PPsychology.com

Theories / How behavior is learned

Classical conditioning

Ring a bell before food often enough and a dog starts to salivate at the bell. Ivan Pavlov spent thirty years on what that simple fact means, and a century later it is still one of the load bearing ideas in psychology.

Published November 3, 2026. 4 sources at the foot of the page.

What it says

Classical conditioning is learning that one event predicts another. A signal that means nothing on its own comes to stand in for something that matters, and the body starts answering the signal.

Four terms do the work. Food is the unconditioned stimulus, because it makes a dog salivate with no training. Salivating at food is the unconditioned response. A bell paired with food becomes the conditioned stimulus. Salivating at the bell alone is the conditioned response.

Four more words cover what happens next. Acquisition is the learning. Extinction is what happens when the bell stops predicting food and the response fades. Spontaneous recovery is the response coming back after a rest, which shows extinction did not erase the learning. Generalization is answering a similar bell, and discrimination is learning which bell counts.

The four terms, in one example

A coffee machine in an office that starts grinding before every free pastry round.

  1. Unconditioned stimulus

    The pastry. It makes your mouth water without anybody teaching you to.

  2. Unconditioned response

    Your mouth watering at the pastry. No learning is involved.

  3. Conditioned stimulus

    The grinding noise, which meant nothing to you in your first week.

  4. Conditioned response

    Your mouth watering at the noise alone, before anything is in front of you.

Where it came from

Pavlov was a physiologist studying digestion in dogs, and he won the Nobel Prize in 1904 for that work, not for conditioning. The salivation at the sound of the attendant was in the way of his real experiment until he turned around and studied it. Conditioned Reflexes, the lectures that set out the whole program, reached English in 1927.

John Watson carried the idea to people. In 1920 he and Rosalie Rayner conditioned a fear response in an 11 month old child. That is the study everybody knows as Little Albert. Watson used it to argue that emotional reactions are built the same way.

What holds up, and what does not

The basic finding is as solid as anything in psychology. It holds across species, it is measured in laboratories every week, and conditioned responses in people are easy to produce and to see.

What changed is the explanation. Pairing two events turns out not to be enough. Robert Rescorla's 1988 paper puts it bluntly: conditioning is not what you think it is. An animal learns when a signal adds information about what is coming, and learns nothing from a signal that tells it what it already knew. Leon Kamin showed this in 1969 with blocking, where a second signal added alongside one that already predicts the outcome is simply ignored.

What it does not mean

It does not explain all learning, and it does not make a person a machine. It also does not mean a learned reaction is permanent or that extinction wipes it out. The response comes back after a rest and in new places, which is why the careful word is new learning rather than erasing.

Check yourself

Three questions on what is above. Each answer says why.

  1. In the coffee machine example, which part is the conditioned stimulus?

    1. The pastry
    2. The grinding noise
    3. Your mouth watering
    Show the answer

    Answer: B. The conditioned stimulus is the signal that meant nothing until it started predicting something that matters. The pastry is the unconditioned stimulus.

  2. A conditioned response fades after the signal stops predicting anything, then reappears a week later. What is that called?

    1. Spontaneous recovery
    2. Generalization
    3. Discrimination
    Show the answer

    Answer: A. Spontaneous recovery is the return of a response after extinction. It is the evidence that extinction is new learning rather than erasing the old.

  3. What did Rescorla and Kamin add to the Pavlovian account?

    1. That a signal has to add information, not merely appear alongside the outcome
    2. That conditioning only works on animals
    3. That extinction erases the original learning
    Show the answer

    Answer: A. Blocking shows that a signal which tells an animal nothing new is ignored, however often it is paired. Conditioning tracks information, not coincidence.

Sources

  1. Pavlov, I. P. (1927). Conditioned Reflexes: An Investigation of the Physiological Activity of the Cerebral Cortex. Translated by G. V. Anrep. Oxford University Press.
  2. Watson, J. B., and Rayner, R. (1920). Conditioned emotional reactions. Journal of Experimental Psychology, 3(1).
  3. Kamin, L. J. (1969). Predictability, surprise, attention, and conditioning. In B. A. Campbell and R. M. Church (Editors), Punishment and Aversive Behavior. Appleton Century Crofts.
  4. Rescorla, R. A. (1988). Pavlovian conditioning: It's not what you think it is. American Psychologist, 43(3).

Practice it on Learn

Learn teaches Pavlov's study in Four more studies that changed the field, the second unit of the Landmark studies track. The unit reads it beside the Little Albert experiment, learned helplessness and the minimal group studies, each against its original paper.

Practice Pavlov's dogsSee what Learn is

Every lesson is free. Plus is optional. For adults 18 and over.

More theories