It depends a bit on whether they are nuisance regressors or not. If they are not nuisance regs, then I would be very careful normalizing them because it could make the regression coefficients hard to interpret when comparing across subjects (you would probably just use the same scale for all subjects). If they are nuisance, then z scoring them is fine.
On 02/12/2017 02:52 AM, Reza Rajimehr wrote:
In a functional analysis, we have a set of external regressors (e.g. mean luminance of stimuli, size of stimuli, etc.). Should we normalize values before including them in the regressor files? If so, what is the recommended way of normalization ... z-scoring?
Thanks, Reza
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer