Social learning theory explains behavior as learned from observing others, while recognizing that thinking processes affect whether observed actions are copied, maintained, or rejected.

Core ideas of social learning theory

Social learning theory was developed by Albert Bandura. It proposes that people learn behavior by observing other people, known as models, and then using that information later. Learning usually happens in a social context, so the behavior of parents, peers, teachers, and media figures can all be influential. Bandura argued that observation alone can produce learning, even if the learner does not copy the behavior immediately.

This means the theory is not limited to visible behavior alone. It also says that mental processing matters. A person notices a behavior, remembers it, judges whether copying it is worthwhile, and only then may perform it. A key idea is the difference between learning and performance: someone may learn a behavior through observation but not show it until the right situation occurs.

Learning from models

In social learning theory, people learn by watching models. In everyday life, models may include:

• Live models, such as parents, teachers, siblings, or peers

• Media models, such as actors, influencers, or game characters

• Symbolic models, including people described in stories or instructions

People are more likely to copy models they see as similar, high-status, attractive, or rewarded. These factors help explain why not all observed behavior is imitated equally.

Imitation

Imitation is a basic process in social learning theory.

Imitation can be direct, where the observed action is copied closely, or more general, where the learner copies the style or goal of the behavior. Social learning theory suggests that children are especially likely to imitate behavior that appears effective or socially approved.

Identification

Identification goes beyond simply copying a single act.

Identification is often stronger when the model is the same sex, a similar age, admired, or has higher status. This helps explain why children may copy not just what a model does once, but broader patterns of conduct over time.

Vicarious reinforcement

Vicarious reinforcement explains why observation can shape behavior even without direct reward.

If a model is praised, admired, or gains some advantage, the observer may expect similar outcomes from copying them. The opposite can also happen: seeing a model criticized or punished can reduce the likelihood of imitation. However, the observer may still have learned the behavior even if they do not perform it.

Mediational processes

Mediational processes are the mental steps between observing behavior and copying it.

Bandura emphasized four important mediational processes:

• Attention: the observer must notice the model and the behavior.

• Retention: the observer must remember what they saw.

• Motor reproduction: the observer must have the physical or intellectual ability to carry out the behavior.

• Motivation: the observer must want to perform the behavior, often because they expect reward or approval.

These processes show that observation does not lead automatically to imitation. A child may watch aggressive behavior, remember it clearly, and still choose not to copy it unless there is motivation to do so.

Bandura’s research

The Bobo doll studies

Bandura’s best-known research investigated whether children would imitate aggressive behavior shown by an adult model. In the classic Bobo doll study, children observed an adult behaving aggressively toward an inflatable doll, for example by hitting it, kicking it, and using aggressive phrases.

A labeled diagram of a Bobo doll’s construction, illustrating the weighted base that makes it self-righting. This helps clarify why the doll was suitable for repeatedly observing and scoring children’s imitative aggressive actions in the lab. Source

The children were then given the opportunity to play in a room containing the Bobo doll and other toys.

Children who had observed the aggressive model showed significantly more aggressive behavior than those who had observed a non-aggressive model or no model. They often copied specific actions and phrases used by the adult. This strongly supported the idea of imitation through observation.

Bandura also found evidence of identification. Boys were generally more physically aggressive than girls, and children were more likely to imitate a same-sex model, especially boys observing a male model. This suggested that similarity between observer and model affects learning.

Later studies by Bandura examined vicarious reinforcement more directly. Children saw a model behave aggressively and then either be rewarded, punished, or experience no consequence. Children were least likely to imitate the model when the model was punished, and more likely to imitate when the model was rewarded or not punished. This supported the claim that observed consequences influence performance.

An important finding came when all children were later offered an incentive to reproduce what they had seen. Children from all conditions were able to show the aggressive acts, suggesting they had learned the behavior even when they had not previously performed it. This is strong evidence for mediational processes and for the distinction between learning and performance.

Bandura’s studies are important because they show that behavior can be acquired through observation rather than direct experience. They also show that imitation is more likely when observers identify with the model and when the model appears rewarded. At the same time, the findings support the role of cognitive mediational processes, because children did not automatically copy everything they had seen.