Hi Nosarti,
> Dear experts,
> Apologies for asking this question again. When I extract values for a
> group difference map (structural MRI) in spm2, what exactly are the
> vaues I obtain for each subject, e.g. the eigenvalues. Could you direct
> me to a reference that explains this ?
>
What you get is the 1st eigenvector across voxels for the different
images you put in.
If you do it per subject, it will be plotted along the scans. If you do
it at the second level, i.e. across subject (like VBM), it will plot the
1st eigenvector across subjects. (for VBM the value being roughly the
mean volume across selected voxels per subject - you can also plot the
mean to see if you have the same result, if all voxels show the same
variability mean = 1st eigenvariate)
You can have a look at the VOI extraction in spm_regions.m which goes like
1 st get the locations
2nd get the voxel coordinates (choosing cluster, sphere, box) and
corresponding values
3rd perform an SVD and returns the 1st eigenvector across voxels for
each image
note that in SPM5 it also allows you to choose the contrast you want and
returns the % of variance explained (cool :-) )
Hope this helps
Cyril
|