The Robbers Cave Experiment

What the study set out to test

The Robbers Cave experiment was conducted in 1954 by Turkish-American social psychologist Muzafer Sherif, together with Carolyn Sherif and colleagues, at Robbers Cave State Park in Oklahoma. It was a field experiment disguised as an ordinary summer camp, designed to study how conflict between groups forms and how it might be reduced.

Sherif's guiding hypothesis became known as realistic conflict theory: that intergroup hostility arises when groups compete for limited resources, and that cooperation toward goals neither group can achieve alone can reduce that hostility. The camp was structured to test this idea in three phases.

Phase one: forming the groups

The researchers recruited 22 boys, around eleven years old, who were similar in background and did not know one another beforehand. The boys were divided into two groups and, at first, were kept unaware of the other group's existence.

Each group spent the first phase bonding through normal camp activities such as hiking, swimming, and pitching tents. Over a few days, each developed its own identity, norms, leaders, and even names: one group called itself the Rattlers, the other the Eagles. This phase showed how quickly a sense of group belonging and structure can form among strangers.

Phase two: competition and conflict

In the second phase, the researchers brought the two groups together for a tournament of competitive games such as tug-of-war and baseball, with prizes for the winners and nothing for the losers. This created exactly the kind of zero-sum competition Sherif wanted to study.

Hostility appeared almost immediately and escalated fast. The groups taunted each other, refused to eat together, raided and ransacked each other's cabins, burned the other group's flag, and came close to physical fights. Negative stereotypes of the other group hardened, while loyalty within each group intensified.

The speed and intensity of the conflict were striking given that the boys were ordinary, well-adjusted children with no prior grievances. The competition itself appeared sufficient to generate genuine intergroup hostility.

Phase three: reducing the conflict

In the final phase, the researchers tried to undo the hostility. Simply bringing the groups together for pleasant activities such as shared meals and a film did not work; these contacts often became occasions for more conflict.

What did work was the introduction of superordinate goals: problems that were important to both groups and could only be solved if the groups worked together. The researchers staged situations such as a breakdown in the camp's water supply, a stuck truck that needed all the boys to pull it, and a jointly funded movie.

As the groups cooperated to solve these shared problems, hostility gradually declined. By the end of the camp, friendships had formed across group lines, and the boys chose to travel home together. The lesson was that contact alone is not enough to heal conflict, but cooperation toward a common goal can.

What the experiment showed

The study is the classic demonstration of realistic conflict theory. It indicated that competition over scarce resources is enough to generate prejudice and hostility between groups, even among people with no history of conflict, and that strong in-group loyalty and out-group hostility can develop together and fast.

It also offered a hopeful counterpart: superordinate goals that require interdependent cooperation can reduce intergroup hostility where mere contact fails. This finding has influenced approaches to reducing prejudice in schools and workplaces, including cooperative learning methods built around shared tasks.

Criticisms and context

The study has important limitations. The sample was small and narrow: 22 boys of similar age, background, and culture, which limits how widely the findings generalize. The researchers were not neutral observers but actively engineered the conditions, raising questions about how much the conflict was provoked rather than spontaneous.

Records also show that Sherif had run an earlier version of the study in 1953 that failed, partly because the boys suspected manipulation and turned against the researchers rather than each other. Critics have argued that this selective reporting flatters the 1954 results. There are ethical concerns too, since the boys were deceived and deliberately placed into conflict without their families' full understanding.

Even with these caveats, the core findings on competition and cooperation have been broadly supported by later research, and Robbers Cave remains a foundational study in the psychology of intergroup relations.