Fig. 4: Prediction performance of optimal linear model (Corrected reward ~ Mixing weight + WM load + Intelligence). | npj Science of Learning