Hi Shan,
> If I understand correctly that a conjunction analysis on individual
> subjects is to test whether the same regions are activated in most
> subjects,
... not quite right but not quite wrong either :) Strictly speaking, a
conjunction shows regions where all subjects activated significantly.
There are variations to this, though, among them being the one you
described. The strict way was not the way that it was implemented
originally within spm, though, and a couple of papers have addressed
this issue in the April, 2005 issue of NeuroImage.
> however I am not sure how the conjunction analysis can be
> performed in SPM. Is the design matrix in conjunction analysis the same
> as in fixed effect analysis?
Yes. The simple, albeit not very intuitive (until you tried :) way to do
his is to select more than one contrast in the contrast manager by
holding down the CTRL key (I think). You will then be asked which type
of conjunction you want to perform. For this, it is a good idea to read
the above papers or to search the archives (when searching for
"conjunction", I stopped counting beyond 250 :)
Best,
Marko
--
=====================================================================
Marko Wilke (Dr.med./M.D.)
[log in to unmask]
Universitäts-Kinderklinik University Children's Hospital
Abt. III (Neuropädiatrie) Dept. III (Pediatric neurology)
Hoppe-Seyler-Str. 1, D - 72076 Tübingen
Tel.: (+49) 07071 29-83416 Fax: (+49) 07071 29-5473
=====================================================================
|