Dear Anderson,
Thanks for your input, it was a great help to set up the design specific for our study. Looking at the results, I think I need to bother you one last time with two follow-up questions:
1. How would I add a covariate (namely the latency of scan) to this model? Is it possible to simply append the demeaned latency as two additional EVs (one per group) to the described design matrix?
2. If I add a covariate, is it also possible to look interactions and compare the slopes using a similar approach as described in the "Two groups with continuous covariate interaction" example in the Glm user guide (https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/GLM#Two_Groups_with_continuous_covariate_interaction)? Would the contrast matrix in the mixed model simply be analog to the example in the user guide?
Thank you again very much for your help with this!
Best wishes,
Bastian
|