Hi Khena, func2roi-sess uses the 2nd method (estimation, then averaging). For the means, it should not matter much since both operations are linear (with the exception of any temporal whitening). You will get a big difference in the variance of course. doug

Khena Swallow wrote:

Hi,

I've been using func2roi-sess for region of interest analyses and was hoping that someone here could answer what might be a pretty basic question. Does anyone know when in the analysis stream the data from voxels in an ROI are averaged? I can think of two possibilities. The first is that the BOLD data are first averaged and then the response of the ROI to different types of events is estimated. The other possibility is that the model estimates are first calculated for each voxel, and these are averaged to produce a single estimate for the ROI. I've read through the func2roi-sess help page and still have some doubt about what the answer is.

Any help on clarifying which of these approaches func2roi-sess uses would be greatly appreciated!

Thanks!

Khena

--

Khena M. Swallow, Ph.D.

Research Associate

Department of Psychology

University of Minnesota

---

---

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