Academic journal article The Psychological Record

Effect of Reward on Subjective Autonomy and Interest When Initial Interest Is Low

Academic journal article The Psychological Record

Effect of Reward on Subjective Autonomy and Interest When Initial Interest Is Low

Article excerpt

Rewards are often used in programs designed to reduce people's dependence on others and increase their autonomy, as in behavioral therapies (Ballard & Stoudemire, 1992; Dunlap & Dunlap, 1989), organizational psychology (Nasanen & Saari, 1987; Overskeid, 1992), and education (Clement, 1981; Olympia, Sheridan, Jenson, & Andrews, 1994). Though often effective, such reward-based interventions are known to sometimes fail and are quite frequently only partially successful (Olympia et al., 1994; Stark, Allen, Hurst, Nash, Rigney, & Stokes, 1989). Among several potential difficulties, the use of rewards in the control of behavior has in itself been pointed out as a problem. The effect of reward for task participation is a particularly interesting issue in this context.

Clearly, in determining the effects of reward for task participation, the nature of the task is relevant. In interesting tasks rewards may decrease participants' interest, most reliably when rewards are expected, salient, and contingent on task engagement (Deci & Ryan, 1987). It has been hypothesized that when task interest is reduced in the presence of rewards, this occurs because rewards have reduced subjective autonomy. Humans are assumed to have a need to feel autonomous. As a means of behavior control, rewards are thought to reduce subjective autonomy, and reduction in subjective autonomy is believed to lead to loss of interest (Deci & Ryan, 1994).

Interesting tasks are often performed to achieve pleasure that is inherent in the task. The pleasurable consequences of performing the task are events that coexist with the reward, discounting external rewards as controlling factors (Kelley, 1972) as well as making rewards less salient. In a boring task, rewards are much more salient as means of behavior control than in an interesting task. If salience of reward is important in reducing subjective autonomy, it might be expected that the less interesting the task, the more a reward will reduce the acting person's sense of autonomy. In other words, because rewards exercise greater control and are more salient when a task is dull than when it is interesting; one might expect a greater reduction in subjective autonomy when boring tasks are rewarded than when interesting tasks are rewarded.

Rewards are, of course, only one of many sources of behavior control. Because subjects participating in experiments are normally instructed to engage in a task, an obvious source of control in this context is instructions. Being instructed to perform a task will by necessity affect a subject's feeling of autonomy in performing that task. However, the observation that being instructed to act may reduce one's sense of autonomy can be accepted without necessarily assuming that behavior control by way of rewards will reduce subjective autonomy. Indeed, the thought that extrinsic rewards should reduce people's sense of autonomy seems somewhat counterintuitive.

Consider the fact that working for no pay is slavery. A slave is not his own master, which means that a slave has little autonomy. Working for a salary is normally associated with a greater degree of autonomy. Thus, an increase, rather than a decrease in autonomy should be a likely finding when engagement in a boring task is rewarded. A test of this hypothesis requires, of course, that the effects of choice and rewards are separated.

Because variation in subjective autonomy has been claimed to affect task interest, the relation between these two variables is theoretically significant. In spite of this, the relationship between subjective autonomy and task interest has not previously been subjected to direct empirical testing. We performed a pilot study that did not indicate any such association. Furthermore, when rewards are offered in exchange for engagement in a boring task, depending on the circumstances, both incentive effects (Calder & Staw, 1975; Loveland & Olley, 1979), dissonance effects (Greenwald & Ronis, 1978), and reverse-incentive effects (Freedman, Cunningham, & Krismer, 1992) have been reported. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.