Rewards bias self-evaluations of ability.

Abstract

How do people learn about their own abilities? Often, people receive rewards that offer information about their performance level. Yet, even when two people perform equivalently on a task, they may receive disparate rewards. In these cases, could rewards still influence self-evaluations of ability? In two behavioral experiments, we asked whether people feel more capable and confident when they receive more rewards, even when their performance is held constant, and they know how they objectively performed. Participants played a perceptual game in which they received trial-by-trial accuracy feedback; a staircase procedure held their objective performance constant. However, participants were assigned to either a high or low-reward condition, which varied the probability of receiving a reward for a correct answer. In Experiment 1 (N = 340), we found evidence that rewards bias overall self-evaluations of ability after the task-particularly estimations of objective accuracy. Next, in Experiment 2 (N = 342), we examined whether reward feedback would inflate participants' trial-by-trial expectations of their own accuracy before each round of the game. Results indicated that participants updated their expectations to a greater extent when a correct response was accompanied with a reward. These findings suggest that rewards enhance how much people integrate accuracy feedback into their dynamic self-beliefs.