Dear SPMers,
I have removed the constant term in a two-sample random-effects analysis
(contrasting two groups) in order to be able to look at conjunctions.
(in accordance with eg Karls mail:
http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind0104&L=spm&P=R8273&I=-1)
Now, when I plot the parameter estimates for a certain region, the
values do not sum up to zero, as is the case when a constant term is
present. I understand that, since the effect of adding a constant term
is that the parameter estimates are scaled to be summing up to zero. My
question is: without the constant term, do the parameter estimates
reflect "true" (de)-activations with respect to the baseline? In other
words, if the parameter estimates for some group or condition are
negative, can one then say that this region is de-activated (with
respect to baseline) for this group/region?
Thanks for any help!
Kind regards,
Floris de Lange
--
Floris de Lange
FC Donders centre for cognitive neuroimaging
PO Box 9101
6500 HB Nijmegen, NL
e-mail: [log in to unmask]
web: www.fcdonders.kun.nl/F.deLange
|