Hello,
it may be the case that the overall FWHM that is a global value is not
matching (i. e. is not corresponding) with the local smoothness.
There is a toolbox written by Satoru Hayasaka that corrects cluster
p-values for local smoothness - one idea would be to try this and see
if under consideration of the smoothness in the cluster area the cluster
is assessed as significant - and is more fitting your impression on the
voxel level.
As far as I know it is not fully released on the SPM page now, so you
probably would have to contact
him directly.
Other, more statistical comments are wellcome of course.
best regards,
Philipp
Max Planck Institute of Psychiatry
NMR Research Group
Kraepelinsr. 2-10
80804 Munich
Mail: [log in to unmask]
Phone: 0049-89-30622-413
|