Dear Michael,
> When focusing in on a small region of the brain, as one does with a
> small volume correction, what threshold is an accepted standard for
> the SPM (in a standard t-contrast)? That is, should I threshold the
> whole map at 0.001 uncorrected, then focus in on my ROI w/ an SVC, or
> do I threshold the whole map at an even lower uncorrected threshold
> (eg 0.01), followed up by an SVC?
There is no such standard. The point about the SVC is that you have
prior knowledge about an activation effect. This can be given by any
source of information, e.g. a hypothesis derived from another
experiment. I'm not sure about the approach Richard Perry talked about
earlier on the SPM helpline
(http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind0012&L=spm&F=&S=&P=4368).
I guess Richard's approach is ok, when the two resulting t-maps are
virtually independent (which is true for very high dfs and orthogonal
contrasts). In other words, one could analyze the two contrasts in two
separate analyses and would still find roughly the same results. What
one definitely shouldn't do is to use information from the same
contrast/t-map.
Stefan
--
Stefan Kiebel
Functional Imaging Laboratory
Wellcome Dept. of Cognitive Neurology
12 Queen Square
WC1N 3BG London, UK
Tel.: +44-(0)20-7833-7478
FAX : -7813-1420
email: [log in to unmask]
|