Bobo Doll Experiment

What was the Bobo doll experiment?

The original Bobo doll experiment studied observational learning: whether children would reproduce an adult’s aggressive actions after the adult was no longer present. Albert Bandura, Dorothea Ross, and Sheila Ross published it in 1961. The adult’s actions involved an inflatable Bobo doll, which children later encountered during a play test.

Bobo doll experiment design and results

The original report describes 72 nursery-school children assigned to aggressive-model, nonaggressive-model, or no-model conditions after matching on prior aggression ratings. Both male and female models appeared in the aggressive and nonaggressive conditions.

Before a later play test, the children experienced a brief frustrating interruption of play. Children exposed to aggressive models reproduced more of the modeled physical and verbal acts. The adult model was absent during the test, but an experimenter remained in the room.

Later variations: learning versus performance

In Bandura's 1965 study, children watched a filmed aggressive model who was rewarded, punished, or received no consequences. In the initial test, children who had seen punishment reproduced fewer of the model's actions than children in the other two groups.

The next step changed the interpretation. Children in all three groups were offered rewards for reproducing the modeled actions. The previously observed differences between groups disappeared. Children who had shown less imitation could nevertheless perform the actions when the incentives changed.

The study therefore separated acquisition—learning an action—from performance—showing it under the current conditions. Seeing less behavior in an initial test did not necessarily mean less had been learned. This was a later experiment about observed consequences, distinct from the 1961 comparison of aggressive and nonaggressive models.

Limitations and ethical questions

The main finding concerns imitation in this experimental setting. It does not establish that every measure of aggression increased or that exposure caused lasting violence outside the laboratory.

Exposing children to an aggressive model also raises an ethical question about risk. The study did not measure lasting harm, so its results cannot resolve that concern.