Dear Andrew,
I would like to know which of the recipies from the random effects kit to
use for the following PET analysis. I am working with 16 subjects in which
we have replicated three conditions twice in each person; AABBCC presented
in random order. I am interesed in BB-AA and CC-AA, where AA is the
control condition. I am planning to use the random effects kit to generate
48 adjusted mean images (from 96 raw images) and follow that by doing two
separate analyses using multi-subject:different conditions in SPM95.
Because we don't have absolute data, I want to do global normalization
across all scans irrespective of subject or condition. Since it looks like
the ANCOVA options in the random effects kit covary the globals within
subject or condition, I was planning to use the "straight mean" recipe
(specifying 48 blocks) and follow with ANCOVA normalization in SPM. Is
this the best way to proceed?
How's SPM98 coming along? We eagerly anticipate its release.
Thanks for your help,
Brenda
Brenda Benson
Biological Psychiatry Branch, NIMH, NIH
Bethesda, MD 20892
Phone: (301)496-6825
Fax: (301)402-0052
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|