Overjustification, competing responses, and the acquisition of intrinsic interest.

Abstract

The overjustification hypothesis proposes that expectation of reward for an inherently interesting activity produces less interest in the activity when reward is subsequently unavailable. Two experiments designed to test this hypothesis are reported. Experiment 1 replicated previous findings that a single trial of noncontingent, promised reward undermines young children's interest in the rewarded play activity. Experiment 2 discontinued the overjustification hypothesis by demonstrating that a multiple-trial, contingent reinforcement procedure did not produce an overjustification effect. Both of these findings are understandable in terms of a competing response hypothesis but not in terms of overjustification. Accordingly, previous findings of an overjustification effect are reinterpreted as having demonstrated that curiosity in a novel play activity can be undermined by presentation of distracting stimuli. The following additional conclusions are drawn: (a) The overjustification hypothesis is conceptually vague, (b) self-perception theory seems to have been concerned with variables relatively unimportant for understanding effects of tangible reward procedures on overt behavior, and (c) recent criticisms that token economies undermine intrinsic interest were based on a failure to appreciate differences between single-trial, noncontingent reward and multiple-trial, contingent reinforcement procedures.Several theorists have proposed that tangible reward procedures undermine intrinsic interest in rewarded activities (e.g., Calder

Extrinsic and intrinsic motivation at 30: Unresolved scientific issues

“…Reiss and Sushinsky (1975), for example, used "extrinsic" reinforcement to teach young schoolchildren to listen to an experimental song. When the children were posttested 2 days later in a different room with different people-the objective criteria Lepper et al (1973) stated were needed for inferring that posttest behavior is intrinsically motivatedthe children listened to the previously rewarded song and, thus, showed an enhancement effect of reward rather than an undermining effect.…”

“…Because the Smith and Pittman (1978) study showed that the first 10 trials of extrinsic reward might produce undermining effects caused by distraction, and because Reiss and Sushinsky (1975) also showed distraction effects, undermining studies with fewer than 10 trials of reward should control for distraction. Undermining effects observed in studies with a single trial of reward may be distraction effects that have nothing to do with cognitive evaluation.…”

“…After the publication of the Smith and Pittman (1978) study, undermining theorists continued to study the effects of novel rewards rather than multiple trials of reward, even though both Reiss and Sushinsky (1975) and Smith and Pittman showed distraction effects. Even if Smith and Pittman had demonstrated a valid undermining effect in the 50-trial reward condition-and had eliminated the possibility that symbolic negative feedback played a role-subsequent studies that used only one or two trials of reward still would have needed to control for distraction.…”

“…If we exclude studies that may have been biased by floor effects, as Deci et al suggested, we also should exclude studies that may have been biased by ceiling effects. Reiss and Sushinsky's (1975) Experiment 2 was among the few studies in this literature in which the methods successfully avoided both ceiling and floor effects. As noted already, the results of this experiment showed an enhancement effect.…”

“…The college athlete competes weekly or more; the professional athlete is paid 12 to 52 times each year. Because we know that single-trial rewards often do not have the same effects as when the same rewards are given repeatedly over a significant period of time (Reiss & Sushinsky, 1975;Smith & Pittman, 1978), Deci and Ryan (1985) and Kohn (1993) made a leap in reasoning when they implied that the results of singletrial reward studies have practical implications for education and athletics. Sushinsky (1975, 1976) repeatedly criticized the use of singletrial reward procedures to demonstrate undermining.…”