Muzafer Sherif, O. J. Harvey and colleagues · 1954–1961

The Robbers Cave experiment

If you split ordinary children into two teams and make them compete, will they turn on each other? And can anything bring them back together?

  • A Boy Scouts camp inside Robbers Cave State Park, southeastern Oklahoma
  • Contested

The question

In the early 1950s, after a world war and at the start of the Cold War, the social psychologist Muzafer Sherif wanted to know where hostility between groups comes from. One popular answer located it inside individuals: frustrated, maladjusted or prejudiced people. Sherif suspected something simpler and more unsettling. Perhaps ordinary, well-adjusted people turn against another group when the two groups want something that only one of them can have. And if that is how hostility starts, what would it take to end it?

How it was done

The research team recruited 22 boys from Oklahoma City. All were finishing fifth grade, with an average age of 11 years and 1 month. They were chosen to be as similar and as “typical” as possible: middle-class Protestant families, both parents at home, with normal school records and no known problems. Only one boy was taken from each school, and boys who already knew someone on the list were dropped, so the boys began as strangers. They were split into two groups of 11, matched on height, weight and athletic skill.

The camp was a fenced Boy Scouts site of about 200 acres, surrounded by Robbers Cave State Park in southeastern Oklahoma. The groups arrived on separate days, 19 and 20 June 1954, and the study ran for about three weeks. The researchers appeared to the boys as ordinary camp staff: a camp director and counsellors. Observers kept notes in private, about 1,200 photographs were taken, and some conversations were recorded with hidden microphones. The boys were not told they were in an experiment about groups.

The plan had three stages:

  1. Forming the groups. For about a week, each group lived on its own, unaware of the other at first. Through hikes, cookouts and shared chores, each developed leaders, habits and a name: the Rattlers (after the rattlesnakes they kept meeting) and the Eagles.
  2. Friction. Staff then let each group discover the other and announced a tournament: baseball, tug-of-war, tent pitching, cabin inspections, a treasure hunt. Only the winning team would get the trophy, medals and pocket knives.
  3. Integration. Finally, the researchers tried to reduce the hostility, first through friendly contact, then through problems that neither group could solve alone.

What came out

Competition produced hostility fast. Name-calling began before the first baseball game was under way, and the insults grew harsher, including a racial slur aimed at one boy. After losing a tug-of-war, the Eagles tore down and burned the Rattlers’ flag. The Rattlers retaliated by raiding the Eagles’ cabin at night, overturning beds. Raids, stolen property and a fist fight followed, and staff stepped in several times to stop the boys hurting each other. The Eagles won the tournament. Each group described itself in glowing terms and the other in negative ones.

Contact alone did not help. The researchers then brought the groups together in pleasant settings: shared meals, a film, fireworks. In all seven of these contact situations, friction stayed or grew. Meals turned into food fights.

Shared goals did. Next came a series of problems that required both groups. Staff secretly blocked the camp’s water supply, and boys from both groups searched for and cleared the blockage together. Staff said the camp could not cover the full cost of renting the film Treasure Island, so the two groups, with help from staff, pooled their money. On a camping trip, the truck bringing food “stalled”, and the boys pulled it together with the same rope they had used for tug-of-war.

The change was gradual. After the water episode, hostility returned at supper. But one shared problem after another, the tone shifted. The groups chose to ride home on the same bus, and the Rattlers spent a $5 prize on malted milks for all 20 remaining boys. Friendship choices crossing the group line rose from 6.4% to 36.4% among the Rattlers and from 7.5% to 23.2% among the Eagles. The share of Eagles rejecting the Rattlers fell from 95% to 47.1%.

Sherif’s conclusion became known as realistic conflict theory: hostility between groups grows out of competition for things only one side can have, and it is reduced by superordinate goals, goals both sides want but can reach only together. He also argued that one or two such episodes were not enough; it took a series.

What really happened (replications and critiques)

There was an earlier camp that “failed”. The 1954 report mentions in passing a 1953 study that was stopped during its conflict stage because of “various difficulties and unfavorable conditions.” It was held at a camp near Middle Grove, New York, with groups called the Panthers and the Pythons. Drawing on archival material and interviews, the writer Gina Perry reconstructed it in The Lost Boys (2018). In her account and in press reports of it, staff stole clothing, cut a flag rope, knocked down a tent and broke a boy’s ukulele, hoping each group would blame the other. Instead, the boys largely refused to turn on each other, came to suspect the staff, and rebuilt the tent together. Sherif ended the study. Unpublished notes by Herbert Kelman, who took part, also describe the boys starting to see the experimenters as the real “other group”. Historians note that Sherif’s later descriptions of this camp shifted over time.

The Robbers Cave conflict was partly staged. The 1954 report itself says that staff kept tournament scores close and in the Eagles’ favour, and that the treasure hunt result was steered by the way routes were plotted. Staff also timed meals so that the Rattlers would find their burned flag first. Perry adds more from interviews and archives: staff supplied the matches for the flag burning, took photographs during raids, and noted bullying without intervening. One researcher, O. J. Harvey, told her the team let the Eagles win because some Eagles were homesick and the group might otherwise fall apart. Perry argues that the study was closer to a demonstration of a theory than a test of it.

These critiques are themselves debated. In his review of Perry’s book, Nick Haslam agreed the deception was troubling, but noted that it was typical of the era and argued that small-scale studies of hostility remain legitimate. Stephen Reicher and Alex Haslam read the 1953 “failure” as a finding in its own right: how people interpret a situation, and who they see as “us”, shapes whether conflict happens. They also suggest that adults standing by during fights may have signalled that hostility was acceptable.

Competition may not be necessary. In the early 1970s, Henri Tajfel and colleagues in Britain found that schoolboys favoured their own group even when the groups were based on something trivial, such as which of two painters they supposedly preferred (in fact the boys were assigned at random). There was no competition and nothing to win for themselves. This “minimal group” work suggests that simply being divided into groups can be enough to start favouritism, a challenge to the idea that real conflicts of interest are the main cause. Social identity theory later grew from these studies.

No full replication. A 1970 monograph by Lutfy Diab describes a similar camp study in Lebanon. Secondary sources report it was abandoned after one group stole the prize knife and threatened a boy from the other group with it; we have not been able to check the original.

Ethics

The boys did not know they were in an experiment. Parents gave permission and paid a fee, and they knew the camp was run by the University of Oklahoma for research on group interaction. But according to Perry, recruitment letters described a wholesome, cooperative camp, not a plan to set children against each other. Parents were told visits would not be allowed. Two boys went home homesick, and there is no record of the boys being debriefed. Former campers Perry traced decades later still did not know the conflict had been engineered. One boy from the 1953 camp told her: “Morally it was the wrong thing to do.”

A modern ethics board would almost certainly not approve this design. Studies with children today require informed parental consent to what will actually happen, the child’s own agreement, protection from harm, the freedom to stop, and a full explanation afterwards.

What it means today

The most durable lesson from Robbers Cave is also the most hopeful: in this camp, simply putting rival groups in the same room did not help, but working together on something both groups needed did. Later research supports a version of this. Meta-analyses by Thomas Pettigrew and Linda Tropp found that contact between groups is generally linked to less prejudice, and that the link is stronger under conditions Gordon Allport described in 1954, such as equal status, shared goals, cooperation and support from authorities.

Careful framing matters, though. This was one camp of 22 boys of the same background, steered by adults, and the conflict and the cure were both partly arranged. It shows what can happen, not what always happens. And the abandoned 1953 camp is a reminder that people do not simply react to competition: they can see through it, and refuse it.

Would you have passed?

Coming soon. You lead one of two rival teams. Choose how to respond to a series of camp events, from a lost game to a broken water tank, and see where your team ends up next to what happened in 1954. One honest caveat: a screen can’t recreate three weeks of living with a group, and the real camp was shaped by adults the boys trusted.

Questions people ask

What was the Robbers Cave experiment?

A 1954 field study by Muzafer Sherif and colleagues. Twenty-two boys at a summer camp in Oklahoma were split into two groups, the Eagles and the Rattlers, set against each other in a tournament, and then given problems that neither group could solve alone.

What are superordinate goals?

Goals that both groups want but neither can reach on its own, so they can only be achieved together. At Robbers Cave these included restoring the camp's water supply, paying for a film and pulling a stalled truck.

Was the Robbers Cave experiment ethical?

By today's standards, no. The boys did not know they were in an experiment, parents were told it was a cooperative camp, staff helped to stoke the conflict, and no debrief of the boys is recorded. Research ethics rules were far less developed in 1954.

Has the Robbers Cave experiment been replicated?

Not in full. Sherif's own 1953 attempt was abandoned when the boys refused to turn on each other and began to suspect the staff. A similar study in Lebanon in the 1960s, published in 1970, is reported to have been stopped. Lab studies such as Tajfel's minimal groups test parts of the theory instead.

Sources

  1. Sherif, M., Harvey, O. J., White, B. J., Hood, W. R., & Sherif, C. W. (1954/1961). Intergroup conflict and cooperation: The Robbers Cave experiment. Institute of Group Relations, University of Oklahoma; 1961 edition: Norman, University of Oklahoma Book Exchange (publisher from a secondary source). Online edition: Classics in the History of Psychology (C. D. Green, York University). psychclassics.yorku.ca/Sherif/index.htm
  2. Sherif, M. (1958). Superordinate goals in the reduction of intergroup conflict. American Journal of Sociology, 63(4), 349–356.
  3. Perry, G. (2014, 26 November). The view from the boys. The Psychologist (British Psychological Society). bps.org.uk/psychologist/view-boys
  4. Perry, G. (2018). The lost boys: Inside Muzafer Sherif's Robbers Cave experiment. Scribe.
  5. Shariatmadari, D. (2018). A real-life Lord of the Flies: The troubling legacy of the Robbers Cave experiment. The Guardian. www.inkl.com/news/a-real-life-lord-of-the-flies-the-troubling-legacy-of-the-robbers-cave-experiment
  6. Reicher, S., & Haslam, S. A. (2014, 27 November). Camps, conflict and collectivism. The Psychologist (British Psychological Society). www.bps.org.uk/psychologist/camps-conflict-and-collectivism
  7. Bazar, J. L. (2026, 19 July). Journey to Robbers Cave. Behavioral Scientist. behavioralscientist.org/journey-to-robbers-cave/
  8. Haslam, N. (2018). Review of The lost boys by Gina Perry. Australian Book Review, no. 402. www.dev.australianbookreview.com.au/259-abr-online/2018/june-july-2018-no-402/5899-nick-haslam-reviews-the-lost-boys-inside-muzafer-sherif-s-robbers-cave-experiment-by-gina-perry
  9. Cherry, F. (1995). The 'stubborn particulars' of social psychology: Essays on the research process. Routledge.
  10. Tajfel, H. (1970). Experiments in intergroup discrimination. Scientific American, 223(5), 96–102. doi:10.1038/scientificamerican1170-96
  11. Tajfel, H., Billig, M. G., Bundy, R. P., & Flament, C. (1971). Social categorization and intergroup behaviour. European Journal of Social Psychology, 1(2), 149–178. doi:10.1002/ejsp.2420010202
  12. Diab, L. N. (1970). A study of intragroup and intergroup relations among experimentally produced small groups. Genetic Psychology Monographs, 82, 49–82.
  13. 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, 922–934. doi:10.1002/ejsp.504