28The What and Why of Goal Pursuits

Edward L. Deci and Richard M. Ryan, Psychological Inquiry 11(4), pp. 227–268, 2000; and Ryan and Deci, Self-Determination Theory, chapter 21 on work and organisations, Guilford, 2017

Linked to 20 chapters: supports 10 · contradicts 6 · mechanism 10 · same claim 8 · pairing 4

Most theories of motivation in 2000 assumed that people act to reach goals they value and expect to attain, and that two goals of equal value and equal odds would produce the same quality of effort. Deci and Ryan's paper argues that this leaves out the thing that gives goals their force. People have three innate psychological needs, for competence, for relatedness and for autonomy, and needs are not wants; they are nutriments, like water to a plant, whose satisfaction can be detected by the flourishing that follows and whose frustration by the decay. From this the paper draws two claims. The why of a goal matters: the same behaviour done because one is made to, or because one would feel ashamed not to, or because one values it, or because one finds it interesting, differs in its quality, its persistence and its effect on health, because these regulations satisfy the needs to different degrees. And the what of a goal matters: aspirations for growth, relationship and community feed the needs, while aspirations for wealth, fame and image do not, and are often substitutes acquired where the needs went unmet. The paper is a review of thirty years of experiments and field studies organised around those claims, and the book chapter written seventeen years later takes them to the place most adults spend most of their waking time, and finds that managers can be trained to support the needs, that pay works through what it signals rather than what it buys, and that the manager's own needs are part of the mechanism.

What a need is

Two traditions had used the word. Hull's drive theory named innate physiological needs, for food, water and sex, whose deficits push an organism to act and whose reduction teaches it what to do; it could not account for curiosity, play and exploration, which have no deficit behind them. Murray named psychological needs but treated them as acquired and so broad that anything that moves a person counts, including abasement, greed and dominance, and his tradition measures individual differences in the strength of achievement, power and affiliation motives. Deci and Ryan take innateness from the first tradition and the psychological level from the second. A need in their sense is an innate nutriment essential for growth, integrity and well-being, and the test is functional: something is a need if conditions that satisfy it produce health and conditions that thwart it produce ill-being, whether or not the person values it. There are three, and psychological health requires all three; one or two are not enough. The theory's starting point is an active organism inclined towards integration, of its own psychic elements into a coherent self and of itself into larger social structures, and these inclinations need the nutriments to operate but are not aimed at them. Children play because it is adaptive to play, not to feel competent; a man at the piano in the evening is not seeking need satisfaction, though he would feel no pleasure if he were coerced to play or unable to master the piece. Because the needs are innate, the authors do not study their strength. An unusually strong wish to be with people, or to be in control, is read not as a strong need but as evidence of past thwarting, the way an unusually strong hunger would be.

Intrinsic motivation

The theory grew out of experiments in the early 1970s in which paying people for an activity they had found interesting left them less interested afterwards than people who had not been paid. Deci and Ryan read this as a shift in the perceived locus of causality from inside the person to outside, and the finding held: a meta-analysis of 128 studies over three decades, published the year before this paper, confirmed that all contingent tangible rewards undermine intrinsic motivation. Threats, surveillance, evaluation and deadlines do the same; offering choice and acknowledging a person's feelings do the opposite, and the same manipulations that raise or lower intrinsic motivation raise or lower creativity, cognitive flexibility and conceptual learning. Autonomy is the need at stake, and the authors are careful to say what it is not. It is not independence, not detachment from others, not an internal locus of control; it is volition, acting from an integrated sense of self and endorsing one's actions, and it can be positively related to relatedness and to depending on others. Competence is the second need at stake. Positive feedback enhances intrinsic motivation and negative feedback undermines it, but only when the person feels responsible for the competent performance; competence is necessary for any motivation and autonomy is what makes it intrinsic. Relatedness plays a more distant role, as a secure base: children ignored by an experimenter lost interest in an interesting task, and students with warm teachers showed more intrinsic motivation, but people hike and play solitaire alone.

Internalisation

The paper's second body of work concerns behaviour that is not interesting and must be done anyway. Extrinsic motivation, the authors argue, is not one thing but a continuum of how far an outside regulation has been taken into the self. At one end is external regulation, behaving for a reward or to avoid a punishment, which stops when the contingency is withdrawn. Next is introjection, a regulation swallowed whole without being digested, in Perls's phrase the paper borrows: the person now administers the contingency to themselves, as pride, guilt or shame, and the behaviour is maintained but at the cost of inner conflict. Then identification, in which the person accepts the value of the behaviour as their own, exercising because they care about their health. Then integration, in which that identification is brought into harmony with the person's other values, which is fully volitional and still extrinsic, because the behaviour is instrumental rather than done for its own sake. Intrinsic motivation stands at the far end as the standard the others are measured against, and amotivation, the absence of any intention to act, stands at the near end. The first two are controlled; the last three are autonomous.

Internalisation is natural but not automatic. It is fed by relatedness to the people who hold the value and by competence in the behaviour, and those two are enough to produce introjection; integration needs autonomy as well, the chance to process and endorse the value rather than be pressed into it. A laboratory experiment isolated three facilitators for an uninteresting task, a meaningful rationale, an acknowledgement that the task was not interesting, and an emphasis on choice rather than control. Each increased later voluntary engagement, and the kind of internalisation differed by count: with two or three facilitators, people who did more of the task also valued and enjoyed it more, the mark of integration, while with one or none the correlation reversed and those who did more felt less free, the mark of introjection. Parents who were interviewed at home and rated as more involved and more supportive of autonomy had children who valued schoolwork more and did better at it; medical students taught by autonomy-supportive instructors still showed the course's values in their behaviour six months later.

Why and what

Across schools from elementary to medical, and across Canada, Japan, Russia and the United States, students who gave more autonomous reasons for their work showed better conceptual understanding, higher grades, more persistence, less dropping out, and better coping, while introjected reasons went with anxiety and poor coping, so that two students who look equally motivated can be on different courses. The same pattern held for religious practice, where the reasons and not the behaviour predicted mental health; for a six-month medically supervised diet, where patients who felt the staff supported their autonomy lost more weight and had kept it off nearly two years later; for diabetic patients' glucose control, for smoking cessation confirmed biochemically at thirty months, for New Year's resolutions, for political engagement and for methadone patients' clean urine tests. At the broadest level, people differ in causality orientation, a general tendency to regulate behaviour by interest and endorsed values, by controls and directives, or by helplessness; the autonomy orientation goes with well-being and integrated personality, the controlled orientation with public self-consciousness and the coronary-prone pattern, the impersonal orientation with depression. Autonomy occupies a special position among the three needs. Competence and relatedness can be satisfied by controlled behaviour, and the athlete driven to beat others may feel competent, but the optimal outcomes appear only when the behaviours are autonomous as well. And parents who make love conditional on compliance pit relatedness against autonomy, and their children feel compelled rather than willing, less loved, and more resentful.

The content of goals matters independently. Aspirations for personal growth, relationships and community contribution are closely tied to the needs; aspirations for wealth, fame and image depend on contingent approval and external signs of worth. Placing high relative importance on financial success predicted lower self-actualisation and vitality and more anxiety and depression, in college samples and in a mixed community sample of eighteen-year-olds assessed by clinical interview, and the pattern replicated in Russia and Germany. Attaining intrinsic aspirations raised well-being; attaining extrinsic ones, once above poverty, did little. Carver and Baird proposed that only the why mattered, and found that autonomous reasons for pursuing wealth did go with better outcomes than controlled reasons, but the relative importance of wealth predicted poorer outcomes even with the reasons controlled for. The paper's cultural argument follows: the needs are universal and the routes to them vary, so that Asian participants showed the highest intrinsic motivation when trusted others chose for them and American participants when they chose themselves, while imposed choices produced the lowest in both. Some cultural values, the authors add, cannot be integrated at all because they contradict the needs, and a culture that transmits them, or transmits its values by control, breeds alienation and is less stable.

When the needs are not met, people do not simply try harder, as they would with hunger; they accommodate. They develop substitute motives, controlled regulatory styles and rigid patterns of behaviour that protect them and that go on frustrating the needs. Teenagers whose mothers were rated cold and controlling, by themselves, by the mothers and by clinical interviewers, placed more relative importance on wealth, and a risk index taken when the children were four predicted that emphasis twelve years later; adolescents with less autonomy-supportive parents held stronger extrinsic aspirations and reported more smoking, drinking and drug use. People high in the controlled and impersonal orientations responded to social predicaments by saving face and blaming others, which frustrates relatedness further. Anorexia is read as control exercised over the one domain left, and the self-defeating behaviours reviewed by Baumeister, from face-saving to helplessness, as controlled regulation and amotivation. The paper closes its argument with a defence of needs in evolutionary terms, against theories that admit only narrow domain-specific mechanisms, and with comparisons to self-efficacy theory, which knows only competence and dismisses autonomy by defining it as total independence; terror management theory, which the authors think mistakes a defensive process for a need; control theory, which cannot tell approach from autonomy; achievement goal theory, whose performance goals can be pursued for autonomous or controlled reasons; flow theory, which has optimal challenge but no concept of autonomy; and attachment theory, whose security they find to vary within a person across partners according to how each partner supports the needs.

At work

The 2017 chapter opens with a sample of American workers who reported, three times a day for three weeks, lower well-being during working hours than outside them, explained by lower satisfaction of all three needs at work. It traces the theory's entry into organisations through Argyris and McGregor, who saw autonomy before other applied fields did, and through the expectancy theories of Vroom and of Porter and Lawler, which assumed that intrinsic and extrinsic satisfactions add, an assumption Deci's early experiments disconfirmed. The anchor study is a field intervention at a Fortune 500 company with branch offices nationwide in a hard year. Managers were trained in three things: taking their subordinates' perspective, giving feedback that was informational rather than critical and hearing feedback in return, and supporting initiative, choice and participation. Trained managers became more autonomy-supportive than controls, their subordinates' perceptions and satisfaction improved over thirteen months, the managers who changed most had subordinates who changed most, and the company's own attitude surveys showed the same. Employees with supportive immediate supervisors trusted top managers several levels above them more. A later analysis put the return on the intervention, in mental-health costs saved, at more than three to one, and a second randomised training study found trained managers more supportive five weeks later and their employees more autonomous and engaged five weeks after that.

The chapter then assembles field studies of autonomous and controlled motivation at work: police officers, teachers in the Gambia, China and Canada, nurses who had been harassed, more than eight hundred teachers under pressure, and more than three thousand Dutch companies, where job autonomy was only weakly related to revenue growth overall and more strongly in young firms. Organisational change met less resistance where managers had given a rationale, some choice about implementation and an acknowledgement of feelings, measured in a telecommunications firm before and thirteen months into a disruptive restructuring; a children's psychiatric hospital that introduced a less restrictive regime in that way found its staff internalising the change and its patients becoming more autonomous in their own treatment. Autonomous motivation predicts proactive behaviour, job offers, call-centre revenue and, in one service study, more than half the variance in service quality; it predicts affective commitment, and the direction runs from motivation to commitment rather than the reverse. The job characteristics of autonomy, feedback and task significance work through need satisfaction, and contact with the people who benefit from one's work supplies significance. Workers who value pay, security and holidays over interesting work are less satisfied with their jobs and their lives, and income does not soften the relation; teams and organisations that endorse growth over status have more engaged and more employable members.

Pay is the chapter's longest argument. What a reward does depends on its functional significance, whether the person experiences it as information about competence or as pressure, and the same performance-contingent reward enhanced intrinsic motivation when given in an autonomy-supportive way and undermined it when given controllingly. Unequal pay for the same task was the most damaging contingency in the meta-analysis. In Norwegian insurance companies, non-contingent salaries predicted autonomous motivation and quarterly or annual sales bonuses predicted controlled motivation, and autonomous motivation predicted effort and staying while controlled motivation predicted wanting to leave. In a telecommunications firm that put sales staff on full commission after two months, three-quarters had left within a year. An attendance award at an industrial laundry improved the punctuality of employees who had been late, was gamed, damaged the motivation of those whose attendance had been excellent, and lowered their efficiency at tasks outside the programme. A meta-analysis of many dozens of studies found intrinsic motivation a moderate to strong predictor of performance, strongest where quality matters, and extrinsic incentives better predictors where quantity matters; a meta-analysis of public-sector pay for performance found it helped on simple tasks and hurt on complex ones. Among senior managers in mergers, individual incentives predicted more in-role performance and fewer of the discretionary acts that correct problems. Level of pay predicted only whether pay seemed fair; autonomy support predicted need satisfaction, intrinsic motivation and procedural justice. The chapter's four conclusions are that pay matters, that it should not be salient as an instrument of control, that it must be equitable, and that supporting the needs does more for performance than pay can.

Goal setting done participatively is internalised; performance evaluation done as joint problem-solving on goals the person helped set is both accurate and motivating. Transformational leadership works through need satisfaction, and in the one experiment competence and relatedness carried most of the effect on performance, and the leaders who could lead that way were those whose own needs were supported: principals with more interpersonal support, more autonomous motivation and more felt competence led more transformationally. The chapter's closing sections treat need support in a shoe factory, a sheltered workshop for psychiatric patients where supervisor-rated need satisfaction predicted hours worked and pay earned, four United Nations agencies, banking, where the authors note that women perceived less autonomy support and received poorer evaluations, animal-shelter volunteers and ten state firms in Bulgaria under central planning, where the same model fitted as in the United States. Even unemployment has a why: among more than four hundred unemployed Belgians, autonomous reasons for seeking work predicted searching and well-being, controlled reasons predicted feeling worthless, and autonomous reasons for not searching predicted life satisfaction, in a country the authors are careful to note has a safety net most of the world lacks.

Since 2000

The paper became the standard statement of the theory and one of the most cited papers in the psychology of motivation, and the theory became the dominant account of motivation in education, health, sport and work, with a 2017 book of seven hundred pages consolidating it. Its central experimental claim has been the most contested. Behaviourists had already answered the rewards meta-analysis with their own, and the dispute over whether rewards undermine interest has continued; the field evidence collected in the 2017 chapter, from insurance bonuses to laundry attendance, is the authors' reply, and behavioural economists reproduced the crowding-out effect on their own terms. The continuum has been confirmed in structure across many domains, though the introjected step has proved the hardest to separate cleanly. The three needs held up as predictors of well-being across cultures in meta-analyses, which is the paper's strongest cross-cultural claim vindicated, while the criticism that autonomy is a Western value has continued in parallel. The work chapter's evidence is, by its own account, largely self-reported and often cross-sectional, drawn heavily from the authors' own network, and light on effect sizes; its two randomised manager-training studies are the exceptions, and the founding one is from the 1980s. What the theory offered a manager in 2000 and still offers is a vocabulary for something the pay debates could not name: that a person doing the right thing under pressure and a person doing it because they own it are not in the same state, that the difference shows up in health and in the quality of the work, and that a manager's ordinary acts, a rationale, a choice, an acknowledgement, are what move a person from one state to the other.