Learning: Classical and Operant Conditioning, Observational Learning
Three paradigms share this domain: classical conditioning (stimulus association), operant conditioning (consequence) and observational learning (modelling). Questions typically test schedules of reinforcement and the distinction between negative reinforcement and punishment.
Classical conditioning
- Pavlov: UCS → UCR; a neutral stimulus paired with the UCS becomes a CS producing a CR.
- Processes: acquisition, extinction, spontaneous recovery, generalisation, discrimination, higher-order conditioning.
- Watson and Rayner's Little Albert: fear conditioned to a white rat and generalised.
- Applications: systematic desensitisation, aversion therapy, conditioned taste aversion.
Operant conditioning
- Thorndike's law of effect; Skinner's operant chamber.
- Positive reinforcement adds a pleasant consequence; negative reinforcement removes an unpleasant one - both increase behaviour. Punishment decreases behaviour.
- Schedules: continuous; fixed ratio; variable ratio (greatest resistance to extinction); fixed interval (scalloped pattern); variable interval (steady rate).
- Shaping, chaining, token economy, biofeedback.
Observational and cognitive
- Bandura's Bobo-doll study: attention, retention, motor reproduction, motivation; vicarious reinforcement.
- Köhler's insight learning in chimpanzees; Tolman's latent learning and cognitive maps.
Worked example
A child screams in a shop; the parent buys chocolate to stop it. The screaming is followed by a pleasant consequence, so it is positively reinforced and will increase. The parent's giving in is negatively reinforced because the screaming stops, so the parent gives in faster next time. Both behaviours strengthen.
Traps
- Negative reinforcement is not punishment; reinforcement always increases behaviour.
- Extinction is not forgetting; the response can spontaneously recover.
- Variable ratio, not continuous reinforcement, produces the greatest resistance to extinction.