Figure 2.

Basic effects of reward magnitude and probability on choice. Alternatives were rewarded with different magnitudes or probabilities over the course of 50 trials as indicated. (a) The alternative with the greater magnitude of reward is exclusively preferred and larger differences in magnitude lead to faster convergence. (b) The alternative with the greater probability of reward is favoured at levels above probability matching and larger differences in probability lead to faster learning (reward r = 4 for both options). Learning curves are averages over 50 simulation runs. (Online version in colour.)