11. INTRINSIC AND EXTRINSIC MOTIVATION. Intrinsic motivation is the performance of a task purely for the joy of doing it. There is no reward except in accomplishing the task the individual set out to complete. Conversely, extrinsic motivation is the performance of a task for some external reward. The most common form in a capitalist society is money. What separates these two forms of motivation is that one is difficult to measure because it is difficult to determine for each individual what it is that motivates him or her. External motivating factors are often visible and tangible. Current research into both areas reveals that individuals can be motivated both intrinsically and extrinsically, separately, and sometimes, simultaneously. Some individuals are more prone to accept extrinsic motivational factors than others. This does not mean that they shun external factors; they just prefer internal ones. Is one more important or powerful than the other? There is no definitive answer to this question. However, intrinsic motivation prompts individuals to strive for goals in the absence of any discernable reward. Externally motivated individuals, in the absence of rewards may become unmotivated unless there is some hope for measurable reward in the end. What does current research into these two unique forms of motivation reveal? The following are an overview of some findings in business, industry, and education. Intrinsic and extrinsic motivation are reviewed simultaneously because both capture the essence of this aspect of human motivation. Boggiano and Barrett (1992) examined whether extrinsic children were more depressed than intrinsic children and whether this effect was more pronounced for girls than for boys. Sixty seven female and 60 male 3rd-grade children completed self-report measures assessing motivational orientation (MO) and depressive symptoms. Subjects with an extrinsic MO (i.e., subjects who performed a task for approval or fear of evaluation) were more depressed than those with an intrinsic MO. Significant gender differences show that girls were more extrinsic than boys and thus more susceptible to helplessness and depression. Fair and Silvestri (1992) reviewed literature on the effects of rewards, competition, and success/failure on intrinsic motivation. Data indicated that intrinsically motivated individuals with an internal desire to excel are superior in both quality and quantity of task performance to externally motivated performers. Studies of children and adults credited praise and positive verbal affirmations with increasing internal motivation, and external rewards and punishment with decreasing internal motivation. Although the sexes were not significantly different in internal motivation, they appeared to vary in their approach to competitive situations. They suggested ways to increase motivation including the need for enough praise and positive affirmation at every step of the way through a process to enhance both children and adults to achieve at maximum levels. Boggiano and Katz (1991) tested the hypothesis that the importance of adult approval and feedback for females relative to males would render girls of elementary school age more likely to develop an extrinsic orientation in comparison to boys. Using Harter's (1981) Scale of Intrinsic versus Extrinsic Motivation, the data supported the hypothesis for Study 1 in which 107 girls and 106 boys in 4th-6th grades participated. Because of the assumed differential importance of controlling feedback from adults for females relative to males, a second study examined 64, 9-11 yr old girls' and boys' preference for challenge as a function of adult controlling feedback and children's motivational orientation. The data supported the hypothesis that girls relative to boys show differential preferences for challenge, depending on the presence and type of adult feedback and motivational orientation in girls. Enzle, Roggeveen, and Look (1991) hypothesized that ambiguous behavior standards coupled with self-administration of rewards would reduce intrinsic motivation whereas clear standards coupled with self-administration of rewards would maintain high preexisting levels of intrinsic motivation. Fifty four university students participated in a study comprised of a 2 (ambiguous vs unambiguous standards) by 2 (self- vs other-administered reward) factorial design and unrewarded control group. Results supported their predictions. Behavior standard clarity was a critical determinant of whether self-administration protected or undermined initially high intrinsic motivation. Lawler, Armstead, and Patton (1991) attempted to integrate the concept of the type A behavior pattern with the research on intrinsic motivation. The hypothesis that type A behaviors focus attention on extrinsic rather than intrinsic motivation was tested by measuring type A behavior and motivational orientation in 95 male college students. Results were consistent with the hypothesis. Type A subjects were also unaffected by instructions designed to increase intrinsic motivation, whereas type B subjects were highly sensitive to such instructions. These data provide an explanation for the paradoxical effect of type A behavior in males (i.e., success accompanied by feelings of dissatisfaction). They suggested more research needed to be conducted to determine what motivated type B individuals. Wiersma (1991) found contrasting predictions from expectancy valence and cognitive evaluation theories while testing his hypothesis when testing using a 2 * 2 factorial design in which intrinsic and extrinsic rewards were manipulated. Several relevant factors were also considered, namely type of extrinsic reward, measure of intrinsic motivation, and use of moderator variables (e.g., higher order need strength). Ninety four undergraduates worked on either an interesting or boring task for a low or high extrinsic reward. The primary measure of intrinsic motivation was the subjects' return rate to a second experimental session. Results supported the expectancy valence hypothesis that intrinsic and contingent extrinsic rewards have an additive effect on motivation. Both types of rewards showed significant main effects. There were no significant moderator effects found. Koestner et al (1991) examined whether motives as assessed from fantasy (seen as implicit needs) were primarily aroused by factors intrinsic to the process of performing an activity, whereas motives obtained through self-report inventories (seen as self-attributed needs) were aroused by social factors that were extrinsic to the process of performing an activity (e.g., the way in which a task is presented by an experimenter). In Experiment 1, with 71 college students, performance on a memory task depended on the interaction of subjects' self-reported motive for achievement with achievement-arousing instructions, whereas performance on a word-finding puzzle depended on the interaction of subjects' fantasy need for achievement with the puzzle's level of intrinsic challenge. Experiment 2, with 54 college students, generalized these findings to the power domain. Sakurai (1990) examined the self-evaluative motivation (SEM) model proposed by Sakurai (1990), using four kinds of extrinsic rewards (verbal, token, verbal and token, and feedback only) among 110 6th graders. According to the SEM model, experimental groups should select more difficult tasks (receiving no rewards or feedback) than the control group because they have higher feelings of competence. The token, verbal and token, and control groups should select more dissimilar tasks than verbal and feedback groups because the former groups have little feeling of self-determination. In the experiment, subjects were assigned lucky-puzzle tasks and given extrinsic rewards or feedback. They were asked which task they wanted to try among four kinds of tasks that were similar or dissimilar to the previous task and easy or difficult. Results supported the behavioral predictions of the SEM model. Tripathi (1991) studied the effects of contingency and timing of reward on intrinsic motivation for subsequent task performance in 65 undergraduates. After completing the mandatory problem session, subjects spent a significantly greater amount of free-activity time solving puzzles under performance-contingent reward conditions, followed by task-contingent reward with feedback, and task-contingent reward conditions, respectively. Performance on problem-solving tasks during a free-activity period yielded a significant effect of reward contingency, implying greater intrinsic motivation under performance-contingent reward, followed by task-contingent reward with feedback, and task-contingent reward conditions, respectively. The effect of timing of reward was not significant. The roles of challenge and competence information facilitated intrinsic motivation for task performance. Diaz-Soto (1989) examined differences in home environment, motivational orientation, and relationships among the home environment and the motivational orientation of 28 higher and 29 lower achieving 5th- and 6th-grade Puerto Rican children. The relative weights of the variables and their ability to predict achievement were examined via a post-hoc multiple regression analysis. Home interviews were conducted using the Family Environment Schedule (Marjoribanks, 1979) and a scale of intrinsic vs extrinsic orientation. Gender differences were also noted. Family involvement accounted for a significant amount of variance with regard to achievement. Home environment differed for the higher and lower achievers' family, with parental aspirations higher for higher achievers. Motivational orientation differed, with higher achievers adopting a more intrinsic orientation and lower achievers adopting a more extrinsic orientation. Rummel and Feinberg (1990) examined the effect of motivational orientation on intrinsic motivation among 40 females. Use of the Jonckheere Test of Order (A. R. Jonckheere, 1954) suggested that the detrimental effect of extrinsic rewards on intrinsic motivation might be explained within the reinforcement paradigm. Intrinsically motivated subjects who received intrinsic rewards showed the highest amount of intrinsic motivation as assessed by free time on the Soma puzzle. Extrinsically motivated subjects receiving intrinsic rewards showed the lowest amount of free time. Rewarding intrinsically motivated subjects negatively affected their motivation while just the opposite was true for the extrinsically motivated ones. Mawhinney (1990) described cognitive and behavior analytic approaches to the study of intrinsic and extrinsic reward effects and considered the implications of differences in these approaches for future study and application. A critique of the research design and data analysis techniques used in Deci's (1971, 1972; Deci and R. M. Ryan, 1985) Deci-type paradigm was presented. A behavioral model of intrinsic reinforcement for use by organizational behavior managers was also described. Evidence contradicting the Deci-type theory indicated that people who are most highly intrinsically motivated by a task are least likely to exhibit any post-extrinsic reinforcement decrement to intrinsic motivation. Flora (1990) proposed that the use of contingent, "extrinsic" reinforcement undermines the rate at which "intrinsically interesting" behaviors occur. A review of the literature suggested that environmental stimuli control the rates of behaviors rather than interest intrinsic to the organism. Reduced rates of behavior typically attributed to the undermining of intrinsic interest are more objectively accounted for by environmental stimuli functions, including instructional control and by Herrnstein's (1970) matching law. When the hypothetical version of intrinsic motivation is contrasted with a physiological version, the hypothetical version makes the opposite prediction of every effect that occurs. Recommendations based on the concept of undermining intrinsic interest are flawed and possibly dangerous, he believed and suggested that extrinsic motivating factors should be removed from the intrinsically motivated behavioral situation or event. Dickinson (1989) wrote that extrinsic consequences have been criticized on the grounds that they decrease intrinsic motivation or internally initiated behavior. Two popular rationales for this criticism, Lepper's (1981) overjustification hypothesis and Deci and Ryan's (1985) motivational theory, were reviewed, and the criticism was then redefined behaviorally. Intrinsically controlled behavior was defined as behavior maintained by response-produced reinforcers. Empirical support for the decrease in the reinforcing value of stimuli caused by extrinsic reinforcement was presented, and possible explanations for the phenomenon were offered. The effect was transient and not likely to occur at all if extrinsic rewards are reinforcing, noncompetitive, based on reasonable performance standards, and delivered repetitively. Dolke and Srivastava (1988) investigated whether the job attitudes of satisfaction, involvement, and intrinsic motivation were conceptually distinct and empirically independent variables, using questionnaire responses from 200 clerical and 118 technical textile mill workers in India. Results provided fairly strong evidence for treating the three variables as separate attitudes. Findings appeared to be generalizable cross-culturally and they were similar to those obtained by Lawler and Hall (1970) with research scientists and Cummings and Bigelow (1968) with blue-collar workers. Anderson and Rodin (1989) studied 50 undergraduates who received cues supporting different causality prior to answering a set of brain-teasers on a computer terminal. Results indicated that mild negative feedback increased intrinsic motivation (IM) when it was associated with environmental cues signaling self-determination. Subjects who were given mild negative feedback but who had a choice of problems to solve, no expectation of evaluation, and private discussion of scores, retained as much (or more) IM as subjects given positive feedback under the same conditions. Subjects in controlling contexts showed less IM. Feldman and Weitz (1988) suggested that a variety of individual, job-related, and organizational factors influence whether employees reach career plateaus. Career plateaus were associated with (1) individual skills, (2) individual needs and values, (3) lack of intrinsic motivation, (4) lack of extrinsic rewards, (5) stress and burnout, and (6) slow organizational growth. They reported that to motivate employees managers must plan strategies to increase individual skills, understand needs and values, support intrinsic motivational patterns, provide extrinsic rewards to those who need them, and increase organizational growth so that employees can strive to fulfill the challenge of promotions being offered by the company. Deci (1987) suggested that excitement as well competence and self-determination underlie intrinsically motivated behavior. He argued that excitement (which is an emotion) belongs in a different class of concepts from competence and self-determination (which are needs). He also suggested that several recent studies (e.g., Baumeister and Tice, 1985) have added confusion to the intrinsic motivation literature by focusing on paradigms (rather than theories) and operations (rather than constructs). If intrinsically motivated individuals are to be supported, he wrote, then organizations must begin to accept that they exist, and create an environment in which intrinsically motivated individuals can fully develop. Tang et al (1987) examined the performance of 131 undergraduates, classified as either high or low in self-esteem, on an anagram-solving task labeled as difficult or easy. In the first work period, subjects in the easy condition set higher goals than subjects in the difficult condition. High self-esteem subjects in the easy condition solved more anagrams than those in the difficult condition. In the second period, high self-esteem subjects in the easy condition set higher goals than those in the difficult condition. In the free-choice period, low self-esteem subjects in the easy condition spent significantly less time on the anagram-solving task than did the average of the other groups. The combination of low self-esteem and an easy label led to the lowest level of intrinsic motivation. Cellar and Wade (1988) studied the effects of behavior modeling (intrinsically vs. extrinsically motivated model) and a symbolic rehearsal intervention (presence vs. absence) on intrinsic motivation, task satisfaction, and script-related recognition. A 2 * 2 factorial design was used with 80 male undergraduates as subjects. As predicted, the motivational orientation of the model affected behavioral measures of intrinsic motivation and script-related recognition. It did not affect self-report measures of interest or task satisfaction. Contrary to expectations, the symbolic rehearsal intervention had no effect on these variables. Regression analysis revealed that the intrinsic script-recognition measures explained incremental variance in intrinsic motivation beyond measures of locus of causality and perceived self-competence, suggesting that a script-processing model may add to existing intrinsic motivation theory. Reeve et al (1987) conducted two experiments with 107 female and 78 male undergraduates that examined individual differences of need for achievement, anxiety, locus of control, and gender to determine their intervening role in intrinsic motivational processes following objective competence information. Subjects were either allowed to win or were made to lose a puzzle-solving contest against a same-gender confederate with the success/failure experience serving as the objective source of competence information. In experiment I, the outcome, locus of control, and the resultant achievement motivation * outcome interaction predicted level of intrinsic motivation, thereby substantiating the claim that individual differences are important in the prediction of intrinsic motivation. Experiment II showed that the need for achievement affected level of intrinsic motivation through the high achievers' favorably biased performance expectancy and heightened positive affect, and, after losing, through both favorable actual and perceived performance relative to low achievers. Kohn (1987) discussed research on intrinsic task motivation, defined as the delight in doing something for its own sake, and examined the relation between creativity, external rewards, and motivation. The work of Amabile (1986) suggested that the intrinsic motivation principle of creativity questions the behaviorists' assumption that any activity is more likely to occur if it is rewarded. Environmental factors that aid creativity (e.g., encouragement, autonomy) were reviewed and added as variables that enhance the intrinsic motivation to be creative, some of which were a trust, encouragement, acceptance of divergent thinking, and support for nonconformity. Most studies of intrinsic motivation under reward and evaluative contingencies have used social comparison criteria to evaluate subjects' performance. In those studies evaluation tended to reduce intrinsic interest. Harackiewicz et al (1987) contrasted normative evaluation against a more task-focused evaluation of subjects' performance on an interesting word game and examined the role of achievement motivation in moderating reactions to performance evaluation. Focus differences were implemented under conditions of performance-contingent reward, anticipated evaluation, and control conditions in which subjects received performance feedback at task conclusion. They predicted that evaluation would reduce interest relative to reward and feedback control groups under a normative focus but not under a task focus and also that a process of competence valuation (Harackiewicz & Manderlink, 1984) would mediate the effects of reward and achievement on interest, especially in normative conditions. The data conformed to these expectations with one exception: evaluation under a task focus increased intrinsic interest. These results were interpreted in the context of a general model that considers the separate effects of situational contingencies, personality factors, performance and motivational processes, and evaluative outcomes on intrinsic motivation. Sinha (1986) administered questionnaires designed to assess job satisfaction, intrinsic motivation, work values, and job involvement to 60 government and 50 private-enterprise employees in India. Questionnaires included items from instruments developed by Lawler and Hall (1970), Blood (1969), the Institute for Social Research at the University of Michigan, and U. N. Agarwala (1976). Analysis of data obtained from the 60 completed questionnaires was accomplished using t -tests and correlations. Quality dimensions correlated positively with satisfaction, and some differences could be detected between private and government workers. Orpen (1986) administered the Pay Satisfaction Questionnaire (Heneman and Schwab; see PA, Vol 73:20996) to 47 managers from a variety of industries. Subjects also completed measures of job involvement, work satisfaction, and internal motivation, and provided self-ratings of performance, absenteeism, and turnover. Results showed that only 2 of the 24 correlations between these outcomes and satisfaction with pay level, raises, benefits, and structure were significant, those between pay level and motivation and pay level and involvement. Results suggested that subjects pay was unrelated to job attitudes that lead to high motivation and performance. Shalley et al (1987) examined the possibility that participative goal-setting methods combined with goal difficulty and expected external evaluation to affect intrinsic motivation, using 96 male undergraduates. Subjects were assigned to one of eight experimental conditions. Results indicated that only the method of setting goals had an effect on intrinsic motivation. Subjects who were assigned goals exhibited significantly higher levels of intrinsic motivation than those who participated in setting goals. In this experiment, Koestner et al (1987) examined the relation between content of praise, type of involvement, and intrinsic motivation. College students were introduced to a hidden-figure task in either an ego-involving (i.e., test-like) or task-involving (i.e., game-like) manner and then received either ability-focused, effort-focused, or no praise for their performance. As predicted, task involvement increased intrinsic motivation relative to ego involvement, and ability praise increased intrinsic motivation relative to effort praise or no praise. Furthermore, praise and involvement interacted so that subjects who received effort praise were relatively more intrinsically motivated under task-involving than ego-involving conditions, whereas those who received ability praise were relatively more motivated under ego-involving than task-involving conditions. The higher levels of intrinsic motivation were accompanied by a choice of a higher level of challenge and better performance at a related but more complex task. Finally, a significant Sex * Praise interaction was found, reflecting that women tended to display more intrinsic motivation in the no-praise condition than in the two praise conditions, whereas men showed the reverse pattern. Reeves et al (1986) asked undergraduates in experiment I to list favorite and least favorite activities and report the predominant affective experience underlying the activities. Results showed that, in a majority of cases, subjects reported excitement rather than either competence or autonomy as the feeling most associated with their favorite activities. In experiment II, the traditional intrinsic motivation paradigm was used with 31 undergraduates. Results showed that excitement had a significant, positive association with each of four indices of intrinsic motivation. It was concluded that excitement functioned as a third reward in intrinsic motivational processes. Sugihara (1985) investigated the hypothesis that test anxiety (TA) was one of the main factors undermining intrinsic motivation. Two hundred and sixty three male undergraduates with high and low TA were assigned to a test or no-test condition. Subjects engaged in solving puzzles in both conditions. Only those in the test condition were told that the task was a kind of intelligence test. The intrinsic motivation of high-TA subjects did not differ significantly between conditions, while that of low-TA's were significantly lower in the test than in the no-test condition. In order to foster intrinsic motivation in learners, text anxiety needs to be diminished so that motivation is enhanced. Sansone (1986) examined whether competence information is the feedback feature that affects intrinsic motivation and whether perceived competence is the process responsible in two studies in which 174 undergraduates compared competence feedback with meaningful task feedback. In study one, positive competence feedback and task feedback were manipulated independently. Findings indicated that although positive feedback resulted in the highest level of perceived competence, both positive and task feedback enhanced interest individually. In study two, an ego-involvement manipulation emphasized competence prior to task engagement. Path-analytic techniques were used to identify two processes that mediated the effects of positive, negative, and task feedback on interest: perceived competence and personal valuation. Results indicated that perceived competence enhanced enjoyment only when performance quality was stressed by the ego-involvement manipulation. When competence was not made salient, subsequent interest depended more on the degree the individual personally valued involvement. Both studies indicated that competence information can affect both perceived competence and personal valuation. Feeling competent itself enhanced intrinsic motivation only if attaining competence was a primary goal of the task. To summarize the research cited above, intrinsic motivation was sometimes negatively influenced by extrinsic factors. If intrinsic motivation is characterized by the pure delight in doing something just for the joy of doing it, then extrinsic factors, when added to the motivation equation, can inhibit the individual from intrinsically pursuing his or her goal. The intrinsic/extrinsic motivational theories do not possess the same amount of complexity that some of the earlier ones reviewed do. They are important because they provide some manner of describing what motivates individuals who appear to be doing something just for the pleasure of doing it. This was a departure from the earlier researchers who theorized that causal relationships existed between D (drive) and motivation. Finally, intrinsically motivated individuals comprise a unique group of individuals. Under normal circumstances, extrinsic motivation seems to be more commonplace because it is something which can be seen, heard, and measured. Intrinsic motivation is difficult to measure because it is centered in the affective rather than the cognitive or psychomotor domains.