In the winter of 1962, Jack William Brehm ran a study with college women and a row of household appliances. Each participant was shown eight products — a toaster, a coffee maker, and six others — and asked to rate how much she liked each one. Then the experimenter presented her with two of the appliances as a gift, and said she could choose one. For half the participants, the choice was straightforward: take either of the two items. For the other half, the experimenter added a complication: one of the two items, he explained, was not actually available — it had been spoken for. The woman could only have the remaining item. Then Brehm asked everyone to re-rate all eight appliances again. The result was sharp and clean. When a participant had her choice eliminated — when one option was simply taken off the table — her rated desirability of that eliminated item shot upward relative to her first ratings. The thing she could not have had become, in her own estimation, more valuable than it was before the restriction arrived. Brehm called this effect what it was: a restoration of perceived freedom through enhanced valuation of the threatened option.
That experiment, reported in Brehm's 1966 monograph A Theory of Psychological Reactance published by Academic Press, launched one of the most empirically productive and counterintuitive research programs in the history of social psychology. The theory Brehm articulated was not simply a name for obstinacy or contrariness. It was a formal, falsifiable account of a motivational state — reactance — with specific antecedents, internal structure, and behavioral consequences. Understanding it fully requires going beyond the boomerang effect and into the architecture: what freedoms are, what threatens them, how reactance varies in intensity, and what restoration looks like across dozens of behavioral domains.
The intellectual stakes were also larger than the appliance study implied. Brehm was arguing, in 1966, that human beings are not passive recipients of influence, instruction, or incentive. They are agents who monitor their autonomy, detect encroachments on it, and respond with motivated counterpressure. Every theory of persuasion, compliance, parenting, health messaging, or public policy that ignores this monitoring function will be systematically wrong — not occasionally, but structurally.
The Formal Structure of the Theory
What Counts as a Freedom
Brehm's first and most important specification was definitional. Reactance is not aroused by the loss of just anything. It is aroused by the loss — or threatened loss — of a behavioral freedom: a behavior that the person believes they have the option to perform or not perform, at their own discretion. The word "believes" is critical. Freedoms are psychological constructs, not legal entitlements. A person who has never considered whether they can freely eat red meat has no freedom about eating red meat in Brehm's sense. But a person who has eaten red meat regularly and regards it as part of their normal life does have that freedom — and a government campaign telling them to stop will trigger reactance, even if the campaign is medically correct.
Freedoms, Brehm argued, are established by past behavior, social norms, and explicit promises. If a person has always had access to a particular option, they develop an implicit expectation of continued access. When that expectation is violated, reactance is the response.
The Four Variables That Determine Reactance Magnitude
The 1966 monograph specified four parameters that control how much reactance any given threat produces.
Importance of the freedom. Trivial freedoms — the freedom to choose between two brand-identical pens — generate minimal reactance when threatened. Freedoms that touch on deeply held values, strong preferences, or behavioral patterns central to one's identity generate intense reactance. The implication is that persuasive campaigns targeting behaviors that people find identity-defining (dietary choices, political affiliations, religious practices) will encounter the strongest counterpressure.
Proportion of freedoms threatened. When a single freedom is threatened among many, reactance is moderate. When multiple freedoms are threatened simultaneously — or when a single threat is interpreted as a sign that many other freedoms are also in jeopardy — reactance intensifies. Brehm called this the "implicational potential" of a threat: a restriction that implies further restrictions to come generates more reactance than an isolated one.
Magnitude of the threat. Partial restrictions produce less reactance than complete eliminations. A law that makes a behavior more difficult is less reactance-inducing than one that makes it illegal. An advertisement that says "consider switching" triggers less reactance than one that says "you must switch."
Social dimension. Threats imposed by social agents — other people, institutions, authorities — produce reactance through the additional channel of perceived illegitimacy. A freedom eliminated by impersonal circumstances (a product goes out of stock) arouses less reactance than the same freedom eliminated by another person's deliberate act (a salesperson tells you the product is reserved for preferred customers only).
Reactance vs. Resistance vs. Compliance: A Structural Comparison
| Dimension | Reactance | Mere Resistance | Compliance |
|---|---|---|---|
| Internal state | Active motivational arousal (aversive) | Passive inertia or skepticism | Neutral or positive engagement |
| Trigger | Perceived threat to a specific, valued freedom | General disagreement or dislike | Request consistent with existing preferences or trust |
| Direction of motivation | Toward the restricted behavior; away from the recommended one | Away from recommended behavior (no clear alternative pull) | Toward recommended behavior |
| Effect on target behavior's value | Increases desirability of the threatened/forbidden option | No systematic change in desirability | May increase desirability through persuasion |
| Effect on attitude toward source | Derogation, counter-arguing, anger | Skepticism, disengagement | Positive regard or neutrality |
| Behavioral signature | Boomerang effect: adoption of the prohibited behavior | Non-compliance without boomerang | Adoption of recommended behavior |
| Conditions that intensify it | Important freedom; strong/explicit threat; social agent as source; implicational scope | Weak argument quality; low trust in source | High argument quality; trusted source; low autonomy salience |
| Restoration mechanism | Performing the threatened behavior; valuing it more; vicarious restoration | Ignoring or avoiding the message | No restoration needed |
The distinction between reactance and mere resistance matters enormously for practice. A health communicator who attributes non-compliance to "resistance" may try harder — louder, more emphatic messaging — and thereby intensify reactance. The person was not simply unconvinced; they were motivated in the opposite direction, and more pressure accelerates that motivation.
The Cognitive Science of Reactance
Brehm's 1966 Foundational Studies
Beyond the appliance experiment, Brehm's 1966 monograph reported a series of studies that established the breadth of the phenomenon. In one study, male participants were told they would have a choice among several essays and could read whichever they preferred. For some participants, the experimenter then said that one essay was not available. Ratings of the restricted essay rose significantly relative to control conditions. The pattern held across product choices, essay topics, and social situations.
Brehm's 1966 framework also distinguished between direct reactance (the person whose freedom is threatened restores it themselves) and vicarious reactance (observing another person's freedom being eliminated arouses reactance in the observer). The vicarious variant has been underappreciated in applications — it explains why authoritarian public health campaigns can create community-level backlash even among individuals who were not personally targeted.
Sharon Brehm and Jack Brehm: The 1981 Reformulation
Jack Brehm's daughter-in-law Sharon S. Brehm co-authored the 1981 book Psychological Reactance: A Theory of Freedom and Control (Academic Press), the definitive updated statement of the theory. The 1981 formulation clarified several ambiguities in the original. First, it addressed the question of legitimacy: some threats to freedom, even from social agents, are accepted as legitimate (a doctor's medical advice, a teacher's classroom rule), and legitimate threats generate less reactance than illegitimate ones. Second, the 1981 book more fully developed the social influence implications — how reactance dynamics shape not just individual behavior but group conformity pressure, negotiation, and therapeutic relationships. Third, Sharon and Jack Brehm elaborated on the concept of freedom importance, noting that this is not simply preference intensity but also involves the person's sense of the freedom as a right, an entitlement, or a defining feature of their self-concept.
Wicklund and the Experimental Program
Robert Wicklund, Brehm's closest collaborator in the early empirical decades, conducted the experiments that confirmed the theory's predictions under controlled conditions and extended its reach. His 1974 synthesis Freedom and Reactance (Lawrence Erlbaum Associates) remains one of the most rigorous treatments of the theory's empirical foundations. Wicklund's most important conceptual contribution was distinguishing between two routes to reactance reduction: direct restoration (doing the forbidden thing) and indirect restoration (restoring a different freedom to compensate for the lost one, or simply asserting autonomy in some unrelated domain). The indirect route explains why people who are told what to do in one area of life sometimes assert independence in an apparently unrelated one — the motivational pressure of reactance is real but not domain-specific.
Dillard and Shen: Measuring the Unmeasurable
A persistent methodological problem for reactance research was that the core construct — the aversive motivational state — was not directly observable. Researchers inferred it from behavioral outcomes, which left the theory open to the objection that "boomerang effects" could have other explanations. James Price Dillard and Lijiang Shen addressed this directly in their 2005 paper "On the Nature of Reactance and Its Role in Persuasive Health Communication," published in Human Communication Research (vol. 32, pp. 207-234). Dillard and Shen proposed that reactance has two measurable components: cognitive reactance (counter-arguing thoughts, negative ideation about the message and its source) and affective reactance (anger and irritation). Using structural equation modeling across a series of health communication experiments, they demonstrated that both components independently predicted non-compliance and message rejection, and that the two components were empirically distinguishable from general attitude toward the topic. The Dillard-Shen model gave the field its first reliable measurement framework and enabled a new generation of experimental work.
Four Named Case Studies
Case Study 1: Worchel and Arnold (1973) — Forbidden Fruit in the Marketplace
Stephen Worchel and C. Arnold published a study in 1973 that became one of the clearest demonstrations of the boomerang effect in a consumer context. Participants heard a recording of a radio commercial for a rock album. In the standard condition, the commercial was an ordinary promotional message. In the reactance condition, the announcer added that the commercial had been banned by the radio station's management and was being broadcast in defiance of that ban. Participants in the ban condition rated the album as significantly more desirable than those in the control condition — without having heard a single note of music. The restriction itself, not the quality of the product, drove increased desirability. Worchel and Arnold's study has been widely cited in marketing literature as evidence for what is now called the "forbidden fruit effect," and it demonstrates that reactance does not require a personal, direct threat — a socially communicated restriction is sufficient.
Case Study 2: Grandpre et al. (2003) — Reactance and Adolescent Tobacco Prevention
Joseph Grandpre, Eusebio M. Alvaro, Michael Burgoon, Claude H. Miller, and John R. Hall published "Adolescent Reactance and Anti-Smoking Campaigns: A Theoretical Approach" in Health Communication (vol. 15, pp. 349-366) in 2003. They measured trait reactance in adolescents (using Hong and Faedda's scale) and then exposed participants to anti-smoking messages that varied in how directive and prescriptive they were. Adolescents with high trait reactance responded to strongly worded anti-smoking messages not with reduced smoking intentions but with increased ones — a clean boomerang. Lower-trait-reactance adolescents showed more conventional responses. The study had significant policy implications: anti-smoking campaigns directed at teenagers, which are often explicitly prohibitive and authority-invoking, may be not merely ineffective but actively counterproductive for the subset of the population most prone to reactance — which, developmental psychology suggests, is a very large subset of adolescents.
Case Study 3: Miller et al. (2007) — Health Messaging and the Intensity Trap
Claude H. Miller, Lindsay T. Lane, Leslie M. Deatrick, Anne M. Young, and Kimberly A. Potts published "Psychological Reactance and Promotional Health Messages: The Effects of Controlling Language, Lexical Concreteness, and the Restoration of Freedom" in Human Communication Research (vol. 33, pp. 219-240) in 2007. They manipulated the controlling language in health messages — comparing phrases like "you should," "you must," and "it is important that you" against low-controlling equivalents. High-controlling language reliably increased reactance (as measured by the Dillard-Shen instrument) and reduced behavioral intentions to follow the health advice. A second experiment showed that adding a "freedom restoration postscript" — a phrase at the end of the message explicitly affirming the reader's right to make their own choice — substantially reduced reactance and improved compliance intentions. This "postscript" technique has since been tested and replicated across health domains including alcohol reduction, exercise promotion, and dietary change.
Case Study 4: Clee and Wicklund (1980) — The Consumer Reactance Review
Meryl Clee and Robert Wicklund published "Consumer Behavior and Psychological Reactance" in the Journal of Consumer Research (vol. 6, pp. 389-405) in 1980. While not a single experiment, this review synthesized the experimental literature and reframed it for marketing science. Their core argument was that high-pressure sales environments — hard-sell tactics, artificial urgency, coercive promotion — systematically trigger reactance in consumers, resulting in negative brand evaluations and reduced purchase likelihood. They documented that the same consumer who would freely choose a product under neutral conditions will actively reject it if the sales context makes the choice feel coerced. Clee and Wicklund also introduced the distinction between perceived and actual restriction: reactance is aroused by what the consumer believes is happening to their freedom, not by the objective structure of their options. This insight is commercially significant — a promotion that is technically not coercive can still trigger reactance if it is perceived as manipulative.
Intellectual Lineage
Jack Brehm's intellectual debts were multiple but clearly identifiable. The most important was Leon Festinger, under whom Brehm trained at the University of Minnesota in the 1950s. Festinger's theory of cognitive dissonance, published in 1957, established the principle that psychological inconsistency creates aversive motivational states that people are driven to reduce. Brehm's reactance theory followed the same structural logic: a perceived inconsistency between one's expected freedoms and one's actual situation creates an aversive state (reactance) that motivates restoration-seeking. Brehm's 1956 dissonance study — showing that post-decision enhancement of chosen alternatives increases after decisions are made — was directly absorbed into the appliance experiment design. Festinger's framework also influenced Brehm's insistence on specifying antecedent conditions formally, rather than treating reactance as a diffuse personality trait.
A second influence was Fritz Heider's balance theory and attribution framework, which oriented Brehm toward thinking about perceived social causation. The key variable in reactance is not simply that a freedom is lost but that another agent is perceived as having eliminated it. Heider's emphasis on perceived intentionality in social perception maps onto Brehm's specification that socially imposed threats generate more reactance than impersonal ones.
Brehm in turn influenced a wide generation of researchers. Robert Wicklund became the theory's most prolific empiricist through the 1970s. Sharon Brehm extended the framework into clinical and therapeutic settings, showing that therapeutic relationships can trigger reactance when therapists are too directive — an insight that reshaped reactance-compatible therapy models such as Motivational Interviewing, developed by William Miller and Stephen Rollnick in the 1980s and 1990s. Dillard, Shen, and Miller represent the post-2000 empirical generation, whose primary contribution has been measurement precision and application to public health.
Empirical Research: What the Evidence Shows
The empirical record on reactance theory is extensive, though not uniform. Several consistent findings have emerged across decades of research.
First, the boomerang effect is real but conditional. Threats to freedom do reliably increase the desirability of the threatened option — but only when the freedom is important to the person and when the threat is perceived as illegitimate or unjustified. Weak threats to trivial freedoms produce negligible reactance. Strong threats to important freedoms from social agents perceived as overstepping their authority produce the strongest boomerang effects.
Second, message framing reliably modulates reactance in persuasion contexts. Studies in health communication, across smoking, alcohol, diet, exercise, and medical adherence domains, have consistently shown that controlling language ("you must," "you should," "it is required") increases reactance and reduces compliance relative to autonomy-supportive language ("you might consider," "some people find," "the choice is yours"). A meta-analysis by Rains and Turner (2007) in Human Communication Research examined 20 studies and confirmed a reliable, moderate-sized effect: high-threat messages produced less attitude change and more counter-arguing than low-threat equivalents.
Third, the trait version of reactance — a stable individual disposition to experience higher reactance across situations — is real and measurable. Merilee Hong and Sonja Faedda published a trait reactance scale in 1996 (Psychological Reports, vol. 79, pp. 1323-1330) that has been widely used since. High-trait-reactance individuals show stronger boomerang effects in response to the same objective messages, are less responsive to authority-based persuasion, and report higher sensitivity to perceived autonomy violations. The scale has moderate test-retest reliability and correlates with related constructs including low agreeableness, high neuroticism, and anti-conformity attitudes.
Fourth, cultural context moderates reactance. Studies comparing reactance effects across individualist and collectivist cultural contexts have found that the effect is generally weaker in high-collectivism cultures, where social obligations and group norms are more salient and personal freedom is weighted differently. Jung and Kellaris (2004), in a cross-cultural study published in Psychology and Marketing, found reactance effects in U.S. samples that did not replicate with equal magnitude in Korean samples under equivalent freedom-threat conditions.
Limits, Critiques, and Nuances
Reactance theory has attracted substantial criticism, and the critiques have sharpened the field's understanding considerably.
The most persistent problem is measurement. Reactance, as Brehm originally defined it, is an internal motivational state — not a behavior, not an attitude, and not an emotion, though it produces all three. Until Dillard and Shen's 2005 framework, researchers were almost entirely dependent on behavioral inference: if the person did the opposite of what was requested, reactance was inferred. This created a serious confound — other mechanisms (simple noncompliance, source derogation, ambivalence) can produce the same behavioral pattern without invoking reactance. Even the Dillard-Shen instrument, which measures cognitive counter-arguing and anger, captures proxies of the state rather than the state itself, and critics have pointed out that anger and counter-arguing can arise from reasons unrelated to freedom threat.
A second critique is that the theory lacks a clear neurobiological grounding. Brehm was a behaviorally oriented theorist writing before neuroimaging was available, and he made no claims about neural mechanisms. More recent researchers have speculated about connections to threat-detection systems in the amygdala and the prefrontal regulation of behavioral inhibition, but these remain speculative. The absence of a neural account makes it difficult to distinguish reactance from related motivational constructs.
Third, the theory assumes that freedoms are psychologically real and perceivable — but in practice, people often do not have stable beliefs about what freedoms they possess. Freedoms are constructed in context, and the same individual may or may not perceive a given behavior as a freedom depending on framing, social context, and recent experience. This contextual variability in freedom perception makes a priori prediction of when reactance will and will not be triggered difficult.
Fourth, the boundary between reactance and reactance-adjacent constructs — autonomy motivation (Deci and Ryan's Self-Determination Theory), negativity bias toward commands, simple non-compliance due to lack of motivation — remains contested. Multiple researchers have argued that what is labeled "reactance" in applied research is often a composite of these phenomena rather than a pure expression of Brehm's motivational state. The Dillard-Shen framework helps, but has not fully resolved the boundary problem.
Finally, individual differences in trait reactance complicate any generalized recommendation. A communication strategy that reduces reactance in low-trait populations may be insufficient for high-trait individuals, who will experience freedom threats even in relatively mild message framings. Public health and policy contexts typically cannot tailor messages to individual reactance levels, which limits the practical applicability of the fine-grained experimental findings.
None of these critiques have invalidated the core phenomenon. The boomerang effect is robust. The manipulation of controlling language reliably affects compliance. Trait reactance predicts differential responses to authority-based persuasion. What the critiques have accomplished is to clarify what reactance theory is and is not: it is a theory of a specific motivational state with specific antecedents, not a theory of all non-compliance or all resistance to persuasion.
References
Brehm, J. W. (1966). A theory of psychological reactance. Academic Press.
Brehm, S. S., & Brehm, J. W. (1981). Psychological reactance: A theory of freedom and control. Academic Press.
Wicklund, R. A. (1974). Freedom and reactance. Lawrence Erlbaum Associates.
Clee, M. A., & Wicklund, R. A. (1980). Consumer behavior and psychological reactance. Journal of Consumer Research, 6(4), 389-405.
Worchel, S., & Arnold, S. E. (1973). The effects of censorship and attractiveness of the censor on attitude change. Journal of Experimental Social Psychology, 9(4), 365-377.
Dillard, J. P., & Shen, L. (2005). On the nature of reactance and its role in persuasive health communication. Human Communication Research, 32(2), 207-234.
Miller, C. H., Lane, L. T., Deatrick, L. M., Young, A. M., & Potts, K. A. (2007). Psychological reactance and promotional health messages: The effects of controlling language, lexical concreteness, and the restoration of freedom. Human Communication Research, 33(2), 219-240.
Grandpre, J., Alvaro, E. M., Burgoon, M., Miller, C. H., & Hall, J. R. (2003). Adolescent reactance and anti-smoking campaigns: A theoretical approach. Health Communication, 15(3), 349-366.
Hong, S. M., & Faedda, S. (1996). Refinement of the Hong Psychological Reactance Scale. Psychological Reports, 79(3), 1323-1330.
Rains, S. A., & Turner, M. M. (2007). Psychological reactance and persuasive health communication: A test and extension of the intertwined model. Human Communication Research, 33(2), 241-269.
Jung, J. M., & Kellaris, J. J. (2004). Cross-national differences in proneness to scarcity effects: The moderating roles of familiarity, uncertainty avoidance, and need for cognitive closure. Psychology and Marketing, 21(9), 739-753.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in human behavior. Plenum Press.
Frequently Asked Questions
What is Reactance Theory?
Reactance Theory, developed by Jack Brehm in his 1966 monograph, proposes that when a person's behavioral freedom is threatened or eliminated, they experience a motivational state called reactance — an aversive arousal that motivates behavior to restore the lost freedom. The theory predicts the boomerang effect: restricted options become more desirable, and pressure to adopt a position can produce attitude change in the opposite direction.
What is the boomerang effect in psychology?
The boomerang effect occurs when persuasion attempts produce attitude or behavior change in the direction opposite to what was intended. In reactance research, this happens when messages are perceived as too controlling or freedom-restricting — the recipient reacts by moving away from the advocated position. Dillard and Shen (2005) formalized this as a combined cognitive and emotional response: reactance involves both counterarguing and anger.
How does reactance affect health communication?
Miller et al. (2007) found that controlling language in health messages ('You must,' 'You have to') triggers reactance and reduces compliance. Grandpre et al. (2003) showed that adolescents high in trait reactance responded worse to anti-tobacco messages. The practical implication: health campaigns should emphasize choice and information rather than directives, minimizing freedom-threat while providing reasons to act.
What is the difference between reactance and mere resistance?
Mere resistance is a passive failure to comply. Reactance is an active, motivational state that drives behavior specifically toward restoring lost freedom — including enhanced desire for the threatened option and sometimes explicit defiance. Reactance predicts not just non-compliance but a specific directional pattern: increased attraction to what has been restricted.
What factors determine reactance magnitude?
Brehm's formal model identifies four variables: the importance of the threatened freedom, the proportion of freedoms threatened (one vs. many), the magnitude of the threat (direct elimination vs. implied threat), and the social dimension (whether the threat affects freedoms shared with others). Greater importance, broader threat, and higher social stakes all amplify reactance intensity.