[pymvpa] spatial normalization for MVPA - is it good or bad?

Matthew Brett matthew.brett at gmail.com
Tue May 6 18:57:56 UTC 2014


Hi,

On Tue, May 6, 2014 at 11:51 AM, Vadim Axel <axel.vadim at gmail.com> wrote:
> Interesting, thanks!
> So, in "Realign" I select only "Estimate". And in this stage nothing
> actually happens and SPM keeps it sort "in mind". I guess it should not be a
> problem to chain the estimates? I mean currently I have after realign after
> co-register with "Estimate" So, I will do Realign:Estimate,
> Co-register:Estimate and Normalization. Makes sense?

Yes, that's what I do normally.  Don't forget to apply the
coregistration estimates to your time points as well as your
structural scans though (using the 'other images') option.

The transformations do get chained because they are just affines, and
they get chained by multiplication...

Cheers,

Matthew



More information about the Pkg-ExpPsy-PyMVPA mailing list