What is classical conditioning?

Classical conditioning is one of the most fundamental forms of associative learning — the process by which a neutral stimulus comes to elicit a response through repeated pairing with a stimulus that naturally produces that response. First systematically studied by the Russian physiologist Ivan Pavlov in the 1890s and early 1900s, classical conditioning provided psychology with one of its most powerful and enduring frameworks for understanding how organisms learn from their environment.

From neutral stimulus to conditioned response.

Classical conditioning unfolds through predictable stages. Understanding where each process sits helps clarify what is being learned — and how learning can be reversed.

Stage 1Before conditioning

Unconditioned Stimulus → Response

A stimulus that naturally and automatically triggers a response without any prior learning — the UCS. Food in the mouth automatically produces salivation (UCR). This reflex is innate and requires no experience. The goal of conditioning is to transfer this response to a previously neutral stimulus.

No learning required · Innate reflex
Stage 2During conditioning

Pairing NS with UCS

A neutral stimulus (NS) — one that initially produces no relevant response — is repeatedly presented just before the UCS. Pavlov rang a bell before presenting food. Over repeated pairings, the neutral stimulus becomes associated with the unconditioned stimulus. Timing matters: the NS must reliably predict the UCS.

Neutral stimulus precedes UCS · Association forming
Stage 3After conditioning

Conditioned Stimulus → Response

Once the association is established, the formerly neutral stimulus becomes a conditioned stimulus (CS) — presenting it alone now elicits a conditioned response (CR) that resembles but is typically weaker than the original unconditioned response. The bell alone makes the dog salivate. Learning is complete.

Learned response · CS alone is enough
Stage 4Later processes

Extinction & Recovery

If the CS is repeatedly presented without the UCS, the conditioned response gradually weakens and eventually disappears — extinction. But learning is not erased: after a rest period, the CR often spontaneously returns (spontaneous recovery), demonstrating that extinction is new learning (suppression of the CR) rather than forgetting.

CS without UCS → CR weakens · Not erased, suppressed

Ivan Pavlov

1849–1936

Russian physiologist who discovered classical conditioning while studying salivary reflexes in dogs. His systematic experimental methodology and precise terminology (CS, UCS, CR, UCR, extinction, generalisation) became foundational to behaviourism and learning theory. He was awarded the Nobel Prize in Physiology or Medicine in 1904 for his work on digestion — before his conditioning research had fully developed.

John B. Watson1878–1958

American psychologist who adopted Pavlov's conditioning framework and extended it to human emotion in the Little Albert experiment. Watson argued that all human behaviour — including emotion, personality, and thought — could be explained through conditioned stimulus-response associations, laying the ground for American behaviourism.

Mary Cover Jones

1896–1987

American psychologist who demonstrated that conditioned fears could be eliminated through counter-conditioning. Her work with Peter — a child afraid of rabbits — used gradual exposure paired with a pleasant experience (food) to eliminate the fear response. Her approach directly foreshadowed systematic desensitisation and is considered the first empirical demonstration of behaviour therapy.

John Garcia

1917–2012

American psychologist whose taste aversion experiments revealed that classical conditioning is constrained by biological preparedness — organisms learn some associations far more readily than others. His findings challenged the behaviourist assumption that any stimulus could be conditioned to any response with equal ease.

Unconditioned stimulus (UCS) and response (UCR)

The UCS is a stimulus that automatically and naturally produces a response without any learning — food, pain, a loud noise. The UCR is the natural, unlearned response to the UCS — salivation, withdrawal, startle. These form the biological foundation onto which conditioning grafts new learned associations.

Conditioned stimulus (CS) and response (CR)

After conditioning, the previously neutral stimulus becomes a CS that elicits the conditioned response (CR) on its own. The CR resembles the UCR but is typically weaker and may differ in some respects — it is a learned anticipatory response shaped by the organism's prediction of the UCS.

Stimulus generalisation

The tendency for stimuli similar to the CS to also elicit the CR, even if they were not part of the original conditioning. The more similar the new stimulus to the original CS, the stronger the CR it elicits. Generalisation is adaptive — if a predator's growl is dangerous, similar growls should also trigger escape — but underlies the spread of phobias and anxieties to related stimuli.

Stimulus discrimination

The opposite of generalisation: learning to respond to the specific CS but not to similar stimuli. Achieved by pairing only the target CS with the UCS and presenting other similar stimuli without the UCS. Discrimination produces more precise, targeted responses and is the basis of many therapeutic applications.

Higher-order conditioning

Once a CS has been established, it can itself serve as a UCS to condition a response to a new neutral stimulus. A bell (CS1) that reliably predicts food can be paired with a light (NS) to make the light a conditioned stimulus (CS2) — even though the light was never paired with food. Higher-order conditioning extends the reach of conditioning into more complex chains of association.

Preparedness and biological constraints

Organisms are not equally conditionable to all stimuli. Garcia and Koelling's taste aversion studies showed that rats readily associate flavour with illness (even after a single trial and a long delay) but cannot associate a light-and-tone cue with illness in the same way. Biological preparedness — the evolutionary predisposition to form certain associations more easily — constrains learning in ways that pure stimulus-response accounts cannot explain.

Conditioned emotional responses

Watson and Rayner's Little Albert experiment (1920) demonstrated that emotional responses — in this case, fear — could be classically conditioned in humans. By pairing a white rat (NS) with a loud noise (UCS), they induced fear of the rat (CR) in an 11-month-old infant. The experiment established classical conditioning as a mechanism of phobia development, though it is now considered deeply unethical.

What is classical conditioning?+

Classical conditioning is a form of associative learning in which a neutral stimulus comes to produce a response after being repeatedly paired with a stimulus that naturally triggers that response. Pavlov demonstrated this with dogs that learned to salivate at the sound of a bell after the bell was consistently paired with food. Classical conditioning is now understood to operate across species and underlies phenomena as diverse as phobias, drug cravings, advertising effectiveness, and some forms of therapy.

What is the difference between a conditioned and unconditioned stimulus?+

An unconditioned stimulus (UCS) naturally and automatically produces a response without any learning — food produces salivation, a loud noise produces startle. A conditioned stimulus (CS) is a previously neutral stimulus that has acquired the ability to produce a response through repeated pairing with the UCS — a bell that has been paired with food now produces salivation on its own. The key difference is that the UCS works by biology; the CS works by learned association.

What is an example of classical conditioning in everyday life?+

Phobias are a clear example: if you were bitten by a dog (UCS producing fear) in a specific park (NS), the park itself may become a CS that triggers fear even with no dog present. Advertising works the same way — products repeatedly paired with attractive people, beautiful settings, or upbeat music acquire positive associations. Drug craving is another: addicts often experience strong cravings (CR) in environments associated with past drug use (CS), even after long periods of abstinence.

What did Pavlov's dog experiment show?+

Pavlov's experiments showed that a neutral stimulus — a bell, a metronome, a light — could come to elicit salivation in dogs after being repeatedly paired with food. This demonstrated that learning could be studied objectively, without reference to mental states or consciousness, simply by measuring stimulus-response associations. Pavlov also systematically described extinction (the CS stops producing the CR when the UCS is withheld), spontaneous recovery, generalisation, and discrimination — establishing the fundamental vocabulary of conditioning that is still used today.

What is extinction in classical conditioning?+

Extinction is the weakening and eventual disappearance of a conditioned response when the conditioned stimulus is repeatedly presented without the unconditioned stimulus. If the bell is rung many times with no food, the dog gradually stops salivating to the bell. Importantly, extinction is not the same as forgetting — the original association is suppressed rather than erased. This is demonstrated by spontaneous recovery: after a rest period following extinction, the conditioned response often returns, showing that the original learning persists.

What is the difference between classical and operant conditioning?+

Classical conditioning involves learning associations between stimuli: a neutral stimulus becomes associated with an unconditioned stimulus through repeated pairing, producing a conditioned response. Operant conditioning (Skinner) involves learning the relationship between voluntary behaviour and its consequences — behaviours are strengthened by reinforcement or weakened by punishment. Classical conditioning is largely automatic and involuntary; operant conditioning depends on the organism's own actions and their outcomes.

How is classical conditioning relevant to clinical psychology?+

Classical conditioning underlies several key clinical phenomena and treatments. Specific phobias are understood as conditioned fear responses that have generalised and resist extinction. Systematic desensitisation (pairing feared stimuli with relaxation) and exposure therapy (repeated non-reinforced exposure to feared stimuli) are effective, evidence-based treatments grounded in conditioning principles. Conditioned drug cravings — where environmental cues associated with drug use trigger strong cravings — reflect classical conditioning and are central to understanding relapse in addiction.

Last reviewed July 2025
  1. 1.

    Pavlov, I. P. (1927). Conditioned Reflexes: An Investigation of the Physiological Activity of the Cerebral Cortex (G. V. Anrep, Trans.). Oxford University Press.

    +About this source

    Pavlov's own account of his conditioning experiments — the foundational text for classical conditioning.

  2. 2.

    Watson, J. B., & Rayner, R. (1920). Conditioned emotional reactions. Journal of Experimental Psychology, 3(1), 1–14.

    +About this source

    The Little Albert experiment: demonstration of conditioned fear in a human infant.

  3. 3.

    Rescorla, R. A. (1988). Pavlovian conditioning: It's not what you think it is. American Psychologist, 43(3), 151–160.

    +About this source

    Influential revision of conditioning theory, demonstrating that contingency — not mere contiguity — drives conditioning.

  4. 4.

    LeDoux, J. E. (2000). Emotion circuits in the brain. Annual Review of Neuroscience, 23, 155–184.

    +About this source

    Reviews the neural substrates of conditioned fear, including the role of the amygdala.