[pymvpa] A couple of questions from a new MVPA user

Raúl Hernández raul at lafuentelab.org
Wed Apr 8 18:37:08 UTC 2015

Hi, I’m new on MVPA so I’m not sure if my interpretation of the results I’m
getting is accurate, I would like to ask you for a little comment just to
make sure that I’m on the right track.

I have an experiment in which a subject is presented with different
stimulus corresponding to 10 different categories. Each category has
different properties which allow the subject to distinguish between them.

As I don’t know where in the brain are this categories distinguish from
each other, I created several mask’s using the brain regions defined on the
Harvard MNI atlas. So I tested my data on each mask getting an estimate of
how good is the classifier using each mask.

My results show that there are around 5-6 masks that can classify my data
on greater than chance levels (estimated using the Monte Carlo method
described in the tutorial).

My questions are,

Am I following a correct logic in the analysis?

Is it correct for me to conclude that those areas are involved in
processing the categories?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/pkg-exppsy-pymvpa/attachments/20150408/3dd987e7/attachment.html>

More information about the Pkg-ExpPsy-PyMVPA mailing list