Gary Price wrote:
> Dear SPMers,
>
> I am trying to reproduce ROb McClure's model for repeated measures ANOVA in
> VBM-3rd May 2005.
>
> http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind05&L=SPM&P=R204978&I=-3
>
> I have set up a multi group conditions and covariates model with 2 groups
> and 2 conditions.
>
> I have a problem with the contrasts suggested for main group effect (-1 -1 1
> 1) as this contrast does not seem to be allowed.
>
>
> Anyone else experience this problem and how did they get around it?
Have you tried the method in Rik's post (preceding the one you cite)?
http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind05&L=SPM&P=177804
And a question for Rik/Will/anyone: for the main group effect, should
the comparison be on the mean images (as Rik suggests), or on the
difference images (this was the impression I got -- perhaps wrongly --
from correspondence with Will)? Or are both potentially interesting?
*Or* is neither method a good substitute for a multilevel model? I
think this is a very interesting and important discussion for
longitudinal VBM.
Many thanks.
Ged.
|