Prejudice Reduction and the Contact Hypothesis: What Social Psychology Got Right, What It Missed, and Why Contact Alone Is Not Enough
Few ideas in social psychology have travelled as far as the Contact Hypothesis. The promise is simple and deeply appealing: bring people from different groups together under the right conditions, and prejudice will fall. It is one of the field’s most famous insights, one of its most hopeful, and one of its most policy-friendly. Yet that popularity has also made it easy to oversell. Contact can reduce prejudice. The evidence for that is strong. But reducing prejudice is not the same thing as dismantling inequality, redistributing power, or producing justice. To understand both the brilliance and the limits of contact theory, we need to ask a harder question than whether people end up liking each other more. We need to ask what kind of change actually matters.
The seductive promise of prejudice reduction
Prejudice reduction sounds like an obviously good thing. In one sense, it is. Fewer hostile attitudes, fewer stereotypes, less suspicion, less fear, and less disgust directed at other groups are all meaningful achievements. Social psychologists have spent decades asking how such change happens, and few ideas have shaped that work more than the Contact Hypothesis. At its most recognisable, the theory suggests that contact between members of different groups can reduce prejudice, especially when the interaction is structured in ways that reduce threat and increase cooperation. That basic idea has had enormous impact across psychology, education, integration policy, and diversity work.
But the phrase prejudice reduction carries a hidden danger. It makes it sound as though the core problem in intergroup conflict is mainly a problem of bad attitudes inside individual minds. Sometimes it is. Often, though, prejudice is braided together with law, institutions, status hierarchies, economic inequality, media narratives, and histories of domination. Once that is recognised, a difficult tension appears. A theory can be highly effective at improving interpersonal attitudes while still being far less effective at challenging the wider systems those attitudes sit inside. This tension sits at the heart of the Contact Hypothesis. It is both one of social psychology’s great successes and one of its most frequently overstretched ideas.
That is why the real question is not simply whether contact “works.” In a narrow and important sense, it often does. The more searching question is what kind of success it produces. Does it lower anxiety? Does it soften stereotypes? Does it produce warmer feelings toward outgroup members? Often yes to all but does it reliably increase support for social change, sharpen awareness of injustice, or alter structures of exclusion? That is a much less comfortable story.
What prejudice reduction is actually trying to reduce
One reason these debates become confused is that several distinct phenomena are too often bundled together. A stereotype is a generalised belief about a group. Prejudice is an evaluative orientation, usually involving dislike, suspicion, fear, contempt, or moral devaluation. Discrimination is behaviour that disadvantages people because of group membership. Structural inequality refers to the patterned distribution of power and resources across institutions and social systems. These are related, but they are not interchangeable. A person can sincerely report lower prejudice while still benefiting from and defending unequal systems. Equally, institutions can reproduce inequality even when many individuals within them regard themselves as fair-minded and tolerant.
This matters because much prejudice-reduction research has historically used changes in explicit attitudes as its main outcome. That is understandable. Attitudes are easier to measure than institutional change. Yet once the outcome variable narrows in this way, success can start to look more complete than it really is. If members of different groups come to feel more comfortable with one another, something valuable has happened. But it does not follow that discrimination has fallen, that unequal structures are being challenged, or that the disadvantaged group is now better positioned.
Allport’s Contact Hypothesis and why it became so influential
The classic starting point is Gordon Allport’s The Nature of Prejudice (1954), where he proposed that contact between groups is most likely to reduce prejudice under a set of favourable conditions. Later summaries, especially Pettigrew’s major review, crystallised these as four core conditions: equal status within the contact situation, common goals, cooperation, and support from authorities or social norms. These conditions matter because contact is not automatically benign. People can meet under competitive, humiliating, threatening, or coercive conditions too, and those encounters may reinforce rather than reduce animosity. The original theory, properly understood, was never a naive celebration of mere exposure. It was a conditional model of social encounter.
The appeal of the theory is obvious. It carries both scientific and moral elegance. If prejudice thrives on ignorance, anxiety, and rigid group boundaries, then meaningful interaction should erode some of the mechanisms that keep it alive. It also offered a practical route forward in societies struggling with segregation, conflict, and difference. Rather than relying only on moral preaching or abstract education, contact theory suggested that social arrangements themselves could be designed to encourage better relations. For policymakers and educators, this made the theory unusually actionable.
Its influence only grew as the evidence base expanded. Pettigrew and Tropp’s landmark 2006 meta-analysis drew on 713 independent samples from 515 studies and concluded that intergroup contact is typically associated with lower prejudice. Crucially, the positive relation was not confined to ideal conditions alone. Allport’s conditions helped, but they were better understood as facilitators than absolute requirements. That is one reason the theory became so central: it was not just intuitive, it was empirically resilient across a wide range of groups, settings, and methods.
How contact is supposed to work
The simple public version of contact theory is that people stop fearing what they get to know. The psychological story is more interesting than that. Pettigrew argued that contact works through multiple processes, including learning about the outgroup, changing behaviour, developing affective ties, and reappraising the ingroup. Later meta-analytic work by Pettigrew and Tropp suggested that the most important mediators are not simply increased knowledge, but reduced intergroup anxiety and greater empathy or perspective taking. In other words, contact often works not because it dumps facts into people’s heads, but because it changes the emotional climate in which the outgroup is encountered.
This is one reason friendship has loomed so large in the literature. Cross-group friendship is especially potent because it combines repeated exposure, mutual self-disclosure, trust, norm change, and emotional investment. Friendship also provides a setting in which outgroup members are harder to reduce to crude category labels. Someone stops being only “one of them” and becomes a person with a history, personality, contradictions, humour, and vulnerability. That shift does matter. It does not solve everything, but it matters.
The broad evidence base for contact
Any serious critique has to begin by being fair: the evidence that contact often reduces prejudice is substantial. Pettigrew and Tropp’s 2006 meta-analysis remains foundational for good reason. The central finding was clear: more intergroup contact is typically associated with less prejudice. The analysis also found evidence that contact effects can generalise beyond the specific individual encountered to the wider outgroup, which matters because otherwise the benefits would be too localised to be socially useful.
Subsequent work has not overturned that picture. A 2023 meta-analytic paper by Van Assche and colleagues examined whether contact’s benefits disappear under conditions of perceived threat or discrimination, two contexts in which critics might expect the theory to struggle most. Instead, the authors found that contact was associated with decreased prejudice in both cross-sectional and longitudinal data, among advantaged and disadvantaged groups, and in both WEIRD and non-WEIRD settings. Indeed, the beneficial effects were at least as strong among those high in perceived threat or discrimination as among those lower on those variables. That does not make contact invincible, but it does strengthen the case that the theory remains robust even under pressure.
There is also evidence that contact-based interventions can work outside tightly controlled laboratory settings. Lemmer and Wagner’s 2015 meta-analysis of direct and indirect contact interventions found that contact-based approaches can reduce ethnic prejudice and reported no evidence that direct contact interventions were categorically more effective than indirect ones. That finding helped validate the wider move toward studying not only face-to-face contact, but also more mediated or symbolic forms.
The rise of indirect contact
One reason contact theory expanded so successfully is that researchers did not restrict it to literal face-to-face encounters. In many real-world settings, especially those marked by residential segregation, social avoidance, or active conflict, direct positive contact may be rare or difficult to arrange. This led to the development of indirect forms such as extended contact, where simply knowing that members of one’s ingroup have close outgroup friends can soften prejudice, and imagined contact, where mentally simulating a positive intergroup interaction is expected to improve attitudes.
Turner and colleagues’ influential 2008 paper provided one of the first tests of the extended contact hypothesis, finding that its effects were mediated by reduced intergroup anxiety, perceived norms, and inclusion of the outgroup in the self. Later meta-analytic work on extended contact covered 248 effect sizes from 115 studies, suggesting that this indirect route has become a major research programme in its own right. The appeal is obvious: if societies cannot immediately produce widespread deep positive contact, perhaps attitudes can still be shifted through vicarious or imagined forms of social connection.
Yet here the story begins to bend toward critique. Paluck and colleagues’ 2021 review noted that in the previous decade, face-to-face contact experiments were relatively few compared with newer simulated forms of contact. In their review, only 28 studies across 27 manuscripts experimentally randomised face-to-face contact, with an average effect size of d = 0.28. By contrast, indirect and imagined contact had become much more common, often in student samples and with immediate self-report outcomes. This is not trivial. It suggests that one of psychology’s most policy-relevant theories has increasingly been tested in forms that are easier to administer than the social realities it is frequently invoked to solve.
The methodological critique: the evidence is real, but often thinner than the public story suggests
This is where many summaries become too flattering. Saying that contact “has a strong evidence base” is true, but incomplete. The deeper question is how that evidence has been produced. Paluck and colleagues’ 2021 review of 418 experiments from 309 manuscripts paints a more sobering picture of the wider prejudice-reduction literature. They found that 76% of studies tested light-touch interventions, the long-term impact of which remained unclear, and they raised concerns about publication bias exaggerating effects. Their conclusion was not that prejudice reduction is impossible, but that much of the research effort is poorly suited to generating actionable real-world recommendations.
That critique matters for contact research because contact theory is often treated as one of social psychology’s most field-ready ideas. Sometimes it is. But many interventions still rely on convenience samples, immediate attitude measures, or environments that are tidier than everyday life. Even within the face-to-face contact studies reviewed by Paluck and colleagues, the majority were conducted in school, university, or similarly structured settings, and only a minority measured outcomes after a delay. If contact is to carry heavy policy weight, it has to survive not just in classrooms and questionnaires, but in messy, unequal, emotionally charged social worlds.
The political critique: interpersonal harmony is not the same as justice
This is the most important critique, and the one most often diluted in softer explainers. Dixon, Durrheim, and Tredoux argued in their 2005 “reality check” that contact research had become overly attached to optimal settings and overly focused on prejudice as the outcome of interest. Their concern was not that positive contact is bad, but that the theory had drifted toward a psychologically tidy model of intergroup improvement that does not map neatly onto real social conflict. In divided societies, inequality is not just sustained by mutual ignorance. It is sustained by power, institutions, material interests, and contested histories.
From that perspective, contact can become politically ambivalent. It may create civility without contestation, tolerance without transformation, or even comfort without justice. People may feel better about one another while the underlying asymmetries remain substantially unchanged. Worse, improved interpersonal relations can sometimes reduce the perceived urgency of structural problems. If the dominant group appears friendly and the immediate encounter feels harmonious, then broader inequality can become easier to minimise, reinterpret, or ignore.
The paradox of integration
This is where the critique becomes especially sharp. Dixon and colleagues’ 2010 study of Black South Africans in post-apartheid society found that harmonious interracial contact was associated with lower perceptions of collective discrimination. That is a striking and uncomfortable result. The very process celebrated for reducing prejudice also appeared linked to reduced recognition of ongoing group-level injustice. The point is not that friendly contact is itself harmful. The point is that contact can operate differently depending on what outcome one values. If your goal is warmer attitudes, the result looks positive. If your goal is sustained recognition of structural inequality, it looks more troubling.
This finding opened up a broader line of work on what contact does to social change motives. Tausch, Saguy, and Bryson examined how intergroup contact affects collective action and individual mobility intentions among disadvantaged-group members, asking whether prejudice-reducing contact may also reduce resistance to inequality. Their work sits squarely inside a growing debate: positive contact may improve interpersonal relations while also dampening anger, reducing injustice appraisals, and weakening willingness to mobilise collectively. Reimer and colleagues pushed this further, showing that positive contact can discourage collective action among disadvantaged groups while sometimes encouraging advantaged-group members to act on their behalf. That is a profoundly mixed picture. It suggests that contact is not simply good or bad, but politically directional in different ways for different groups.
For a critical Simply Put Psych piece, this is the hinge point: contact theory succeeds most clearly as a theory of interpersonal thawing and least clearly when stretched into a theory of emancipation. That does not diminish its achievements. It situates them.
The negative-contact correction
Another major correction to older, more optimistic readings is that contact is not only positive. Bad contact matters, and recent evidence suggests it may matter more than good contact in some circumstances. Paolini and colleagues’ 2024 meta-analysis synthesised 238 independent samples with 936 nested effects and a total sample of 152,985. They found that positive contact was associated with lower prejudice and negative contact with higher prejudice, but crucially the detrimental effect of negative contact was significantly larger than the beneficial effect of positive contact. This negativity bias was especially pronounced where people could self-select out of contact, where avoidance motivation was high, and where the outgroup was low status or stigmatised.
This matters for both theory and practice. It means contact is not a one-way machine producing tolerance. In unequal societies, one humiliating, threatening, or hostile encounter may carry more weight than several mildly pleasant ones. If that is true, then contact interventions cannot just aim to create more interaction in the abstract. They have to care intensely about quality, context, power, norms, and the possibility of backlash. Quantity without structure may be ineffective, and in the wrong conditions it may be counterproductive.
So is contact theory wrong?
No. That would be too easy, and it would ignore decades of valuable work. Contact theory was right about something fundamental: group boundaries are not maintained only by ideology or misinformation, but also by segregation, unfamiliarity, affective distance, and normative separation. Contact can change those things. It can reduce anxiety, humanise the outgroup, shift norms, and create the sort of cross-group relationships that make blanket devaluation harder to sustain. Those are not small achievements.
What the theory got wrong, or at least what its public reception often got wrong, was the temptation to confuse one kind of success with all others. Lower prejudice does not necessarily mean lower inequality. Better feelings do not necessarily mean better institutions. Friendly encounters do not necessarily produce redistributive politics, anti-racist practice, or durable solidarity. Contact can help create the psychological conditions in which those things become more possible, but it cannot guarantee them, and in some contexts it may even soften the very perceptions that would motivate collective change.
What a mature prejudice-reduction agenda should look like
A stronger modern approach would keep contact theory, but stop idolising it. It would recognise that prejudice reduction needs multiple levels of intervention: interpersonal, normative, institutional, and structural. Contact can still play a vital part, especially when it is sustained, cooperative, normatively supported, and embedded in settings that do not simply paper over unequal power relations. But it should work alongside broader changes in representation, law, education, workplace norms, housing, status relations, and political voice.
It would also take outcomes more seriously. Instead of asking only whether people like each other more after an intervention, researchers and practitioners should ask whether the intervention changes behaviour, policy attitudes, support for equality, recognition of discrimination, and willingness to oppose unfair systems. Those are harder outcomes to measure, but they are closer to the actual social problems prejudice-reduction efforts are often invoked to address.
Finally, a mature agenda would stop pretending that all contact is inherently good. Quality matters. Valence matters. Power matters. History matters. Contact between groups meeting as nominal equals inside a cooperative setting is not the same as contact shaped by surveillance, status threat, dependency, or routine humiliation. The theory is most useful when it stays concrete about that. It becomes least useful when turned into a vague liberal slogan that mere proximity will save us.
Simply Put
The Contact Hypothesis remains one of social psychology’s genuinely important ideas. It deserves that status. The evidence that intergroup contact can reduce prejudice is substantial, and the theory has illuminated crucial processes involving anxiety, empathy, friendship, and social norms. But its very success has encouraged overreach. Contact is too often discussed as though it were a universal solvent for intergroup conflict, when in fact it is better understood as a partial and conditional tool. It can soften hostility. It can humanise. It can improve relations. What it cannot do, by itself, is carry the full weight of social justice.
That is not a disappointment unless we asked the theory to do more than it ever reasonably could. Contact theory is strongest when treated as a theory of how people come to see one another differently. It is weakest when treated as a substitute for politics, structure, or collective struggle. The mature lesson is not to abandon contact, but to place it in its proper role: valuable, evidence-based, often humane, but never enough on its own.
references
Allport, G. W. (1954). The nature of prejudice. Addison-Wesley.
Lemmer, G., & Wagner, U. (2015). Can we really reduce ethnic prejudice outside the lab? A meta-analysis of direct and indirect contact interventions. European Journal of Social Psychology, 45(2), 152-168. https://doi.org/10.1002/ejsp.2079
Paluck, E. L., & Green, D. P. (2009). Prejudice reduction: What works? A review and assessment of research and practice. Annual Review of Psychology, 60, 339-367. https://doi.org/10.1146/annurev.psych.60.110707.163607
Paluck, E. L., & Green, D. P. (2021). Prejudice reduction: Progress and challenges. Annual Review of Psychology, 72, 533-560. https://doi.org/10.1146/annurev-psych-071620-030619
Paolini, S., Gibbs, M., Sales, B., Anderson, D., & McIntyre, K. (2024). Negativity bias in intergroup contact: Meta-analytical evidence that bad is stronger than good, especially when people have the opportunity and motivation to opt out of contact. Psychological Bulletin, 150(8), 921-964. https://doi.org/10.1037/bul0000439
Pettigrew, T. F., & Tropp, L. R. (2006). A meta-analytic test of intergroup contact theory. Journal of Personality and Social Psychology, 90(5), 751-783. https://doi.org/10.1037/0022-3514.90.5.751
Pettigrew, T. F., & Tropp, L. R. (2008). How does intergroup contact reduce prejudice? Meta-analytic tests of three mediators. European Journal of Social Psychology, 38(6), 922-934. https://doi.org/10.1002/ejsp.504
Table of Contents
