Hi all,
I have a set of 3D statistical parametric maps derived from an fMRI experiment
in which we have two levels of dependency.
- All subjects were scanned twice.
- All subject are dizygotic (non-identical) twins.
For every single scan (and thus statical map) we have a behavioural measure,
which we would like to regress against the set of maps, while correcting for
the dependency between scans and ideally also between twins.
In neuroimaging software like SPM and FSL it is not possible to have a
within-subject error term as well as a between-subject error term. The advice I
got from multiple people on the FSL and SPM mailinglists was to just average
the spatial maps over sessions and possibly also twins. However, that would
mean also averaging over the behavioural measure which can actually be quite
different over sessions as well as twins, in this way greatly reducing power.
I was told that it might be possible to analyse these statistical parametric
maps within R with a mixed procedure. Can anyone confirm that this is the case
and perhaps point me in the right direction, e.g. with a publication or other
helpful document, webpage or package?
Cheers,
Diederick
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
