Decades of research confirm that meaningful contact between members of different groups reduces prejudice, and the effect holds across a wide range of group types and settings. A landmark meta-analysis of over 500 studies found that intergroup contact typically lowers bias, with more rigorous studies actually showing larger effects than weaker ones.1PubMed. A meta-analytic test of intergroup contact theory But the path from proximity to genuine harmony is far less straightforward than that headline suggests. The conditions under which contact occurs, how people perceive group boundaries, and what kind of threat they feel all shape whether mixing produces warmth or friction.
What Makes Contact Actually Work
The idea that bringing different groups together can reduce hostility traces back to the mid-twentieth century, and the accumulated evidence is now enormous. Across hundreds of independent samples spanning racial, ethnic, religious, age-based, and other group divisions, the pattern is remarkably consistent: intergroup contact tends to reduce prejudice, and these improvements in attitudes generalize beyond the specific individuals involved to the broader outgroup.2PubMed. A meta-analytic test of intergroup contact theory If you have a pleasant interaction with someone from a group you previously avoided, you tend to feel warmer not just toward that individual but toward the group as a whole.
That said, not all contact is created equal. The classic framework emphasizes certain conditions: equal status between the groups, cooperative goals, and institutional support. When those elements are present, contact is more likely to shift attitudes. And one key mechanism appears to be the way contact changes how people mentally sort others into “us” and “them.”
Redrawing the Boundaries of “Us”
One of the most influential ideas in this space is that harmony improves when people begin to see members of another group as part of a shared, larger group rather than as outsiders. Cooperative interaction helps people shift from thinking in terms of “us versus them” to a more inclusive “we.”3European Review of Social Psychology. The Common Ingroup Identity Model: Recategorization and the Reduction of Intergroup Bias This recategorization doesn’t require people to forget their original group membership. It just adds a layer of shared identity on top of it.
This model has been tested in real-world settings. College roommates from different racial and ethnic backgrounds who started the semester perceiving more commonality between their groups maintained stronger cross-group friendships over time, while those who didn’t perceive that shared identity saw friendships decline.4Journal of Experimental Social Psychology. Superordinate identity and intergroup roommate friendship development The sense of belonging to a common group acted as a buffer against the natural drift apart that often happens in cross-group relationships.
There is an interesting nuance here, though. Some research suggests that the lowest levels of bias emerge not when people focus only on a shared overarching identity but when both the shared identity and the original subgroup identities remain visible at the same time. In one experiment, bias was actually strongest when the overarching category alone was emphasized, and lowest when participants could hold both levels of identity in mind simultaneously.5Personality and Social Psychology Bulletin. Subgroup Relations: A Comparison of Mutual Intergroup Differentiation and Common Ingroup Identity Models of Prejudice Reduction The implication is that erasing difference entirely isn’t the goal. People respond best when they can maintain a distinct group identity while also feeling part of something larger.
Perspective-Taking and Its Complications
Deliberately trying to see the world through another group’s eyes is widely recommended as a bridge to harmony. And there is real evidence for it. Perspective-taking has been linked to more favorable evaluations of other groups, more positive body language during cross-group interactions, greater willingness to help outgroup members, and reduced reliance on stereotypes.6Social and Personality Psychology Compass. Perspective‐Taking as a Strategy for Improving Intergroup Relations: Evidence, Mechanisms, and Qualifications
But the picture gets messier when you look closely. Perspective-taking without mindful awareness can sometimes backfire, amplifying rather than reducing implicit bias. In one study examining attitudes toward elderly people, perspective-taking on its own was associated with greater negative implicit bias. It was only when perspective-taking was combined with a mindfulness component that some of the negativity decreased.7PubMed Central. The Impact of Mindfulness and Perspective-Taking on Implicit Associations Toward the Elderly: a Relational Frame Theory Account This suggests that imagining another group’s experience, if done carelessly or from a place of anxiety, can reinforce the very stereotypes it is meant to dissolve.
When Contact Breeds Friction Instead of Friendship
Harmony research has an uncomfortable finding that proponents of simple “just get people together” approaches sometimes underplay: contact can make things worse. When people approach cross-group encounters already feeling anxious, that anxiety tends to amplify threat perception, anger, and hostile impulses. In a series of experiments, intergroup anxiety caused people to interpret outgroup-initiated contact as more threatening, feel angrier, and respond more offensively.8PubMed. Testing the limits of tolerance: how intergroup anxiety amplifies negative and offensive responses to out-group-initiated contact The anxiety essentially shrank people’s tolerance for ambiguous or mildly uncomfortable social situations.
The type of threat matters, too. Research on Hindu-Muslim relations in India found that the threats predicting prejudice differed between the majority and minority groups. For the majority Hindu group, symbolic threat (the feeling that the other group’s values and beliefs are incompatible with your own) was a stronger driver of prejudice. For the minority Muslim group, realistic threat (concerns about economic competition and physical safety) was more central.9Journal of Community & Applied Social Psychology. The relationships between contact, status and prejudice: An integrated threat theory analysis of Hindu–Muslim relations in India This asymmetry means that interventions aimed at fostering harmony need to address different fears depending on which group they’re working with.
The Irony of Positive Relations
Here is a paradox that complicates the entire field: positive intergroup contact can actually reduce disadvantaged groups’ motivation to push for equality. Research has found that when members of disadvantaged groups have more positive contact with the advantaged group, they come to see the dominant group as fairer, which in turn predicts decreased support for social change.10PubMed. The irony of harmony: intergroup contact can produce false expectations for equality Harmony, in this reading, can function as a kind of sedative that makes real structural inequality feel less urgent to the people most affected by it.
This does not mean contact is bad. It means that warm feelings between groups are not the same thing as justice between groups, and the two can sometimes work at cross-purposes. A society where everyone gets along but vast inequalities persist hasn’t achieved meaningful harmony; it has achieved comfortable complacency. The researchers who work in this area increasingly emphasize that genuine intergroup harmony has to be paired with honest acknowledgment of power differences.
How Shared Threats Can Unite or Divide
The folk wisdom that a common enemy or shared crisis brings groups together turns out to be only half right. Whether a shared threat promotes cooperation depends on how that threat is framed and on the pre-existing relationship between the groups. When a crisis is framed as a safety threat (everybody’s physical well-being is at risk), cooperation tends to increase. But when the same crisis is framed as a resource threat (there isn’t enough to go around), cooperation drops. Across multiple scenarios involving wars, air pollution, and climate change, this pattern held consistently.11Social Psychological and Personality Science. Global Crisis Frame Matters for Unity: Resource Threat Hampers While Safety Threat Promotes Intergroup Cooperation
The pre-existing relationship adds another layer. When groups already have negative relations, resource threat framing makes things even worse, while safety threat framing can still improve cooperation. When groups already get along, both framings tend to increase cooperation.12Social Psychological and Personality Science. Opposing and Promoting Effects: Existing Intergroup Relations Shape the Divergent Effects of Common Threats on Intergroup Cooperation The practical takeaway is that how leaders and media describe a shared crisis matters enormously. Telling hostile groups they’re competing for scarce resources is a recipe for escalation. Telling them they face a shared danger to their safety opens a window.
Digital Contact and Parasocial Bonds
You don’t necessarily need to sit across a table from someone to build intergroup harmony. Digital intergroup contact, whether through social media, online forums, or video platforms, shows a small but consistent effect on prejudice reduction. A meta-analysis of 88 independent samples found a modest effect in favor of reduced prejudice from digital contact.13PubMed. Does Digital Intergroup Contact Reduce Prejudice? A Meta-Analysis These findings align with earlier work showing that online intergroup contact moderately reduces prejudice, consistent with what has been documented for face-to-face contact.14International Journal of Intercultural Relations. Allport meets internet: A meta-analytical investigation of online intergroup contact and prejudice reduction
Virtual reality offers a more immersive version of this. A review of 64 studies found that VR contact has potential for improving intergroup relations, but also a cautionary note: under certain conditions, VR contact can increase prejudice.15PubMed Central. The use of virtual reality in studying prejudice and its reduction: A systematic review The technology isn’t inherently helpful or harmful; its effect depends on design and context.
Even completely one-sided media exposure can shift attitudes. Parasocial relationships, the emotional bonds people form with characters on TV or YouTube, function as a form of extended contact. One study used YouTube videos to successfully reduce explicit prejudice and anxiety toward people with mental health issues, with the strength of the parasocial bond mediating the effect.16PubMed Central. Parasocial relationships on YouTube reduce prejudice towards mental health issues Similarly, longitudinal work with television characters has found that viewers can develop emotional connections with outgroup characters that influence their real-world attitudes and behaviors in much the same way that direct personal contact does.17Communication Research. The Development and Influence of Parasocial Relationships With Television Characters: A Longitudinal Experimental Test of Prejudice Reduction Through Parasocial Contact
Why Standard Diversity Training Disappoints
Given how much organizations spend on diversity training, you might expect a robust evidence base behind it. The reality is less encouraging. A sweeping review of hundreds of articles on the topic concluded that the literature does not allow decisive conclusions about best practices. Researchers noted that the enthusiasm for, and monetary investment in, diversity training has outpaced the available evidence that these programs achieve their goals.18PubMed Central. Diversity Training Goals, Limitations, and Promise: A Review of the Multidisciplinary Literature One persistent problem is that studies of diversity training frequently measure proxy outcomes like awareness or stated attitudes rather than the consequential behaviors the training is supposed to change.
A controlled experiment reinforced this skepticism. One-off online diversity training successfully shifted attitudes among groups that were already relatively less supportive, but it did not change their actual behavior. Meanwhile, groups that were already supportive before training showed some behavior change but didn’t need the attitude shift.19PubMed Central. The mixed effects of online diversity training The people whose behaviors organizations most want to change are the least likely to change them from a single training session. This doesn’t mean all training is useless, but it means that short, standalone workshops are unlikely to produce lasting intergroup harmony on their own.
Neighborhoods, Segregation, and Trust
Where you live shapes how you feel about other groups, but the relationship is more complex than “diverse neighborhoods equal more trust.” Research in Great Britain found that neighborhood ethnic diversity was associated with lower trust in neighbors, but only when those diverse neighborhoods were embedded in wider communities that were highly segregated. In areas where the broader region was also mixed, diversity at the neighborhood level didn’t depress trust nearly as much.20PubMed Central. Wider-community Segregation and the Effect of Neighbourhood Ethnic Diversity on Social Capital: An Investigation into Intra-Neighbourhood Trust in Great Britain and London The implication is that a diverse block surrounded by segregated areas creates a feeling of boundary and threat that undermines trust. Diversity needs to be more than a pocket surrounded by homogeneity.
The Evolutionary Baggage We Carry
Some of the difficulty in achieving intergroup harmony has deep evolutionary roots. The hypothesis of parochial altruism proposes that cooperativeness toward your own group and hostility toward outsiders may have co-evolved in humans, driven by violent conflict between groups.21PubMed Central. The evolutionary interplay of intergroup conflict and altruism in humans: a review of parochial altruism theory and prospects for its extension The traits that make us generous within our circles and suspicious of outsiders may be two sides of the same coin.
Experimental evidence from Papua New Guinea illustrates how deeply this runs. When observers were given the chance to punish people who violated social norms, they punished far more harshly on behalf of victims from their own group than victims from another group, regardless of who the violator was. And norm violators expected more leniency from punishers who belonged to their own group, which made them more willing to violate norms in the first place when the punisher was an ingroup member.22Nature. Parochial altruism in humans Our sense of fairness, at a deep level, is group-flavored.
Oxytocin, sometimes called the “bonding hormone,” reinforces this duality. Rather than making people universally warmer, oxytocin promotes ingroup favoritism and, to a lesser extent, outgroup hostility.23PubMed Central. Oxytocin promotes human ethnocentrism The same neurochemical system that makes you love your friends can make you distrust strangers. Similarly, brain imaging research shows that empathic responses are bounded by group membership. Neural activity in regions associated with empathy can differ depending on whether the person you’re watching is from your own group or another, and in some cases the brain responds to an outgroup member’s good fortune in ways that predict willingness to harm them.24PubMed Central. Bounded empathy: neural responses to outgroup targets’ (mis)fortunes
Cultural Variation in How Groups Manage Harmony
Not all cultures handle group boundaries the same way. Across 19 countries, researchers found that collectivistic cultures, which place a high value on group harmony, react more harshly to people who break social norms. In individualistic cultures, norm violators were seen as more powerful and provoked less moral outrage. In collectivistic cultures, it was the reverse: norm violators were seen as less powerful and sparked greater outrage.25PubMed Central. Cultural Collectivism and Tightness Moderate Responses to Norm Violators: Effects on Power Perception, Moral Emotions, and Leader Support Cultures that prize harmony enforce it more vigorously, which can look like social cohesion from the inside and conformity pressure from the outside.
Despite these cultural differences, some patterns in how people view ingroups and outgroups appear to be universal. A cross-cultural study found that across four distinct societies, people consistently viewed their ingroups, often family, as conventionally moral and caring, while perceiving various outgroups as hedonistic and self-serving.26PubMed. In- and out-groups across cultures: Identities and perceived group values The specific outgroups varied across societies, but the moral logic was the same everywhere: “we” are virtuous, “they” are selfish.
How Group Biases Take Root in Childhood
Stereotyping and prejudice emerge remarkably early in life, which means that efforts to foster intergroup harmony need to start young. Research in developmental psychology has identified the causal ingredients: children naturally categorize people into groups, and they readily absorb environmental cues about which groups matter and what those groups are supposedly like. A theoretical framework called developmental intergroup theory argues that these biases are largely under environmental control, meaning they can be shaped through educational and social policy.27Current Directions in Psychological Science. Developmental Intergroup Theory
This is encouraging in one respect: if the environment drives the formation of group bias, then changing the environment can change the bias. Schools that facilitate cross-group friendships, curricula that present diverse groups in complex rather than stereotypical ways, and families that model positive intergroup attitudes all have the potential to intervene before biases harden. The challenge, of course, is that children absorb messages from every direction, and contradictory signals from peers, media, and neighborhoods can undermine what any single institution tries to teach.
Moving in Sync
One of the more unexpected findings in intergroup research involves synchronized movement. Dancing, marching, or moving together in time appears to signal social bonds and group cohesion to observers. In experiments using virtual avatars engaged in group dance, researchers found that groups moving in complete unison were rated as both more socially close and more formidable than groups whose movements were misaligned.28PubMed. Signals through music and dance: Perceived social bonds and formidability on collective movement The effect only appeared when movements were temporally aligned; when timing was off, the perception of closeness vanished.
This finding connects to a broader phenomenon across cultures. Shared rituals, from religious ceremonies to national celebrations to workplace team-building, involve coordinated physical activity, chanting, singing, or dancing. The research suggests these practices aren’t just symbolic performances of unity. They actively generate perceptions of social bonding in participants and observers alike, which may explain why every known human culture has developed some form of collective ritual movement.