Observational and Cognitive Learning

Learning by watching

Bandura's Bobo doll study showed children who watched an adult hit an inflatable doll later hit it themselves, in the same style. No reinforcement reached the child at all — modelling was enough. Vicarious reinforcement (seeing the model rewarded) raises imitation; vicarious punishment lowers it. Models are copied most when they are admired, similar to the observer, or successful. The same mechanism produces prosocial modelling: a parent who actually reads is a stronger cause of a reading child than a parent who tells the child to read.

Learning you cannot see yet

• Latent learning — Tolman's rats explored a maze with no reward and looked like they had learned nothing. The day a reward appeared, they ran it almost perfectly. They had built a cognitive map all along: learning is not the same as performance.
• Insight learning — Köhler's chimpanzees solved a problem in a sudden reorganisation, not by gradual trial and error.
• Intrinsic motivation comes from the activity itself; extrinsic motivation from an outside reward. The overjustification effect: paying someone for something they already enjoyed can convert the reason into the payment and reduce the behaviour once the payment stops.

Biology sets the limits

Garcia found that rats associate nausea with taste after a single pairing and a delay of hours, yet will not associate nausea with a light or sound. This taste aversion breaks two supposed rules of classical conditioning — that pairings must be repeated and nearly simultaneous. Preparedness means animals are biologically primed to learn associations that mattered for survival. The Brelands' instinctive drift showed trained animals sliding back into innate behaviour. Seligman's learned helplessness: after inescapable shocks, dogs stopped trying even when escape became possible — a model later applied to depression and to a pessimistic explanatory style.