While research that demonstrates a reward not reinforcing a behavior seems to contradict the tenets of behaviorism, operant psychologist have their own explanations for the phenomenon, perhaps the most compelling of which belongs to Eisenberger and Cameron (19 ). They claim that some post-reward decrements in behavior can be explained in part by Seligman’s learned helplessness theory. Eisenberger and Cameron suggested that “when people receive rewards that are not dependent on the quality of their performance… they learn that they cannot control the receipt of rewards, so they become helpless…
Further… when rewards are dependent on the quality of the performance… the rewards facilitate learned industriousness: (Deci, 1999). This model would explain the research demonstrating differences in post reward behavior for informational verses controlling rewards. An informational reward would facilitate learned industriousness, which would then result in increased levels of the rewarded behavior. A controlling reward, one given just for participation regardless of quality of performance, would facilitate learned helplessness resulting in decreased levels of the rewarded behavior.
This is only one among several possible behavioral explanations for the data being accumulated on the subject of rewards. Operant psychologists have also criticized the methodology of much of the research on this topic. Carton (1996) accounts for the data by pointing to three confounds that taint the research done on motivation. The first confound is “temporal contiguity”. Typical research has involved at least two groups, one promised and then given a tangible reward and one that is given praise as the reward.
However, in the praise conditions(informational reward), the praise is delivered immediately while in the tangible reward condition(controlling reward), the tangible reward is generally given after the post-treatment session. Operant researchers would predict the same outcome (decrease in interest following reward and increase in interest following praise) based solely on temporal contiguity. When someone is rewarded quickly, the behavior is more likely to increase-it does not necessarily have anything to do with extrinsic or intrinsic motivation, locus of causality, or perceived competence.
Another confound that Carton addresses is the number of reward administrations. In most studies, the tangible reward was given only once while praise was given repeatedly. Once again operant psychologist have simpler explanation for the results. They would likely explain the results merely reflect more reinforcement as having had more of an effect of behavior. The last confound that Carton addresses involves the presence of discriminative stimuli in studies. For example, tangible reward subjects were told specifically before the post-treatment sessions that they would not be rewarded.
Carton explains that tangible rewards were associated with clear stimuli that signaled their availability. The presence of these stimuli most likely decreased the chance that the behavior in the treatment session would generalize to the post-treatment session. On the other hand, in the praise condition, the lack of stimuli being present to signal availability of the reward increased the likelihood that the target activity would generalize form treatment to post-treatment conditions.
Carton also claims that the operant theory has several advantages over the other theories surrounding motivation. First, he claims that operant explanations are more parsimonious. Second, he claims that operant theories are falsifiable. Thirdly, he claims that operant explanations work best in applied settings, and finally, he claims that operant explanations do not depend on definitional distinctions between “intrinsic” and “extrinsic” motivation A Brief Summary
To summarize briefly, a plethora of research demonstrates a phenomena whereby praise (informational rewards) generally cause increases in interest in a rewarded behavior while tangible rewards (controlling rewards) generally cause decrease in interest for the rewarded behavior. Cognitive Evaluation Theory explains that when rewards are controlling, a person feels less autonomous which results in decreased motivation. However, when rear4ds are informational they enhance intrinsic motivation because this provides a person with satisfaction of their need for competency.
Self-perception theory’s “overjustification” hypothesis has another explanation for this phenomenon: a controlling reward given conditionally upon completion of a task, regardless of quality of performance would cause the person rewarded to perceive the task as only a means to an end, thereby stripping it of any inherent interest. An unexpected informational reward would not have a detrimental effect relative to the controlling reward because it would never effectively become a bribe. And finally, operant psychologists such as Esenberger and Cameron have another explanation for this phenomenon.
They claim that controlling rewards (when given regardless of quality of performance) would facilitate learned helplessness which would result in decreased levels of behavior while informational rewards would facilitate learned industriousness resulting in increased levels of the behavior. In essence, when either controlling (tangible rewards) or informational reward (praise) is introduced into a situation the three theories presented here make basically the same behavioral predictions.