Peter -
That's an interesting question. I guess it would be ok to use fMRI priors for just those subjects whose MEEG model evidence suggests they help (particularly if this reflects different quality fMRI data for different subjects). If you want to do inference on the resulting parameters (source strengths), it might be unconventional to have used different models (ie sets of priors) for different subjects, but it is not obvious to me how this could bias any such inference over the group of subjects (particularly if comparing parameters of two or more conditions inverted with same model).
BW,R
---------------------------------------------------------------
Professor Richard Henson
Assistant Director for Neuroimaging
MRC Cognition & Brain Sciences Unit
15 Chaucer Road
Cambridge, CB2 7EF
England
EMAIL: [log in to unmask]
URL: http://www.mrc-cbu.cam.ac.uk/people/rik.henson/personal
TEL +44 (0)1223 355 294 x501
FAX +44 (0)1223 359 062
MOB +44 (0)794 1377 345
---------------------------------------------------------------
-----Original Message-----
From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]] On Behalf Of Peter Goodin
Sent: 20 March 2013 01:54
To: [log in to unmask]
Subject: [SPM] Use of source priors on some, but not all participants?
Hi SPM list (and apologies if this is a double post)
I'm doing source localisation on my MEG data (MSP currently as it gives better log evidence compared to IID and COH) and have fMRI data that can be used as a prior. I've noticed that for some participants when using a prior the log evidence increases quite dramatically, but for others (co-incidentally enough those who show less activation in fMRI...), the evidence decreases and can be worse than simply using MSP alone.
My question is, it "acceptable" to use priors for some participants but not others?
Thanks for any help,
Peter.
|