It is not possible to do this kind of statistics with a single image
per condition. You should either enter all your single trials into the
design rather than averages or do your analysis across subjects.
On Wed, Nov 11, 2009 at 6:10 PM, Kambiz Tavabi <[log in to unmask]> wrote:
> Dear SPMers,
> I am trying to conduct a sensor space analysis of my MEG data using a 2x2x2
> factorial design. Attached are SPM8 outputs for design/cell specifications
> for single subject using the 'specify 2nd level' procedure dialog. I receive
> the following: 'Warning: Please check your data: There are no significant
> voxels' when attempting to estimate the model parameters, and cannot proceed
> with contrasts.
> I am essentially adapting what is in the SPM8 manual for conducting t-tests
> in MEG sensor space to the factorial approach. I wonder if the procedure is
> correct for single subject analysis? At what point does the data for all my
> subject enter into the analysis?