Ivan Pavlov demonstrating an experiment on a dog to students in a lecture theatre of the Military Medical Academy in St Petersburg
Unit 04 · All masterclasses · 60 min

Learning and Conditioning

Pavlov, Skinner, and the rules by which experience changes behaviour
Image: Wellcome Collection, CC BY 4.0

How experience changes behaviour, from Pavlov's dogs to the treatment of a phobia. The unit works through classical conditioning with Pavlov's own data, runs the Rescorla-Wagner rule by hand to show why animals learn predictions rather than pairings, reads Skinner's schedules of reinforcement from their cumulative records, and ends with learning by observation and the uses of all this in therapy, in education, and in advertising.

Preview: the beginning of the unit

A dog that salivates at the sound of a metronome, a rat that presses a lever for food, a child who hits a doll after watching an adult hit it, and a patient who can hold a spider after a few sessions with a therapist are four scenes from one subject. In each of them experience changes behaviour in a lawful way, and the laws were found with simple apparatus and careful counting, most of them between 1898 and 1972. This unit presents them with the original data, from Pavlov’s drops of saliva to Bandura’s tally of punches, and ends with their uses, because the rules that make a phobia are the rules that remove one.

1 What learning is

Psychologists define learning as a lasting change in behaviour, or in the capacity for behaviour, that results from experience. The definition excludes changes that come from growth (a child walks at about one year because its body has matured), from fatigue, and from drugs, and the word lasting excludes a momentary reaction. It says nothing about awareness, because much of what this unit describes happens without the learner being able to report it, and much of it was discovered in animals, which cannot be asked.

The simplest form of learning is habituation, the decline of a response to a stimulus that is repeated and has no consequence. A sudden noise makes a person flinch, and the tenth repetition at the same volume does not. The same decline can be followed down to the cells. The sea slug Aplysia californica pulls in its gill when it is touched, and the withdrawal shrinks over a dozen touches; the group of Eric Kandel traced this decline to a smaller release of neurotransmitter at one synapse, of the kind that Unit 2 describes. Habituation is specific to the stimulus, so a louder or a different noise brings the full response back, which is how habituation is told apart from a muscle or a sense organ that is merely tired. Its opposite is sensitization, an increase of the response to many stimuli after one strong or painful stimulus, as when the same slug withdraws its gill more vigorously to a light touch after a shock to its tail. Both are called non-associative forms of learning, because the animal learns about one stimulus and not about a relation between two. The rest of the unit is about associative learning, in which the learner acquires a relation, either between two stimuli or between a behaviour and its consequence.

The full unit is part of the program

Want to see a complete unit first? Unit 3, Sensation and Perception, is free to read.