Albert Bandura

Introduction

Albert Bandura (4 December 1925 – 26 July 2021) was a Canadian-American psychologist who demonstrated that most human learning does not happen by trial and reward. People watch what other people do, remember it, and reproduce it — and they can acquire a complete novel behaviour without ever performing it, and therefore without ever being reinforced for it.

That claim was a direct challenge to the behaviourism dominant when Bandura began. B.F. Skinner's account required that a behaviour be emitted and then shaped by its consequences; Bandura's Bobo doll studies showed children learning an elaborate sequence of aggressive actions from a single viewing, with no reinforcement of any kind. Learning, he argued, and performance, are different things.

His second major contribution was the concept he introduced in 1977 to explain why knowledge doesn't convert into action: self-efficacy, a person's belief in their capacity to execute the behaviour required to produce a given outcome. It is the most frequently cited construct in the history of social psychology and it appears in nearly every domain where behaviour change is studied.

Bandura spent his entire fifty-seven-year academic career at Stanford University. In a 2002 survey published in the Review of General Psychology he was ranked the fourth most eminent psychologist of the twentieth century, behind Skinner, Piaget and Freud, and he was among the most cited psychologists alive in that period.

Early Life and Education

Bandura was born in Mundare, Alberta, a hamlet of roughly four hundred people on the Canadian prairies, the youngest child and only son of immigrant parents — his father from Kraków, his mother from Ukraine. Neither spoke English and neither had formal education. The town had one school with, he recalled, two teachers and a handful of students; the resources were so thin that students largely taught themselves from whatever was on hand.

The experience left a permanent mark, and it is the source of his later emphasis on self-regulation: he wrote that while the content of a course is perishable, the ability to direct your own learning is not, and that his own education had been largely self-directed out of necessity.

He enrolled at the University of British Columbia and, by his own account, chose psychology by accident — he needed a course that fitted his timetable, picked the introductory class from the catalogue, and found the subject. He finished the degree in three years, graduating in 1949 with the Bolocan Award in psychology.

For graduate work he went to the University of Iowa, which in the early 1950s was the centre of the American learning-theory establishment. He took his MA in 1951 and his PhD in 1952, under Arthur Benton, in a department shaped by Clark Hull and by Robert Sears's work on identification and the family origins of social behaviour. The intellectual inheritance was behaviourist; the research question he took from it was not. His dissertation was on the psychodynamics of a religious cult.

Academic and Professional Career

Bandura took a clinical internship at the Wichita Guidance Center, then joined the Stanford faculty in 1953, where he remained for the rest of his career. He was made a full professor in 1964 and in 1974 was given the endowed chair he held until his death: David Starr Jordan Professor of Social Science in Psychology.

His early work, much of it with Richard Walters, his first doctoral student, was on aggression in adolescents — the subject of Adolescent Aggression (1959) and later Aggression: A Social Learning Analysis (1973). The Bobo doll studies followed in 1961 and 1963, and the theoretical framework that would become social learning theory was set out in Social Learning and Personality Development (1963) and, definitively, in Social Learning Theory (1977).

The 1970s produced the turn to self-referent processes. Working with snake phobics, Bandura found that the mechanism of change in modelling-based treatment was not fear extinction but the patient's growing belief that they could do the thing they feared — and that this belief predicted behaviour change better than the patient's actual fear level. He renamed the framework social cognitive theory in Social Foundations of Thought and Action (1986), and consolidated the self-efficacy programme in Self-Efficacy: The Exercise of Control (1997).

He was elected president of the American Psychological Association in 1974 and a fellow of the American Academy of Arts and Sciences in 1980. His later honours include the APA Award for Outstanding Lifetime Contribution to Psychology (2004), the American Psychological Foundation Gold Medal (2006), the Grawemeyer Award for Psychology (2008), appointment as an Officer of the Order of Canada (2014), and — in 2016 — the National Medal of Science, presented by President Obama.

In the last two decades of his life Bandura worked on moral disengagement: the set of cognitive manoeuvres — euphemistic labelling, displacement of responsibility, dehumanisation, advantageous comparison — by which people do harm while continuing to regard themselves as decent. He applied it to terrorism, to corporate conduct, and, late in his life, to climate change and the tactics by which industries sustain environmentally destructive practice. He died at his home in Stanford on 26 July 2021, of congestive heart failure, at ninety-five.

Major Contributions to Psychology

Observational learning and the Bobo doll studies

Bandura's best-known experiment is also one of the most reproduced images in psychology. Between 1961 and 1963, in the studies conducted with Dorothea Ross and Sheila Ross, nursery-school children were exposed to an adult model who either attacked a five-foot inflatable Bobo doll — hitting it, sitting on it, striking it with a mallet, and using distinctive novel phrases — or played quietly with other toys. The children were then mildly frustrated and left alone with the doll.

The children who had seen the aggressive model reproduced its behaviour in detail, including the specific verbal and physical sequences they had never been reinforced for. Children who saw the model rewarded showed more imitation; children who saw the model punished showed less. But the learning had occurred in all three conditions — when the punished group was later offered an incentive, they performed the behaviour at the same rate as the others. Learning and performance had come apart, which is exactly the point.

Bandura specified four conditions that govern observational learning:

Attention — the observer must notice and accurately perceive the behaviour. Model characteristics (status, similarity, warmth) and observer characteristics matter here.
Retention — the behaviour must be encoded and remembered, usually as imagery and verbal codes.
Reproduction — the observer must be physically capable of translating the remembered code into action, which typically requires practice and corrective feedback.
Motivation — the observer must have reason to perform it, supplied by direct reinforcement, vicarious reinforcement (seeing the model rewarded), or self-reinforcement (judging one's own performance against internal standards).

The addition of vicarious and self-reinforcement is what distinguishes Bandura's account from Skinner's. A person does not have to perform a behaviour and be reinforced for it in order to learn it; watching someone else be reinforced is sufficient.

Reciprocal determinism

Bandura rejected both one-way models of causation: the behaviourist view that behaviour is produced by environment, and the trait view that it is produced by internal disposition. Reciprocal determinism (or triadic reciprocity) holds that three sets of factors continuously interact — personal factors (cognition, belief, affect, biology), behaviour, and the environment — each influencing and being influenced by the others. People both shape and are shaped by their circumstances, and the weight of each component varies by person and situation.

Self-efficacy

Self-efficacy is not self-esteem, and not a measure of actual competence. It is a domain-specific belief about capability: the judgement that one can organise and execute the course of action required to manage a prospective situation. Bandura's argument was that people routinely know what to do and fail to do it, and that the intervening variable is the belief that the doing will work.

He specified four sources:

Mastery experiences — actual performance accomplishments; the strongest source by a wide margin, and the reason genuine success matters more than encouragement
Vicarious experience — seeing similar others succeed, which is most effective when the observer doubts their own capability
Verbal persuasion — credible encouragement; the weakest source, and fragile, because it is undone by failure
Physiological and affective state — how arousal is interpreted; the same elevated heart rate read as anxiety lowers efficacy, read as readiness raises it

Self-efficacy has turned out to predict effort, persistence, resilience after failure, stress, and achievement across an unusually wide range of domains — education, health behaviour, sport, organisational performance, and clinical treatment. It is also frequently misapplied: Bandura insisted it must be measured at the level of the specific task, and general "self-efficacy" scales that ignore that requirement produce weak and uninterpretable results. See Self-Efficacy.

Self-regulation

Extending the same logic, Bandura described self-regulation as a cybernetic loop: self-observation, judgement against a standard, and self-reaction. People set goals, monitor their own performance, compare it to the standard, and respond to themselves — with satisfaction or dissatisfaction — in ways that then affect subsequent behaviour. Self-regulation fails, he noted, in two directions: through lax standards and through standards so rigid that ordinary shortfall produces demoralisation.

Collective efficacy and moral disengagement

In his later work Bandura extended efficacy beyond the individual. Collective efficacy is a group's shared belief in its capacity to act together, and it predicts neighbourhood cohesion, organisational performance and political mobilisation independently of the actual resources available.

Moral disengagement addresses the opposite case: how people who hold ordinary moral standards nonetheless do harm. Bandura identified the mechanisms by which self-sanctions are selectively switched off — moral justification, euphemistic labelling, advantageous comparison, displacement and diffusion of responsibility, distortion of consequences, dehumanisation, and attribution of blame. The construct has become central to research on aggression, bullying, corruption and organisational wrongdoing.

Try It Yourself

How do children learn by watching others? Step through a simulation inspired by Bandura's famous Bobo doll study:

Try the Bobo Doll Experiment →

Major Publications

Adolescent Aggression (1959, with Richard Walters) — the origins of the modelling programme
Social Learning through Imitation (1962) — observational learning as a general mechanism
Social Learning and Personality Development (1963, with Walters) — the first full statement of social learning theory
Principles of Behavior Modification (1969) — the clinical application
Psychological Modeling: Conflicting Theories (1971) — a comparison of modelling accounts
Aggression: A Social Learning Analysis (1973) — aggression as acquired rather than instinctual
Social Learning Theory (1977) — the mature statement of the theory
Social Foundations of Thought and Action: A Social Cognitive Theory (1986) — social cognitive theory and reciprocal determinism
Self-Efficacy: The Exercise of Control (1997) — the consolidated account
Moral Disengagement: How People Do Harm and Live with Themselves (2016) — his last major work

Impact and Legacy

Bandura's contribution is unusual in being both theoretically decisive and directly usable. He is the main figure in the shift from behaviourism to cognitive social learning, and the constructs he introduced are still the working vocabulary of applied psychology.

The applications are spread unusually widely. Education uses self-efficacy and observational learning as core mechanisms, and teacher efficacy is among the better predictors of student achievement. Clinical psychology uses modelling in the treatment of phobias, and modelling is a component of most CBT packages. Health psychology treats self-efficacy as a standard predictor in behaviour-change models. Media and communication took observational learning directly: Miguel Sabido built the "entertainment-education" method — telenovelas with deliberate prosocial models, designed for family planning and literacy — explicitly on Bandura's principles, and it has been used in dozens of countries. Bandura's own testimony on televised violence informed Federal Trade Commission deliberations on advertising standards.

Two of his constructs have since moved in directions he did not live to take them. Moral disengagement has been extended from the individual to the organisation. Recent work treats it as a property of a group's culture rather than only of a person's reasoning, and a 2024 study found that employees' perception of their organisation's moral disengagement predicted both unethical pro-organisational behaviour and employee silence — that is, that the mechanisms Bandura catalogued can be embedded in how an institution operates and not only in how an individual justifies himself. The framework has also been carried into a setting he never wrote about. Work on generative AI has begun to apply it to unethical AI use and to AI-enabled deception such as deepfakes, and above all to the case his scheme predicts best: responsibility displaced onto an automated system, and the resulting harm treated as less morally significant because a machine was the instrument. That it transfers so easily is a point in the theory's favour — it was always an account of how self-sanctions are managed, not of any particular kind of act.

Self-efficacy has reached its fiftieth year with the four sources intact and under renewed argument. Interventions continue to be built explicitly on them: a game-based programme for trainee teachers, for example, gives players classroom decisions with visible consequences (mastery), feedback (persuasion) and modelled alternatives (vicarious experience), and has raised teaching self-efficacy among trainees in the United Kingdom and Australia. But the list of sources is now contested. James Maddux has proposed a fifth — imagined experiences, in which a person visualises themselves succeeding — which is the mechanism that guided imagery in sport psychology has long relied on. The cross-cultural evidence raises the larger question. Self-efficacy beliefs tend to be lower in non-Western samples, and there are indications that for some groups collective efficacy, the belief of the group rather than of the individual, predicts better than self-efficacy does. Bandura introduced collective efficacy and applied it to neighbourhoods and movements, but never developed it to anything like the depth he gave the individual construct; it is now the most unfinished part of the framework, and the one most likely to matter in the places where his individualistic framing has been criticised.

The critiques are principally about the Bobo doll studies and about the boundaries of self-efficacy. On the experiments: the Bobo doll is designed to be struck, which weakens the claim that the children learned aggression rather than a game; the studies have been criticised for demand characteristics and for measuring behaviour in an artificial setting; and the extrapolation from a laboratory doll to real-world violence has never been straightforward. The media-violence debate that Bandura's work opened is still unresolved, though meta-analytic evidence supports a small-to-moderate effect.

On self-efficacy: the construct is sometimes treated as a general personality trait — "high self-efficacy" as a characteristic of a person — which is precisely the reading Bandura rejected. Measured at the level of specific tasks, it is one of the most robust predictors in applied psychology; measured globally, it becomes a loose synonym for confidence and loses most of its predictive power.

What is not in dispute is the central insight. Behaviourism needed the organism to be reinforced in order to learn; Bandura showed that it only needs to watch. That single finding, and the self-efficacy construct he added to explain why observed behaviour sometimes fails to transfer, account for most of what he is cited for.

Associated Figures

B.F. Skinner — the behaviourist position Bandura's work displaced; the target of the argument that learning requires reinforcement
Richard Walters — his first doctoral student and co-author on the early aggression work
Robert Sears — the Iowa influence behind his interest in identification and family antecedents of social behaviour
Neal Miller and John Dollard — whose Social Learning and Imitation supplied the vocabulary he then transformed
Miguel Sabido — the television producer who turned observational learning into the entertainment-education method

Learn More

References

Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191-215.
Bandura, A. (1977). Social Learning Theory. Prentice Hall.