Hello SPM mailinglist,
I am doing a VBM study in approximately 350 healthy elderly participants. I segmented all images using the VBM8 toolbox, following the manual to create smoothed modulated for non-linear effects only grey matter volumes (sm0wrp1).
While doing the VBM analysis I saw some strange effects of gender in the population. Attached you will find the results of a two sample T test model comparing males to females with age as a covariate.
I am worried about the strange distribution, especially for males > females; which makes me wonder whether something has gone wrong.
For the analysis I used sm0wrp1 images. Adding TIV as an extra covariate reduced the effects but the strange distribution pattern did not disappear. Making a customized-DARTEL template of the population did not do anything either.
What do you think about these effects? Is this normal?
|