Chat with us, powered by LiveChat thought question Journal of Pta’sonality and Social Psychology Copyright 1987 by the American Psyc | Office Paper

thought question

Journal of Pta’sonality and Social Psychology Copyright 1987 by the American Psychological Association, Inc.
1987, Vol. 53, No. 6, 1024-1037 0022-3514/87/$00.75

The Support of Autonomy and the Control of Behavior

E d w a r d L . D e c i a n d R i c h a r d M . R y a n
University o f Rochester

In this article we suggest that events and contexts relevant to the initiation and regulation of inten-
tional behavior can function either to support autonomy (i.e., to promote choice) or to control
behavior (i.e., to pressure one toward specific outcomes). Research herein reviewed indicates that
this distinction is relevant to specific external events and to general interpersonal contexts as well as
to specific internal events and to general personality orientations. That is, the distinction is relevant
whether one’s analysis focuses on social psychological variables or on personality variables. The
research review details those contextual and person factors that tend to promote autonomy and those
that tend to control. Furthermore, it shows that autonomy support has generally been associated
with more intrinsic motivation, greater interest, less pressure and tension, more creativity, more
cognitive flexibility, better conceptual learning, a more positive emotional tone, higher self-esteem,
more trust, greater persistence of behavior change, and better physical and psychological health
than has control. Also, these results have converged across different assessment procedures, different
research methods, and different subject populations. On the basis of these results, we present an
organismic perspective in which we argue that the regulation of intentional behavior varies along a
continuum from autonomous (i.e., self-determined) to controlled. The relation of this organismic
perspective to historical developments in empirical psychology is discussed, with a particular empha-
sis on its implications for the study of social psychology and personality.

For several decades A m e r i c a n psychology was d o m i n a t e d by
associationist theories. A s s u m i n g t h a t behavior is controlled by
peripheral mechanisms, these theories held t h a t the initiation
o f behavior is a function o f stimulus i n p u t s such as external
contingencies o f r e i n f o r c e m e n t (Skinner, 1953) or internal
drive stimulations (Hull, 1943) and t h a t the regulation o f behav-
ior is a function o f associative b o n d s between i n p u t s and behav-
iors t h a t develop through r e i n f o r c e m e n t processes. W i t h t h a t
general perspective, the central processing o f i n f o r m a t i o n was
n o t p a r t o f the e x p l a n a t o r y system, so concepts such as in-
tention were considered irrelevant to the d e t e r m i n a t i o n o f be-

D u r i n g the 1950s and 1960s, associationist theories gave way
to cognitive theories in which the processing o f i n f o r m a t i o n was
assumed to play an i m p o r t a n t role in the d e t e r m i n a t i o n o f be-
havior. O n the basis o f this assumption, the initiation o f behav-
ior was theorized to be a function o f expectations a b o u t behav-
i o r – o u t c o m e contingencies a n d o f the psychological value o f
outcomes (e.g., Atkinson, 1964; Tolman, 1959; Vroom, 1964),
and the regulation o f behavior was seen as a process o f c o m p a r –
ing o n e ‘ s c u r r e n t state to a s t a n d a r d (i.e., the desired outcome)
and then acting to reduce the discrepancy (e.g., Kanfer, 1975;
Miller, Galanter, & P r i b r a m , 1960). Thus, the cognitive per-
spective shifted the focus o f analysis from the effects o f past con-

This work was supported in part by Grant HD 19914 from the Na-
tional Institute of Child Health and Human Development to the Human
Mofivaton Program at the University of Rochester.

Correspondence concerning this article should be addressed to Ed-
ward L. Deci or to Richard M. Ryan, Human Mofivaton Program,
Department of Psychology, University of Rochester, Rochester, New
York 14627.

sequences o f behavior to expectations a b o u t future conse-
quences o f behavior. The concept o f i n t e n t i o n a l i t y (Lewin,
1951) b e c a m e i m p o r t a n t because behavior, whether i m p l i c i t l y
o r explicitly, was understood in t e r m s o f p e o p l e ‘ s i n t e n t i o n s to
act in a way t h a t would yield certain outcomes.

W i t h i n the concept o f intentionality, however, a further dis-
t i n c t i o n can usefully be made. S o m e i n t e n t i o n a l behaviors, we
suggest, are i n i t i a t e d a n d regulated t h r o u g h choice as an expres-
sion o f oneself, whereas other i n t e n t i o n a l behaviors are pres-
sured a n d coerced by intrapsychic and e n v i r o n m e n t a l forces
a n d thus d o n o t r e p r e s e n t t r u e choice (Deci & Ryan, 1985b).
The fdrmer behaviors a r e c h a r a c t e r i z e d b y a u t o n o m o u s initia-
tion and regulation and are referred to as self-determined; the
latter behaviors are c h a r a c t e r i z e d b y h e t e r o n o m o u s initiation
and regulation and a r e referred to as controlled, l

We shall argue t h a t the d i s t i n c t i o n between self-determined
and controlled behaviors has r a m i f i c a t i o n s for the quality o f ac-
tion a n d experience and is relevant t o the study o f b o t h social
contexts and personality.

I n t e n t i o n a l i t y a n d A u t o n o m y

A n i n t e n t i o n is generally u n d e r s t o o d as a d e t e r m i n a t i o n to
engage in a p a r t i c u l a r behavior (Atkinson, 1964). In the cogni-
tive theories o f motivation and a c t i o n (e.g., Heider, 1960;
Lewin, 1951; Tolman, 1959), which have their r o o t s in G e s t a l t
psychology, having an i n t e n t i o n i m p l i e s personal causation and
is equivalent to being m o t i v a t e d to act. I n t e n t i o n s are said to

t Like most dichotomies in psychology, being self-determined versus
controlled is intended to describe a continuum. Behaviors can thus be
seen as being more or less self-determined.



derive from one’s desire to achieve positively valent outcomes
or avoid negatively valent ones.

Using an intentional perspective, psychologists working in a
neo-operant reinforcement tradition have emphasized that peo-
ple’s beliefs about whether certain behaviors are reliably related
to desired outcomes are o f central import. An abundance o f
research has shown, for example, that when a situation is struc-
tured so that outcomes are independent of behaviors (Seligman,
1975) or when people have a generalized belief that behaviors
and outcomes are independent (Rotter, 1966), nonintentional-
ity and maladaptation are likely to result. However, believing
that behaviors are reliably related to outcomes is not enough to
ensure a high level o f motivation and adaptation. People must
also believe that they are sufficiently competent to execute the
requisite behaviors (e.g., Bandura, 1977). Indeed, the expecta-
tion o f incompetence, like the expectation o f behavior-
outcome independence, has been shown to result in low motiva-
tion and maladaptation (Abramson, Seligman, & Teasdale,
1978). In sum, the cognitive perspective maintains that when
people believe that desired outcomes will follow reliably from
certain behaviors and that they are competent to execute those
behaviors, they will display intentionality and experience per-
sonal causation (Heider, 1958).

Our organismic approach diverges from the cognitive ap-
proach by distinguishing between those intentional behaviors
that are initiated and regulated autonomously and those that
are controlled by intrapersonal or interpersonal forces.
Whereas the cognitive approach equates the concepts o f inten-
tion and choice (Lewin, 1951), the organismic approach re-
serves the concept of choice for those intentional behaviors that
are autonomously initiated and regulated, and it uses the con-
cept of control for those intentional behaviors that are not au-
tonomous. Thus, although having perceived control over out-
comes (i.e., perceiving behavior-outcome dependence and com-
petence) promotes intentionality, it does not ensure that the
intentional behavior will be initiated and regulated autono-

The concept o f autonomy is a theoretical rather than empiri-
eal one, though it has clear empirical consequences. Autonomy
connotes an inner endorsement of one’s actions, the sense that
they emanate from oneself and are one’s own. Autonomous ac-
tion is thus chosen, but we use the term choice not as a cognitive
concept, referring to decisions among behavioral options (e.g.,
Brehm & Brehm, 1981), but rather as an organismic concept
anchored in the sense of a fuller, more integrated functioning.
The more autonomous the behavior, the more it is endorsed by
the whole self and is experienced as action for which one is re-

Let us clarify this point through some examples. First con-
sider the behavior of an anorexic person abstaining from food.
Clearly, there is intentionality, yet the person would not appro-
priately be described as acting autonomously (or through
choice), for the experience is one o f compulsion (Strauss &
Ryan, 1987). In a similar vein, the behavior o f someone who is
desperately seeking approval or avoiding guilt is intentional, but
it is not autonomous. The person is compelled to engage in the
behavior and would not experience a sense of choice. Finally, a
person who follows a therapist’s suggestion not out o f an inte-
grated understanding but rather out o f deference to the thera-

pist’s authority is behaving intentionally, but until the action is
self-initiated and grasped as one’s own solution it would not be
characterized as autonomous.

When autonomous, people experience themselves as initia-
tors o f their own behavior; they select desired outcomes and
choose how to achieve them. Regulation through choice is char-
acterized by flexibility and the absence of pressure. By contrast,
being controlled is characterized by greater rigidity and the ex-
perience of having to do what one is doing. There is intention,
but lacking is a true sense of choice. When controlled, people
are, in the words o f deCharms (1968), “pawns” to desired out-
comes, even though they intend to achieve those outcomes.

Initiation a n d Regulation o f Behavior

When someone engages in a behavior, there are generally as-
pects o f the context that play a role in the initiation and regula-
tion of that behavior. We have argued (Deci & Ryan, 1985b)
that these contextual factors do not, in a straightforward sense,
determine the behavior. Instead, the person gives psychological
meaning (what we call functional significance) to those contex-
tual factors, and that meaning is the critical element in determi-
nation o f the behavior.

O f central concern to the issue of autonomy and control in
human behavior is whether people construe contexts as sup-
porting their autonomy (i.e., encouraging them to make their
own choices) or controlling their behavior (i.e., pressuring them
toward particular outcomes). Thus, this review will consider
varied social-contextual factors that have a functional signifi-
cance of being either autonomy supportive or controlling, ~ and
it will relate each type o f functional significance to the quality
o f people’s experience and behavior. However, dispositional or
person factors are also relevant to the study o f autonomy and
control. There are evident individual differences in the func-
tional significance people give to contextual factors. Further-
more, individual difference measures of autonomy and control
orientations have been used to predict people’s experience and
behavior directly, without reference to contextual factors. The
current review is intended to give substance to the theoretical
concepts of autonomy and control by examining research on
both contextual and person factors that are relevant to that dis.
tinction. In addition, it will compare this organismic perspec-
tive to other perspectives within empirical psychology.

C o n t e x t u a l Factors

There are two broad sets ofstudies, generally considered to be
in the province of social psychology, that focus on the autonom$
supportive versus controlling distinction. The first set explored
specific environmental events–things like task-contingent re.
wards, positive feedback, or imposed deadlinesmthat tend tr
promote either self-determined or controlled behaviors and the
qualifies associated with each. The second set of studies focused

2 According to cognitive evaluation theory (Deci & Ryan, 1985b), in
puts can also have an amotivating functional significance. These input
signify or promote incompetence at reliably obtaining desired out
comes. They are not relevant to this discussion, however, as they pro
mote nonintentional responding and impersonal causation.


on interpersonal or social contexts, showing not only that gen-
eral contexts can have either an autonomy-supportive or a con-
trolling functional significance, but also that this varied func-
tional significance has predictable effects on people’s experi-
encc, attitudes, and behavior within those settings.

When the a u t o n o m y supportive versus controlling distinc-
tion was initially made (e.g., Deci & Ryan, 1980; Deci,
Schwartz, Sheinman, & Ryan, 1981 ), it was hypothesized that
autonomy-supportive events and contexts would maintain or
enhance intrinsic motivation and that controlling events and
contexts would undermine intrinsic motivation. Because in-
trinsic motivation has been so widely explored as the dependent
variable in studies o f autonomy-supportive versus controlling
events and contexts, the effect o f an event or context on intrinsic
motivation can be used as one criterion for classifying whether
that event or context tends to be experienced as a u t o n o m y sup-
portive or controlling. Thus, within the reviews o f research on
external events and on interpersonal contexts, we will first pres-
ent studies that used intrinsic motivation as a dependent vari-
able, so as to specify the average functional significance o f par-
ticular events or contexts. Then, within each o f the two reviews,
we will move on to studies that have explored the relation o f
those factors to other variables so as to explicate empirically
the concomitants and consequences o f self-determined versus
controlled behavior.

External Events.” Autonomy Supportive or Controlling

The term event refers to a specifiable occurrence or condition
relevant to the initiation and regulation o f behavior. The offer
o f a reward, for example, is an event, as is an instance o f compe-
tence feedback, a demand, a deadline, and an opportunity for
choice. The most frequently studied events have been rewards,
though many others have also been explored. In this section,
studies o f the effects o f various events on intrinsic motivation
will be reviewed so as to allow each event to be classified as
tending to be either a u t o n o m y supportive or controlling.

Rewards. Dozens o f studies have explored the effects o f re-
wards on intrinsic motivation. These have included monetary
payments (Deci, 1971), good-player awards (Lopper, Greene, &
Nisbett, 1973), food (Ross, 1975), and prizes (Harackiewicz,
1979). In general, rewards have been found to undermine in-
trinsic motivation. When people received rewards for working
on an interesting activity, they tended to display less interest in
and willingness to work on that activity after termination o f the
rewards than did people who had worked on the activity without
receiving a reward. This phenomenon, labeled the undermining
effect (Deci & Ryan, 1980), has been most reliably obtained
when rewards were expected (Lopper et al., 1973), salient (Ross,
1975), and contingent on task engagement (Ryan, Mims, &
Koestner, 1983).

Ryan et al. (1983) pointed out that when rewards are differ-
ently structured, they have discernibly different effects. The au-
thors provided a t a x o n o m y o f reward structures and related it to
reward effects. Their review indicated that task-noncontingent
rewards–those that are given independent o f task engage-
m e n t – – w e r e least likely to undermine intrinsic motivation be-
cause the reward is not given for doing the activity and thus is
not salient as a control. Task-contingent rewards–those made

contingent on doing the activity–have been consistently and
reliably shown to undermine intrinsic motivation, presumably
because their controlling function is salient. The effects o f per-
formance-contingent rewards–those given for attaining a spec-
ified level o f good p e r f o r m a n c e – – a r e more complicated. Be-
cause they inherently provide positive competence feedback,
the appropriate comparison condition is one that conveys the
same feedback without a reward. When such comparisons have
been made, performance-contingent rewards have generally
been found to undermine intrinsic motivation, although they
have sometimes been shown to maintain or enhance intrinsic
motivation when the controlling aspect is minimized and com-
petence cues are emphasized (Harackiewicz, Manderlink, &
Sansone, 1984).

To summarize, m a n y studies have shown that rewards, on
average, undermine people’s intrinsic motivation. It appears,
therefore, that rewards tend to be experienced as controlling,
which o f course makes sense, as rewards are typically used to
induce or pressure people to do things they would not freely do.
When people behave in the presence o f reward contingencies,
the rewards tend to have a functional significance o f control,
thus representing an external event that restricts self-determi-
nation, although under certain circumstances they can be used
to support self-determination.

Threats and deadlines. Using a modified avoidance condi-
tioning paradigm, Deci and Cascio (1972) found that subjects
who solved interesting puzzles to avoid an unpleasant noise
demonstrated less subsequent intrinsic motivation for the activ-
ity than did subjects who solved the puzzles without the threat
o f noise. Amabile, DeJong, and Lopper (1976) found that the
imposition o f a deadline for the completion o f an interesting
activity also decreased subjects’ intrinsic motivation for that ac-
tivity. It appears, therefore, that these events, like rewards, tend
to be experienced as controlling and thus to diminish people’s

Evaluation and surveillance. Other experiments have indi-
cated that the mere presence o f a surveillant or evaluator, even
without rewards or aversive consequences, can be detrimental to
intrinsic motivation and thus, we suggest, t o self-determination
more generally. Lopper and Greene (1975), for example, found
that surveillance by a video camera undermined the intrinsic
motivation o f children, and Plant and Ryan (1985) found the
same result for college students. Pittman, Davey, Alafat, Weth-
erill, and K r a m e r (1980) reported that in-person surveillance
also undermined intrinsic motivation.

Harackiewicz et al. (1984) found that subjects who were told
that their activity would be evaluated displayed less subsequent
intrinsic motivation than did subjects who were not told this,
even though the evaluations were positive. Smith (1974) found
the same results for intrinsic motivation to learn. Similarly,
Benware and Deci (1984) and Maehr and Stallings (1972) have
found that learning in order to be tested or externally evaluated
has detrimental effects on intrinsic motivation for learning.

The effects o f evaluation and surveillance are not surprising,
as both are integral to social control. These events tend to limit
self-determination and thus reduce intrinsic motivation even
when they are not accompanied by explicit rewards or punish-

Choice. Autonomy-supportive events are defined as those


that encourage the process o f choice and the experience o f au-
tonomy. The one type o f event that both fits the definition and
has been shown, on average, to enhance intrinsic motivation is
the opportunity to choose what to do.

Zuckerman, Porac, Lathin, Smith, and Deci (1978) found
that when college student subjects were given a choice about
which puzzles to work on and about how much time to allot to
each, they were more intrinsically motivated during a subse-
quent period than were no-choice subjects in a yoked compari-
son group. The provision o f choice enhanced their intrinsic mo-
tivation. Swarm and Pittman (1977) reported similar results in
an experiment with children.

Positive feedback. The event o f positive competence feed-
back has been widely studied as it relates to intrinsic motiva-
tion. 3 Several studies have found that it increased intrinsic mo-
tivation (Blanck, Reis, & Jackson, 1984; Boggiano & Ruble,
1979; Vallerand & Reid, 1984), although this has occurred only
under certain circumstances (Fisher, 1978; Ryan, 1982) or for
certain kinds o f people (Boggiano & Barrett, 1985; Deci, Cas-
cio, & Krusell, 1975; Kast, 1983). Taken together, the studies
indicate that positive competence feedback neither supports au-
t o n o m y nor controls behavior per se. It can enhance intrinsic
motivation by affirming competence (e.g., Harackiewicz,
Manderlink, & Sansone, in press) because intrinsic motivation
is based in the need for competence as well as the need for self-
determination, although it will do so only when the sense o f
competence is accompanied by the experience o f self-determi-
nation (Fisher, 1978; Ryan, 1982). 4 But it can also undermine
intrinsic motivation by being experienced as a form o f interper-
sonal control (Ryan et al., 1983). The Harackiewicz, Abrahams,
and Wageman (1987) article in this special section focuses on
the issue o f competence, whereas our article focuses on self-

Effects and Correlates o f Autonomy-Supportive Versus
Controlling Events

The studies just reported used intrinsic motivation as the pri-
m a r y dependent variable and were used to help classify events
as tending to be either a u t o n o m y supportive or controlling. It is
interesting to note that more o f the events manipulated in these
experiments were experienced as controlling than as a u t o n o m y
promoting. This makes sense, however, because a u t o n o m y must
emanate from oneself and can therefore only be facilitated by
contextual events, whereas control is something that can be
done to people by contextual events and is therefore more easily
evidenced. We shall now address additional effects o f these au-
tonomy-supportive versus controlling events to begin explicat-
ing the qualities o f self-determined versus controlled behaviors.

Interest-enjoyment. Along with the free-choice measure o f
intrinsic motivation, self-reports o f interest are often obtained.
Ryan et al. (1983) reported a correlation o f .42 between the
behavioral measure o f intrinsic motivation and self-reports o f
interest, and Harackiewicz (1979) reported a correlation o f . 4 4
between intrinsic motivation and expressed enjoyment. Al-
though research has not always found these strong correlations
(see Ryan & Deci, 1986), self-reports o f interest-enjoyment do
appear to be related to intrinsic motivation. Furthermore, nu-
merous studies that have not used the free-choice, behavioral

measure have found that postexperimental interest-enjoyment
is higher following autonomy-supportive events than following
controlling events (e.g., Enzle & Ross, 1978).

Creativity. Amabile (1979) reported that subjects who were
told that their work would be evaluated produced artistic col-
lages that were rated as less creative than those produced by
subjects who did not expect evaluations. Similar effects were
found for surveillance (see Amabile, 1983). Furthermore, when
children competed for a reward, they produced less creative col-
lages than those produced in a noncompetitive condition
(Amabile, 1982), and when children contracted for rewards
they were also less creative (Amabile, Hennessey, & Grossman,
1986). Additionally, Kruglanski, Friedman, and Zeevi ( 1971)
found that when subjects who wrote stories were rewarded with
the opportunity to engage in an interesting activity in the future,
their stories were judged to be less creative than the stories o f
subjects who were not rewarded. In sum, events that are typi-
cally controlling appear to affect creativity negatively, whereas
events that are more a u t o n o m y supportive seem to promote

Cognitive activity. Results similar to those for creativity have
been reported for cognitive flexibility. M c G r a w and McCullers
(1979) found that monetarily rewarded subjects had a more
difficult time breaking set when doing Luchins-type (1942)
water-jar problems than did nonrewarded subjects. Benware
and Deci (1984) reported that evaluative tests impaired college
students’ conceptual learning in addition to undermining their
intrinsic motivation. Grolnick and Ryan (1987) found impair-
ments in conceptual learning o f fifth-grade subjects who
learned material under a controlling-evaluative condition rather
than an autonomy-supportive one. It appears that when cogni-
tive activity is controlled, it is more rigid and less conceptual,
perhaps with a more narrow focus, than when it is self-deter-

Emotional tone. Garbarino (1975) studied fifth- and sixth-
grade girls who were rewarded with movie tickets for teaching
younger girls how to do a sorting task. He reported that the
rewarded tutors were more critical and demanding than were
nonrewarded tutors. In a complementary study, children in-
duced to interact with another child in order to play with a nice
game had less positive impressions o f that other child than did
children who had not been focused on the incentive (Boggiano,
Klinger, & Main, 1985). Controlling events, it seems, tend to
induce a negative emotional tone and a less favorable view o f
others in that situation.

Maintenance of behavior change. Rewards have also been
studied as they relate to the persistence o f behavior change fol-
lowing the termination o f treatment conditions. A study by
Dienstbier and Leak (1976) o f a weight-loss program, for exam-
ple, indicated that although rewards facilitated weight loss, their
termination led to much o f the lost weight’s being regained.

3 Negative feedback has also been studied and has been found to re-
duce intrinsic motivation; however, we interpret these decreases as re-
sulting from the feedback’s being experienced as amotivating rather
than controlling.

4 In cognitive evaluation theory (Deci & Ryan, 1985b), inputs that
both alarm competence and promote self-determination are referred to
as informational.


When behavior is controlled by events such as rewards, the
behavior tends to persist only so long as the controlling events
are present. In terms of effective behavior change in therapeutic
settings, the implication is that behavior change brought about
through salient external controls is less likely to persist follow-
ing the termination o f treatment than is change that is brought
about more autonomously. Behavior and personality change
will be maintained and transferred, we have argued, when the
change is experienced as autonomous or self-determined (Deci
& Ryan, 1985b).

To summarize, behavior undertaken when the functional sig-
nificance o f events is autonomy supportive has been related to
greater interest, more creativity, more cognitive flexibility, bet-
ter conceptual learning, a more positive emotional tone, and
more persistent behavior change than has behavior undertaken
when the functional significance o f events is controlling. Thus
far, research has related these motivationally relevant depen-
dent variables primarily to the events of rewards and evalua-

Interpersonal Contexts: Autonomy Supportive
Versus Controlling

In the preceding discussion we described research on specific
events relevant to the initiation and regulation o f behavior. Nu-
merous other studies have focused on interpersonal contexts
rather than specific events. For example, in interpersonal situa-
tions the general ambience can tend either to support autonomy
orto control behavior. We now turn to that research on interper-
sonal contexts. We begin, of course, with studies in which in-
trinsic motivation was the dependent measure, because those
are the ones that we use to establish the usefulness o f the distinc-

Studies of autonomy-supportive versus controlling contexts
have been of two types. Some are correlational field studies in
which the functional significance of the context is measured and
related to motivationally relevant variables o f people in those
contexts. The others are laboratory experiments in which
events such as rewards or feedback are administered within ex-
perimentally created …

error: Content is protected !!