I have the following experimental paradigm:

1. patient and healthy control groups
2. each subject has two separate scanning sessions on 2 days
3. on each day subject has fMRI task followed by a)active substance OR b)placebo 
administration and repeat fMRI task

I wanted to confirm proper design matrix construction to test the question of whether 
activation deltas (post - pre substance administration) differ among group A and B

My thought is to use a variation of the Multi-Session & Multi-Subject (Repeated Measures 
- Three Level Analysis) scenario on the FEAT webpage but instead of running the initial 
fixed effects analysis on subject mean COPEs, run it instead on COPEs generated by a 
fixed effect analysis of before and after condition for each subject.

Not sure if this is correct or how to design the matrix. Any advice would be much