please correct any errors in my understanding. :)

mri_vol2surf --projfrac 0.5 --interp nearest --surf white This would select a single value closest to halfway up the normal vector from the WM vertex.

mri_vol2surf --projfrac 0.5 --interp trilin --surf white This would interpolate from multiple nearby values halfway up the normal vector from the WM vertex.

mri_vol2surf --projfrac-avg 0.25 0.75 0.25 --surf white Average from two values a quarter of the way up and down?

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