Have you noticed how much of your behavior is based on anticipated outcomes
P A R T F O U R
External Sources of Motivation
Have you noticed how much of your behavior is based on anticipated outcomes? Is behavior motivated to make some consequences happen and other consequences not happen? These consequences refer to the anticipated incentives or goals that a person’s behav- ior brings. Incentives (Chapter 10) and goals (Chapter 11) motivate either approach or avoidance behavior. Positive incentives, such as a high GPA or a raise in pay, motivate approach behavior, which is designed to make these consequences happen. Negative incentives, however, such as failing grades or a cut in pay, motivate avoidance behavior, which is designed to prevent these consequences. People commit to goals that they plan to achieve. An incentive provides extra motivation on the way toward achieving a goal. A uni- versity degree, for example, is a goal for students, while course grades are incentives to motivate students along the way.
What determines the motivational strength of incentives and goals? The answer is linked to the concept of value and utility. Value refers to quantity while utility refers to usefulness. Thus, $100 has greater value than $10 but also has greater utility, since more services and goods can be bought with $100. As incentive or goal value increases, motiva- tion increases. In addition, positive and negative values motivate opposing behaviors at different intensities. Loss has a stronger impact than does an identical gain. People are motivated more to prevent negative events like a loss, failure, or lower status.
Is behavior always performed for extrinsic reasons—that is, for anticipated conse- quences? Can’t behavior also simply occur for its own sake without external rewards? Yes, sometimes behavior is motivating in its own right, which is referred to as intrinsic motiva- tion (Chapter 10). Recreation, hobbies, and leisure time activities occur because these behaviors are enjoyed for their own sake.
How is motivation affected by the delay interval between the present and a future goal? The delay interval reduces the motivating effects of future incentives compared to im- mediate ones. For example, an exam one day away provides more motivation for study than does an exam one week away. Also, do you ever estimate the likelihood that you can achieve your chosen goal? Motivation is also affected by the likelihood that a goal will be achieved. As likelihood increases, motivation increases. Among goals of equal value, likely goals are more motivating than unlikely ones.
Motivation: Biological, Psychological, and Environmental, Third Edition, by Lambert Deckers. Published by Allyn & Bacon. Copyright © 2010 by Pearson Education, Inc.
238 P A R T F O U R / External Sources of Motivation
Do some goals seem more expensive to achieve than others in terms of response, time, and energy costs? Chapter 12 (Economics of Motivation) describes how various costs affect goal achievement. Incentives and goals are like the items available at the store. Some are more expensive than others just like some goals are more difficult to achieve than others. They require more behavior, time, and effort. Usually, as the cost of goals increase, the demand for them decreases. Fewer people try to achieve expensive goals, such as a 4.00 GPA, an Olympic gold medal, or the presidency. However, people have motivation resources with which to achieve their goals, such as their behavior, time, and energy.
Motivation: Biological, Psychological, and Environmental, Third Edition, by Lambert Deckers. Published by Allyn & Bacon. Copyright © 2010 by Pearson Education, Inc.
C H A P T E R
Extrinsic and Intrinsic Motivation
A thing is worth whatever the buyer will pay for it. —Publius Syrus, 50 B.C.
Let not the enjoyment of pleasure now within our grasp be carried to such excess as to incapacitate you from future repetition.
—Seneca, 4 B.C.–65 A.D.
■ In the push/pull metaphor of motivation, motives are within the person and push while incen- tives are outside the person and either pull or repel. What qualities determine their strength of attraction or repulsion? How does this strength wane with temporal distance after which the incentive becomes available? These considerations are rephrased in the following questions for you to think about:
1. How do incentives differ from reinforcers and punishers in motivating behavior?
2. What characteristics influence the value of an incentive?
3. How does incentive value affect the motivation of behavior?
4. Are some behaviors freely chosen—that is, not coerced by incentives?
5. Can behavior be motivating in its own right—that is, be intrinsically motivating?
6. How do extrinsic and intrinsic sources interact to motivate behavior?
Extrinsic Motivation and Incentive Value According to the familiar axiom of “wanting more,” to have something desirable is good, and to have more of it is better, and to have still more is better yet; to have less is worse. One corollary of the axiom of wanting more is that the more frequently a desirable event occurs, the better it is, and the less frequently it occurs the worse it is. A second corollary is that the sooner a desirable event occurs the better it is, and the later it occurs the worse it is. The reverse, then, is likely to be true: to have something undesirable is bad, to have more of it is worse, and to have still more is worse yet; to have less is better. A corollary of this axiom is that the more frequently an undesirable event occurs, the worse it is, and the less frequently it occurs the better it is. A second corollary is that the sooner an unpleasant event occurs, the worse it is and the later it occurs the better it is.
Motivation: Biological, Psychological, and Environmental, Third Edition, by Lambert Deckers. Published by Allyn & Bacon. Copyright © 2010 by Pearson Education, Inc.
These axioms, or self-evident truths, describe some major features of extrinsic moti- vation: often more and larger incentives are preferred—though not in the case of negative incentives—and are more motivating than fewer and smaller incentives.
The purpose of this section is to describe what characteristics determine the value of an incentive and how this affects a person’s choices and behavior.
Reinforcers and Punishers versus Incentives How does a student know which behaviors result in good grades? How does a worker know how to earn a year-end bonus? Why does the prospect of good grades or a bonus motivate certain behaviors and not others? These questions address the difference between rein- forcers and punishers, on the one hand, and positive and negative incentives, on the other. The difference is based on the effects of past events versus the anticipation of future events. In the past, attending class, studying, or working diligently resulted in a good grade or year-end bonus. When anticipating the future, a good grade or year-end bonus motivates behavior.
Selecting versus Motivating Behavior. Learning what to do and actually doing it illus- trate the separate effects of reinforcers and incentives (Cofer & Appley, 1964; Tolman, 1955; Tolman & Honzik, 1930). Learning what to do results from the action of reinforcers. These are stimuli that select appropriate behaviors and make them more likely to occur in a situation (see Skinner, 1938, 1953; Staddon & Simmelhag, 1971). Learning what not to do results from the action of punishers that select against behaviors and make them less likely to occur (Skinner, 1953). Thus, attending class and studying are selected for by good grades (reinforcers), while skipping classes and not studying are selected against by fail- ing exams and courses (punishers). Incentives are the external stimuli that motivate or in- duce behavior to occur (Bolles, 1975; Logan & Wagner, 1965). A positive incentive motivates the behavior that is instrumental in attaining the incentive. For example, the incentive of a good grade motivates studying. A negative incentive motivates avoidance behavior, which is instrumental in averting or preventing the incentives from happening. For example, the negative incentive of failing a course motivates a student to avoid skip- ping classes.
Past versus Future. Reinforcers and punishers are the actual consequences of behavior, whereas positive and negative incentives are the anticipated consequences. Incentives influ- ence behavior based on their anticipation (Bolles, 1975; Karniol & Ross, 1996). Both mech- anistic and cognitive accounts have been provided as explanations. According to one mechanistic explanation, anticipation of an incentive consists of minuscule responses that resemble the final consummatory response of the incentive (Hull, 1952). For instance, a person salivates driving to a restaurant and then, when there, he salivates more intensely while eating. Or a student smiles in anticipation of receiving congratulations from family members at graduation. According to a cognitive explanation, people visualize themselves in a future incentive situation (Markus & Nurius, 1986). For instance, a student cognitively represents the incentive of a university degree as “my earning a B.A. or B.S.” The cognitive representation bridges the present to the future and motivates a person’s behavior. Thus, a student may form a mental image of her professor giving an exam, which in turn motivates the student to study for the exam.
240 P A R T F O U R / External Sources of Motivation
C H A P T E R T E N / Extrinsic and Intrinsic Motivation 241
TABLE 10.1 Distinction between Reinforcers and Punishers versus Positive and Negative Incentives
Outcomes of Behavior: Past Events
Anticipated Outcomes of Behavior: Future Events
Positive Reinforcer: Positive Incentive: Stimulus Behavior is selected for
and becomes more likely. Motivates approach behavior
to attain the incentive Example: Past passing grade,
parental approval, and pride Example: Anticipated passing grade,
parental approval, and pride
Negative Punisher: Negative Incentive: Stimulus Behavior is selected against
and becomes less likely. Motivates avoidance behaviors
to prevent occurrence of incentive Example: Past failing grade,
parental disapproval, and shame
Example: Anticipated failing grade, parental disapproval, and shame
Some incentives are derived from their association with reinforcement and punish- ment. If a response consistently results in a reinforcer, then that reinforcer becomes a positive incentive; it will be sought out or approached. If a response consistently results in a punisher, then that punisher becomes a negative incentive; it will be avoided or pre- vented (Bolles, 1975; Logan, 1960). In some instances, however, prior experience seems unnecessary for a stimulus to act as an incentive. A new car, for example, serves as an incentive to save money even if an individual has never owned a new car. A person also avoids touching a hot stove even if she has never been burned by one before. Table 10.1 summarizes the distinctions between reinforcers and punishers and positive and negative incentives.
Progress through a university illustrates the distinctions in Table 10.1. Studying and attending class produce passing grades, parental approval, and a feeling of pride. These con- sequences are reinforcers, provided they increase or maintain those behaviors. As antici- pated consequences these outcomes serve as positive incentives that motivate a student to attend class and study. Not attending class and not studying produce failing grades, parental disapproval, and perhaps a feeling of shame. These consequences are classified as punish- ers, provided they decrease those behaviors. As anticipated consequences these outcomes are negative incentives if they motivate a student to avoid them by not missing class and not neglecting to study (see Table 10.1).
Objective and Subjective Incentive Value An incentive refers to the motivational properties of a reinforcer. The value of an incentive determines its preference and motivational strength. Terms like more, bigger, and better usually indicate increased value of a positive incentive, while less, smaller, and worse usu- ally indicate decreased value. Incentive value refers to the attractiveness of an incentive and is based on objective properties like the number or the amount. Objective incentive
value refers to physical properties of an incentive, while subjective incentive value refers to an individual’s appraisal of the objective value. The distinction between the objective and subjective value of economic goods serves as a case in point. According to economists, subjective value is synonymous with utility, which refers to the satisfaction, pleasure, or usefulness of an economic good. A pencil has a certain objective value based, say, on how much it costs, but it also has a great deal of utility. It is a very useful instrument (try attend- ing school without one). Utility rather than objective value provides a better understanding of why economic goods such as cars, clothes, and computers are satisfying.
What is the nature of the relationship between the objective and subjective value or utility of an incentive? One answer was provided by Fechner (1860/1966), who explored the relationship between stimulus intensities and corresponding psychological sensations. He showed that equal increases in stimulus intensity produce smaller and smaller increases in sensations. For instance, equal increases in a tone’s loudness produce increases in the sen- sation of loudness but in diminishing amounts. This relationship became known as Fechner’s law. Several economists in the late 1800s noted that the relationship between amount of money and its utility followed Fechner’s law (Stigler, 1950). Even earlier, in 1738, the French mathematician Bernoulli made a similar proposal about money and its utility (Stevens, 1972). Figure 10.1 shows such a relationship: as the number of dollars increases, the utility of those dollars increases but in diminishing amounts. (Utility is measured in utils, units employed at one time by economists to indicate the utility of economic goods.) Thus in Figure 10.1, the first $10 corresponds to 10 utils of satisfaction, whereas the next $10 corresponds to approximately 9 more utils of satisfaction. Although $20 is twice as much as $10, $20 does not have twice as much utility as $10. Equal increments in
0 10 20 30 40 50 60 70 80
FIGURE 10.1 Dollars and Utility. The relationship between the number of dollars and their utility is such that as the dollar amount increases, utility increases, but in smaller and smaller amounts. The graph is only an approximation of the true dollar-utility relationship.
242 P A R T F O U R / External Sources of Motivation
C H A P T E R T E N / Extrinsic and Intrinsic Motivation 243
dollar value lead to smaller and smaller increases in utility. Perhaps the most general state- ment we can make about money and its utility is that more is better.
Incentives as Losses and Gains Suppose there is very little time remaining in a very exciting game and your team has the ball. Will your team score and win the game or fail to do so and lose? The outcome of any game ends with a change in status for each team. One team attains a positive incentive or win, while the other team fails to prevent a negative incentive or loss. Which outcome has the greater psychological effect? The win or the loss?
Losses Loom Larger Than Gains. Rather than considering incentives as absolutes, think of them as gains and losses (Kahneman & Tversky, 1979, 1982). With this thought, positive incentives become gains, and negative incentives become losses. For instance, a B grade earned in a course is a positive incentive if it increases a student’s GPA but a neg- ative incentive if it decreases GPA. Furthermore, the impact of incentives as gains and losses, although opposite, are not equal. Losses loom larger than gains, which means losses are more dissatisfying than gains are satisfying (Kahneman & Tversky, 1979, 1982). For instance, a $100 loss produces a greater decrease in subjective value than a $100 gain produces an increase. In other words, a $100 loss is more dissatisfying than a $100 gain is satisfying.
Buyer and Seller Experiments. A technique for determining the effects of losses and gains involves using identical measuring units to assess both outcomes. Money is a good measure because it reflects how much a person pays to gain (buy) an item and how much a person receives for the loss (sale) of an item. Using this approach, Kahneman and associ- ates (1990) conducted a marketing experiment during which they randomly gave half the students in a class some attractive coffee mugs imprinted with the name of the university. The new owners were informed that the mugs were theirs to keep or they could sell them. The nonowners were told to examine their neighbors’ mugs and that they could offer them any price in order to buy one. Owners (potential sellers) had to decide at what price they were willing to sell their mugs. Nonowners (potential buyers) had to decide how much they were willing to pay for the mugs. The purpose of the individual price setting was to deter- mine the subjective value of the loss of a mug when sold and the subjective value of the gain of a mug when bought. To summarize:
Willingness-to-accept price � Subjective value of a loss of the mug Willingness-to-pay price � Subjective value of a gain of the mug
If losses are more dissatisfying than gains are satisfying, then the sellers’ prices should be higher than the buyers’ prices. In other words, the decrease in value from losing (selling) a mug would be greater than the increase in value from gaining (buying) a mug. To make these experiments as realistic as possible, Kahneman and associates (1990) also let students barter for pens and folding binoculars in addition to coffee mugs. Students had been told to bring their own money to class in order to make any purchases, and provisions were made for extending credit and for making change. The results of several experiments are shown in Figure 10.2. For all commodities, a typical seller’s willingness-to-accept price exceeded a typical prospective buyer’s willingness-to-pay price (Kahneman et al., 1990).
Mugs Pens Binoculars
FIGURE 10.2 Loss/Gain and Buyer/Seller Behavior. Typical median willingness-to-accept prices were greater than median willingness-to-pay prices for mugs, pens, and binoculars. When seller and buyer agreed on price, money and goods changed hands.
Source: Adapted from “Experimental Tests of the Endowment Effect and the Coase Theorem” by D. Kahneman et al., 1990, Journal of Political Economy, 98, table 2, p. 1332, and table 3, p. 1334.
This difference in price indicates that the loss of a particular item was more dissatisfying than the gain of the same item was satisfying.
Section Recap Prior reinforcers and punishers provide the knowledge and anticipated incentives provide the motivation for what behavior to carry out in a particular situation. Reinforcers select a particular behavior, making it more likely to occur in the appropriate situation. Punishers, however, select against a particular behavior, making it less likely to occur. Incentives are external stimuli that induce or motivate the behavior to occur in a particular situation. People are motivated toward positive and away from negative incentives.
The degree an incentive motivates behavior depends on its incentive value, which is based on objective properties such as number or quantity. Subjective value or utility is based on the pleasure, satisfaction, or usefulness of the incentive. As objective incentive value increases, subjective value or utility increases but in smaller and smaller amounts, as described by Fechner’s law. Positive incentives can be interpreted as gains and negative incentives as losses. According to the view that losses loom larger than gains, the loss of an incentive is more dissatisfying than the gain of an incentive is satisfying.
Factors That Affect Incentive Value Storewide Clearance: 40% to 50% off original prices One Day Sale: Fantastic Storewide Savings Up to 50% Off Sale Every car, truck, and van in stock MARKED DOWN
244 P A R T F O U R / External Sources of Motivation
C H A P T E R T E N / Extrinsic and Intrinsic Motivation 245
Sale notices bring customers into stores and motivate them to buy. If a shirt has a sign marked “33% off the original price,” then shoppers are more likely to buy it than if it is listed for the same price but without the sale sign. For example, they are more likely to buy a shirt advertised at “$20, marked down from $30” than to buy it when simply priced at $20. Is it just the lower price or also the contrast between the original and the sale price that moti- vates shoppers to buy?
The purpose of this section is to describe some of the properties of incentives that motivate behavior.
Amount An important feature of incentive motivation is incentive amount, which refers to the quan- tity or number of incentives. Crespi (1942) demonstrated the effect of incentive amount on how fast rats would run down a straight alley to a goal box holding either 16, 64, or a whop- ping 256 food units. His rats ran fastest for 256 food units and slowest for 16 units. Kraeling (1961) provided rats with 2.5, 5.0, or 10.0% sucrose solutions as incentives for running down an alley. Her results indicated that rats ran consistently faster for the 10.0% sucrose concentration and slower for the 2.5% sucrose concentration, with an intermediate speed for the 5.0% concentration. Clearly, running, which is instrumental in obtaining the incen- tive, increases with an increase in the amount of the incentive.
Incentive value also affects the motivation for behavior not to occur. For instance, in- centives have been used to reduce smoking in university students. Correia and Benson (2006) recruited smokers in order to examine the motivating effects of incentive amount on smoking abstinence. Participants reported to the laboratory twice per day for one week and were paid in cash if their carbon monoxide level was below the criterion amount. Carbon monoxide is obtained from a person’s breath and indicates the amount a person has smoked recently along with reports of the number of cigarettes smoked and hours of not smoking. The amount paid defined the incentive value for not smoking: up to $40 versus $80 for the week. The results indicated that the larger incentive was more effective in reducing smok- ing than the smaller incentive. Carbon monoxide levels and the number of cigarettes smoked were lower for smokers who received the larger incentive. In addition, they also re- ported a greater number of hours of not smoking.
Rate of Reinforcement The frequency of an incentive or reinforcer also affects motivation. Rate of reinforcement is studied in situations where an animal is given a choice between two responses, one of which is reinforced more frequently. For example, pigeons are presented with two response keys to peck. Pecking one key yields 30 reinforcers per hour while pecking the other key yields 15 or 45 reinforcers per hour. If reinforcement rate affects motivation, then the pigeon should peck whatever key produces the higher rate. To verify this prediction, Herrnstein (1961) varied the rate of reinforcement between the left and right response keys. For example, pecking the left response key yielded 27 reinforces per hour while the right key yielded 13 reinforcers per hour. The results showed that pigeons chose the left response key more than the right one, since the former provided a higher rate of reinforcement. Humans are also affected by the rate of reinforcement. Neef and associates (1992) reinforced spe- cial education students for working on math problems from one of two stacks situated on
the students’ right and left. When students worked on one stack of problems they were reinforced with a nickel or token every 30 seconds, on average, compared to every 120 seconds from the other stack. The rate of reinforcement affected their choice. Students spent more time working math problems from the stack that provided a higher rate of rein- forcement. Thus, animals and humans shift to activities that provide higher rates of reinforce- ment (de Villiers, 1977). Melioration refers to a shift toward an activity that is more lucrative or provides a greater rate of reinforcement (Herrnstein, 1990). For example, if the rate of reinforcement declines for one activity, then a person switches to another activity that pro- vides a higher rate. To illustrate, a person may switch credit cards to one that offers a lower interest rate, go to another restaurant because the food is better, or change into a clean shirt.
One implication of melioration is that a faster rate of responding will produce a faster rate of reinforcement. Imagine a hungry rat in a Skinner Box that is reinforced with food each time it presses a lever. Faster rates of responding result in faster rates of reinforcement. A similar relationship exists in the “real world” between the amount of responding and the magnitude of the reinforcer. For example, enrolling in more courses results in quicker graduation, studying more results in higher grades, and working more hours produces larger paychecks. Lippman (2000) provided experimental demonstrations of this relationship. In order to earn points (reinforcers), participants pushed a button. All that was required for a fixed amount of reinforcement was to push the button once at the conclusion of a 15-second interval. However, additional button presses during the 15 seconds increased the size of the reinforcer in one condition while it decreased the size of the reinforcer in another condition. As might be expected, response rate increased when it provided for a larger rein- forcer and decreased when it produced smaller reinforcers. Lippman’s (2000) experiments showed that the amount of the anticipated incentive determines the rate of responding.
Contrast Effects How effective is showing the sale price alongside the old price in motivating shoppers to buy? How motivating is a wage of $10 per hour? Well that depends. Did the worker just re- ceive a raise from $8 per hour or a pay cut from $12 per hour?
Contrast of Incentive Amount. The phenomenon of contrast effects means that the ability of an incentive to motivate behavior depends on how it differs from prior incentives. Positive incentive contrast refers to an upward shift in value, which increases behavior above what is considered normal. For example, after receiving a substantial raise, workers perform more efficiently. Negative incentive contrast refers to a downward shift in value, which decreases behavior below what is considered normal. For instance, after their salaries were cut, workers performed less efficiently. Evidence for the effects of incentive contrast comes mainly from animal research. Pieper and Marx (1963) made either a 4, 11.3, or 32% sucrose solution available to rats in their cages. The sucrose concentration became the rat’s neutral point to which subsequent sucrose concentrations were contrasted. In the next phase of the experiment, rats were required to lever press in order to earn an 11.3% sucrose rein- forcer. For the rats accustomed to a 4% solution, 11.3% was an increase in incentive amount. They lever pressed at a faster rate than the rats accustomed to the 11.3% concen- tration. For the rats accustomed to a 32% sucrose solution, 11.3% was a decrease in incen- tive amount. They lever pressed slower than the 11.3% group. Thus, the motivating effects of an 11.3% sucrose solution depended on its contrast with the previous amount of sucrose.
246 P A R T F O U R / External Sources of Motivation
C H A P T E R T E N / Extrinsic and Intrinsic Motivation 247
Increases in incentive amount increased motivation, and decreases in incentive amount decreased motivation. However, although negative contrast has been observed consistently, the reliability of positive contrast has been questioned (Bolles, 1975; Dunham, 1968). For instance, if an animal is responding at close to maximum effort, then it would be difficult to show an increase with an upward shift in incentive amount. Similarly, if a salesperson has been working very hard already, it may not be possible to show an increase in sales with a pay raise.
Hedonic Contrast. Results of incentive contrast experiments bear a striking resemblance to the law of hedonic contrast, which Beebe-Center (1932/1965) attributed to the German psychophysicist Fechner (1876). According to this law, the pleasure a stimulus gives will be greater if it contrasts with sources of lesser pleasure or displeasure. Similarly, the displeasure a stimulus gives will be greater if it contrasts with sources of greater pleasure or less dis- pleasure. Contrast effects, however, also depend on how those other sources are categorized. Are they from the same or a different category? Zellner and coresearchers (2003) investi- gated the effects that category membership had on the hedonic evaluation of fruit drinks. Three groups of participants evaluated the pleasantness of eight fruit-flavored drinks followed by two diluted test drinks. The test drinks were either from the same or a different category than the prior eight. One group was told that all 10 solutions were commercial drinks from England—that is, the two test drinks were from the same category as the first eight drinks. Another group was informed that only the last two solutions were drinks from England—that is, the last two drinks were from a different category. A control group only evaluated the two diluted test drinks.
Contrast effects were apparent in two ways. First, the two diluted drinks were rated lower in pleasantness when they followed the eight full-strength solutions as compared to being the only two drinks. In addition, the test drinks were rated as more unpleasant when they were labeled as being from the same category compared to a different category. Zellner and coresearchers (2002) found similar results for coffee and beer. The contrast effect of ordinary coffee compared to gourmet coffee was greater when all coffees were la- beled as belonging to the same compared to different categories. Contrast effects were also greater for beers from the same category than from different categories. These category effects imply that people’s stimulus preferences can be manipulated through comparison with other stimuli. The contrast effects will be greater if people are told the comparison stimuli are from the same category rather than a different category.
Contrast effects also depend on a person’s expectations. For instance, students have experienced changes in their feelings as a result of the contrast between the exam grade they expected and the grade they received (Shepperd & McNulty, 2002). Imagine a class where you have taken an exam and that you expected an A. When the professor returned the exam, you found that you earned an A (or C). How does that make you feel? Or, imagine that you expected a C on the exam and found that you earned a C (or A). How does that make you feel? More precisely, how unhappy or happy would you feel about expecting an A and receiving an A or C and expecting a C and receiving a C or A? Use the following scale: 1 � Very unhappy to 7 � Very happy. The hedonic contrast effects found in such an experi- ment as this are presented in Figure 10.3 (Shepperd & McNulty, 2002). An A grade produces the greatest happiness if contrasted with an expected C rather than an expected A. And a C grade produces the least happiness if contrasted with an expected A rather than an expected C.
Expected A Expected C
Received C H
FIGURE 10.3 Happiness and Expected Grades. Happiness ratings were highest when the received grade was higher than the expected grade (expected C, received A). Happiness ratings were lowest when the received grade was lower than the expected grade (expected A, received C). Ratings were intermediate when the received grade matched the expected grade, although an A produced greater happiness than a C.
Source: Adapted from “The Affective Consequences of Expected and Unexpected Outcomes” by J. A. Shepperd and J. K. McNulty, 2002, Psychological Science, 13, table 1, p. 86.
Temporal Motivation Theory You have known for two weeks that there will be an exam on this material but you have not started studying in earnest until two days before. Why is that? One answer is that incentives are based in the future. This feature has tremendous negative impact on incentive value. Although incentive value increases with amount, it decreases with delay. How do these two features combine to determine the value or utility of a future incentive? Remember that value refers to the amount of an incentive while utility refers to its usefulness. Thus, $50 has greater value than $10 but it also has greater utility, since more goods and services can be purchased with $50.
Incentive Utility from Amount and Delay. Several mathematical formulas express how the utility of an incentive changes with delay (Green & Myerson, 2004; Mazur, 1987; Steel & König, 2006). As part of their temporal motivation theory, Steel and König (2006) provide the following formula:
Utility � Expected incentive value
(1 � Delay interval)
Temporal motivation theory integrates how incentive utility changes temporally (with time) (Steel & König, 2006). Utility in this formula refers to the ability of a positive
248 P A R T F O U R / External Sources of Motivation
C H A P T E R T E N / Extrinsic and Intrinsic Motivation 249
incentive to motivate approach behavior and a negative incentive to motivate avoidance behavior. The term expected in the formula refers to the likelihood that the incentive will occur. The expectation is that the incentive is promised or likely to happen although not guaranteed. For example, a job interview may result in a job and diligent studying may result in an A. However, the job or the A are not guaranteed. Finally, the denominator of the formula represents the temporal distance to the incentive—that is, the incentive delay interval.
This interval refers to the time between current behavior and the availability of a future incentive. For example, there is a delay between a telephone call for a date and the actual date or between studying and taking an exam. The Saturday night date (the incentive) occurs some time after (the delay) the telephone call. The exam is scheduled some time after a student has studied. As a consequence of this delay, incentives lose value. A future incentive is represented in the present at a marked-down value by a process known as delay discounting or temporal discounting (Green & Myerson, 2004; Myerson & Green, 1995). The amount of discounting increases with the length of the delay interval. For instance, the value of a Saturday night date or an exam depends on how far in the future they are. The later they occur, the lower their value in the present. However, as their due time approaches, their value increases. As Saturday night approaches, the value of the date increases, and as exam time approaches, the value of the exam increases.
According to the utility formula, as the incentive delay interval decreases (approaches 0), the utility of the incentive increases for both positive and negative incentives. Figure 10.4 illustrates this relationship. The increase in utility is the result of a decline in delay discounting. When the incentive is reached, the delay interval equals zero (delay � 0), and the incentive value equals the amount of the incentive. At this point, a positive incentive becomes a reinforcer, and a negative incentive becomes a punisher (Rachlin, 1989). However, their values are not equal. The absolute value of a negative incentive is greater than the absolute value of an identical positive incentive. The reason for this is based on the concept losses loom larger than gains, which was described earlier. Thus, the negative- incentive curve in Figure 10.4 declines more steeply than the positive-incentive curve rises (Knetsch & Sinden, 1984; Miller, 1959).
Some experiments illustrating delay discounting require participants to make a choice between two monetary values. For example, they might be asked to choose between $800 right now or $1,000 after six months and between $600 right now or $1,000 after one year. By examining the pattern of choices, researchers have determined the discounted value of $1,000 and $10,000 after delay periods ranging from zero to 25 years (300 months) (Green et al., 1994a; Myerson & Green, 1995; Rachlin et al., 1991). The curve of discounted values across delay periods resembles the positive-incentive curve in Figure 10.4. In other words, as the delay interval of receiving $1,000 or $10,000 increases, the discounted value of that money decreases. Delay discounting indicates that participants are more likely to accept a smaller amount immediately rather than larger amounts after a delay. In other words, we want our money immediately rather than waiting, even if it means receiving less.
Preference Reversal. Have you ever planned to do one thing only to change your mind later? For example, you planned to get up early to study for an exam but when the alarm went off you decided to remain in bed. This example illustrates a reversal of incentive or
9 8 7 6 5 4 3 2 1 010
� Punisher = –$100
Reinforcer = +$100
Positive incentive Negative incentive
Number of Incentive Delay Intervals
FIGURE 10.4 Incentive Utility and Delay. The values of both a positive incentive (+$100) and a negative incentive (�$100) increase as their delay intervals decrease. When the incentives are reached (zero delay), the positive incentive equals the reinforcer (gain $100), and the negative in- centive equals the punisher (lose $100). The subjective absolute value of the negative incentive is greater than that of the positive incentive.
250 P A R T F O U R / External Sources of Motivation
reward preferences. Generally, people prefer large incentives over small incentives and immediate incentives over delayed incentives. When amount and delay are combined, how- ever, a person may prefer a small immediate incentive (sleep) over a larger delayed incen- tive (good grades). Then, as the delay interval lengthens, preference shifts from the small immediate incentive to the large delayed incentive. This process is known as preference reversal (Green et al., 1994a; Green & Myerson, 2004; Steel & König, 2006). For example, the night before a student prefers a good grade (the larger, more delayed incentive) over extra sleep (the smaller, less delayed incentive). However, the next morning a student prefers extra sleep (the smaller incentive) because it is available immediately (no delay). Figure 10.5 illustrates preference reversal between time x and time y. During time x, the larger delayed incentive has a higher value, but as the delay interval decreases, the smaller immediate incentive attains a higher value during time y. Thus, preference reversal occurs because of the change in value between two incentives over time.
In a demonstration of preference reversal, participants were presented with choices between different dollar amounts that were available after various delays (Green et al., 1994a). The choices pitted smaller against larger dollar amounts, such as $20 versus $50, $100 versus $250, and $500 versus $1,250. The choices also pitted shorter delay intervals
Decreasing Incentive Delay Interval
Time x Time y
Smaller incentive Larger incentive
FIGURE 10.5 Changes in Incentive Value or Utility Over Time. During time x, the larger incentive has a greater value than the smaller incentive. As the delay interval decreases, the smaller incen- tive attains the higher value during time y.
against longer delay intervals ranging from zero to 20 years. As an illustration, decide between each of the following six choices:
Choice 1. $25 now versus $50 now Choice 2. $25 now versus $50 in 2 weeks Choice 3. $25 now versus $50 in 4 weeks Choice 4. $25 now versus $50 in 6 weeks Choice 5. $25 now versus $50 in 8 weeks Choice 6. $25 now versus $50 in 10 weeks
Assume that a hypothetical individual preferred $50 on choice 1 and also preferred $50 on choices 2 through 4—that is, he preferred the larger delayed reward over the smaller imme- diate reward. If on choices 5 and 6 the person selected $25, then he reversed and preferred the smaller immediate reward over the larger delayed reward. This preference reversal occurred between choices 4 and 5 because the larger reward of $50 was discounted at the longer delay intervals so that its value fell below the value of an immediate $25.
Procrastination. Waiting until the last hour to file a federal tax return, not studying for an exam until the last opportunity, or finishing a paper until late the night before it is due are examples of procrastination. However, is procrastination merely putting off until later what can be done now? One perspective is that “to procrastinate is to delay an intended course of action despite expecting to be worse off for the delay” (Steel, 2007, p. 66). This definition indicates that procrastination depends on the temporal distance of the important activity and the state of being “worse off.” For example, an essay that is due in one week has high positive utility and being “worse off ” is also high. Not writing the essay in this case would be an instance of procrastination. Steel (2007) presents a hypothetical illustra- tion of student procrastination, which is summarized in Figure 10.6. A student is assigned an essay on September 15 that is due on December 16, a 92-day delay. The expected
C H A P T E R T E N / Extrinsic and Intrinsic Motivation 251
Socializing December 3rd
Oct. 08 Oct. 31
Nov. 23 Dec. 16
FIGURE 10.6 Changes in Utility of Socializing versus Essay Writing. This graph shows a stu- dent’s utility estimation of socializing versus writing an essay over the course of the semester. The utility of socializing remains constant, since that activity is always available. The utility of writing an essay increases as the essay’s due date approaches—that is, the delay interval decreases.
Source: From “The Nature of Procrastination: A Meta-Analytic and Theoretical Review of Quintessential Self- Regulatory Failure” by P. Steel, 2007, Psychological Bulletin, 133, figure 1, p. 72. Copyright 2007 by American Psychological Association. Reprinted by permission.
incentive value of the essay is based on its letter grade and the likelihood that the student’s efforts will produce it. However, how high should the utility of the essay be so that a student will work on it compared to other activity-related utilities? In Figure 10.6, socializing is always available either down the hall, across the campus, or with a phone call. So a student socializes because its utility is higher than essay writing. However, on December 3, 12 days before the essay is due, its utility exceeds that of socializing and the student begins to work.
The formula [Utility � (Expected incentive value) / (1 � Delay interval)] helps show what factors will affect procrastination. First, the lower incentive value of a task, the more people will procrastinate on it. For example, students are more likely to procrastinate on as- signments they consider less interesting (Ackerman & Gross, 2005). Second, as indicated in Figure 10.6, when procrastinating on a task individuals are doing something else, such as socializing. Thus, procrastination should increase for alternative activities that have greater value. For example, students procrastinate with activities that they consider more pleasant, less stressful, and less difficult than their assignments (Pychyl et al., 2000a). They procras- tinate by watching TV, sleeping, talking, playing, eating, or working.
What are some examples of being “worse off ” that result from procrastination? One ob- vious illustration is that procrastination results in less time in which to accomplish one’s goal, assignment, or project. One way of defining procrastinators is to use various available psycho- logical scales such as Lay’s (1986) General Procrastination Scale. Students identified as pro- crastinators, for example, tend to predict that they will study less for an exam and do so. In addition, they are more likely than nonprocrastinators to cram by squeezing more of their study
252 P A R T F O U R / External Sources of Motivation
C H A P T E R T E N / Extrinsic and Intrinsic Motivation 253
time into the last three days (Pychyl et al., 2000b). In general, investigations have concluded that procrastination does make a person worse off. Rothblum and associates (1986) found that students identified as procrastinators had lower GPAs for the semester, delayed taking self- paced quizzes, and had more anxiety or anxiety-related symptoms that accompanied their pro- crastination. Tice and Baumeister (1997) also showed that procrastinators as defined by Lay’s (1986) General Procrastination Scale turned in their papers later or late, earned lower scores on their papers, and earned lower exam scores in the course. In addition, procrastination took its toll on the procrastinators at the end of the semester. Procrastinators reported more stress, more symptoms, and more health center visits than did nonprocrastinators (see Figure 7.5).
➣ Lay’s (1986) General Procrastination Scale is available at http://www.yorku.ca/rokada/ psyctest/prcrasts.pdf
Section Recap Incentive amount refers to the objective quantity or number of stimuli that serve as incentives. Usually, as incentive amount increases, indicators of motivation also increase. Incentives also affect choice behavior, as shown by the fact that an animal or person will shift to a better in- centive if one is available. According to the concept of melioration a person keeps shifting to alternatives that provide a higher rate of reinforcement than the current one. The phenomenon of contrast effects means that the motivational effects of an incentive depend on the organism’s experience with prior incentives. Positive incentive contrast refers to a sudden increase in the incentive that results in a sudden upward shift in motivation. Negative incentive contrast, however, refers to a sudden decrease in the incentive that results in a sudden downward shift in motivation. The law of hedonic contrast refers to the subjective feelings that accompany incentive contrasts. Positive contrast produces pleasant feelings, while negative contrast produces unpleasant feelings.
According to temporal motivation theory, the utility (usefulness) of an incentive de- pends on the value of the expected incentive and when it becomes available in the future. This relationship is expressed in the formula: Utility � (Expected incentive value) / (1 � Delay interval). The incentive delay interval represents the time span between the present and the availability of the incentive in the future. Delay discounting is the process by which the future incentive is represented in the present but at a reduced (discounted) value. Thus, longer delays result in greater discounting, which can result in preference reversal. This means that initially an individual prefers incentive A over B, but as the delay interval de- creases, preferences reverse and incentive B is now preferred over A. Procrastination is a form of preference reversal. The important task has been discounted to the extent that an immediate alternative activity is preferable.
Intrinsic Motivation Not all behavior is done for extrinsic incentives. Many activities are done for no apparent external reasons. To illustrate, consider author John Irving’s motivation for writing:
The unspoken factor is love. The reason I can work so hard at my writing is that it’s not work for me. Or, as I said before, work is pleasure to me. I work, and always have, quite obsessively. I can’t just write for four or five hours and then turn the book off. I wake up in the middle of the night and I’m writing it. (Amabile, 1989, p. 56)
Consider the motivation for dancing to rock-and-roll music expressed by a dancer:
When you feel [the music] resonating through you, it really helps. Cause when it’s loud, like when you’re dancing to a rock group, you can really feel yourself vibrate almost. And also, the louder it is, the more it blocks out other noises, so it’s more of a total immersion in the music, which is also a very good sensation and is also conducive to just dancing and being part of the music—almost incorporating it. (Csikszentmihalyi, 1975, p. 105)
In these examples, the motivation for writing or dancing seems inherent in the activity. External incentives are not apparent.
The purpose of this section is to examine differences between motivation for an activ- ity that seems inherent or intrinsic to that activity and motivation that is external or extrinsic to the activity. Another purpose is to describe how these two types of motivation interact.
Differences between Extrinsic and Intrinsic Motivation For what reasons are you reading these pages? To earn a good grade on the exam or to please your parents? To feel good about yourself or to reduce anxiety about a forthcoming exam? To learn the material so that it can be put to future use? To act in accord with your self- concept as a student? To satisfy curiosity or because the material is interesting?
The source of motivation defines whether behavior is extrinsically or intrinsically motivated. Extrinsic motivation comes from an external source, such as money, good grades, or the approval of others. Intrinsic motivation, in contrast, does not come from an external source but instead is inherent in the activity being performed. Extrinsically moti- vated behavior is coerced (forced) by environmental contingencies, while intrinsically motivated behavior is freely chosen (Deci & Ryan, 1985). Money forces a person to work. A passing grade on an exam forces a student to study. Intrinsically motivated behavior, how- ever, is not forced by environmental events. The activity is performed because the person seemingly wants to experience the activity for its own sake. Thus, individuals may play ten- nis, bowl, or watch a movie simply for the sake of doing that activity and not because they were forced to do it in order to obtain something. A distinction also should be made between psychological needs (Chapter 8) and intrinsic motivation. For instance, a person could be motivated to play tennis to satisfy needs of achievement, affiliation, and power. Tennis played for these reasons would be extrinsically motivated, however, since satisfying these psychological needs is external to playing the game. However, playing tennis for the enjoy- ment that the game provides would be an instance of intrinsic motivation.
Could a student who enrolled in a course because it was required end up liking it? Is it possible that although the original reason for doing an activity is no longer relevant the person still performs the activity anyway? Behavior that began for extrinsic reasons later can be performed for its own sake—that is, for intrinsic reasons. This shift in motivation de- scribes Allport’s (1937a) concept of the functional autonomy of motives. Allport charac- terizes the idea as “what was a means to an end has become an end in itself ” (Allport, 1937a, p. 150). Allport uses the example of doing fine work. An individual may feel compelled to do an excellent job even though job security or pay are no longer dependent on performance as was originally the case. In the case of college work, a student may enroll in a course to please her parents or because it comes at a convenient hour. Yet the course may prove to be so interesting that the student decides to major in the field and make it her life’s profession.
254 P A R T F O U R / External Sources of Motivation
C H A P T E R T E N / Extrinsic and Intrinsic Motivation 255
Purpose of Intrinsically Motivated Behavior Is intrinsically motivated behavior an end in itself? Or is it linked to other sources of moti- vation? For instance, intrinsically motivated behavior may be in service to satisfying curiosity and developing competence and self-determination. Curiosity provides the moti- vation to learn about the environment (Woodworth, 1958), while motives for competence and self-determination provide reasons to control the environment (White, 1959).
Curiosity. Can you avoid eavesdropping on a conversation about someone’s sexual es- capades at the next table? Can you put off opening a first-class letter you received or not bother answering your telephone? These questions illustrate the power of curiosity to intrin- sically motivate behavior (Loewenstein, 1994). A stimulus situation can evoke an intense feeling of curiosity. For example, a suspenseful mystery novel is difficult to put down. Curiosity also makes people act impulsively in their desire to seek information, even against their better judgment. For example, a person might eavesdrop on the conversation at the next table at the risk of being rude and being embarrassed if caught. Furthermore, curios- ity can make people act against their own self-interest. For example, a person may experi- ment with an illegal and potentially dangerous drug. Although curiosity can be classified as an instance of intrinsic motivation, it is linked to a person’s desire to know. Curiosity re- sults when the amount a person wants to know about a situation exceeds his current knowl- edge. A situation evokes curiosity to the extent it has the capacity to resolve this gap in knowledge (Loewenstein, 1994). The conclusion of the mystery novel provides the “who done it.” The letter provides information about its content. By answering the telephone or using “caller ID,” a person will discover the identity of the caller. Thus, satisfying one’s curiosity is the motive for exploring and learning about the environment.
Effectance Motivation. A person plays games, pursues hobbies, or spends time with friends simply because these activities are pleasurable or intrinsically motivating. However, intrinsic motivation is also in service to the development of competence. Intrinsically motivated behavior guides an animal or child to act effectively on its surroundings—that is, to become competent (White, 1959). The increase in competence is gained from such activities as exploring novel objects, playing, crawling, walking, speaking, writing, and thinking. These activities result from effectance motivation, which is the motive to actively interact and control one’s environment (White, 1959). For example, Piaget (1951/1976) refers to a child’s mastery play resulting from the mere joy of conquering a particular be- havior. A child plays to make things happen and gain control over those happenings.
A simple experiment illustrates effectance motivation in infants who were eight weeks old at the start of the two-week experiment (Watson, 1972/1976). One thing the infants could do at that age was to press their heads back against their pillows, and so this response was used in the experiment. Watson (1972/1976) devised a mobile that was placed above the infants in their crib at home. Some mobiles moved and others did not when the infants pressed their heads back against their pillows. In the contingent condition, pressing against the pillow caused the overhead mobile to turn. In the noncontingent condition, the pillow pressing response had no effect on turning the mobile. Although the mobile moved on occasion, it did so independent of any pillow pressing by the infant. In the stable condi- tion, the mobile remained stationary regardless of what the infant did. Thus, in the contin- gent condition the infant caused the mobile to move, while in the noncontingent condition,
256 P A R T F O U R / External Sources of Motivation
the infant did not cause the mobile to move; it moved on its own. Being able to have an ef- fect on the mobiles motivated the infants. They increased their frequency of pillow press- ing in the contingent condition only, not in the noncontingent and stable conditions. In addition, infants in the contingent condition began cooing and smiling as if they were enjoying their mastery over the movements of their mobiles. Infants in the other conditions cooed and smiled much less at their mobiles. Thus, infants as young as 8 to 10 weeks are intrinsically motivated to have an effect on their environments.
Flow. Competence is also gained by pushing oneself successfully to the limits of one’s capability. To illustrate, on more than one occasion I have heard a student say, “I work hard in a course I like.” Couldn’t the reverse also be true? That is, “I like a course because I work hard in it.” The fit between a person’s efforts and success at an activity produces flow (Csikszentmihalyi, 1975, 1988; Csikszentmihalyi & Rathunde, 1993). Flow refers to the desirable subjective state a person experiences when completely involved in some challenging activity that matches the individual’s skills. The activity has a clear goal and provides immediate feedback regarding the caliber of one’s performance. In addition, a person is enjoyably concentrating all attention on the activity so that time, fatigue, and everything else is disregarded (Csikszentmihalyi, 1975, 1988; Csikszentmihalyi & Rathunde, 1993). Flow also has been used as a metaphor to describe the feelings experi- enced by artists, athletes, composers, dancers, scientists, and others when engaged in their favorite activities of doing something for the sake of doing: “they are in their flow.”
What happens when personal capabilities exceed or fall below the demands of a task? The answer is that individuals are now out of their flow. People are aware of opportunities that challenge them as well as their capabilities of handling that challenge. When the challenges of the task exceed the person’s capabilities, then stress is felt as anxiety and worry. However, when the challenges of the task are easily met, then a person feels bored. Flow, however, is experienced when the individual’s skills fit with the demands of the task (Csikszentmihalyi, 1975, 1988; Csikszentmihalyi & Rathunde, 1993). To illustrate, canoeing, skiing, and play- ing chess can be performed at various levels of difficulty. However, whether any of these activities produce flow depends on the match between the difficulty level and the person’s capability. If the water is too swift, the slope too steep, or the opponent too skilled, then the canoeist, skier, or chess player experiences stressful reactions like anxiety and worry. However, when the water is too slow, the slope too flat, and one’s opponent less skilled, then the person generally experiences boredom. These activities, however, provide maximum enjoyment and flow occurs when there is a match between the challenge of the activity and the person’s capability level (Csikszentmihalyi, 1975, 1988; Csikszentmihalyi & Rathunde, 1993). The personality psychologist Gordon Allport (1937a) also noted the posi- tive relationship between ability and interest. He remarked, “A person likes to do what he can do well” (p. 150).
Interaction between Extrinsic and Intrinsic Motivation What happens when extrinsic and intrinsic motivation combine? When amateur athletes be- come professionals, does money make them enjoy their sports more or less? Do retired pro- fessional athletes still play their sport for fun? The combination of extrinsic and intrinsic motivation affects the enjoyment and performance of behavior differently (Tang & Hall, 1995; Wiersma, 1992).
C H A P T E R T E N / Extrinsic and Intrinsic Motivation 257
Extrinsic Reward and Intrinsic Motivation. According to one major league outfielder, “I used to enjoy playing baseball until I started getting paid for it” (Arkes & Garske, 1977, p. 251). This quote previews some findings regarding the effects of extrinsic rewards on intrinsically motivated behavior (Deci, 1971, 1972a, 1972b). In his research, Deci tested the effects of extrinsic rewards on the intrinsic motivation for solving interesting and challenging three-dimensional puzzles. In general, the research involved three different reward condi- tions. In the contingent payment condition, participants were told they would be paid for each puzzle completed. In the noncontingent payment condition, participants were unexpectedly paid after the puzzle-solving session ended. The amount paid, however, did not depend on the number of puzzles solved. In the no-payment condition, participants did not receive any money for solving the puzzles.At the end of the puzzle session, the experimenter left the room on some pretext for several minutes, during which time participants were given a choice: they could smoke, read magazines, do nothing, or continue working on the puzzles, although now no money was given. During this time the participants were unobtrusively observed to see what they would do. Deci reasoned that participants would continue working on the puzzles during this time to the extent such activity was intrinsically motivating. The results confirmed this supposition. Participants in the contingent payment condition spent less time working on the puzzles than participants in the noncontingent payment and no-payment conditions. The conclusion was that when payment was contingent on performance, intrinsic motivation de- creased, whereas noncontingent or no payment did not affect intrinsic motivation. Deci (1972a, 1972b; Deci & Ryan, 1985) interpreted these findings in terms of cognitive evalua- tion theory. According to this theory, a person’s intrinsic motivation for a task depends on what is perceived to be the reason for the behavior. If the reason is perceived as external, as in the case of being paid for a task, then a person perceives the reward as the basis for doing the task. Consequently, when payment ceases, the behavior ceases. In this case, intrinsic motiva- tion declines, since the individual reasons that his behavior is externally rather than internally controlled. However, when no contingent external rewards are perceived, then the person concludes that her behavior is under self-control, which means it is freely chosen and is intrinsically motivating. Extrinsic rewards can also undermine the intrinsic enjoyment in chil- dren. For instance, Lepper and associates (1973) found that children who had been rewarded for drawing would spend less time drawing on a future occasion compared to children who had not been rewarded.
Extrinsic Reward, Intrinsic Motivation, and Performance. The decline in intrinsic motivation from extrinsic rewards comes from experiments in which a person is given a choice at the end of the contingency payment session. Besides choice, motivation is also reflected in performance measures of behavior. Although extrinsic reward undermines in- trinsic interest, it facilitates performance measures of motivation (Wiersma, 1992). For ex- ample, what would happen to performance on an intrinsically motivated task if the person started receiving contingent pay? Such a question was asked by Harackiewicz and associ- ates (1984, study 2) in their investigation of students who engaged in the intrinsically motivating task of playing pinball. The expected-reward group was offered and received a reward (movie pass) if its pinball playing performance was at the 80th percentile level. The unexpected-reward group was not told anything about a reward (movie pass) and unexpect- edly received it anyway at the end of the game. A no-reward group neither expected nor re- ceived any kind of reward for its playing. At the conclusion of one game of pinball, the
258 P A R T F O U R / External Sources of Motivation
FIGURE 10.7a Extrinsic Reward and Interest. The graph shows mean number of pinballs played during a free period. Fewer balls were played dur- ing the expected-reward condition, which indicates that extrinsic reward undermined intrinsic interest of the game.
Source: Adapted from “Rewarding Pinball Wizardry” by J. M. Harackiewicz et al., 1984, Journal of Personality and Social Psychology, 26, p. 294.
FIGURE 10.7b Extrinsic Reward and Perfor- mance. The graph shows scores earned playing pinball during a free period. Pinball Performance was greatest in the expected-reward condition, which in- dicates that an extrinsic incentive can enhance per- formance based on intrinsic motivation.