[pymvpa] question about cross-subject analysis

[John Magnotti]

On Wed, Jan 18, 2012 at 11:10 AM, Yaroslav Halchenko
<debian at onerussian.com> wrote:
>> This sometimes works surprisingly well and is fairly straightforward.
>
> yeap -- but I thought that the main question behind was on how could we
> account for areas where nice (linear) alignment is difficult so
> that union (pretty much your scenario) of voxels across subjects would
> not be ideal either.
the main issue I'm worried about is that if I flatten and then
vstack() timeseries from multiple participants will there be enough
consistency across subjects to make the columns (voxel at a given
time) meaingful.

John