Dear SPM experts,
I have read the discussion about global scaling and global normalisation
in fMRI on the list.
My gist of it was that there is no need to do global scaling on either the
first or second level analysis.
My question is, what about global normalisation ?
Can I still use the Global normalisation -> ANCOVA option
on the second level of an fMRI analysis ? Is it not recommended or does it
lead to false results ?
And does the conclusion of Aguirre et al., (1998) still hold
that
"This implies that regions that
evince a positive relationship with a task covariate in
the absence of a global signal covariate will not necessarily
evince a positive relationship with that task covariate
in the presence of a global signal covariate. Furthermore,
when a global signal covariate is present, it is
inappropriate to automatically infer that regions with a
negative relationship with the task have decreased
signal values during control compared to experimental
periods (for example), as this statistical circumstance
may arise if the signal in a voxel is simply less
correlated with the task than is the global signal itself."
Am I correct in concluding, that including the global signal as a
covariate using the ANCOVA option on the second level is more conservative
in terms of positive task effects (less sensitive, but more specific) but
might give rise to negative task effects that are artefacts ?
Thanks for any help on this ....
Best,
Roland
Roland Zahn, Dr. med.
NIH / NINDS
Cognitive Neuroscience Section
Building 10, 5C206
10 Center Drive, MSC 1440
Bethesda, MD 20892-1440
Tel.: (+1)-301-402-6392
Fax.: (+1)-301-480-2909
|