Hi,
it seems some intermediate checks should be made on this data set, that may
include:
- import/first segmentaton successful (orientation etc.)
- template generations 1 to 6 looking okay?
- all individual results of warpings looking okay in terms of position?
The initial threshold (analysis threshold) does not affect multiple test
procedures per se, but it of
course have an influence on the mass of voxels analysed. The mask should be
checked; if the mask.img is
plausible, i. e. covering the areas you want to test, then still a result
in which "all is showing up"
is not convinving.
You then state "FWE corrected at p>0.0001" -
If that indeed is the corrected threshold then in a VBM analysis in 99.5%
of cases NOTHING should show up;
also a VBM analysis that is robust to FWEcorr<0.05 is a generally a very
strong result, so it is quite
sure that something in the pipeline has gone wrong.
If data are modulated, whole brain / skull size effects are contained, of
course. Think of using appropriate
covariates to correct for that. Other covariates such as age and gender
might depend on specific features
of your study and question.
Hope this helps,
Philipp
Max Planck Institute of Psychiatry
NMR Research Group
Kraepelinstr. 2-10
80804 Munich
Mail: [log in to unmask]
Phone: 0049-89-30622-413
|