Dear spm-users,
Let me try to clarify my previous post...
I would like to run a 2x2 ANOVA to study the effects of an intervention (Brainstim) on rs-fMRI.
Short overview of my data:
- 4 rs-fMRI datasets per subject
- 2 types of brainstim (sham or active, factor 1)
- 2 time points (pre- and post-intervention, factor 2)
For every subject, I ran 2 single level analyses comparing rs-fMRI data before vs after brainstim (so separately for sham and active). Every analysis gives me the beta images and the contrast ([-1 1] for after vs pre stim).
Then I would like to investigate effects of both factors (so stimulation type (2 levels sham and active) and time (2 levels pre and post) on the group level. For this I am using a flexible factorial design and I am loading the 4 beta images from every subject.
If I understand correctly, the designs should be:
- [1 1 1 1] to see overall effects
- [1 1 -1 -1] and [-1 1 -1 1] for the main effects
- [1 -1 -1 1] for my interaction between factor 1 and 2.
I guess I have this all up and running. However, I also have information about the strength of the stimulation for every subject and would like to incorporate this in my model as a between-subject factor. Can anyone guide me on how I can do this? I assume I should add a third factor to my flexible factorial design. Where can I put the values for the between-subject factor? Should they be in the model as covariate? Is this the right way to investigate the 3-way interaction between 3 factors (type of stim, time, and stimulation strength)?
Thank you very much for you help.
Best,
Debby
|