Bon jour Lucile,
Off the top of my head it sounds like you are doing a group analysis
but are using the within subject model specification. The within
subject analysis (first level) expects time series data and I am
pretty sure the Vi is the autocorrelation estimation part. So if you
use one image per subject to do a group analysis you need to use the
second level modeling and not the first level. Please make sure that
you are using second level modeling and if so, then I am wrong and am
sorry to waste your time.
Best,
Jason
On Wed, Feb 27, 2013 at 4:29 PM, Lucile Rapin <[log in to unmask]> wrote:
> Hi SPMers,
>
> I m occuring some trouble when i run a model estimation for a one-sample t test. It happens for 2 groups separately.
> This is what i get in Matlab:
>
> Initialising parameters : ...done
> Plane 46/46 , block 1/1 : ...done
> Temporal non-sphericity (over voxels) : ...ReML estimation
> Failed 'Model estimation'
> Reference to non-existent field 'Vi'.
> In file "/Applications/spm8/spm_spm.m" (v4191), function "spm_spm" at line 868.
> In file "/Applications/spm8/config/spm_run_fmri_est.m" (v4022), function "spm_run_fmri_est" at line 33.
>
> The following modules did not run:
> Failed: Model estimation
>
> On the odd part is that when i run two-sample t-test with these 2 groups involved, the model estimation runs fine.
> Can anyone help me understand the error and find a way to fix it?
>
> Thanks a lot,
>
> Lucile Rapin, Ph.D
> Post-Doctorate Researcher,
> UQAM
|