PsychOS
The Robbers Cave Experiment: How Muzafer Sherif Turned Ordinary Boys Into Warring Tribes in Eleven Days
Twenty-two well-adjusted eleven-year-olds went to a normal-looking summer camp. Within a week and a half, they were raiding each other's cabins, burning flags, and calling each other cheaters and s...
Explore PsychOS
PsychOS combines psychology education with structured self-reflection tools. Explore assessments, articles, experiments, and related resources while treating results as educational signals rather than diagnosis.
In the summer of 1954, two buses dropped off two separate groups of eleven-year-old boys at a 200-acre Boy Scouts of America camp in the Robbers Cave State Park in Oklahoma. The boys didn't know each other. They didn't know the other group existed. Each bus had been carefully screened in advance: all the boys were white, Protestant, from stable two-parent middle-class homes, with no prior behavioral problems and average or above-average school performance. On paper, there was no meaningful difference between the two busloads of children.
Within one week, each group had named itself, the Eagles and the Rattlers, and had started acting like it was at war with the other. Within eleven days, the boys were raiding each other's cabins in the dark, burning the opposing team's flag, calling each other cheaters, sneaks, and worse, and in at least one instance had to be physically separated by adult staff before a real fight broke out. The researchers hadn't told them to fight. They had, however, given them exactly one thing: a reason to see the other group as competition.
The man who orchestrated all of this was Muzafer Sherif, a Turkish-American social psychologist trying to answer a question that felt, at the time, dangerously relevant. The world had just emerged from a war built substantially on the machinery of group hatred. Sherif wanted to know how much of that hatred required deep-seated prejudice or ideology, and how much could be manufactured, quickly and reliably, using nothing more than an ordinary human tendency to form groups and compete over scarce things.
THE HYPOTHESIS
Sherif's central hypothesis, later formalized as Realistic Conflict Theory, proposed that intergroup hostility doesn't require pre-existing differences in race, religion, ideology, or personality. It can emerge directly from real or perceived competition over limited resources, once two groups have been established and placed in a situation where one group's gain functionally requires the other group's loss.
This was a deliberately uncomfortable hypothesis, because it implied that prejudice wasn't necessarily rooted in deep character flaws or specific hateful beliefs. It could be manufactured, in ordinary, carefully selected children, using nothing more than a summer camp schedule and a tournament bracket.
THE METHODOLOGY
Sherif and his colleagues, including Carolyn Sherif, designed the study to unfold in three distinct phases, each specifically engineered to isolate a different stage of group conflict formation.
In the first phase, lasting about a week, the two groups of boys were kept entirely separate and unaware of each other's existence. Each group spent this time on ordinary camp activities, hiking, swimming, cooking, building group cohesion naturally through shared experience. It was during this phase that each group spontaneously named itself and developed its own internal norms, hierarchy, and sense of identity, entirely without contact with any outside group to compare itself against.
In the second phase, the researchers introduced the two groups to each other and immediately placed them into a series of competitive tournament games: baseball, tug-of-war, a treasure hunt, with the winning team receiving a trophy, medals, and pocketknives, while the losing team received nothing. This wasn't a subtle manipulation. The competition was explicit, scored, and had real, visible stakes for eleven-year-olds who cared very much about winning.
Hostility didn't require weeks to develop. Name-calling started almost immediately. Within days, each group began raiding the other's cabin, stealing or burning flags, and forming increasingly negative stereotypes of the other side, even though, again, there was no actual meaningful difference between the two groups beyond which color shirt they'd been assigned.
The third and final phase was the part of the study most often left out of popular retellings, and arguably its most important contribution. Sherif's team didn't simply document how conflict formed; they tried to reverse it. They first attempted simple contact, bringing the two groups together for pleasant, non-competitive shared activities like eating meals together or watching a movie. This largely failed. Shared meals frequently turned into food fights. Proximity alone didn't reduce hostility; it sometimes gave the boys more opportunities to express it.
What did work was the deliberate introduction of what Sherif termed superordinate goals: problems that affected both groups simultaneously and that neither group could solve alone. The researchers staged a series of engineered crises, a water supply "failure" that required both groups to work together to locate and fix a blockage, and a stalled supply truck that needed all the boys pulling together on a rope to get moving again. These weren't small requests. They required genuine cooperative effort toward a shared outcome neither group could reach independently.
THE EXECUTION
The shift produced by the superordinate goals was significant and, notably, took considerably less orchestrated effort to achieve than the original conflict had taken to create. After several rounds of cooperating on shared tasks, hostility between the Eagles and the Rattlers measurably declined. Name-calling dropped. Friendships began forming across group lines. By the end of the camp, the boys voted to ride home on the same bus together, and one group used its own prize money to buy the other group refreshments for the trip, an outcome that would have seemed implausible during the height of the flag-burning phase just days earlier.
The pattern that emerged across the full three-phase study was clear and, at the time, genuinely novel in experimental social psychology: group identity formed rapidly and spontaneously even among near-identical individuals; intergroup hostility escalated sharply once real competition with visible stakes was introduced; simple positive contact between hostile groups was insufficient to reduce that hostility on its own; and cooperation toward genuinely shared goals, where success required both groups' effort and neither group could take credit alone, was the intervention that actually worked.
It's worth being direct about the boundaries of what this study actually demonstrated. The sample consisted entirely of white, Protestant, socioeconomically similar American boys in the 1950s, a narrow demographic that limits how confidently the findings generalize to conflicts rooted in genuine historical, racial, or ideological difference, rather than an artificially constructed rivalry between two groups with no prior history. Sherif also ran an earlier, less successful pilot version of the study in 1953 that he largely abandoned and rarely publicized, a detail that has led some historians of psychology to note the published 1954 study may have benefited from refined methodology developed specifically because the first attempt didn't produce the intended result.
THE REVELATION
The Robbers Cave study became one of the founding pieces of evidence for Realistic Conflict Theory, and its central finding reframed how psychologists think about the origins of prejudice and intergroup hostility. Conflict, Sherif's data suggested, doesn't require deeply held bigotry or ideological difference to get started. It requires structure: two groups, a scarce resource, and a scoring system.
This connects directly to the mechanism explored in Groupthink Explained , where Irving Janis described how group cohesion, once established, can override a member's willingness to voice dissent or question the group's direction. The Robbers Cave study demonstrates the flip side of that same cohesion: the speed and ease with which an in-group identity forms is precisely what makes an out-group feel threatening once competition enters the picture. Sherif's Eagles and Rattlers didn't need pre-existing animosity. They needed only to first become a group, and then be told there was another group standing between them and a trophy.
It also extends the dynamics covered in Conformity Explained , since much of the boys' escalating hostility was reinforced socially within each group; a boy expressing doubt about calling the other team "cheaters" would have been going against an increasingly unified group norm, the same social pressure Solomon Asch would later demonstrate in a very different experimental context.
The third phase of the study, the part most often left out of shorter retellings, arguably carries the most practical relevance today. Sherif's finding that mere contact between hostile groups doesn't reliably reduce conflict, but shared, cooperative work toward superordinate goals does, has since influenced conflict resolution strategies far beyond summer camps, including organizational team-building, school desegregation policy discussions, and international peace-building frameworks that emphasize joint projects over simple proximity.
See Your Own In-Group Instincts More Clearly
Reading about the Eagles and the Rattlers is different from noticing how quickly you personally form a sense of "us" and "them," at work, in a friend group, online, and how much that division shapes who you're willing to extend trust or suspicion toward.
→ Take the Self Awareness Test
Free. No account required. Tests how clearly you recognize your own patterns around group loyalty, competition, and social identity.
Curious how this connects to your broader communication style and how you handle conflict or competition?
→ Explore the full PsychOS Assessment Lab
Frequently Asked Questions
What was the Robbers Cave Experiment?
It was a 1954 field study by Muzafer Sherif in which two groups of unfamiliar eleven-year-old boys at a summer camp were placed into competitive tournaments, rapidly developing intergroup hostility, which researchers later reduced through cooperative tasks requiring both groups to work together.
What is Realistic Conflict Theory?
It's the theory, developed from the Robbers Cave findings, proposing that intergroup hostility can emerge from real or perceived competition over limited resources alone, without requiring pre-existing prejudice, ideological difference, or personality-based hatred.
Did simple contact between the two groups reduce their hostility?
No. Sherif's team found that bringing the hostile groups together for shared, non-competitive activities like meals or movies largely failed to reduce conflict, and in some cases created more opportunities for open hostility.
What actually worked to reduce the conflict between the groups?
Superordinate goals, shared problems that affected both groups and required their combined effort to solve, such as a staged water supply failure, were the intervention that measurably reduced hostility and rebuilt cooperation between the two groups.
Are there limitations to how far the Robbers Cave findings generalize?
Yes. The study used a narrow, demographically similar sample of white, Protestant, middle-class American boys in the 1950s, which limits confidence in generalizing the findings to conflicts rooted in genuine historical, racial, or deep ideological division.
Want to see how PsychOS handles psychology research, sources, limitations, and interpretation? Read Editorial Standards