Dear Varun,
the designs look fine. Given the scans are properly assigned to the correct subject and cell of the designs there is no obvious error in the specification. However, the interaction contrast would be a one-line F- or t-contrast like [1 -1 -1 1]. for an interaction between two groups. If you want to test any interaction between three groups, you would need an F-contrast matrix like [1 -1 -1 1 0 0; 1 -1 0 0 -1 1].
Hope this helps
Volkmar
|