The fun and function of uncertainty: Uncertain incentives reinforce repetition decisions

Abstract

This research studies repetition decisions—namely, whether to repeat a behavior (e.g., a purchase) after receiving an incentive (e.g., a discount). Can uncertainty drive repetition? Four experiments, all involving real consequences for each individual participant, document a counterintuitive reinforcing-uncertainty effect: individuals repeat a behavior more if its incentive is uncertain than if it is certain, even when the certain incentive is financially better. This effect is robust; it holds in both lab and field settings and at both small and large magnitudes. Furthermore, the experiments identify two theory-driven boundary conditions for the reinforcing-uncertainty effect: the effect arises (a) only if the uncertainty is resolved immediately and not if the resolution of uncertainty is delayed, and (b) only after, not before, one has engaged in repetitions. These results support a resolution-as-reward account and cast doubt on other explanations such as reference-dependent preferences. This research reveals the hidden value of uncertain incentives and sheds light on the delicate relationship between incentive uncertainty and repetition decisions.