Classical conditioning, first systematically studied by Ivan Pavlov in the 1890s, is the process by which an organism learns to associate a neutral stimulus with a biologically significant event. Pavlov's dogs learned to salivate at the sound of a bell that had been repeatedly paired with food. This simple form of learning is far from trivial — it underlies emotional responses, phobias, drug tolerance, immune responses, and advertising effects, and its study launched the scientific investigation of learning.
Basic Phenomena
In the standard procedure, an initially neutral conditioned stimulus (CS, e.g., a tone) is repeatedly paired with an unconditioned stimulus (US, e.g., food) that naturally elicits an unconditioned response (UR, e.g., salivation). After sufficient pairings, the CS alone elicits a conditioned response (CR, e.g., salivation to the tone). Acquisition follows a negatively accelerated learning curve, with the CR growing stronger over trials but at a decreasing rate.
Key Phenomena
Extinction occurs when the CS is repeatedly presented without the US — the CR gradually weakens. But extinction does not erase the original learning: spontaneous recovery (return of the CR after a rest period), renewal (return of the CR in a different context), and reinstatement (return of the CR after US-alone presentations) all demonstrate that the original association persists beneath the extinguished behavior.
Stimulus generalization produces CRs to stimuli similar to the original CS, with the response strength declining as similarity decreases. Stimulus discrimination develops when one stimulus is paired with the US and a similar stimulus is not. Higher-order conditioning occurs when a second CS is paired with the first CS rather than directly with the US.
John Watson and Rosalie Rayner's (1920) "Little Albert" study demonstrated that fear could be classically conditioned. By pairing a white rat (CS) with a loud noise (US), they conditioned 11-month-old Albert to fear the rat. This fear generalized to other furry objects. The study (despite its ethical problems) established the principle that emotional responses are learned through association — a foundation for understanding phobias and for the development of behavioral therapies such as systematic desensitization.
Modern Understanding
Modern theories view classical conditioning not as simple stimulus-response association but as the learning of predictive relationships. The Rescorla-Wagner model formalized this by showing that conditioning depends on prediction error — the discrepancy between expected and actual outcomes. Blocking (a pre-trained CS prevents conditioning to a new CS added to the compound) demonstrates that mere contiguity is insufficient; the CS must provide new information about the US.