Prejudice and intergroup relations (2): reducing prejudice
The Social Psychology of Democracies in Transition
Overview of the lecture
Last week’s lecture developed the social-psychological theory of prejudice and argued that post-communist CEE is distinctive for producing patterns of prejudice that operate largely without the intergroup contact that ordinary prejudice requires: anti-Semitism without Jews, anti-Muslim prejudice with almost no Muslims, anti-migrant sentiment with few migrants. Today’s lecture asks what can be done about prejudice, with particular attention to the central claim of the field that intergroup contact under the right conditions is the single most effective intervention for reducing it. The lecture surveys the classical contact hypothesis (Allport 1954), its four conditions, its difficulties when contact cannot be organised, the indirect alternatives (extended, vicarious, imagined, parasocial contact), and the three social-identity-based strategies (decategorisation, recategorisation, mutual differentiation). The second half of the lecture applies the apparatus to Islamophobia in CEE, using Pickel and Öztürk’s 2018 analysis Islamophobia Without Muslims? to show that the CEE pattern of abstract prejudice is precisely the pattern contact theory predicts would arise in societies with little direct out-group experience, and to explore what this implies for prejudice reduction in contexts where organising contact is difficult.
Why information alone is not enough
Prejudice feels ubiquitous enough that the question sometimes arises whether it is inevitable. The social-activist tradition of the early twentieth century held that the answer was to educate: prejudice is typically based on false information, so correcting the information should reduce the prejudice. The record has been disappointing. When people are presented with one or two examples that seem to refute their existing stereotype, most do not change the underlying belief; they treat the examples as exceptions, which do nothing to disturb the rule. In one experimental paradigm, participants shown disconfirming evidence actually strengthened their stereotypical belief because the challenge forced them to generate additional arguments for the prejudice, entrenching it further. The emotional and identity-based components of prejudice are not reachable through factual correction alone, and the intervention portfolio has had to reach further than the original activists anticipated.
The contact hypothesis
The central framework for prejudice reduction in social psychology is the contact hypothesis, articulated most clearly in Gordon Allport’s 1954 The Nature of Prejudice. The hypothesis holds that social interaction between members of different groups will, under the right conditions, reduce prejudice. Seventy years of laboratory and field research have supported the claim across a remarkable range of contexts: young people’s attitudes toward the elderly, healthy people’s attitudes toward the mentally ill, non-disabled children’s attitudes toward disabled peers, heterosexual people’s attitudes toward LGBTQ+ communities, and many inter-ethnic and inter-religious relationships. Meta-analyses (Pettigrew and Tropp 2006, and subsequent updates) find that the effect is robust across studies, moderated by but not dependent on the specific conditions Allport specified.
The barriers to contact
The classical contact hypothesis requires each individual to directly experience intergroup contact, and this is a demanding condition. Many people live in homogeneous areas with little exposure to people who are different from them. Rural Slovakia has very few Muslims; small-town Poland has almost no Jewish residents; most CEE villages are ethnically and religiously monocultural. For these populations, direct contact is essentially unavailable, and yet it is precisely in these populations that some of the strongest prejudice is found. The contact hypothesis therefore has to be supplemented with forms of indirect contact, or it fails to address the most important cases in CEE.
Indirect contact
Extended contact is the finding that the mere knowledge that an in-group member has a close relationship with an out-group member can improve attitudes toward the out-group as a whole. Parasocial contact extends the logic to media: emotional connection to an out-group character or celebrity, formed through repeated media exposure, can produce attitude change similar in direction (if smaller in magnitude) to direct contact. Vicarious contact covers news and entertainment vignettes showing intergroup contact occurring. Imagined contact is the more recent addition: deliberate mental simulation of positive intergroup interactions can produce attitude change before any actual contact has occurred.
All four forms operate through three channels. They are informational (the in-group member’s friendship with the out-group signals that the out-group is not as threatening as the stereotype suggests). They are normative (the in-group member’s behaviour models what is acceptable in the group, and seeing successful intergroup interactions signals that engaging with out-groups is socially acceptable). And they are anxiety-reducing (imagined or vicarious experience of successful interaction reduces fear of actual future interactions). Indirect contact is particularly valuable in settings where direct contact is rare, such as most CEE societies’ relationship with Muslim populations.
Intergroup anxiety
Intergroup interactions are typically characterised by mistrust and anxiety, which are core reasons that people avoid interacting with out-group members in the first place. The discomfort runs deep: people show physiological threat patterns (elevated heart rate, elevated cortisol) during interaction with people who have stigmatised identities, even when their behaviour is outwardly friendly. One useful counter-finding is that people’s expectations of intergroup interaction are typically worse than the interactions actually turn out to be: the gap between anticipated and realised experience is a target for intervention. Programmes that reduce pre-contact anxiety (through imagined contact, through advance information, through structured introductions) reliably produce more positive contact outcomes than programmes that do not.
When contact backfires
The largest challenge to the classical hypothesis is that contact sometimes makes things worse. In situations of intense intergroup violence, mere contact does not reduce prejudice and may increase it. Even in violent contexts, however, high-quality contact (cross-group friendship, sustained cooperation) still predicts less prejudice and greater willingness to reconcile. The distinction between quantity and quality of contact matters: brief, superficial, or involuntary contact can reinforce negative stereotypes, while sustained, voluntary, cooperative contact reliably improves attitudes. This finding motivated the classical specification of conditions that make contact effective.
Allport’s four conditions
Allport specified that contact reduces prejudice only when four conditions are met. The groups must have equal status during the contact. They must pursue common goals and be aware of shared interests. The contact must involve intergroup cooperation rather than mere co-presence. And the contact must be supported by institutions, law, and custom. Superficial contact without these conditions is unlikely to reduce prejudice, and may reinforce existing hierarchies. Contact must be of sufficient frequency, duration, and closeness to allow true acquaintance between members of the different groups.
Equal status
Equal status during the contact is historically the most commonly violated condition. Contact between high-status and low-status groups typically takes place in settings that confirm rather than disturb the hierarchy: majority employers interviewing minority applicants, majority landlords screening minority tenants, teachers addressing pupils in segregated schools. In such contacts, the negative stereotype of the minority is confirmed rather than challenged, because the preconditions of the stereotype (subordinate role, dependent position) persist in the contact itself. Equal-status contact is needed to disconfirm negative stereotypes of the minority group.
Common goals and cooperation
Cooperation toward a common goal increases the likelihood of overcoming prejudice, and Muzafer Sherif’s Robbers Cave experiment remains the textbook demonstration. Two groups of boys at a summer camp, initially brought into conflict through competitive games, had their animosity dissolved only when they were given problems (a broken water supply, a broken camp bus) that neither group could solve alone. Creating willingness to cooperate with the out-group is especially difficult in situations of enduring intergroup conflict, where both parties are suspicious of each other’s motives for cooperation. In post-conflict reconciliation contexts (Bosnia, Rwanda, Northern Ireland), the design of cooperative activities that are meaningful to both sides has been one of the central challenges of intergroup intervention.
Institutional support
Contact works best when supported by law, custom, and authority. Without institutional backing, contact can generate more anxiety and less tolerance, because participants read ambiguous signals from the surrounding environment as potentially hostile. Social norms matter: the amount of prejudice a person expresses is strongly predicted by the perceived level of prejudice of other people in that context, especially by the perceived prejudice of in-group members. Institutional support for contact — through anti-discrimination law, through formal integration programmes, through public signalling that intergroup relations are a priority — creates the normative environment in which contact can succeed.
Islamophobia and the contact hypothesis: Pickel and Öztürk (2018)
The paradox
Gert Pickel and Cemal Öztürk’s 2018 article in the Journal of Nationalism, Memory and Language Politics examines the contact hypothesis as an explanation for anti-Muslim attitudes across Europe, placing CEE countries in comparative perspective with Western Europe. The study is situated in the aftermath of the 2015 refugee crisis, in which the perception of Islam and Muslims as an aggressive menace to Western societies — a perception that had gained importance after 9/11 — became politically dominant across the region. The article begins with a puzzle: Islamophobia appears to be more widespread in Eastern Europe than in Western Europe, even though Muslim communities are virtually nonexistent in most Eastern European societies.
Is this empirically true, and, if yes, how can it be explained? Pickel and Öztürk hypothesise that Islamophobia is more prevalent in Eastern Europe than in Western Europe, and that the pattern is caused by a mixture of more pronounced ethnocentric nationalism and fewer contacts with Muslims. Islamophobia itself is defined as negative attitudes toward individuals based on their perceived religious (Muslim) background; because even non-practising Muslims face discrimination for their ethnocultural characteristics, it is the ascribed group identity rather than actual religious practice that drives the prejudice.
The empirical pattern
The empirical pattern is striking. There is a mismatch between the prevalence of anti-Muslim statements made by public figures in CEE and the factual size of Muslim minorities in those societies. Eastern European citizens are, on average, more opposed to Muslim immigration than their Western European counterparts. At the country level, the smaller the actual Muslim minority in a society, the higher the average support for a “Muslim ban”: the societies with the least actual contact have the most prejudice. Czechia, Hungary, Estonia, Lithuania, and Poland cluster at the top of the prejudice distribution, with Muslim populations typically below 1% of the total. France, Belgium, the UK, and the Netherlands cluster at the bottom of the prejudice distribution, with Muslim populations in the 4–7% range.
Hypotheses
Pickel and Öztürk test the contact mechanism against two families of alternative explanation. The full set of hypotheses falls into three groups.
| H | Family | Claim |
|---|---|---|
| 1 | Contact | Individual contact with other ethnicities reduces Muslim-ban support |
| 2 | Contact | Smaller factual Muslim minority → higher societal support for Muslim ban |
| 3 | Contact | Lower frequency of contacts with immigrants → higher societal support for Muslim ban |
| 4 | Identity | Strong national identification → higher Muslim-ban support |
| 5 | Identity | Religious individuals → higher Muslim-ban support |
| 8 | Identity | Ethnocentric worldview → higher Muslim-ban support |
| 6 | Threat | Realistic threat perception (physical, material) → higher Muslim-ban support |
| 7 | Threat | Symbolic threat perception (cultural values) → higher Muslim-ban support |
| 9 | Threat | Perceived collective deprivation → higher Muslim-ban support |
The analytical question is whether the contact family dominates the identity and threat families once all three are entered into the model.
Design
The study uses European Social Survey 2014 data covering 18 EU member states plus Switzerland, Norway, and Israel. The CEE sample is limited to six countries: Czech Republic, Estonia, Hungary, Lithuania, Slovenia, and Poland. The dependent variable is a binary measure of support for a “Muslim ban” (allow no Muslims to come and live in the country). The analysis uses multilevel modelling to combine individual-level and country-level predictors.
Results at the individual level
At the individual level, the contact hypothesis receives clear empirical support in all CEE countries except Hungary. Respondents who maintain contact with migrants, make friends with people from other ethnicities, and perceive these contacts as pleasant are less inclined to support a Muslim ban. Much less empirical support emerges for the hypothesis that identification with an in-group — whether through religion or through nationalism — is a fundamental precondition of anti-Muslim prejudice. Islamophobia rests on quite similar social-psychological underpinnings across European societies: a mixture of threat perceptions and ethnocentrism drives anxiety toward Muslims. Fear of terrorism combined with a perception of sharp cultural differences produces both realistic and symbolic threat perception.
Results at the societal level
At the societal level, contacts with migrants dominate as an explanation of Islamophobic attitudes. The effect of threat perceptions and ethnocentrism diminishes substantially when contact with migrants is included in the model. The authors conclude that the absence of Muslim communities in Eastern Europe leads to fewer contacts, fewer friendships with immigrants, and more sceptical perceptions of the intergroup contacts that do occur. Less contact is the core driver of “Islamophobia without Muslims” in CEE. The finding is precisely what the contact hypothesis would predict: in the absence of the contact that would disconfirm stereotypes, the stereotypes are free to be shaped by media, elite discourse, and political mobilisation, without any corrective from lived experience.
Contact as antivenom
The conclusion is that there are good reasons to consider contact with migrants as an “antivenom” to Islamophobic attitudes. Individuals who maintain contact with out-groups, make cross-ethnic friendships, and perceive these contacts positively are less likely to feel prejudice toward Muslims. But because Muslim communities are nearly nonexistent in most CEE countries, most citizens have no opportunity to adjust their negative stereotypes through real-life experience.
Pickel and Öztürk identify several reasons for CEE’s special vulnerability. Terrorist attacks and the media’s tendency to portray Islam as an “unintegratable religion” shape the image of all Muslims, regardless of whether they are refugees, migrants, or native-born Europeans. CEE countries have fewer opportunities to adjust negative stereotypes through face-to-face encounter. Since the 2015 refugee crisis, right-wing populist parties have been highly successful across Europe, and citizens’ unease about the crisis has been exploited by politicians who portray Muslims as a threat and position themselves as defenders of Christian identity. In the absence of direct contact, CEE citizens increasingly accept anti-Muslim political positions, speeches, and activities as legitimate. The political consequence is that anti-Muslim prejudice is maintained by media and elite discourse in the absence of the lived experience that would challenge it. Intervention must therefore operate either through indirect contact mechanisms (parasocial, extended, imagined) or through shifting the media and political environment that sustains the stereotypes.
Contemporary evidence
Subsequent evidence has largely confirmed Pickel and Öztürk’s analysis while adding temporal nuance. The peak of anti-Muslim sentiment in CEE came during 2015–2016, at the height of the refugee crisis, when Hungary and Poland constructed fences, the Visegrád-4 refused the EU refugee quota system, and public opinion hardened across the region. Between 2016 and 2020, Islamophobic attitudes remained elevated in CEE but ceased their upward trajectory, while Western attitudes gradually softened as contact extended. The 2019 Pew Global Attitudes Survey shows Hungarian unfavourable views of Muslims at 58% (down from 72% in 2016), Italian at 55% (down from 69%), and most European countries showing declining unfavourable views over the three-year period. The pattern is consistent with the contact mechanism: where contact is possible, contact is happening, and attitudes are softening.
The 2022 Ukrainian refugee crisis demonstrates, on a compressed timescale, that contact-driven attitudinal change can also run in the other direction. A very large Ukrainian population arrived in Poland, Czechia, Slovakia, Hungary, Romania, and the Baltic states; the population was framed as culturally similar and victimised by a shared enemy; attitudes toward Ukrainians shifted rapidly in a positive direction; and the shift was sustained (though it declined somewhat with mobilisation fatigue, as discussed in Lecture 9). CEE societies appear to be capable of significant contact-driven attitude change, given the right framing and the right target population, even in the short run. The 2022 case is a proof of concept that the mechanism is active, not absent.
Paluck, Green and colleagues’ 2021 meta-analysis of prejudice-reduction interventions confirms that contact-based interventions remain among the most effective. Short, one-off “diversity training” interventions typically do not produce durable change; sustained, structured contact with institutional support remains the gold standard. The Allport framework has aged well; the principal weaknesses are in implementation rather than in theory.
Conclusions
The contact hypothesis, in its classical and extended forms, is the main tool social psychology has for reducing prejudice. It works well in principle and has substantial empirical support across seven decades of research. It faces a structural challenge in CEE, where the populations most prejudiced against out-groups (Muslims, Jews) are often those with the least direct contact with them, because the out-groups are largely absent. Pickel and Öztürk’s analysis shows that the absence of contact is the strongest country-level predictor of Islamophobia, not religiosity, nationalism, or threat perception in themselves.
Indirect contact — extended, vicarious, parasocial, imagined — is a partial substitute, but it is weaker than direct contact and is vulnerable to counter-framing by the media and political environments that sustain the prejudice. Institutional signalling against prejudice (anti-discrimination law, explicit elite condemnation of prejudiced rhetoric, educational curricula that model intergroup cooperation) reinforces the norms in which prejudice is less easily expressed; the reverse, in which political elites reward prejudice electorally, produces the opposite trajectory. The practical CEE lesson is that prejudice-reduction is achievable but politically contested, and the contest is more about which framings of out-groups dominate political discourse than about any psychological intervention on individual attitudes.
Questions for discussion
- If direct contact is the main tool for reducing prejudice but the most prejudiced populations have no out-group present to contact, is the social-psychological prescription structurally inadequate for CEE contexts? Do indirect forms of contact provide a sufficient substitute, or do they require conditions (a media environment willing to represent out-groups positively) that CEE often lacks?
- Pickel and Öztürk wrote before the 2022 Ukrainian refugee influx. That episode shows that contact can produce rapid attitudinal change under the right framing. What conditions would generalise the Ukrainian case to other out-groups, and what specifically about the Ukrainian case (cultural similarity, shared enemy, humanitarian framing) is not replicable for Muslim or Roma populations?
- Right-wing populism has made Islamophobia, anti-Roma prejudice, and anti-LGBTQ+ hostility electorally profitable in parts of CEE. Can prejudice-reduction interventions succeed against a political environment that actively rewards prejudice? What would successful intervention in such an environment look like, and who could deliver it?