Skip to content
Psych Dictionary

Bobo doll experiment

  • Concept · Learning theory
  • Mixed or contested
  • Published October 19, 2026
  • 10 sources
  • 6 min read

In short

The Bobo doll experiment was a 1961 study by Albert Bandura, Dorothea Ross and Sheila Ross in which preschool children who had watched an adult attack an inflatable doll reproduced the adult’s distinctive acts. Its findings on imitation are clear; what they show about aggression is disputed.
Related in this topic
On this page

Key facts

Field
Learning theory
Originators
Albert Bandura
First described
1961, Stanford University Nursery School
Also known as
Bobo doll study; Bobo doll experiments; Bandura’s Bobo doll experiment

State of the evidence

Mixed or contested

The 1961 study and its successors consistently found that young children reproduce novel acts they have seen an adult perform, without reinforcement and in a new setting[1], and a 1965 study showed that a model’s consequences affect whether children perform the acts but not whether they learn them[2]. Whether the findings show aggression is contested: the target was a toy, the test followed deliberate frustration and lasted 20 minutes, the procedure has been criticized for demand characteristics and permissive cues[3], and an imitation effect measured soon after exposure had disappeared a week later in one replication[4].

Evidence labels summarise the research literature; they are not recommendations. How labels are assigned

The Bobo doll experiment was a study by Albert Bandura, Dorothea Ross and Sheila A. Ross, published in 1961 as “Transmission of aggression through imitation of aggressive models”. Preschool children watched an adult either attack a large inflatable doll or play quietly, and were later observed alone with a smaller doll of the same kind. Children who had seen the attack reproduced many of its distinctive acts and words; children in the other conditions rarely did[1]. It became one of the founding studies of social learning theory, which describes learning by observation.

The name also covers the follow-up studies with filmed, rewarded or punished models. The theory they supported has its own entry, as does Albert Bandura.

Participants and conditions

The participants were 36 boys and 36 girls from the Stanford University Nursery School, aged 37 to 69 months, with a mean age of 52 months. A male and a female adult served as models, and one female experimenter ran the study[1]. There were eight experimental groups of six children and a control group of 24. Half the experimental children saw an aggressive model and half a subdued, nonaggressive one, and within each condition half saw a model of their own sex. The control group saw no model[1].

To make the groups comparable, the experimenter and a nursery school teacher rated each child’s aggressiveness in the nursery school on four five-point scales, covering physical aggression, verbal aggression, aggression toward objects and aggression anxiety. Children were then matched in triplets on the composite score and assigned at random to the three conditions[1].

What the children saw

Each child sat at a table in one corner of a room, making pictures with potato prints and stickers. In the opposite corner the model had a tinker toy set, a mallet and a 5-foot inflated Bobo doll. The nonaggressive model assembled the tinker toys quietly and ignored the doll. The aggressive model began with the tinker toys but after about a minute turned on the doll for the rest of the 10-minute session[1].

The authors wanted acts that children were unlikely to produce unless they had seen them. Besides punching the doll, which they expected children to do without any demonstration, the model laid the doll on its side, sat on it and punched it in the nose, struck it on the head with the mallet, tossed it in the air and kicked it about the room, saying “Sock him in the nose”, “Pow” and, twice, the nonaggressive remarks “He keeps coming back for more” and “He sure is a tough fella”[1].

Frustration and the test session

Before the test, every child was mildly frustrated. In an anteroom with attractive toys, the experimenter let the child play for about two minutes, then said these were her very best toys and that she had decided to reserve them for the other children[1]. The authors gave two reasons: watching aggression might reduce the observer’s own instigation to aggression, and low aggression after real instigation would point to inhibition[1].

The test room, in a building the children took to be off the nursery school grounds, held a 3-foot Bobo doll, a mallet and peg board, two dart guns and a tether ball with a face painted on it, along with nonaggressive toys such as a tea set, crayons, dolls and farm animals. Each child spent 20 minutes there while observers behind a one-way mirror scored behavior in 5-second intervals[1].

The procedure of the 1961 Bobo doll experimentA flow read from top to bottom, with page locators from the 1961 report. Participants: 36 boys and 36 girls from the Stanford University Nursery School, aged 37 to 69 months, page 575. Rated and matched: aggressiveness rated on four five-point scales; children matched in triplets on the composite score and assigned at random to the three conditions, page 576. The flow then splits into three conditions. Aggressive model: after about a minute, attacked a 5-foot Bobo doll for the rest of a 10-minute session, page 576. Nonaggressive model: assembled tinker toys quietly and ignored the doll, page 576. Control: 24 children, no model, pages 575 to 576. A bracket under the two model conditions notes eight experimental groups of six, with half of each model condition seeing a model of their own sex, pages 575 to 576. The three conditions converge on mild frustration, for every child: about two minutes with attractive toys, then told they were reserved for other children, pages 576 to 577. Last, the test session, 20 minutes: a 3-foot Bobo doll among aggressive and nonaggressive toys, with observers behind a one-way mirror scoring behavior in 5-second intervals, page 577.THE 1961 PROCEDURE, IN ORDERParticipants36 boys and 36 girls from the Stanford University Nursery School,aged 37 to 69 months (p. 575)Rated and matchedAggressiveness rated on four five-point scales; children matchedin triplets on the composite score and assigned at random to thethree conditions (p. 576)Aggressive modelAfter about a minute,attacked a 5-footBobo doll for therest of a 10-minutesession (p. 576)NonaggressivemodelAssembled tinker toysquietly and ignoredthe doll (p. 576)Control24 children; no model(pp. 575–576)Eight experimental groups of six; in eachmodel condition, half saw a model of theirown sex (pp. 575–576)Mild frustration, for every childAbout two minutes with attractive toys, then told they werereserved for other children (pp. 576–577)Test session, 20 minutesA 3-foot Bobo doll among aggressive and nonaggressive toys;observers behind a one-way mirror scored behavior in 5-secondintervals (p. 577)

Figure 1The 1961 study matched nursery school children on rated aggressiveness, assigned them to an aggressive model, a nonaggressive model or no model, mildly frustrated every child, and then observed each one for 20 minutes in a room with a 3-foot Bobo doll. Numbers and page locators from Bandura, Ross and Ross (1961).

Text description

A flow read from top to bottom, with page locators from the 1961 report. Participants: 36 boys and 36 girls from the Stanford University Nursery School, aged 37 to 69 months, page 575. Rated and matched: aggressiveness rated on four five-point scales; children matched in triplets on the composite score and assigned at random to the three conditions, page 576. The flow then splits into three conditions. Aggressive model: after about a minute, attacked a 5-foot Bobo doll for the rest of a 10-minute session, page 576. Nonaggressive model: assembled tinker toys quietly and ignored the doll, page 576. Control: 24 children, no model, pages 575 to 576. A bracket under the two model conditions notes eight experimental groups of six, with half of each model condition seeing a model of their own sex, pages 575 to 576. The three conditions converge on mild frustration, for every child: about two minutes with attractive toys, then told they were reserved for other children, pages 576 to 577. Last, the test session, 20 minutes: a 3-foot Bobo doll among aggressive and nonaggressive toys, with observers behind a one-way mirror scoring behavior in 5-second intervals, page 577.

Findings reported in 1961

Children who had watched the aggressive model reproduced a good deal of its physical and verbal aggression, while those in the nonaggressive and control groups showed virtually no imitative aggression and did not differ from each other. About a third of the children in the aggressive condition also repeated the model’s nonaggressive remarks[1]. Exposure to the aggressive model also increased nonimitative aggression compared with the nonaggressive condition, but the conditions did not differ in how often children punched the doll or played aggressively with the guns[1].

Boys imitated more physical aggression than girls but not more verbal aggression, and the male model influenced the children more than the female model[1]. The authors concluded that observation alone can elicit responses whose original probability is very low, and that imitation may short-cut the acquisition of new behavior without reinforcement of successive approximations[1].

Film, consequences and the 1965 study

Later studies extended the design. A 1963 study compared children who saw real-life aggressive models with children who saw the same models on film[5], and a later summary of the series reports that children who watched filmed people or cartoons behaving aggressively also tended to imitate them[6]. A second 1963 study of 80 children showed a televised sequence in which an aggressive model was either rewarded or punished; according to a published summary of the paper, imitation depended on the consequences the model received[7].

In 1965 Bandura tested whether a model’s consequences affect learning or only performance. Children who saw a filmed model punished reproduced fewer of the model’s acts than children who saw the model rewarded or left without consequences. When all the children were then offered attractive incentives for reproducing the acts, the differences disappeared: the three groups had learned the same amount[2]. Reward or punishment observed being delivered to someone else is the subject of vicarious reinforcement[7]. The learning–performance distinction is also central to latent learning.

Methodological criticisms

The first criticism concerns the target. The doll was an inflatable toy, and the authors themselves classed punching it as something children would do without a demonstration[1]. The experiments show that children copy specific acts directed at a toy; whether this is aggression in the sense of an intent to harm is a separate question. A psychoanalytic opinion article has read the children’s inventive use of the toys as evidence of the playful character of their imitation, while acknowledging that its formulations are not supported by research data[6].

The second concerns the situation. A 2005 review of laboratory aggression paradigms included the Bobo doll modeling paradigm among the classic designs and argued that criticisms first set out in 1996 still applied, among them demand characteristics, permissive cues and the availability of non-aggressive options[3]. The 1961 authors noted a related point: to the extent that seeing an adult act aggressively communicates permission to do so, it may weaken children’s inhibitions[1]. Scoring was also not fully blind. The male model scored every session, and children who had seen the aggressive model could be readily identified by their distinctive behavior[1].

The third concerns time. Children were tested immediately after exposure, for 20 minutes[1]. In a 1970 study, children who had already played in the setting shown in an aggressive film imitated less than children meeting it for the first time, and when the children returned a week later the groups no longer differed[4]. The experiments have also been criticized on ethical grounds for exposing young children to violence with unknown short- and long-term consequences[6].

The media-violence debate

The experiments helped to inspire research on children’s exposure to violence, including violence in the media[6]. Their relevance to that question is contested and should be kept separate from what they showed. A 1986 study, presented by its authors as moving beyond the Bobo doll, found that men’s reported childhood television viewing did not by itself predict violent crime; heavy viewing was related to violent crime only in combination with parental abuse[8].

Research on violent video games divides along similar lines. A 2010 meta-analysis, which drew its predictions from social-cognitive models, concluded that exposure is a causal risk factor for aggressive behavior, cognition and affect[9]. A 2015 meta-analysis of 101 studies of children and adolescents found the association with aggression to be minimal (r = .06) and reported continuing problems with publication bias[10]. The laboratory finding that children copy an adult’s acts toward a toy does not decide between these readings.

Sources10

Numbered in order of first citation.

  1. 1.Bandura, A., Ross, D., & Ross, S. A. (1961). Transmission of aggression through imitation of aggressive models. The Journal of Abnormal and Social Psychology, 63(3), 575–582. https://doi.org/10.1037/h0045925
  2. 2.Bandura, A. (1965). Influence of models' reinforcement contingencies on the acquisition of imitative responses. Journal of Personality and Social Psychology, 1(6), 589–595. https://doi.org/10.1037/h0022070
  3. 3.Ritter, D., & Eslea, M. (2005). Hot Sauce, toy guns, and graffiti: A critical account of current laboratory aggression paradigms. Aggressive Behavior, 31(5), 407–419. https://doi.org/10.1002/ab.20066
  4. 4.Kniveton, B. H., & Stephenson, G. M. (1970). The effect of pre-experience on imitation of an aggressive film model. British Journal of Social and Clinical Psychology, 9(1), 31–36. https://doi.org/10.1111/j.2044-8260.1970.tb00634.x
  5. 5.Bandura, A., Ross, D., & Ross, S. A. (1963). Imitation of film-mediated aggressive models. The Journal of Abnormal and Social Psychology, 66(1), 3–11. https://doi.org/10.1037/h0048687
  6. 6.Galanaki, E., & Malafantis, K. D. (2022). Albert Bandura's experiments on aggression modeling in children: A psychoanalytic critique. Frontiers in Psychology, 13. https://doi.org/10.3389/fpsyg.2022.988877
  7. 7.Bandura, A., Ross, D., & Ross, S. A. (1963). Vicarious reinforcement and imitative learning. The Journal of Abnormal and Social Psychology, 67(6), 601–607. https://doi.org/10.1037/h0045550
  8. 8.Heath, L., Kruttschnitt, C., & Ward, D. (1986). Television and violent criminal behavior: Beyond the Bobo doll. Violence and Victims, 1(3), 177–190. https://doi.org/10.1891/0886-6708.1.3.177
  9. 9.Anderson, C. A., Shibuya, A., Ihori, N., Swing, E. L., Bushman, B. J., Sakamoto, A., Rothstein, H. R., & Saleem, M. (2010). Violent video game effects on aggression, empathy, and prosocial behavior in Eastern and Western countries: A meta-analytic review. Psychological Bulletin, 136(2), 151–173. https://doi.org/10.1037/a0018251
  10. 10.Ferguson, C. J. (2015). Do Angry Birds make for angry children? A meta-analysis of video game influences on children’s and adolescents’ aggression, mental health, prosocial behavior, and academic performance. Perspectives on Psychological Science, 10(5), 646–666. https://doi.org/10.1177/1745691615592234

Further reading

  • Bandura, A., Ross, D., & Ross, S. A. (1961). Transmission of aggression through imitation of aggressive models. The Journal of Abnormal and Social Psychology, 63(3), 575–582. https://doi.org/10.1037/h0045925
  • Ritter, D., & Eslea, M. (2005). Hot Sauce, toy guns, and graffiti: A critical account of current laboratory aggression paradigms. Aggressive Behavior, 31(5), 407–419. https://doi.org/10.1002/ab.20066
  • Galanaki, E., & Malafantis, K. D. (2022). Albert Bandura's experiments on aggression modeling in children: A psychoanalytic critique. Frontiers in Psychology, 13. https://doi.org/10.3389/fpsyg.2022.988877

Cite this entry

  • APA 7

    Psych Dictionary. (2026, October 19). Bobo doll experiment. In Psych Dictionary. https://www.psychdictionary.com/concepts/bobo-doll-experiment

  • MLA 9

    “Bobo Doll Experiment.” Psych Dictionary, 19 Oct. 2026, www.psychdictionary.com/concepts/bobo-doll-experiment.

Written by Editorial team. First published October 19, 2026. Editorial standards

Educational reference material. Nothing on this site is a diagnosis, a treatment recommendation or a substitute for care from a qualified clinician.