Dear Experts,
I am running a first level analysis on SPM using Reward Prediction Errors as the parametric modulation parameter.
I have 1 within condition with 2 levels Opponent 1 and Opponent 2 (scans belonging to 2 separate runs).
In my design matrix, the columns are:
1st column – Outcome (at the time reward feedback was received), [Opponent1]
2nd column - Prediction error (Parametric modulator) (at the time reward feedback was received)
3st column – Outcome (at the time reward feedback was received), [Opponent2]
4nd column - Prediction error (Parametric modulator) (at the time reward feedback was received)
Q1) Regarding the polynomial expansion field specification, should I use a 1st order modulation or a higher one? (given the parametric modulator is a reward prediction error)
Q2) Do I have to orthogonalize the prediction error?
I think it is not needed, but I just want to be sure that I am doing it correctly.
Q3) If I want to test the effect difference of Opponent 1 vs Opponent 2, which value should I attribute to the parametric modulator? Could it be correct to indicate the following contrast [0.5 0.5 -0.5 -0.5]?
Best regards,
João Simões
|