Dear Danai,
Just to add to Marta's answer, that you can always check whether your model comparison is valid by enabling the 'Verify data identity' option in the Bayesian Model Selection' tool.
VladimirSent using BlackBerry® from Orange
Date: Fri, 6 May 2011 13:41:19 +0100
Subject: Re: [SPM] Comparing MEG-DCM models with different number of volumes of interest
Dear Danai,
Yes, you can compare models with different number of sources - see http://www.ncbi.nlm.nih.gov/pubmed/19261714
When you do this though, make sure that all your models contain all the sources of the most complete model, and then turn off the connections you don't need (for simpler models).
Hope this helps!
Marta
2011/5/6 DANAY DIMA
<[log in to unmask]>
Hi everyone
I just wanted to check if we can compare directly MEG-DCM models with different number of volumes of interest?
Thanks in advance to anyone that answers!
Best wishes,
Danai
--
Dr. Marta I. Garrido
Research Associate
Wellcome Trust Centre for Neuroimaging
University College London
12 Queen Square
London
WC1N 3BG
United Kingdom
Tel: +44(0)20 7833 7472 (ext: 4366)
Fax: +44(0)20 7813 1420
Email: [log in to unmask]
www: http://www.fil.ion.ucl.ac.uk