 |
 |
Dear SPM experts, I have a T1 dataset of patients at 3 time points and 2 groups. I ran a longitudinal voxel-based morphometry analysis and a flexible factorial model to examine groupxtime interactions. In this analysis, I found some significant effects of groupxtime for gray and white matter anatomical regions that I was expecting. Now my question is whether the effects in these regions can be explained by changes in the scores of a clinical questionnaire across time points. However, I´m not sure of which model to build in order to test for this. Two ideas come to mind: 1) Set up the same flexible factorial but including a covariate with the questionnaire's scores at the 3 time points (centering). But in this case, I would just be testing whether the effect holds once the variance from the questionnaire is allocated to a separate regressor, right? 2) Set up a multiple regression analysis. In this case, since I have 2 groups and 3 time points, how can this be taken account into a single model? As separate covariates? Maybe I should build a separate model for each time point and specify 2 covariates of interest (questionnaire and group)? Any help will be greatly appreciated. Thank you in advance. Best, Noelia