Dear Joelle,
One would want to have two separate conditions A and B within the same model then, one accounting for the first five trials and the other one for the last five trials. Alternatively, you could go with a single condition A and a 1st order parametric modulator with the values [-1 -1 -1 -1 -1 1 1 1 1 1].
With parametric modulators you could also go with something like [-4.5 -3.5 -2.5 -1.5 -0.5 0.5 1.5 2.5 3.5 4.5], thus testing for linear changes over trials, not just first half vs. second half. But this depends on the hypotheses.
Best
Helmut
|