The Power of Reinforcement

4. THE MYTH THAT “EXTRINSIC REWARDS UNDERMINE INTRINSIC INTEREST”

39 4 THE MYTH THAT “EXTRINSIC REWARDS UNDERMINE INTRINSIC INTEREST” In addition to making behavior that produces reinforcement “valuable behavior,” conditioned, contrived reinforcement programs can and do have many powerful and beneficial effects; however, the use of conditioned reinforcement and contrived reinforcement contingencies to shape and maintain behavior has led to the most damaging and widespread myth regarding reinforcement: The myth that rewarding behavior undermines, or eliminates, any “intrinsic interest” in the behavior (interest in the activity “for its own sake”). The myth, (see the introduction in this book), holds that if I am given contrived reinforcers of any type (money, a game of PingPong with Dad, or even simply praise) for reading, then I will no longer want to read for the sake of reading. I will view reading as only a means to an end, and reading will no longer be a reinforcing activity in and of itself. This myth is most damaging to the effective use of planned reinforcement procedures because it argues that reinforcement will ultimately produce the opposite effect of its intended one. The myth holds that if I want Susan to read more, praising, hugging, or giving her a treat for reading may make her read then and there. But ultimately the reinforcement will make Susan read as little as possible, and she will have an expressed dislike of reading. Hugging her for reading will “undermine her intrinsic motivation” for reading. Likewise, according to the myth, if I pay Sam for bringing home good grades, he will learn to dislike school and will possibly drop out as soon as possible. Every believer in the myth that reward undermines interest is lost to any opportunity to help people through positive reinforcement. Tragically, this myth is part of the dogma of the United States’ educational establishment. “The expectation of reward can actually undermine intrinsic motivation and creativity,” argues a primer by the National Education Association (NEA) under a heading titled “How to Kill Creativity.” “A wide variety of rewards have now been tested” continues the primer, “and everything from good-player awards to marshmallows produces the expected decrements in intrinsic motivation” (Tegano et al., 1991, p. 119). Although hundreds of studies have shown the powerful, beneficial, and lasting effects of reinforcement across a vast range of human situations and behaviors, the NEA and others came to believe the myth that reinforcement undermines intrinsic interest based on a few studies in the 1970s that claimed to demonstrate “detrimental effects of reward” (See Cameron et al., 2001, for the most recent objective review of these studies). Typically in these studies, children or college students were given a simple puzzle or other easy task to do; then they were given a sticker, praise, or trinket after doing the task; later the task was made available to them again. Some results have shown that some subjects engaged in the task less at the end than they did during their initial exposure, and some subjects report that they like the task less. Inferences from these results have snowballed from claiming that in some situations rewards may have “detrimental” effects on task enjoyment to claiming that rewards “turn play into work” to Alfie A. Kohn’s (1993) polemic book Punished by Rewards that argues that all rewards (by which he means reinforcers) in all situations are inherently evil, ineffective, detrimental and should be avoided at almost all costs. What is the truth? Are reinforcement programs the most effective tool for behavior analysts, educators, and psychologists, or are rewards really “punishers” as Kohn argues? (By definition reinforcers can’t be punishers and punishers can’t reinforce. But “rewards” that have nontechnical meaning, can be reinforcers or punishers.) Should rewards be used or avoided? In an attempt to answer these questions, several “meta-analyses” have been conducted on the effects of rewards on intrinsic interest. (Metaanalyses are statistical analyses of the statistical analyses of many studies pooled together.) Not surprisingly, when opponents of the use of rewards have conducted the meta-analysis they have found that rewards have detrimental effects (e.g., Deci et al., 1999). But when proponents of the use of rewards have conducted the meta-analysis, they have found beneficial effects of rewards (e.g., Cameron et al., 2001; Cameron & Pierce, 1994; Eisenberger et al., 1999). (Meta-analyses of the same subject by different investigators may produce different results because the studies, manipulations, control conditions, and other characteristics of the previous studies on which the meta-analyses are based may be...