In seed-based resting state analyses, it is common to include mean CSF signal as a covariate in the first level analysis. You could try doing that and see what happens.
________________________________________
From: SPM (Statistical Parametric Mapping) [[log in to unmask]] on behalf of Andreas Lidström [[log in to unmask]]
Sent: Monday, February 24, 2014 6:12 AM
To: [log in to unmask]
Subject: [SPM] CSF activation in second level analysis
Hello,
I am running resting state analysis between two groups, but when I put my .con images from each individual into a two-sample t-test at the 2:nd level I see activation in the ventricles.
I have checked the group images for each group (t-test) and the individual 1:st level images and I cannot se this activation in either of these. (I use individual CSF regressors in the 1:st level GLM)
Any insight is much appreciated.
/Andreas
|