Actor-critic temporal-difference routines in Matlab to simulate activity-contingent reinforcement findings
