See whether the interaction survives correction for multiple comparisons. If it does not, then proceed with doss. If it does not, see whether the interaction voxels overlap with your contrast of interest; if not, then proceed with doss.
On 9/6/2019 11:28 AM, Barletta, Valeria wrote: So when I use doss (and the contrast -1 +1 0), I have very similar results to those that I got when I completely remove age: [cid:part1.A2C83656.0A5EE751@mgh.harvard.edu]
________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edumailto:freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edumailto:freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Greve, Douglas N.,Ph.D. DGREVE@mgh.harvard.edumailto:DGREVE@mgh.harvard.edu Sent: Friday, September 6, 2019 10:38 AM To: freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Fw: question on GLM
Did you try using DOSS?
On 9/6/2019 9:51 AM, Barletta, Valeria wrote: Dear Doug, As you suggested yesterday, I launched the GLM to check for interactions group-age with the contrast: 0 0 +1 -1 And I obtained this image: [cid:part2.6ACC220B.1598BE28@mgh.harvard.edu] Then I re-launched my model including age as a covariate and I obtained very similar results: [cid:part3.FE061024.68F4E7AB@mgh.harvard.edu]
Finally, I re-launched the model excluding age and I obtained a very different thing, which is actually more consistent with my data: [cid:part4.B3E8792E.485B0EE5@mgh.harvard.edu]
How shall I interpret this? Shall I exclude age from my analysis? (I do not really expect it to have an influence on my variable y)
Thank you, Vale ________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edumailto:freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edumailto:freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Greve, Douglas N.,Ph.D. DGREVE@mgh.harvard.edumailto:DGREVE@mgh.harvard.edu Sent: Wednesday, August 28, 2019 6:03 PM To: freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Fw: question on GLM
It is showing b = g2 - g1 so negative means that g1 > g2
On 8/28/19 1:59 PM, Barletta, Valeria wrote:
Forgot to report my contrast matrix:
-1 1 0 0
Thanks, Valeria
*From:* Barletta, Valeria *Sent:* Wednesday, August 28, 2019 1:02 PM *To:* Freesurfer support list freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu *Subject:* question on GLM Good afternoon! I have a question on the Freesurfer GLM. The command:
mri_glmfit-sim --glmdir ${glm_dir}/${hemi}_contrast-${contrast}_${surfname_cme} --cache 1.3 neg
is giving me blue clusters (see attached picture). Do blue clusters mean that my variable y is LOWER in the FIRST or the SECOND group listed in the FSGD file?
Thanks, Valeria
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edumailto:Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edumailto:Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edumailto:Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edumailto:Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
