Dear William,
Indeed, spm_mfx.m is written in a way it assumes that first level models
are from fMRI data.
It might be possible to relax this assumption but, beforehand, could you
let us know whether you specified one model per subject or if you
already have a large fixed-effect model (it might help to see the design
matrix). Also, what is your rationale for moving away from the summary
statistics approach?
Best regards,
Guillaume.
On 16/03/15 13:17, W.R.Jones wrote:
> Dear all,
>
> I was wondering if it is currently possible to perform the mixed effects
> analysis that is available for fMRI on M/EEG images files? In
> particular, the functions that are called seem to use parameters that
> are specific to fMRI data - and I can't really see a way around this at
> the moment.
>
> Best,
> William
--
Guillaume Flandin, PhD
Wellcome Trust Centre for Neuroimaging
University College London
12 Queen Square
London WC1N 3BG
|