Prospective Control by Reinforcers in Human Choice: Misallocation and Bias, but Not Time Orientation, Influences Reinforcer Control

Abstract

Research with nonhuman animals and children suggests reinforcer control is prospective; that is, behaviour is organised according to the likely future, as extrapolated from past experience. The present experiment extended this work to adult choice, and also explored the effects of individual differences in past, present, and future time-orientation on reinforcer control. Participants (N = 163) responded in a concurrent schedule in which the probability of a reinforcer at the same location as the previous reinforcer varied across conditions. Choice tracked these probabilities, albeit imperfectly, indicating that reinforcers controlled behaviour prospectively. Deviations in choice from reinforcer probabilities were well-captured by a quantitative model assuming that such deviations arise because reinforcers are misallocated to the wrong alternative and because of biases towards one alternative. This replicates previous findings in pigeons and children, hence demonstrating the cross-species generality and developmental continuity of prospective reinforcer control. Individual differences in time orientation appeared not to influence reinforcer control, although further work is needed to explore the conditions under which time orientation modulates prospective reinforcer control.