There is not a way to do this using an FSGD and a DODS model. It will work if you change it to DOSS (if that is appropriate). Otherwise, you will need to create your own design matrix and feed it to mri_glmfit with --X
doug


On 9/13/13 10:21 PM, ye tian wrote:
Dear Freesurfers,
 
I am studying how the practice of meditation affects cortical thickness. I have two types of meditations, and I also know for how many years each mediator has meditated. Therefore, ideally I would like my categorical factor to be meditation (control, meditation1 and meditation2), and continuous factor to be the number of years of meditation. This design, of course, has a problematic column of zeros, which corresponds to the slope of the control (a non-meditator has meditated for 0 years).
 
Does freesurfer has a way out of this problem? For example, I have came across pseudoinverse methods in literature?
 
Any suggestion is highly appreciated.
 
Thank you very much!
Sincerely,
Ye


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