Social Cognitive
Theory
We watch others. Then we decide what to try.
Explore how models, capability beliefs and the environment shape learning and action. Connect Bandura’s theory with classrooms, media demonstrations and everyday decisions.
“I saw how it works.”
“I think I can do it.”
“I have a chance to practise.”
Learning, perceived capability and opportunity are connected. They are not the same achievement.
In this chapter · Open the contents
Explain how a demonstration becomes a possible action.
- Trace attention, retention, production and motivation in a media-learning example.
- Distinguish self-efficacy, outcome expectations, intention and performance.
- Explain reciprocal influence, self-regulation and the social support for agency.
- Design a learning study that measures more than viewing or confidence.
Watching can open a route to action.
A student watches a journalist trace a quotation to its original source. The journalist explains each decision, spots a misleading edit and corrects an initial mistake. Later, the student tries the method on a different claim. What connects the demonstration with the student’s action?
Social Cognitive Theory (SCT) explains human functioning through the interplay of personal factors, behaviour and environment. Albert Bandura’s account examines how people learn through their own actions and through models, while also planning, evaluating and regulating what they do. Media supply a large symbolic environment in which such models can be encountered.
Think of watching a musician demonstrate a phrase. Seeing and hearing it can help you learn a pattern, but performance also depends on remembering it, developing the skill, having an instrument and deciding to practise. A model offers information and possibilities; it does not press a button that guarantees the same response in every observer.
The theory grew from Bandura’s social-learning research and developed into a broader account of cognition and agency, consolidated in Social Foundations of Thought and Action. Its mass-communication application examines how symbolic models can inform, enable and motivate action across dispersed audiences. Bandura, 1986; Bandura, 2001b.
Pause and think: Must someone receive a reward to learn from a model?
No. Observational learning can occur without the observer receiving a direct reward. Consequences can influence attention, expectations and the decision to perform what was learned. Learning and its visible expression must be investigated separately.
People act within conditions they can also change.
Bandura calls the relationship among personal factors, behaviour and environment triadic reciprocal causation, also widely discussed as reciprocal determinism. Each can influence the others. The influences need not have equal strength or operate at the same moment. “Determinism” here does not mean that society mechanically fixes every action. Bandura, 1986; Bandura, 2001a.
Personal factors
Beliefs, knowledge, skills, feelings and biological characteristics.
Behaviour
What the person does, practises, avoids or changes.
Environment
Tools, opportunities, people, rules and material conditions.
Follow the relationships through one example.
In an invented classroom case, Riya doubts her ability to verify a quotation. She avoids checking one during an assignment. A tutor then models the process and provides an accessible source document. Riya completes a small check successfully, asks for a harder example and helps establish a peer-checking group.
Her belief can affect her effort. Practice and feedback can reshape the belief. The tutor and source access affect what she can attempt. Her own actions also change the learning environment. This is the logic of reciprocal influence; a real study would need evidence about the particular links.
Try it: A learner stops watching because the captions are missing. Which component matters?
The immediate obstacle concerns the environment and access. It may then affect behaviour, such as stopping or seeking another source, and personal responses, such as frustration. Explaining everything as low motivation would miss the observed barrier.
Four processes connect a model with a possible performance.
Observational learning involves more than noticing that an action occurred. Four familiar process groups help identify what an observer needs to attend to, remember, produce and find worth doing. They guide investigation, rather than form a rigid staircase that operates identically for everyone. Bandura, 1986.
Which relevant actions and decisions does the observer notice? Clarity, prior knowledge, interest and the model’s perceived relevance can matter.
How is what was observed represented and remembered? Verbal summaries, images and rehearsal can support later use.
Can the observer translate that representation into action? Practice, component skills and informative feedback help reveal and correct gaps.
What makes the person choose to perform? Expected consequences, valued goals and personal standards can affect the decision.
Use the verification example.
- Attention: the lecturer highlights the step where the quotation is checked against the original document, rather than merely displaying a finished answer.
- Retention: the learner writes a short sequence: locate the original, read the surrounding passage, compare the wording.
- Production: the learner applies that sequence to a fresh quotation and receives feedback on a missed qualification.
- Motivation: the learner values accuracy and expects the check to improve the assignment, making the effort worth attempting.
These are proposed teaching applications. A learner might attend carefully but forget a step, remember the procedure but lack practice, or possess the skill but see no reason to use it. Do not infer which process failed solely from an unfinished task.
Models can demonstrate thinking.
A model need not display only a physical movement. A teacher can explain why a source is doubtful, show a decision rule or correct an error aloud. Learners can then use the principle in a new setting. Copying the model’s exact words is different from learning the strategy.
Pause and think: Is an admired presenter always an effective learning model?
No. Admiration may attract attention without making the relevant process visible or understandable. Inspect what the learner notices, can explain, remembers and can perform. Attractiveness or popularity is not a substitute for those measures.
“Can I do this?” differs from “What will happen?”
Self-efficacy concerns a person’s belief in their capability to carry out a specified course of action. It can affect whether they attempt a task, how they approach difficulty and whether they persist. It is more precise than a general instruction to “be confident.” Bandura, 1977.
| Construct | Illustrative statement | What it concerns |
|---|---|---|
| Self-efficacy | “I can trace this quotation to its original source.” | Perceived capability for the task. |
| Outcome expectation | “Checking the original will help me avoid a misleading quotation.” | The consequences expected from the action. |
| Intention | “I plan to check the original before submitting.” | A planned action. |
| Demonstrated skill | The learner locates and compares the source correctly. | Observed performance, assessed by stated criteria. |
A student may believe verification produces better work but doubt their own ability to do it. Another may feel capable while expecting that a rushed assessment will reward speed instead. Both differ from a student who feels capable and sees a worthwhile outcome.
Where can efficacy beliefs come from?
Interpreting one’s own successes and difficulties. Completing progressively demanding checks can supply evidence of capability.
Learning from others’ performances. Watching a comparable learner work through a difficulty can make a route to success visible.
Credible, specific feedback about capability. Encouragement is more useful when connected to achievable action and support.
How people interpret tension, fatigue or emotional arousal. Feeling tense need not mean being incapable.
Efficacy beliefs can be well calibrated, too low or too high. A high rating does not establish expertise. Pair the rating with a task when the educational aim is demonstrated understanding. Self-efficacy is also distinct from self-esteem: confidence about a particular task and a person’s overall sense of worth are different matters.
Measure the capability you mean.
Ask about a defined action under specified demands. “How certain are you that you can find the original report when the post provides no link?” is more informative for this lesson than “Are you a confident person?” A measure should cover the relevant domain and levels of difficulty. Bandura, 2006.
Try it: “I will verify every claim” is a self-efficacy item. Correct?
It expresses an intention. A capability item asks what the person believes they can do. Neither should be used as a direct substitute for observing whether they actually perform a verification task well.
Consequences can change what people show.
Vicarious consequences are consequences observed happening to someone else. A model’s success, reward, criticism or punishment may inform the observer’s expectations. Direct consequences follow the observer’s own action; self-produced consequences include satisfaction or self-evaluation against personal standards.
In an invented newsroom exercise, a student sees a peer praised for correcting an inaccurate caption. The observation may make correction seem worthwhile. It need not guarantee action: the student must still notice the procedure, retain it, have the skill and opportunity, and value the outcome.
What did the Bobo-doll studies contribute?
| Study | Research focus | Lesson to retain |
|---|---|---|
| Bandura, Ross, & Ross · 1961 | Children encountered different adult modelling conditions; subsequent behaviour was observed. | The study supplied evidence of imitation following exposure to aggressive models. It did not establish a lifelong effect of all media violence. Read the study. |
| Bandura · 1965 | Children observed a filmed model whose behaviour was rewarded, punished or met with no stated consequence; later tests included incentives to reproduce the actions. | Observed consequences affected displayed imitation. The incentive test helped distinguish acquisition from performance. Read the study. |
The important distinction is that a person can learn a response without immediately displaying it. In the 1965 study, later incentives revealed learning that was not fully visible in the initial performance conditions. “Did not perform” therefore cannot automatically be translated as “did not learn.” Bandura, 1965.
Modelling can affect more than acquisition.
A model can help someone learn a new strategy, encourage an already learned action or inhibit it by showing adverse consequences. It can also change emotional reactions and the goals an observer considers. Specify which proposed effect you are studying instead of calling every change “imitation.”
People plan, monitor, judge and act together.
SCT treats people as agents who can contribute to their circumstances, while operating within social structures. Bandura describes agency through intentionality, forethought, self-reactiveness and self-reflectiveness: forming plans, anticipating possibilities, regulating action and examining one’s own functioning. These are connected capacities, not a claim that everyone controls every condition of life. Bandura, 2001a.
Self-regulation: turn an intention into a revisable practice.
Consider an invented learner who plans to check one claim before sharing it. They record what they did, compare it with a chosen standard and adjust the next attempt. Self-observation, judgment and self-reaction can help explain that process. Standards and goals can also be revised. Bandura, 1991.
A practical routine might be: state the verification goal, observe your procedure, inspect the result and identify the next correction. Rehearsal and feedback matter. A checklist left unopened provides little evidence that self-regulation occurred.
Agency has more than one mode.
- Personal agency: a learner performs the source check directly.
- Proxy agency: the learner asks a librarian or administrator who can obtain a needed document or change access conditions.
- Collective agency: classmates coordinate a verification desk, share responsibilities and establish checking procedures together.
Collective efficacy concerns a group’s shared belief in its capability to act together. It is not simply the average of several people saying they feel confident. Coordination, shared tasks and the group’s conditions need examination. Bandura, 2001a.
Moral agency and disengagement
Bandura’s account includes personal standards and responsibility. Moral disengagement examines ways people disconnect those standards from harmful conduct—for example, minimising a harm or diffusing responsibility. Bandura, 1999.
“Everyone forwarded it” could diffuse responsibility for spreading a damaging false claim. That interpretation is a hypothesis about reasoning, not a diagnosis from one sentence. Ask how the person understands the harm and their role. SCT does not require the conclusion that society determines conduct while individuals bear no responsibility.
The same tutorial can leave different gaps.
Three learners watch a demonstration of tracing a quotation to its original source. Their difficulties differ. Select a case, identify the evidence and propose a targeted response before revealing the analysis.
Compare the interventions.
Why would the same encouragement message be inadequate for all three learners? Write one question and one form of support for each. Explain what evidence would show whether the support addressed the actual difficulty.
Symbolic models can travel. Support must travel too.
Media can present models through speech, writing, images, drama and demonstrations. A symbolic model can show a method, a way of speaking, a social role or a response to difficulty. Observers can learn a general principle and adapt it rather than reproduce an exact movement. Bandura, 2001b.
Two pathways of media influence
Bandura distinguishes a direct pathway, through which communication informs, enables, motivates or guides people, and a socially mediated pathway, through which communication connects people with social networks and settings that provide support and opportunities for action. The pathways can work together. Bandura, 2001b.
In a fictional university project, a tutorial demonstrates how to check a source. A linked peer workshop then supplies practice, feedback and help with access. Watching provides one part of the learning environment; the workshop provides others. A high viewing count cannot establish that the full process occurred.
Design a demonstration that exposes the decisions.
- Make the relevant action visible. Highlight what the model checks and why.
- Show a recoverable difficulty. A reasoned correction can reveal a strategy that a flawless final answer conceals.
- Give the observer something to try. Provide a comparable task with enough variation to test understanding.
- Support feedback and access. Ensure the learner can practise under suitable conditions.
These are teaching applications to examine, not guarantees that one video format always works best. In a Bengali or Hindi lesson, the language of the explanation and the accessibility of source materials may affect what learners can follow and attempt. Test the materials with the intended audience.
Influencers, rewards and imitation
A creator may demonstrate an action and display approval, status or financial success. Ask what the viewer attends to, which outcomes they expect and whether the action is feasible or valued. Popularity alone does not demonstrate the proposed learning mechanism.
AI demonstrations: inspect the modelled procedure.
An AI-generated worked example can present symbolic material from which a learner tries to learn. Check whether the steps and sources are correct, whether uncertainty is visible and whether the learner can apply the method independently. A convincing presentation can model an error as readily as a sound procedure.
This is a proposed application of SCT to the human learner. It does not establish that the system itself has human intentions or efficacy beliefs. Measure what the learner understands and does, alongside their confidence.
Measure the process you claim to improve.
A useful classroom question is: Does a narrated source-checking demonstration, followed by supported practice, improve independent verification performance? Specify the learners, task and comparison. Avoid defining success only as liking the video or reporting confidence.
- Specify a constructive skill and a comparison.
Use prepared source materials and adult volunteers. For example, compare a narrated demonstration plus practice with the same written procedure plus equivalent practice time. Decide what differs and what is held comparable. Random allocation, where feasible, can strengthen a causal comparison.
- Measure the starting point.
Assess prior knowledge, a brief baseline task and task-specific efficacy before instruction. A post-test alone cannot show an individual’s improvement. Explain recruitment, consent and any limits of the sample.
- Separate the proposed processes.
Ask learners to identify and recall the checking steps. Use a new task to assess production. Ask capability questions separately from expected benefits and intention. A single item called “engagement” cannot measure every component of SCT.
- Observe performance independently.
Set a rubric before inspecting group results: locating the source, checking context, comparing wording and justifying the conclusion. Use equivalent tasks and, where possible, assessors unaware of condition. Check scoring consistency.
- Record the environment and support.
Note access failures, language difficulties, practice time, feedback and assistance. If groups receive different support, it may be part of the intervention rather than an incidental detail. Distinguish independent work from performance achieved with help.
- Report effects and limits in proportion.
Examine uncertainty, baseline differences and missing data. A convenience group cannot represent every university student. A package comparison estimates the effect of that package under its conditions; it does not isolate every mechanism within it.
Draft efficacy items
Ask: “How confident are you that you can do each action now?” Keep the materials and demands clear. The examples below are original teaching items, not a validated scale.
| Demand | Draft item | What to examine |
|---|---|---|
| A linked source | Find the original report when a post provides a working link. | A relatively supported task. |
| An unlinked source | Locate the original report when the post supplies only its title. | A more demanding search task. |
| A misleading extract | Explain how omitted context changes the meaning of a quotation. | Interpretation as well as retrieval. |
A performance rubric you can adapt
| Criterion | Points | Evidence |
|---|---|---|
| Source identification | 2 | Locates the relevant original document rather than only a repost. |
| Context | 2 | Reads enough surrounding material to identify a qualification or omission. |
| Comparison | 2 | Accurately compares the claim or quotation with the source. |
| Justification | 2 | Explains the conclusion using specific evidence and acknowledges remaining uncertainty. |
Why does higher confidence plus better performance not prove the mechanism?
Both may change after the same instruction, or success may raise confidence. A simultaneous post-test association does not establish that efficacy change caused the performance improvement. A mechanism claim needs appropriate timing, design and assumptions; randomising the instructional package alone does not randomise the mediator.
Read the theory critically.
SCT’s breadth helps connect learning, agency and context, but also makes an undifferentiated test difficult. Define which relationships are being tested. Self-report, selective exposure, reciprocal causal directions and transfer from controlled settings all require care. Biological characteristics and social structures belong within the wider account, although a particular study may examine them inadequately.
Continue your methods work: Research Methodology · Content Analysis · Ethnography.
Watch the lecture with a case beside you.
Main lecture · Dr. USPSocial Cognitive Theory of Mass Communication Explained
Use the video alongside this chapter. Start with one media example and identify a model, an observer, the relevant action and the conditions under which the action could occur.
A viewing guide
- Start at 0:00 · Learning through action and modelling.
- Start near 1:44 · Attention and retention.
- Start near 2:39 · Production, motivation and the decision to act.
- Start near 3:32 · Consequences and self-efficacy.
- Start near 5:07 · Reciprocal influence and symbolic modelling.
These entry points are approximate guides based on the supplied transcript.
The Bobo Beatdown
This introductory lesson discusses observational learning and the Bobo-doll research. Compare its examples with the original studies rather than treating an engaging summary as the full account of the evidence.
Viewing task: identify a claim about learning, a claim about displayed behaviour and one limit on generalising from the experiment.
Explain the response, not just the label.
Choose an answer and write your reason before opening the explanation. These questions are for self-study; responses are not submitted or automatically scored.
01. Which explanation best represents triadic reciprocal causation?
- Environment alone fixes behaviour.
- Personal factors, behaviour and environment can influence one another.
- All three influences are always equal and simultaneous.
- A person’s beliefs determine every external condition.
Show answer and explanation
Answer: B. SCT examines reciprocal relationships without requiring equal strength or complete personal control. Identify the particular links a study can support.
02. A learner remembers a procedure but cannot yet carry it out accurately. Which process most directly needs examination?
- Behavioural production and the skills or feedback supporting it.
- Attention alone, regardless of the evidence.
- The presenter’s popularity.
- The learner’s general self-worth.
Show answer and explanation
Answer: A. Remembering and producing are different achievements. Check the component skills, practice conditions and feedback rather than assuming the learner did not attend.
03. Which statement is a task-specific self-efficacy belief?
- Checking sources usually improves accuracy.
- I intend to check this claim tomorrow.
- I can locate the original report when no link is provided.
- I am a worthwhile person.
Show answer and explanation
Answer: C. The statement concerns perceived capability. The other statements concern an outcome expectation, an intention and a broad self-evaluation.
04. An observer sees a model punished and does not imitate the action. What can be concluded without another test?
- The observer definitely learned nothing.
- Punishment always prevents acquisition.
- Every observer will respond identically.
- The observer did not display the action under those conditions; acquisition remains a separate question.
Show answer and explanation
Answer: D. Displayed performance does not exhaust what may have been learned. The distinction was important in Bandura’s research on models’ consequences.
05. A capable learner cannot complete the task because the source is inaccessible. What is the strongest response?
- Treat it as proof of low motivation.
- Investigate and address the access barrier, then reassess performance.
- Increase the learner’s confidence rating and declare success.
- Ignore environment because SCT concerns only cognition.
Show answer and explanation
Answer: B. An environmental constraint can prevent action despite skill and efficacy. A measure of confidence alone would miss the observed obstacle.
06. Which example most clearly shows mastery experience as efficacy information?
- Successfully completing progressively harder tasks and interpreting those achievements.
- Watching someone receive an unrelated prize.
- Hearing only a general instruction to be positive.
- Intending to practise without making an attempt.
Show answer and explanation
Answer: A. Mastery information comes from interpreting one’s own performance. It is not identical to intention, generic encouragement or observing another person.
07. A tutorial links learners to a peer workshop that provides practice and feedback. Which mass-communication pathway is illustrated?
- A claim that all viewers learn immediately.
- A process in which social support is irrelevant.
- A socially mediated pathway, working alongside the instructional message.
- Proof that a viewing count measures skill.
Show answer and explanation
Answer: C. Bandura’s socially mediated pathway connects communication with networks and settings that support action. Investigate what the workshop adds rather than counting views alone.
08. A study finds confidence and performance are positively associated at the post-test. Which conclusion is justified by that fact alone?
- Confidence caused performance to improve.
- Performance could not influence confidence.
- The instructional mechanism is fully established.
- An association was observed; causal direction and the proposed mechanism require stronger evidence.
Show answer and explanation
Answer: D. Both may share an influence, and performance can change efficacy beliefs. Timing and design matter for a causal mechanism claim.
Applied task · Write 400–600 words
Design a five-minute demonstration of a constructive media skill, such as checking a quotation or explaining a data display. Specify the intended learner and identify what must be noticed, remembered and practised.
Describe a possible efficacy barrier, an outcome expectation and an environmental condition. Add one form of feedback and a transfer task. Explain how you would distinguish enjoyment, confidence, recall and independent performance.
| Criterion | Points | Look for |
|---|---|---|
| Conceptual precision | 3 | Separate learning processes, efficacy, outcome expectations and environment. |
| Teaching design | 3 | Show the relevant decisions, enable practice and provide informative feedback. |
| Assessment | 2 | Measure a new performance as well as what learners report. |
| Limits and alternatives | 2 | Identify another explanation and avoid treating the proposed design as guaranteed to work. |
Ready to move on?
Look for the learner, the action and the setting.
SCT connects what people observe with what they think, attempt and regulate within particular environments. A useful explanation identifies the modelled process, the observer’s capabilities and beliefs, the consequences they anticipate and the opportunities available. Then it asks which evidence supports each proposed link.
Quick glossary
- Observational learning
- Acquiring information, strategies or capabilities through observing models.
- Symbolic modelling
- Modelling presented through symbolic media, including language, images and demonstrations.
- Reciprocal causation
- The theoretical interplay of personal factors, behaviour and environment.
- Acquisition / performance
- What has been learned / what is displayed in a particular situation.
- Self-efficacy
- A belief about capability to carry out a specified action under relevant conditions.
- Outcome expectation
- A belief about consequences likely to follow an action.
- Vicarious consequence
- A consequence observed happening to a model rather than directly to the observer.
- Self-regulation
- Processes through which people observe, evaluate and adjust their action in relation to goals and standards.
- Proxy agency
- Acting through others who can secure an outcome or alter conditions on one’s behalf.
- Collective agency
- Coordinated action through which people pursue shared outcomes.
- Moral disengagement
- Processes that disconnect moral self-regulation from harmful conduct.
Read in three steps
Start with Bandura’s mass-communication article. Pair the 1965 experiment with the 1977 efficacy paper to clarify acquisition, performance and capability beliefs. Add the agency account and measurement guide before designing a study.
The links lead to scholarly or bibliographic records. Some full texts may require institutional access. Publication years refer to the original works, not later website migration dates.
- 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. Read the scholarly record ↗
A foundational modelling experiment. - Bandura, A. (1965). Influence of model’s reinforcement contingencies on the acquisition of imitative responses. Journal of Personality and Social Psychology, 1(6), 589–595. Read the scholarly record ↗
Acquisition, performance and observed consequences. - Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191–215. Read the scholarly record ↗
Efficacy beliefs, their sources and behavioural change. - Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Prentice-Hall. Read the scholarly record ↗
The foundational book · bibliographic record. - Bandura, A. (1991). Social cognitive theory of self-regulation. Organizational Behavior and Human Decision Processes, 50(2), 248–287. Read the scholarly record ↗
Self-observation, judgment and self-reaction. - Bandura, A. (1999). Moral disengagement in the perpetration of inhumanities. Personality and Social Psychology Review, 3(3), 193–209. Read the scholarly record ↗
Moral standards, agency and disengagement. - Bandura, A. (2001a). Social cognitive theory: An agentic perspective. Annual Review of Psychology, 52, 1–26. Read the scholarly record ↗
Personal, proxy and collective agency. - Bandura, A. (2001b). Social cognitive theory of mass communication. Media Psychology, 3(3), 265–299. Read the scholarly record ↗
The central communication-theory reading. - Bandura, A. (2006). Guide for constructing self-efficacy scales. In F. Pajares & T. Urdan (Eds.), Self-efficacy beliefs of adolescents (pp. 307–337). Information Age Publishing. Read the scholarly record ↗
Task-specific efficacy measurement · publisher-hosted chapter record.