This mini-lesson covers Edexcel Topic 4 — Learning theories: classical conditioning (Pavlov, 1927), operant conditioning (reinforcement, punishment, schedules, shaping), social learning theory and Bandura's Bobo doll studies, how learning theories explain and treat phobias, the classic study (Watson and Rayner, 1920 — Little Albert), observational research and the chi-squared test.
Three ways of learning — and their application to phobias and their treatment.
Work through each screen, answer the questions as you go and collect ⭐ stars. Press Start when you are ready.
Classical conditioning
Pavlov (1927) and learning by association
Classical conditioning is learning through association. Pavlov found that dogs which naturally salivated to food began to salivate to a bell that had been repeatedly paired with it.
UCS (food) → UCR (salivation): an unlearned reflex.
NS (bell) paired repeatedly with the UCS.
The bell becomes the CS, producing salivation as a CR.
Also know: extinction (the CR fades if the CS is repeatedly presented without the UCS), spontaneous recovery (the CR reappears after a rest), and stimulus generalisation (similar stimuli also trigger the CR).
Evaluate it: highly controlled and replicable with objective measurement — but extrapolating from dogs to humans is risky, and classical conditioning cannot explain voluntary behaviour or the role of thought.
Operant conditioning
Reinforcement, punishment and schedules
Operant conditioning is learning through consequences — behaviour is shaped by what follows it (Skinner).
Positive reinforcement — behaviour is followed by something pleasant, so it increases.
Negative reinforcement — behaviour removes something unpleasant, so it increases. This is not punishment.
Primary reinforcers satisfy a basic need (food); secondary reinforcers acquire value by association (money, tokens).
Schedules: continuous reinforcement produces fast learning but fast extinction. Variable ratio schedules (reward after an unpredictable number of responses) produce the highest, most extinction-resistant response rates — which is why gambling is so persistent.
Shaping: reinforcing successive approximations to the target behaviour.
Quick check
Reinforcement or punishment?
?A student feels intense anxiety in a lift, escapes, and the anxiety immediately disappears. She now avoids all lifts. What has kept the avoidance going?
Social learning theory
Observation, imitation and vicarious reinforcement
Bandura argued we also learn by watching others. SLT keeps behaviourist principles but adds mediational (cognitive) processes.
Modelling and imitation: we copy a model, especially one who is similar to us, high status or nurturing.
Vicarious reinforcement: we do not need to be rewarded ourselves — seeing the model rewarded is enough to make imitation more likely (and seeing them punished makes it less likely).
The four mediational processes: attention → retention → reproduction → motivation (reinforcement).
Learning vs performance: Bandura (1965) showed children who had seen the model punished imitated less — but when they were later offered an incentive, they reproduced the aggression just as accurately. They had learned it all along; the consequences only affected whether they performed it.
Sort it
Which type of learning?
Tap a description, then tap the type of learning it illustrates.
🔔 Classical
🎯 Operant
👀 Social learning
Bandura
The Bobo doll experiments (1961, 1963)
Bandura, Ross and Ross (1961) tested whether children would imitate an aggressive adult model.
Sample: 72 children aged roughly 3 to 6 from the Stanford University nursery, matched on pre-rated aggressiveness.
Design: children saw an adult behave aggressively towards a Bobo doll (hitting it with a mallet, distinctive verbal comments) or non-aggressively, or saw no model. They were then mildly frustrated and observed in a room containing a Bobo doll.
Findings: children who saw the aggressive model reproduced far more imitative physical and verbal aggression, including the specific novel acts. Boys showed more physical aggression than girls, especially after a male model — evidence for same-sex modelling.
Evaluate it: controlled and replicable, with a clear operationalised DV — but Bobo dolls are made to be hit (low ecological validity and possible demand characteristics), it measures only short-term imitation, and there are ethical concerns about exposing children to aggression.
Quick check
Learning or performance?
?In Bandura (1965), children who saw the model punished imitated less — until they were offered a reward, when they reproduced the aggression accurately. What does this show?
Phobias
How learning theories explain phobias
The two-process model combines both types of conditioning.
Acquisition — classical conditioning: a neutral stimulus is paired with something that naturally produces fear, and becomes a conditioned stimulus for fear.
Maintenance — operant conditioning: avoiding the phobic stimulus removes the anxiety, which is negatively reinforcing, so the avoidance (and therefore the phobia) persists and is never disconfirmed.
Social learning can also play a part: a child may acquire a fear by watching a parent react fearfully (vicarious acquisition).
Limitation: not everyone with a phobia can recall a traumatic pairing, and some phobias are far more common than others. Seligman's idea of biological preparedness — that we are evolutionarily prepared to fear snakes and heights but not plug sockets — shows the learning account is incomplete.
Quick check
Explaining phobias
?Which is the strongest criticism of the pure learning explanation of phobias?
Treatments
Systematic desensitisation and flooding
If a phobia is learned, it can be unlearned.
Systematic desensitisation: (1) the client is taught deep relaxation; (2) they build an anxiety hierarchy from least to most feared; (3) they work up the hierarchy, staying relaxed at each step. It works by reciprocal inhibition — you cannot be relaxed and anxious at the same time — and by counter-conditioning the stimulus to a new response.
Flooding: immediate, prolonged exposure to the most feared stimulus with no escape. The fear response cannot be sustained physiologically, so it extinguishes. It is quicker and cheaper, but far more traumatic and has higher drop-out.
Evidence: Capafóns et al. (1998) — one of the contemporary study options — used systematic desensitisation with people who feared flying and found reductions in self-reported fear and in physiological measures compared with a control group. Limitation: behavioural therapies treat the symptom, not any underlying cause — a criticism the psychodynamic approach presses hard.
Match it
Match the term to its meaning
Tap a term on the left, then its correct definition on the right.
Term
Meaning
Classic study
Classic study: Watson and Rayner (1920) — Little Albert
Aim: to show that an emotional response (fear) could be conditioned in a human infant.
Participant: a single infant, about 11 months old at conditioning, described as healthy and unemotional. Baseline testing showed he was not afraid of a white rat.
Procedure: each time he reached for the rat (NS), the researchers struck a steel bar behind his head (UCS), producing fear and crying (UCR). This was repeated across sessions.
Findings: Albert came to show fear of the rat alone — the rat had become a CS producing a CR. His fear generalised to other furry white stimuli, including a rabbit, a dog, a fur coat and a Santa Claus mask.
Evaluate it: a controlled demonstration that phobias can be conditioned — the foundation of behavioural therapy. But it is a single case (no generalisability), the fear responses were rated subjectively, and it is ethically indefensible by modern standards: an infant could not consent, was harmed, and was never deconditioned.
Quick check
Little Albert
?In Watson and Rayner (1920), what was the unconditioned stimulus?
Methods · observation
Observational research and animal experiments
The Topic 4 method is observation, as used in the Bobo doll studies.
Recording: use operationalised behavioural categories and tally them; use event sampling (record every time a behaviour occurs) or time sampling (record what is happening at fixed intervals).
Reliability: use two observers and check inter-rater reliability. Observer bias is the main threat to validity.
Content analysis applies the same logic to media and documents, turning qualitative material into quantitative frequencies.
Animal research is justified by control and by extrapolation to humans, but is limited by species differences and is governed by the Scientific Procedures Act (1986) and Home Office regulations (licences, the 3Rs: replacement, reduction, refinement).
Quick check
Which test?
?An observer tallies how many boys and how many girls show imitative aggression (yes/no) after seeing a male model. Which inferential test does Edexcel specify for this topic?
Quick check
Contemporary study
?Becker et al. (2002) studied ethnic Fijian adolescent girls before and after the arrival of television. What was found?
Exam focus
How Topic 4 is assessed
Learning theories is examined in Paper 1: Foundations in psychology and is compulsory.
The commonest question types: apply the three types of learning to a novel scenario; explain the acquisition and maintenance of a phobia; describe and evaluate a treatment.
Expect data-response on an observation (behavioural categories, tallying, inter-rater reliability) and on a chi-squared analysis.
Know Watson and Rayner (1920) and your chosen contemporary study in full, with ethics.
The classic error: confusing negative reinforcement (behaviour increases because something unpleasant is removed) with punishment (behaviour decreases). Examiners test this every year.