Appendix B
WHAT IS INTRINSIC MOTIVATION?

 

1. For the purpose of analyzing the effects of rewards, it may be enough to define intrinsic motivation as the desire to engage in an activity for its own sake — that is, just because of the satisfaction it provides.1 This offers a nice, crisp contrast to extrinsic motivation, which means that one takes part in the activity because of some other benefit that doing so will bring. However, if we look at the question more closely, things become a good deal more problematic and, for some of us anyway, more interesting. I want to explore a few of the questions, disputes, and complications that arise in thinking about intrinsic motivation (IM).
2. The very idea of IM is controversial in some quarters because of its implicit affirmation that what people do isn't always initiated by forces outside the self. While such forces can explain some of our behavior, it's also possible, and even necessary, to appeal to other motivational systems, which focus on what is inside of us. Behaviorists, not surprisingly, have not looked kindly on discussions of IM because its premise challenges the core of their belief system. They have reacted either by denying its existence or, what comes to the same thing, trying to collapse it into their own framework. If something looks intrinsic, they insist, it's just because we haven't yet figured out the real (extrinsic) causes. (For more on the behaviorists' responses, see Appendix C.)²
3. Apart from this dispute, there are quite a few controversies among those who take the concept of IM seriously but can't seem to agree on exactly what it means. To begin with, we need to decide whether it is to be defined negatively (specifically, as that which is present when an individual does something without expecting a reward), positively, or both. The negative definition is convenient for conducting experiments, and it is used explicitly or implicitly by a number of researchers: they record how often or how long subjects engage in an activity when no extrinsic benefits from doing so are expected. This technique, as I will explain later, has raised a number of troubling issues.
4. [p. 291] Defining IM positively, which is favored by more theorists, opens up a Pandora's box (not to say a Skinner box) of difficulties. The major question is whether we understand the concept in terms of a desire to engage in a particular task or in terms of certain qualities and more general motivations that define human beings. If the latter is chosen, we naturally will want to know what those qualities are. The nominees include a desire "to feel good,"3 "an orientation toward learning and mastery, "4 and a need for competence and self-determination5 as well as, perhaps, to relate to and be engaged with others .⁶
5. Whether any or all of these needs can be shown to be innate or universal, and if so, whether they are indeed the most fundamental human motivators are questions for another day. What interests me now is the relation between any such characteristics and the topic of intrinsic motivation. It would seem that the connection depends largely on the question we are asking. If we want to know what it means to say that humans are intrinsically motivated organisms, or why they want to do so many things, it might indeed be useful to try to identify some primary drives or needs. But if all we want to know is what it means to say that people are (or this person is) intrinsically motivated to pursue this particular task, then it may be enough to answer in terms of the appeal that this task holds. The definition offered in the first sentence of this essay, which doesn't bother to postulate, say, a basic need for humans to be challenged, might be sufficient for this purpose.
6. Once our definition of IM goes beyond someone's desire to perform a particular activity, we begin to run into other problems. One is that the wider human goals designated as intrinsic (such as exploring the environment or expressing oneself) may actually interfere with an individual's focus on a specific task. I may have to choose between satisfying my basic curiosity and attending to the job I'm doing at the moment. The two approaches to understanding IM may, in other words, tug in opposite directions .⁷
7. Another problem is that it is sometimes unclear whether a given characteristic defines IM or is only empirically associated with it. Either intrinsically motivated people turn out to be autonomous (or vice versa), or else autonomy is part of what we mean by the phrase intrinsically motivated. If we try to have it both ways, our argument becomes circular. Even people who have written extensively about the topic occasionally seem confused about whether IM entails task involvement by definition or whether IM (defined some other way) promotes task involvement.
8. [292] Finally, we will have to decide whether to build situational elements into our understanding of IM. To take a concrete example, people often lose interest in a task when they keep doing it over a period of time.8 Does this mean that someone was not really intrinsically motivated to work on that task after all? Or is IM itself partly a function of novelty? Since interest is so variable, must we avoid attributing motivational properties to activities? And what about attributing the motivation to perform an activity to individuals, since that too depends on the circumstances?

 

9. Let us put these conceptual problems aside and look at a very practical issue: how IM in its task-specific sense is measured. Beginning with Deci's early experiments, there have basically been two techniques: asking people how they feel about the activity and watching to see how much time they spend on it (in the absence of extrinsic factors) when given a choice. These are two appealingly straightforward ways of getting at the idea of IM — or so it would seem.
10. In fact, neither technique is without problems. Self-report measures, while undeniably useful, raise the question of whether people are describing their feelings accurately — or in some cases, whether they even know what they feel (which unleashes a host of philosophical questions). Do experimental subjects exaggerate how much they enjoyed what they were asked to do because this is the answer they think the researcher wants to hear? Do some kinds of people do this more often than others, raising the possibility that a measure of IM is actually gauging something else entirely?
11. The "free choice" measure — secretly observing people to see whether they return to an activity when they're not obliged to do so — carries its own difficulties. The length or proportion of time spent on the target task varies depending on a range of situational and dispositional factors, including how appealing the available alternative activities happen to be to the individual. The researcher wants to know whether you will continue to play a game when left alone with it for five minutes or whether you will read a magazine. But the answer may say as much about which magazines are in the room as it does about your interest in the game.⁹ In experiments where subjects are not left alone, moreover, they may be inclined to keep playing the game partly out of a desire to please the experimenter.
12. As if to emphasize the risk of putting our faith in either or both of these techniques for quantifying IM, one study after another has found that the two may not point in the same direction: the correla-
13. [293] tion between self-report and behavioral measures is often negligible.10 This fact led Ryan and Deci to reflect on what is implied by a situation where someone keeps working on an activity in the absence of extrinsic inducements. Is this in itself a valid indicator of IM? The answer, they decided, is probably not, and with this conclusion comes a bundle of new questions about motivation.
14. All of us are familiar with people who drive themselves mercilessly to achieve, who approach their work with a compulsiveness that led to the introduction of the term workaholic. What is interesting for our purposes about this style of task engagement is that it does not depend on the anticipation of receiving rewards or punishments from the environment. The pressure is internally generated and yet devoid of "the genuine interest, enjoyment, and excitement that phenomenologically define intrinsic motivation."11 Could this description fit some of the experimental subjects who report low interest in an activity while nevertheless continuing to work at it on their own time?
15. Ryan and his colleagues set out to answer that question. They told some subjects that the task they were being given to do would reveal how intelligent they were; others were just encouraged to become involved in the task without feeling that their egos were on the line. It turned out that there was no match between the self-report and behavioral measures of interest (what they said about the activity and whether they continued working on it) for the former group. If people felt anxious about whether they were any good at the task, they were more likely to keep at it, presumably more to "preserve their self-worth" than because it was intrinsically motivating.  "Although free-choice behavior is a reflection of intrinsic motivation" when people are encouraged just to explore the activity, "it is more of a reflection of internally controlling regulation in conditions of ego involvement," they concluded.12 Internal does not always imply intrinsic.
16. This conclusion offers a direct challenge to psychological theories that distinguish only between what is inside a person and what is outside. It forces us to reconsider not only how we measure IM but what the concept really means. And it raises questions, as I suggested in chapter 12, about whether it is enough just to get children to "internalize" norms and values. After all, feeling controlled from the inside isn't much of an improvement over feeling controlled from the outside.
17. This last point reminds us that IM, for all its importance, is sometimes irrelevant to the questions that matter to us as parents and teachers. We want children to put their own needs aside sometimes
18. [294] and do what is in the interest of a larger group of people, to respect the rights of others even when this involves inconveniencing themselves. Such behavior is not really analogous to, say, reading, since it is not something in which one develops or maintains an intrinsic interest (although it is sometimes described as intrinsically valuable). In his description of life in an American classroom, Philip Jackson observed that "it is hard to imagine that the students will ever find anything intrinsically satisfying about being silent when they wish to talk," a fact that suggests "the notion of intrinsic motivation begins to lose some of its power"13 — or at least its pertinence to nonacademic issues.
19. If, however, we see the world only in terms of intrinsic versus extrinsic motivation, we will be inclined to think that whatever is not described by the former must be described by the latter. Practically speaking, that means we may resort to extrinsic motivators to induce children to act responsibly. But this dichotomous view overlooks the possibility of helping them internalize a commitment to such actions, and doing so in such a way that they come to feel a sense of self-determination about the matter and ultimately are able to decide for themselves what kind of people they want to be.14

 

20. The need to introduce a concept such as internalization implies that the intrinsic/extrinsic dichotomy is not exhaustive. In fact, we can come up with other real-life situations that can't readily be classified as one or the other. Consider a scientist whose prime motive is to contribute to her field, or a labor organizer who is interested in fighting injustice and helping working people. Or imagine a student in a classroom where learning often takes place in cooperative groups: he finds the activity extremely engaging, but mostly because of the pleasure derived from working with others.
21. In these examples, people are not seeking what we would ordinarily call extrinsic rewards, yet neither are they motivated by the tasks themselves. The scientist may not particularly enjoy the laboratory work, the organizer may not be enthusiastic about making phone calls and attending meetings, and the student may not be delighted with the math assignment per se.
22. I think what explains our frustration in deciding how to categorize these people is a quality of intrinsic motivation that has gone largely unnoticed by psychologists: it is a concept that exists only in the context of the individual. The scientist, the organizer, and the student are all motivated by social concerns, and these don't easily fit into
23. [295] such a paradigm. Forced to choose, we would have to say that these motives are extrinsic to their tasks rather than intrinsic. But the fact that we have to lump such motives with the quest for money or grades reveals a limitation in the idea of IM. It is an idea that was never intended to apply to something beyond the needs of separate selves. That this point has rarely even been raised among researchers and theorists may say something about the pervasiveness of an individualistic framework in psychology.15
24. On the other hand, I am not convinced that we ought to discourage people from being intrinsically interested in what they are doing. For example, writers who love the act of writing are not necessarily being self-indulgent and oblivious to larger social concerns. Even less does it follow from the importance of living in a world with others that we should promote the use of extrinsic motivators. We depend on each other for emotional support and for feedback about what we have done, but neither of these has to take the form of rewards.
25. Very similar to the concern about IM's focus on the individual is the idea that someone intrinsically motivated is caught up only in the process of what he is doing, to the exclusion of the product. To be sure, our society encourages a preoccupation with the product, the bottom line, the practical result. Thus, Csikszentmihalyi finds it refreshing "when experience is intrinsically rewarding [because then] life is justified in the present, instead of being held hostage to a hypothetical future gain."16 However, it was also Csikszentmihalyi whom I quoted in chapter 10 offering the reminder that "intrinsic rewards are not an ultimate standard to strive for. One still must ask: What are the consequences of this particular activity?"17 It is important to consider the content of our work, what it means beyond the pleasure it may provide.
26. The relative emphasis we ought to give to process and product considerations is a topic too ambitious for this discussion. My question is limited to whether encouraging an intrinsic orientation threatens to exclude product concerns. The answer depends on whether we equate intrinsic with process, and extrinsic with product. I am not sure this equation is warranted. Clearly that which pertains to the process of doing something is not always intrinsic to the task: the student who loves cooperating is an example of that. Conversely, it may be possible for one's purely intrinsic motivation as an artist to be geared as much to the product that results as from the creative act that preceded it.18 Satisfaction in the doing is different from satisfaction in having done, but both might reasonably be classified as intrinsic. The
27. [296] latter is more similar to the former than it is to doing something for a reward.
28. Here again, of course, everything depends on how we define our terms. My point is that it is not at all obvious what is meant by the phrase intrinsic motivation. What appears at first blush an uncomplicated idea reveals itself as a tangle of possibilities, all of which have substantive implications for what we counterpose to the use of rewards.