Skip to content
Psych Dictionary

Social learning theory

  • Concept · Learning theory and social psychology
  • Moderate evidence
  • Published October 19, 2026
  • 20 sources
  • 7 min read

In short

Social learning theory is Albert Bandura’s account of how people learn behavior by observing others and the consequences others receive, without performing it or being reinforced themselves. It introduced observational learning, vicarious reinforcement and self-efficacy, and became social cognitive theory in 1986.
Related in this topic
On this page

Key facts

Field
Learning theory and social psychology
Originators
Albert Bandura
First described
1941–1986, United States
Also known as
Observational learning theory; Social cognitive theory

State of the evidence

Moderate evidence

Controlled experiments with young children show that novel behavior can be learned by observation alone and that consequences to the model change whether it is performed, not whether it is learned[1][2]. Self-efficacy, the theory’s best-studied construct, correlates with work performance (r = .38 across 114 studies), and in a meta-analysis of experiments that raised it relative to a control condition, health behavior changed by d = .47[3][4]. The limits are substantial: the classic experiments used a toy as the target and brief tests, and imitation fell when children were familiar with the setting[1][5]; whether self-efficacy drives later performance within the same person is disputed[6][7]; and the theory’s extension to violent media rests on meta-analyses that disagree[8][9].

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

Social learning theory is the account, developed chiefly by Albert Bandura, of how people acquire new behavior by observing other people and the consequences that follow for them. Its central claim is that a behavior can be learned without being performed and without being reinforced. The observer represents what the model did and can reproduce it later[1][10]. This separates it from operant conditioning, in which a response must first be emitted before its consequences can strengthen it[1].

Observational learning is not the same as copying. Bandura described modeling as more than behavioral mimicry: observers can extract the rule behind a model’s actions and use it to produce behavior they have never seen[10]. Nor is learning the same as performance. What a person has learned by observation may not appear until there is an incentive to perform it[2][10].

The name is also used for Julian Rotter’s theory of personality, within which Rotter placed his construct of internal versus external control of reinforcement[11]. That is a separate theory, covered in the entry on locus of control.

Imitation as reinforced behavior in Miller and Dollard

Neal Miller and John Dollard used the term social learning in Social Learning and Imitation (1941)[12]. In their procedure, as Bandura later summarized it, a model made a choice and was rewarded. The observer was rewarded whenever the observer’s choice matched the model’s[1]. Imitation, on that view, is a learned habit. It is acquired because matching has been reinforced, and it generalizes to some extent to other adults[1].

Bandura and his colleagues argued that this treated imitation as a special case of discrimination learning. The model’s behavior served only as a cue for responses the observer could already make, and a light or a sound could have served as well[1]. Their own experiments asked a different question. Could a child acquire a new pattern of behavior by watching it, with no chance to perform it and no reinforcement for anyone[1]?

The Bobo doll experiments

In the 1961 study, Bandura, Dorothea Ross and Sheila Ross tested 72 children, 36 boys and 36 girls aged 37 to 69 months, from the Stanford University Nursery School[1]. Each child sat at a picture-making activity. In the opposite corner of the room, an adult either assembled toys quietly or attacked a 5-foot inflated Bobo doll for about ten minutes; a control group saw no model[1]. The aggressive models performed distinctive acts unlikely to occur by chance. They sat on the doll and punched its nose, struck it with a mallet, tossed it in the air and kicked it, saying “Sock him in the nose” and “Pow”[1].

Each child was then mildly frustrated. In an anteroom, the experimenter let the child begin playing with attractive toys and, after about two minutes, said they were reserved for other children[1]. The child then spent 20 minutes in a different room with a 3-foot Bobo doll, a mallet, dart guns and nonaggressive toys, observed through a one-way mirror[1]. Children who had watched the aggressive model reproduced many of its physical and verbal acts in close to their original form. About a third also repeated the model’s nonaggressive remarks. Children in the other two groups rarely did either, and did not differ from each other[1]. Boys imitated more physical aggression than girls, and the male model had more influence than the female, especially on physical aggression[1].

A 1963 study compared live models with the same models shown on film[13]. A later summary of the series reports that children who watched filmed people or cartoons behaving aggressively also tended to imitate them[14]. A second 1963 study showed 80 nursery-school children a short televised sequence in which an aggressive adult was either rewarded or severely punished. How much the children imitated depended on which consequence the model had received[15].

Limits of the Bobo doll evidence

The studies show that young children can reproduce a novel sequence of acts after watching it once, in a new setting, without reinforcement. They do not show that the children became more aggressive toward people. The target was an inflatable toy, the children had been deliberately frustrated beforehand, and the test lasted 20 minutes[1]. The male model scored every session. The authors also noted that children who had seen an aggressive model could be readily identified by their distinctive behavior, so scoring could not be fully blind[1]. Treatment conditions did not change how often children punched the doll, the act the authors expected children to perform without any demonstration[1].

Novelty mattered. Kniveton and Stephenson let some children play in the setting shown in an aggressive film before they saw it. Those children imitated less than children meeting the setting for the first time, and a week later the groups no longer differed[5]. A 2022 opinion article notes that the experiments have been criticized for exposing children to violence with unknown consequences. It offers a psychoanalytic reading in which arousal from separation, strange rooms and deliberate frustration, not observation alone, drove the children’s behavior, and its authors acknowledge that this reading is not supported by research data[14].

Four processes in observational learning

Bandura set out the theory in book form as Social Learning Theory (1977)[16]. In a later statement he divided observational learning into four subfunctions[10]:

  • Attentional processes determine what is selected from modeled events and what information is extracted. They depend on the observer’s skills and preferences, the salience and value of the activity, and which models are within reach[10].
  • Retention processes transform what was observed into symbolic codes and rules held in memory, aided by rehearsal. Recall is a reconstruction, not a simple retrieval[10].
  • Production processes translate those representations into action. Each attempt is compared with the remembered conception and corrected[10].
  • Motivational processes decide whether what has been learned is performed[10].
The four processes of observational learningA vertical sequence of six boxes joined by downward arrows. Modeled events, behavior observed in other people or in media, come first. Attentional processes determine which modeled events are observed and what is extracted from them. Retention processes represent what was observed in memory, and recall reconstructs it. Production processes turn the representation into action, adjusted toward the conception. Motivational processes determine whether what was learned is performed, given incentives that are direct, vicarious or self-produced, shown as three labels inside that box. The sequence ends in matching performance, behavior that matches the modeled pattern.Modeled eventsBehavior observed in other people or in mediaAttentional processesWhich modeled events are observed and what is extractedfrom themRetention processesRepresenting what was observed in memory; recallreconstructs itProduction processesTurning the representation into action, adjusted toward theconceptionMotivational processesWhether learning is performed depends on incentives:DirectVicariousSelf-producedMatching performanceBehavior that matches the modeled pattern

Figure 1In Bandura’s social cognitive account, learning from observed behavior passes through attentional, retention, production and motivational processes. Motivation decides whether what was learned is performed, so learning and performance can come apart. After Bandura (2001).

Text description

A vertical sequence of six boxes joined by downward arrows. Modeled events, behavior observed in other people or in media, come first. Attentional processes determine which modeled events are observed and what is extracted from them. Retention processes represent what was observed in memory, and recall reconstructs it. Production processes turn the representation into action, adjusted toward the conception. Motivational processes determine whether what was learned is performed, which depends on incentives that are direct, vicarious or self-produced, shown as three labels inside that box. The sequence ends in matching performance, behavior that matches the modeled pattern.

In this account, observation fails to produce imitation for different reasons at each step: an observer may not attend, may not retain what was seen, may lack the component skills, or may have no incentive to perform[10].

Vicarious consequences and the learning–performance distinction

Bandura identified three kinds of incentive that govern whether learned behavior is performed. Direct consequences follow the observer’s own actions, as analyzed in schedules of reinforcement and negative reinforcement versus punishment. Vicarious consequences are seen to follow for others. Self-produced consequences arise from the observer’s own standards[10]. Vicarious consequences work, in his account, through outcome expectations: seeing a model rewarded creates the expectation that the same action will be rewarded[10].

The 1965 experiment separated learning from performance. Children watched a filmed model behave aggressively and then be rewarded, punished or left without consequence. Children who had seen the model punished reproduced fewer of the model’s acts. All the children were then offered attractive rewards for reproducing what the model had done, and the differences disappeared. The three groups had learned equally; the model’s consequences had affected only what they chose to perform[2].

Self-efficacy

In 1977 Bandura proposed that psychological treatments, whatever their form, work by altering the level and strength of self-efficacy[17] . He distinguished an efficacy expectation, the conviction that one can carry out the behavior needed to produce an outcome, from an outcome expectancy, the estimate that the behavior will lead to that outcome. A person may believe that an action would produce a result and still not act, because of doubt about being able to perform it[17]. Efficacy expectations, he hypothesized, determine whether coping behavior is begun, how much effort is spent and how long it is sustained against obstacles. They derive from four sources of information: performance accomplishments, vicarious experience, verbal persuasion and physiological states[17].

The evidence is extensive but not uniform. A meta-analysis of 114 studies found a weighted mean correlation of .38 between self-efficacy and work-related performance[3]. Experimental evidence is more direct. A meta-analysis of 204 experiments that changed attitudes, norms or self-efficacy relative to a control condition found that raising self-efficacy changed intentions by d = .51 and health behavior by d = .47[4]. Within-person studies are more contested. Vancouver and colleagues reported that manipulating self-efficacy on an analytic game was negatively related to performance on the next trial, and linked higher self-efficacy to overconfidence and logic errors[18]. Bandura and Edwin Locke replied that nine meta-analyses and the goal-setting literature contradict the claim that self-efficacy is self-debilitating[7]. A later meta-analysis of within-person data concluded that self-efficacy is primarily a product of past performance, not a driver of future performance. Its effect on later performance was near zero (ρ = .06) once trends over time were controlled[6].

Social cognitive theory

In Social Foundations of Thought and Action (1986), Bandura formulated what he called social cognitive theory and gave more weight to less observable processes such as self-regulation, self-efficacy and expectations[19][14]. The theory explains functioning by triadic reciprocal causation, in which personal factors (cognitive, affective and biological), behavior and environmental events all influence one another[10]. It treats people as agents who act through intention and forethought, regulate themselves by their own standards and reflect on their capabilities. It distinguishes personal agency, proxy agency exercised through others, and collective agency[20].

Observational learning, vicarious motivation and self-efficacy all remain in the later theory, where people are described as self-organizing, proactive, self-reflecting and self-regulating, not just reactive organisms shaped by environmental events or inner forces[10].

Extension to media violence

The Bobo doll studies inspired research on children’s exposure to violence, including violence in the media[14], and the question of media effects remains contested[9]. A 2010 meta-analysis, which used social-cognitive models to generate its predictions, concluded that exposure to violent video games is a causal risk factor for aggressive behavior, cognition and affect and for reduced empathy[8] . 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 of publication bias in studies of aggression[9]. The laboratory finding that children copy a model’s acts toward a toy does not settle which of these readings is correct.

Sources20

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.Stajkovic, A. D., & Luthans, F. (1998). Self-efficacy and work-related performance: A meta-analysis. Psychological Bulletin, 124(2), 240–261. https://doi.org/10.1037/0033-2909.124.2.240
  4. 4.Sheeran, P., Maki, A., Montanaro, E., Avishai-Yitshak, A., Bryan, A., Klein, W. M., Miles, E., & Rothman, A. J. (2016). The impact of changing attitudes, norms, and self-efficacy on health-related intentions and behavior: A meta-analysis. Health Psychology, 35(11), 1178–1188. https://doi.org/10.1037/hea0000387
  5. 5.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
  6. 6.Sitzmann, T., & Yeo, G. (2013). A meta‐analytic investigation of the within‐person self‐efficacy domain: Is self‐efficacy a product of past performance or a driver of future performance?. Personnel Psychology, 66(3), 531–568. https://doi.org/10.1111/peps.12035
  7. 7.Bandura, A., & Locke, E. A. (2003). Negative self-efficacy and goal effects revisited. Journal of Applied Psychology, 88(1), 87–99. https://doi.org/10.1037/0021-9010.88.1.87
  8. 8.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
  9. 9.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
  10. 10.Bandura, A. (2001). Social cognitive theory of mass communication. Media Psychology, 3(3), 265–299. https://doi.org/10.1207/S1532785XMEP0303_03
  11. 11.Rotter, J. B. (1975). Some problems and misconceptions related to the construct of internal versus external control of reinforcement. Journal of Consulting and Clinical Psychology, 43(1), 56–67. https://doi.org/10.1037/h0076301
  12. 12.Miller, N. E., & Dollard, J. (1941). Social learning and imitation. Yale University Press.
  13. 13.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
  14. 14.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
  15. 15.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
  16. 16.Bandura, A. (1977). Social learning theory. Prentice-Hall.
  17. 17.Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191–215. https://doi.org/10.1037/0033-295X.84.2.191
  18. 18.Vancouver, J. B., Thompson, C. M., Tischner, E. C., & Putka, D. J. (2002). Two studies examining the negative effect of self-efficacy on performance. Journal of Applied Psychology, 87(3), 506–516. https://doi.org/10.1037/0021-9010.87.3.506
  19. 19.Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Prentice-Hall.
  20. 20.Bandura, A. (2001). Social cognitive theory: An agentic perspective. Annual Review of Psychology, 52(1), 1–26. https://doi.org/10.1146/annurev.psych.52.1.1

Further reading

Cite this entry

  • APA 7

    Psych Dictionary. (2026, October 19). Social learning theory. In Psych Dictionary. https://www.psychdictionary.com/concepts/social-learning-theory

  • MLA 9

    “Social Learning Theory.” Psych Dictionary, 19 Oct. 2026, www.psychdictionary.com/concepts/social-learning-theory.

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.