The Effects of Contingent and Noncontingent Rewards and Controls on Intrinsic Motivation
All of the theories of work motivation which consider both kinds of rewards assume that the effects of the two are additive. This paper examines that assumption by reviewing a program of research which investigated the effects of external rewards and controls on intrinsic motivation:
Are piece-rate payments or other extrinsic reward systems which tie rewards (especially money) to performance cornpatible with participative management, which focuses on intrinsic motivation? That is, will a person’s intrinsic motivation to do a job remain unaffected by external rewards?
It was reported that a person’s intrinsic motivation to perform an activity decreased when he received contingent monetary payments, threats of punishment for poor performance, or negative feedback about his performance. Noncontingent monetary payments left intrinsic motivation unchanged, and verbal reinforcements appeared to enhance intrinsic motivation
Optimal amount of positive verbal feedback?
How else can intrinsic motivation be increased?
Is extrinsic motivation more powerful than intrinsic motivation (ie. Is sacrificing intrinsic motivation by using extrinsic motivation a worth-while choice)?
Insights, lessons learnt:
Punishment -> reduces intrinsic motivation
Reward contingent on performance -> reduces intrinsic motivation
Reward regardless of performance -> no effect on intrinsic motivation
Positive verbal feedback -> increases intrinsic motivation (effect may decrease or reverse if used too frequently)
piece-rate payments (or wage incentives) have been commonly used for motivating employees. Such systems (which date back to preindustrial times) tie a person’s financial rewards directly to his performance by paying him a set rate for each unit of output which he produces. The motivational assumption underlying piece-rate payments–as well as sales commissions, bonus plans, etc.–is that a person will perform effectively to the extent that his rewards are made contingent upon effective performance (cf. Vroom & Deci, 1970).
This approach to employee motivation utilizes externally mediated rewards, i.e., rewards administered by someone other than the employee himself. In so doing, the management is attempting to control the employee’s behavior so he will do what he is told, in the way and in the amount that he is told
More recent approaches to management (e.g., Likert, 1961; Argyris, 1957; McGregor, 1960) have assumed that man can be intrinsically motivated to perform effectively; that is, they assume that individuals can be motivated by the job itself and can derive satisfaction from doing the job well. These approaches focus on higher-order needs where the rewards are mediated by the person himself, and they stress the importance of getting the worker ego-involved in his work and committed to doing it well.
Employee participation in decision making is stressed as a means of getting employees more ego-involved. The employees are given a voice in decisions which affect them, and they are given greater latitude in the way they do their jobs. There is less reliance on authority as a control mechanism, and employees are judged by their results. These newer theories also suggest that jobs should be enlarged or enriched so as to be more challenging
Comparisons were then made between the experimental and control groups on the amount of free choice time they spent working on the puzzles in the first and third sessions. The results indicated that there was a significant decrease in the experimental subjects’ intrinsic motivation from sessions I to III relative to the controls’ intrinsic motivation. In other words, the experimental subjects had lost intrinsic motivation (i.e., they spent less free choice time working on the puzzles) as a result of their experience of doing the activity for pay.
These results were replicated in a controlled field experiment (Deci, 1971) which took place over a 16-week period in a college newspaper. Subjects were staff members who wrote headlines for the college newspaper and who were unaware that an experiment was being performed. As in the laboratory study, subjeets who were paid for their performance ($.50 per headline written) showed a decrease in intrinsic motivation which was evident as much as eight weeks after the payments had stopped.
As mentioned earlier, using this paradigm Deci (1972) replicated the finding that subjects who were paid one dollar per puzzle solved showed a decrease in intrinsic motivation. Further, Deci and Caseio (1972) reported that when subjects were threatened with punishment for poor performance, their intrinsic motivation also decreased. These experimental subjects were told that if they were unable to solve a puzzle within the ten minutes allotted, a buzzer would sound indicating that their time was up. They were then given a brief exposure (about one second) to the buzzer so they would know that it was truly noxious. Hence they were performing because they were intrinsically motivated and because good performance would allow them to avoid a punishment (the buzzer). The results indicated that those who solved puzzles under threat of punishment were less intrinsically motirated during the free-choice period than subjects who solved the same puzzles with no threat of punishment.
Deci has suggested a cognitive evaluation theory to explain this change in intrinsic motivation. It concentrates on a person’s perception of why he is doing the activity. When he is intrinsically motivated, the perceived locus of causality (Heider, 1958) of that behavior is within himself. He is doing it because it provides him with some sort of internal satisfaction. However, when he performs the activity for external reinforcements such as money, he comes to perceive that he is doing it for the money. The perceived locus of causality changes from within himself to the environment; that is, he cognitively reevaluates the activity as one which he does because it provides him with external rewards. In other words, the first process by which intrinsic motivation can be affected is a change in perceived locus of causality
eived locus of causality. On the other hand, Deei (1971, 1972) has reported that verbal reinforcements do not decrease intrinsic motivation; in fact, they appear to enhance it. In the Deci (1971) study, the three-session paradigm described 224 EDWARD L. DECI earlier was used. In this study, the experimental subjects were rewarded with verbal statements such as, “That’s very good, it’s the fastest anyone has solved this one,” each time they solved a puzzle. The control group received no rewards. The results indicated that subjects who received verbal rewards were more intrinsically motivated following that experience than subjects who received no rewards. These results were replicated for males (Deci, 1972) using the one-session methodology described above. The essential difference between money and verbal rewards is that verbal rewards may not be phenomenologically distinguishable from the feelings of satisfaction which the person gets for doing the activity. Hence, the verbal reinforcements strengthen his intrinsic motivation because they provide additional positive value which becomes associated with the activity; so the subject is more likely to perform the activity in the absence of external rewards
In another experiment using the one-session paradigm, Deci and Cascio (1972) showed that negative feedback resulting from bad performance on an intrinsically motivated activity caused a decrease in intrinsic motivation. According to the cognitive evaluation theory, the psychological process underlying this decrease is the same as the process by which positive feedback enhanced intrinsic motivation. The negative value associated with the failure and the resulting threat to the person’s sense of competence offsets some of the positive value associated with the activity, thereby causing a decrease in intrinsic motivation
Interpreting these results in relation to theories of work motivation, it seems clear that the effects of intrinsic motivation and extrinsic motivation are not additive. While extrinsic rewards such as money can certainly motivate behavior, they appear to be doing so at the expense of intrinsic motivation
Given these results, it becomes interesting to ask whether it is the money per se, or the money administered contingently which causes the decrease in intrinsic motivation. The next experiment was designed to investigate that question. Does money affect intrinsic motivation when it is administered for an activity in a way that is not contingent upon performance?
Subjects in the control condition who received no money for solving puzzles spent an average of 190.2 seconds out of the 480 seconds of free-choice time working on the puzzles, whereas the experimental subjects who received $2 for participating spent an average of 192.8 seconds. Clearly there is no difference. On the other hand, in the contingent-payment experiment (Deci, 1972) there was a significant difference. The control subjects spent an average of 208.4 seconds while the experimental subjects, who were paid $1 for each puzzle they solved, spent only 108.6 seconds.