Bandura vs Skinner: Why Observation Beats Reinforcement

Bandura's Social Learning Theory (Bandura, 1977) explains how people learn from others. We learn by watching, copying, and acting like the people around us. Older behaviourist theories focused only on rewards and punishments, but Bandura suggested a different idea.

This connects to the wider context of fundamental theories of learning in modern classroom practice.

‍

‍

He argued that we learn by seeing and copying others in our social world. This matters for teachers because it explains why children copy both good and bad behaviours. They learn from friends, teachers, and media figures.

The theory relies on shared influence: a person's behaviour, environment and thoughts shape each other. Learning is not something that just happens to us (Bandura, 1986); it is an active relationship with the social world. Teachers can use social learning theory to plan clearer routines, model the behaviours they want to see, and shape the classroom environment on purpose.

Social Learning Theory Definition

Social learning theory was formalised by Albert Bandura in his 1977 book, after earlier experiments in the 1960s. It says that learners pick up new behaviours by watching models, noticing what happens next, and copying what they have seen.

The theory builds on behaviourism, but adds a cognitive layer. Learners do not just respond to reinforcement. They also evaluate, imitate, and self-regulate. In UK classrooms, this means learning is often stronger when teacher modelling, peer examples, and guided practice make expert thinking visible.

Bandura found four key steps in this learning process. These steps control how we learn by watching and copying others:

‍

1. Attention

For learning to occur through observation, a learner must first pay attention to the model's behaviour. In classroom settings, this means teachers must capture learners' interest and make their demonstrations clear and compelling. A teacher explaining a maths concept whilst scribbling notes distractedly will not be an effective model. However, if that same teacher uses think-aloud strategies, makes eye contact, and emphasises key points, learners are more likely to attend to and encode the information.

‍

2. Retention

Watching a behaviour only helps if the learner can remember it. To keep it, the brain stores the action in memory, using mental pictures and words as Bandura described.

In class, teachers can support this by saying their thoughts aloud while they work. They can also repeat key steps during a model, then ask learners to practise or sum up what they saw.

‍

3. Reproduction

The learner must be able to reproduce the behaviour they have watched. This is where practise becomes essential.

A learner may listen to and remember a teacher's explanation of a writing technique. However, reproduction means actually trying to write using that technique.

This often involves trial and error. Learners may make mistakes at first as they refine their imitation of the model, so teachers should provide guided practice with scaffolding and feedback.

‍

4. Motivation

Finally, learners need a reason to copy the behaviour they see. Bandura grouped motivation around three reinforcement routes, rather than finding them all from scratch. Skinner (1953) had already made direct reinforcement central to behaviourism. This means receiving a consequence for your own action.

Vicarious reinforcement means watching what happens to someone else. Self-reinforcement means the private satisfaction of meeting a standard. In school, learners may copy praised peers. They also copy actions that bring competence, belonging or personal achievement.

‍

Reciprocal Determinism in Classrooms

Reciprocal determinism in classrooms describes the ongoing interaction between personal factors, behaviour and environment in shaping learning. He later described this as triadic reciprocal causation within Social Cognitive Theory (1986). In simple terms, what a learner thinks and feels affects how they act, their actions change the classroom around them, and that classroom then influences what they think and do next. For teachers, this matters because behaviour is rarely caused by one factor alone.

A child who avoids answering questions, for example, may have low self-efficacy, show quiet or withdrawn behaviour, and sit in a classroom culture where only the quickest learners speak. Each part reinforces the others. If a teacher changes the environment by using think time, predictable turn-taking, and warm feedback, the learner is more likely to attempt an answer. That small behavioural success can then improve confidence, which makes future participation more likely.

This idea gives teachers practical ways to intervene. One useful strategy is to model mistakes calmly and correct them out loud, showing learners that error is a normal part of learning rather than a threat. Another is to use carefully chosen peer models, pairing learners with classmates who demonstrate clear routines, thoughtful discussion, or perseverance. A third approach is to adjust the environment through seating, group norms, and visual supports, so that productive behaviour is easier to repeat.

Reciprocal determinism also gives leaders a way to examine staff culture. A staffroom can get trapped in the same loop as a classroom: low trust shapes defensive behaviour, defensive behaviour makes meetings feel unsafe, and the environment confirms the belief that change is pointless. Headteachers can interrupt the cycle by modelling calm problem-solving, making norms explicit, and designing meetings where staff see credible peers test new routines. This is one reason Bandura's work remains useful in education: it treats behaviour as responsive to beliefs, actions and setting, not as fixed.

‍

Self-Efficacy and Learner Outcomes

Self-efficacy means learners' beliefs about their own competence. These beliefs shape effort, persistence and attainment. In school, they often affect whether learners try a challenge, keep going when work gets hard, and recover after mistakes.

Bandura linked self-efficacy to human agency in 'Self-efficacy mechanism in human agency' (1982), and later described intentionality, forethought, self-reactiveness and self-reflectiveness as core properties of human agency in his agentic account of social cognitive theory (Bandura, 2001/2006). This also links social learning theory to Zimmerman's work on self-regulated learning, where learners plan, monitor and evaluate their own strategy use (Zimmerman, 2000). For teachers, attainment is not only about what learners know. It is also about what they believe they can do with that knowledge.

One of the most effective ways to build self-efficacy is through mastery experiences, which Bandura described as the strongest source of confidence. In practice, this means structuring tasks so that learners experience genuine success rather than empty praise. A teacher introducing fractions, for example, may begin with concrete apparatus, move to guided examples, and then set one carefully chosen independent question that most learners can complete correctly. When success is visible and linked to effort, learners begin to see improvement as something they can control.

Self-efficacy also grows through vicarious experience and verbal persuasion. Learners are more likely to believe they can succeed when they watch a similar peer model the process, hear a teacher think aloud, or receive feedback that points to a useful strategy.

‍

◆ Structural Learning

Bandura's Social Learning Theory Study Notes

Study notesOne-page revision sheet

Download a one-page study note for Bandura's Social Learning Theory, with the key ideas, limitations and classroom links in one place.

Send me the study notes →

Something went wrong - please try again.

Ready. Download the study notes now.

‍

In a writing lesson, a teacher might share an anonymous example from a learner. The teacher can show why it works, then ask the class to copy one feature in their own paragraph. Feedback such as, "Your opening improved because you used precise vocabulary and revised it twice," is stronger than general praise because it links progress to action.

Classroom climate matters as well, because anxiety can quickly weaken a learner's sense of competence. Low-stakes quizzes, rehearsal time before public answers, and clear routines can reduce threat and help learners approach tasks with greater confidence. Brief reflection activities, such as asking learners what helped them improve today, can reinforce the link between strategy, effort, and success. Over time, these small decisions help create learners who are more willing to participate, more resilient in the face of difficulty, and better prepared to learn independently.

‍

The Bobo Doll Experiment: The Landmark Study

The Bobo doll experiment is Bandura's famous study. It shows that children copy violent acts they see adults perform. In Bandura, Ross and Ross's 1961 study, nursery children aged about three to five observed a live adult model behaving aggressively towards an inflatable Bobo doll; a later 1963 study tested filmed and cartoon models.

The adult hit, kicked, and shouted at the doll. Later, the children played with the same doll. Children who saw the violent adult were much more likely to copy the violence, often using the same actions and words.

Crucially, children who had not observed the aggressive model showed almost no aggressive behaviour towards the doll. This gave important evidence that children do not only learn through direct reinforcement (reward or punishment). They also learn through observational learning, by watching and imitating what others do.

The study raised serious questions about media violence and its impact on children's behaviour. This concern still matters in today's digital age, where children see large amounts of media content.

The Bobo doll experiment showed something new. It proved that people can learn by watching, even without a direct reward. No one rewarded the children for copying the aggressive acts, yet they still copied them.

This challenged the main behavioural theories of that time. It showed that learning is a mental process involving focus, memory, and the choice to copy actions. Learning is not just a simple reaction to a trigger.

‍

Social Learning Theory vs. Behaviourism: Key Differences

Social Learning Theory and behaviourism explain learning in different ways. They disagree about thinking, observation and rewards. Use it as a starting point for professional discussion: identify the learner's current need, record evidence from more than one lesson, and agree the next classroom adjustment with the SENCO or family.

Behaviourism, associated with Watson's 1913 manifesto and later Skinnerian operant behaviourism, prioritised observable behaviour and environmental contingencies; Skinner's operant conditioning specifically emphasised reinforcement and punishment. By contrast, social learning theory includes both the environment and personal thoughts.

Behaviourism treats the mind as a "black box": a trigger goes in, an action comes out, and internal mental processes get little attention.

Social learning theory opens that black box. It shows that what happens in the learner's mind matters a great deal. A child may be punished for a behaviour, so a behaviourist would expect that behaviour to decrease. But if the child sees a peer or role model rewarded for the same behaviour, social learning theory predicts the child may still copy it.

This point matters for classroom management and discipline. Punishment alone is not enough if learners can see peers modelling the behaviour.

‍

Social Learning Theory in the Classroom: 8 Practical Strategies

Teachers can use Social Learning Theory in the classroom. This means using teaching methods based on modelling, watching, and working with peers. These methods help to shape learner learning. The following strategies turn Bandura's theory into real teaching actions:

‍

1. Think-Aloud Modelling

One of the most direct applications of social learning theory is thinking aloud while demonstrating a skill. When you solve a problem, write an essay, or approach a complex task, narrate your internal thought process so learners can observe how an expert thinks.

For example, a teacher reading a complex text may say: "I notice this paragraph is dense, so I'm going to slow down and re-read it. I see the word 'paradox' which I may need to check. Let me read the context to infer the meaning first before looking it up." By making thinking visible, you create a model that learners can observe, retain, and reproduce in their own learning.

‍

2. Deliberate Peer Demonstration and Collaborative Learning

Peer models often have more impact than teacher models. This is especially true for learners who do not relate easily to adults. For this reason, you should pair or group learners with care and let skilled learners model good behaviours for their peers.

A learner may show the class a strong way to solve a problem. Other learners will watch and may want to copy this method, especially if the peer receives praise for their work. Structured collaborative learning is a practical way to use this without leaving group work to chance.

‍

3. Video-Based Learning and Expert Modelling

Videos of experts or exemplars showing skills can give learners useful models to watch again. For example, a short video might show a scientist carrying out an experiment, an author explaining their writing process, or a professional practitioner solving a real-world problem. This gives learners a vicarious learning experience.

Teachers should pause, rewind and discuss the model's decisions. This helps learners notice the thinking, not just the surface behaviour. Avoid using mirror neurons as simple biological proof that video makes learners imitate automatically. Heyes (2010) and Hickok (2014) show that imitation has contested neural explanations, so classroom modelling still needs attention checks, explanation and guided practice.

‍

4. Role Modelling Positive Behaviours and Attitudes

Teachers are the most obvious role models in class. Learners watch how you teach, but they also watch your behaviour, grit, and attitude to learning. You can model curiosity by saying, 'Let us find out together'.

You can show grit by trying a new way to solve a hard problem. When you show respect, learners take on these values. They may also copy impatience or worry, so you need to choose the attitude you model.

‍

5. Explicit Success Criteria and Exemplar Analysis

Showing learners exemplar work is a form of modelling. This might be a high-quality learner essay, a well-executed piece of art, or a solved maths problem with clear working. When teachers pair exemplars with clear success criteria, they become strong guides for learner reproduction.

Learners study the product, analyse how it meets the criteria, and then try to create similar work. Using varied exemplars shows several ways to complete the same task. This helps learners see that there are many valid ways to succeed, which can reduce anxiety about one "correct" approach.

‍

6. Immediate and Vicarious Reinforcement

Use praise and recognition carefully. Publicly praising one learner so others copy them can create vicarious reinforcement, but it can also produce performative compliance: learners act well only when the teacher is watching or when status is available. Kohn (1993) and Ryan and Deci (2020) both warn that external approval can weaken autonomy when it replaces purpose, competence and belonging. Name the strategy, effort or kindness you want copied, then give learners private chances to practise it without turning the class into an audience.

‍

7. Guided Peer Observation and Reflection

Create structured opportunities for learners to observe each other. During peer assessment activities or gallery walks, learners observe peers' work and approaches. Pair this observation with reflection prompts: "What strategies did your peer use that you may try?" or "Whose approach did you find most helpful, and why?" This formalises the observational learning process and helps learners consciously transfer peer models into their own practise. Peer observation also raises awareness of diverse problem-solving approaches, broadening the repertoire of strategies learners can draw upon.

‍

8. Creating a Culture of Modelling and Observation

Set classroom norms that make watching and learning from each other feel valued. Use phrases like "We learn from each other" or "Let's see how another classmate approached this." Celebrate a range of models, not only the quickest or highest-achieving learners. Include learners who show different strengths, perseverance, creativity, or kindness.

When learners see that learning is social and based on observation, they pay closer attention to peer models. They also become more willing to be models themselves.

‍

Common Misconceptions About Social Learning Theory

The main misconception is that social learning is just copying or reward. Bandura's theory is wider: learners notice models, remember what happened, judge whether it feels worth copying and then test the behaviour in practice. Teachers need to shape the model, the context and the reflection.

‍

1. The Direct-Reinforcement Trap

Teachers often assume learning needs a reward or punishment to stick. Bandura's Bobo doll work showed that learners imitate behaviour after only watching a model, with no reward needed (Bandura, Ross and Ross, 1961). This is called vicarious reinforcement: the learner sees what happens to the model and adjusts behaviour accordingly. In a classroom, this means modelling effort and persistence in front of the class is itself a teaching move, not just a personality trait.

‍

2. Treating the Mind as a Closed Box

It is tempting to manage behaviour by tweaking incentives and consequences alone. Bandura (1986) argued that internal cognition, particularly self-efficacy, drives whether a learner attempts a task at all. A learner who believes they are "no good at maths" will not engage, no matter how rich the reward system. Effective social learning practice therefore addresses self-belief, not just observable behaviour.

‍

3. The Teacher as Consequence Dispenser

Reward charts and sanctions matter, but they are not the main lever. Social learning theory casts every adult in the room as a model whose tone, persistence and problem-solving steps get copied. Peer modelling matters too: a struggling reader watching a slightly stronger peer think aloud often gains more than a teacher's direct instruction (Bandura, 1977). The implication is that what teachers and peers do in front of learners carries more weight than what they say.

‍

Criticisms and Limitations of Social Learning Theory

Social learning theory is useful, but it can overstate what learners gain from observation alone. Attention, prior knowledge, culture, motivation and explicit teaching still matter. A model may be visible and still fail if learners cannot interpret the steps or practise them safely.

‍

Individual Differences in Observation

Social learning theory can assume that attention, social status and praise work in similar ways for all learners. That assumption is risky.

An autistic learner may avoid eye contact, miss hidden social cues, or feel pressure to mask distress when public praise sets the model for everyone else. Milton's double empathy problem (2012) reminds teachers that social misunderstanding can run both ways: the classroom may fail to read the learner as much as the learner fails to read the model.

ADHD, anxiety, language needs and prior knowledge can also make learning by watching harder. For this reason, the model needs to be explicit, paced and supported.

‍

The Role of Scaffolding and Explicit Instruction

Social learning theory stresses learning by watching and copying others. Cognitive psychology adds an important caution. Sweller's (1988) cognitive load theory says teacher modelling works best as a worked example.

A worked example reduces needless searching, shows each step, and protects working memory. Kirschner et al. (2006) make a similar case against minimal guidance for novices. For new or hard tasks, learners usually need clear explanation, guided practice and fading support before observation becomes independent performance.

‍

Cultural Context and Modelling

Experts built this theory mostly in Western cultures that focus on the self. For that reason, it may not explain how modelling works in every culture.

In some groups, learning by watching and copying is vital. In others, clear spoken teaching matters more.

The kinds of models that inspire learners can also change. A strong role model in one culture may not work in another.

‍

Media and Digital Modelling

Bandura created his theory before algorithmic feeds and AI companions. This does not mean the theory fails online. The limit is that teachers can no longer assume which model has social currency, or status with learners.

Learners may copy creators, streamers or chatbot personas before they copy adults in school. For this reason, digital modelling needs clear discussion, source checking and comparison with trusted classroom models.

‍

Social Learning in the Age of AI

In the age of AI, social learning includes digital role models as well as teachers and classmates. Learners do not just copy people in the room. They also watch TikTok creators, YouTube explainers, gamers, and chatbot interfaces. Then they imitate the language, confidence, and shortcuts they see.

Ofcom's 2025 children's media report shows that algorithmic feeds now compete for the attention and motivation stages of Bandura's model. Parasocial familiarity can also make a weak model feel more trustworthy than the teacher.

This matters because Generative AI does more than supply answers. It offers virtual modelling: a visible example of how to explain, summarise, argue and solve problems. Learners may also form parasocial relationships with influencers or AI personas that feel familiar and trustworthy, even when the content is weak or biased. UNESCO argues that schools need explicit teaching about bias, privacy and human oversight when using Generative AI (Miao and Holmes, 2023).

A practical response is to model critical digital literacy in public. A Year 8 history teacher may project an AI-written explanation of the causes of the English Civil War and say, "Watch how I test this. I’m checking which claims need evidence, which words sound certain without proof, and whose perspective is missing."

Learners then annotate the text as accurate, doubtful or unsupported, and redraft it using the textbook and one trusted source. That is cognitive modelling in action: the teacher is not only correcting errors but showing the thinking process learners can copy.

For classroom practice, the key shift is simple. Treat AI outputs and algorithmic feeds as models to examine, not as authorities to accept. The UKCIS Education for a Connected World framework already asks schools to teach how online content shapes behaviour and judgement (UKCIS, 2020).

The Department for Education guidance on generative AI in education makes the same point for safe and effective AI use in education. In Bandura’s terms, teachers still shape attention, retention, and imitation. But some of the competition for these processes is now non-human.

‍

‍

‍

‍

‍

◆ Structural Learning

Social Learning Theory: Bandura's Modelling for Classrooms

Classroom-readyWhat the theory means in practice

Social Learning Theory in practice — a classroom-ready briefing you can use this week.

Send me the slides →

Something went wrong — please try again.

✓ On its way. Download the slides now.

‍

‍

Frequently Asked Questions About Social Learning Theory

‍

Simple Definition

Social learning theory explains how people learn by watching and copying others. Learning does not only come from direct rewards or punishments. It also comes from observing models, such as teachers, friends, or family, and then copying their behaviour, skills, and attitudes.

Bandura said this type of learning has four steps. First, learners pay attention to the model, and then they remember what they saw. Next, they need to be able to copy the action, and finally, they need a good reason to do it.

‍

Four Stages

There are four core processes in social learning. Attention means noticing the model's behaviour, while retention means remembering what you saw. Reproduction means practising and doing the behaviour you observed.

The fourth process is motivation. This means having a reason to copy the action. For example, you may see the model get a reward, or you may feel personal pride.

‍

Bobo Doll Experiment

The Bobo doll experiment (1961) was a famous study by Bandura, Ross, and Ross. In the study, children watched a video of an adult model behaving aggressively towards an inflatable doll. Later, when given the same doll, children who had seen the aggression copied it. Children who had not seen the model showed no aggressive behaviour.

This showed that children can learn behaviour through observation, even without direct reinforcement.

‍

Classroom Use

Teachers use social learning theory in many ways. They model good skills and behaviours, such as using think-aloud strategies and showing problem-solving. They also set up peer learning through group work and peer tutoring.

Teachers use videos and study good examples with learners. They reward good behaviour on purpose, which motivates other learners to copy it. The theory supports many teaching methods, including worked examples, expert modelling, and guided teaching.

‍

Difference from Behaviourism

Behaviourism says the environment shapes all actions. Skinner and Watson focused on external rewards and punishments. Bandura's social learning theory is different, although it agrees that the environment is important.

His theory also includes thinking factors in the learner's mind. These include attention, memory, beliefs, and expectations. Bandura said people can learn just by watching, without a direct reward. Learners actively think about what they see, rather than just blindly reacting.

‍

Main Criticisms

There are four main criticisms of the theory. First, it does not fully explain how individual differences affect learning. Second, it relies too much on observation alone, even though learners often need direct teaching and support.

Third, researchers created the theory in Western cultures. This means it may not explain how role models work in other cultures. Fourth, it was made before the digital age, so it does not cover learning from hidden online users or fake news.

‍

Aggression and Observed Behaviour

Social learning theory explains aggression as learned behaviour. Children can acquire it through observation and imitation. The Bobo doll experiment showed this: children who observed an aggressive model behaved more aggressively towards a doll.

The theory suggests that aggression is reinforced when the model receives reward, admiration, or status for aggressive behaviour. It also suggests that exposure to aggressive models in media or peer groups makes children more likely to imitate that aggression. This is why research on violent video games and media violence is often framed within a social learning perspective.

​

​

‍

‍

​

​

​

​

References

• Bandura, A. (1977). Social Learning Theory. Englewood Cliffs, NJ: Prentice Hall.
• Bandura, A. (1986). Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, NJ: Prentice Hall.
• Bandura, A., Ross, D., & Ross, S. A. (1961). Transmission of aggression through imitation of aggressive models. Journal of Abnormal and Social Psychology, 63(3), 575-582.
• Geddes, H. (2006). Attachment in the Classroom: The Links Between Children's Early Experience, Emotional Well-Being and Performance in School. London: Worth Publishing.
• Education Endowment Foundation. (2018). Collaborative Learning: Guidance Report. Retrieved from https://educationendowmentfoundation.org.uk

‍