Sometimes rewards are not rewarding. Although it may be counter intuitive, sometimes rewarding a person for an inherently interesting activity actually results in decreased levels of interest in the very activity that the reward exists to encourage. Some researchers attribute this post-reward lack of interest to decreased intrinsic motivation while others hold behavioral explanations. While various researchers disagree as to exactly why the phenomenon exists, few dispute that it does.
Rewards that have a detrimental effect on interest in an activity are typically controlling in nature and studies demonstrating the effect include: Anderson, Manoogian, Reznick, 1976; Deci, 1971; 1972; Deci, Koestner, & Ryan, 1999; Fabes et al. , 1989; Garbarino, 1975; Greene & Lepper, 1974; Green, Lepper, & Sternberg, 1976; Harackiewicz, 1979; Kruglanski, Friedman & Zeevi, 1971; Lepper & Greene, 1973; 1975; Loveland & Ollely, 1979; McCullers, Fabes, & Moran III, 1987; McGraw & Fiabla, 1982.
A reward often ends up decreasing a person’s interest in the very behavior the reward exists to encourage. Conversely, when informational reward (i. e. praise) is used, it generally increases internal motivation which is demonstrated by increased interest in the praise activity. This experiment is concerned with the topic of motivation, specifically extrinsic motivation and how the introduction of the former might affect the later. Intrinsic motivation supposedly exists within a person.
According to Deci, Benware, and Landy (1974) “A person is intrinsically motivated to engage in a behavior if he does it because the behavior itself is rewarding. ” On the other hand, extrinsic motivation originates outside of a person; Deci et al. continues: “… a person is extrinsically motivated if he performs the behavior in order to achieve some external reward. ” pg. 652. There is a great deal of research showing that informational rewards seemingly enhance intrinsic motivation resulting in raised levels of interest in an inherently interesting activity.
Verbal informational rewards such as praise almost always resulted in increased interest in an activity (Anderson, Manoogian, ; Reznick, 1976; Cameron ; Pierce, 1994; Deci, 1972; Deci, Ryan, ; Koestner, 1999; Rosenfield, Folger, ; Adelman, 1980) while tangible informational rewards tend to have less of a negative impact on motivation than non-informational tangible rewards(controlling rewards).
In an experiment done by Karniol and Ross ( ) children age four to nine were assigned to one of three conditions: (1) a performance-relevant condition in which they were rewarded with marshmallows according to how well they played a game; (2) a performance irrelevant condition in which the number of marshmallows awarded had no relation to the quality of their performance; and (3) a no-reward condition which is self-explanatory.
They found that subjects who performed well and received performance-irrelevant rewards showed decreased interest in the activity during a subsequent free-play period as compared with subjects who received performance-relevant rewards or no rewards. The exception to the rule (that extrinsic rewards tend to undermine intrinsic motivation) seems to lie in multi trial experiments where the reward becomes a reinforcer over time. Riess and Sushinsky (19 ) conducted two experiments, the first of which confirmed a single trial promised reward undermining intrinsic motivation.
However, in the second experiment he demonstrated that a multi-trial contingent reinforcement procedure (much like procedures used in behavioral modification situations) actually increased subsequent interest in the target activity. In a review of the usefulness of token reinforcement programs in the classroom, O’Leary and Drabman (1979) claims that in spite of many methodological considerations in the implementation of reinforcement programs, token reinforcement programs are very powerful and they enjoy generally positive results.
In “Tangible Reinforcers: Bonuses or Bribes”, O’Leary, Poulos and Devine (1972) warn that tangible extrinsic reinforcers should only be employed after other natural reinforcers have been tried. They continue: “However, where other methods fail, a token reinforcement programs may prove very valuable, and if tangible reinforcers are used as primers of behavior, they may prove a bonus to mental health personnel who find their clients “unmotivated”, “lazy”, or “resistant to treatment. ” pg. 7.
An finally, in First Course in Applied Behavior Analysis, Paul Chance summarizes a study done by Donald Dickson in which the academic and social performance of fifty eighth-graders who had gone through a reward program in the fifth and sixth grades was examined. What Dickinson found was that those students who had once earned rewards by behaving well and working hard had made greater progress in the two years following the programs (that is, when the rewards were no longer available) than had students who had never received rewards.
But, there seems to be an exception even to the exception. Greene, Sternberg, and Lepper (1976) found that even with multiple-trial contingent reinforcement procedures, intrinsic motivation was undermined once neither social nor tangible rewards were no longer available. In an effort to organize and make sense of all the research and theories surrounding the topic of motivation, several meta-analyses have been done. Unfortunately even the meta-analysis cannot seem to agree on the effects of extrinsic rewards on motivation.
According to Deci, Ryan, and Koestner (1999), a meta-analysis done by Rummel and Feinberg (1988), a meta-analysis done by Wiersma (1992) and a meta-analysis done by Tang and Hall (1995) all found that extrinsic rewards did have an undermining effect on intrinsic motivation. Deci, Ryan, and Koestner’s (1999) meta-analysis demonstrated that controlling (tangible) rewards very significantly undermined free-choice intrinsic motivation while positive feedback enhanced it.
On the other hand, a meta-analysis conducted by Cameron and Pierce (1994) showed that “overall, reward does not decrease intrinsic motivation. ” p. 363. They also found that praise did produce an increase in intrinsic motivation. According to their study, the only situation that resulted in a slight detrimental effect on intrinsic motivation was when rewards were given to subjects simply for doing a task. Their analysis of within-subjects designs indicated that “reinforcement does not harm and individuals’ intrinsic motivation. ”
Before we proceed to explore some theoretical explanations for some of the effects of rewards on intrinsic motivation, a working definition of intrinsic motivation may be in order. For most cognitive researchers intrinsic motivation, or IM, is simply a desire to engage in some kind of activity for its own sake. Intrinsic motivation (defined negatively) is present when an individual does something without the expectation of a reward. If I play a game without expectation of reward, I would be said to be intrinsically motivated or the game would be said to be intrinsically motivating.
This negative definition is convenient for conducting experiments and dependent measure which usually involves timing how long a person engages in activity without the expectation of reward ( Kohn, ). Not all researchers, however, accept the idea of IM. Operant psychologist believe that all behavior is motivated by external stimuli and they often claim that if something looks intrinsic, it is only because we have not yet figured out what the real (extrinsic) causes are (Kohn, p.
270). For the purpose of this study, intrinsic motivation will be defined by the traditional way (negatively ) but I leave the proof of its existence to those far more philosophical. I intend to examine the effects of informational verses controlling rewards on future behavior. Whereas I may speculate on the effects of rewards on intrinsic motivations as measured by a behavior, I will go no father.