Dear James,
> I was wondering if someone could point me to some information about how FWE correction is implemented in SPM (i.e., when viewing Results, selecting FWE when asked for 'p value adjustment to control'). I'm specifically looking for the kind of details that would be appropriate for a thorough Methods section (like whether the FWE correction is cluster-based). Thanks in advance for any information!
As a quick answer, when selecting FWE and 'p value adjustment to
control', this is a voxelwise correction. The cluster-based FWE
correction is based on the extent of the cluster and is displayed in
the results table (you can limit your display to clusters of a certain
size so that only significant clusters are shown after determining the
cluster size threshold from the results table).
For technical details, the "RFT" (random field theory) section of the
SPM bibliography is a great place to start:
http://www.fil.ion.ucl.ac.uk/spm/doc/biblio/Keyword/RFT.html
and personally I have always found this a helpful article in
understanding random field theory:
Worsley KJ (1996) The geometry of random images. Chance 9:27-39.
Hope this helps!
Best regards,
Jonathan
--
Jonathan Peelle, PhD
Assistant Professor
Department of Otolaryngology
Washington University in St. Louis
Office: (314) 362-9044
http://peellelab.org || http://jonathanpeelle.net
|