A form of learning in which an animal forms an association between two stimuli, such that one stimulus comes to predict the other and produce a response on its own.

The rattle of the feed bin predicting dinner is the classic example. The sound of a leash being lifted predicting a walk is another. The sight of the float predicting either a pleasant or unpleasant journey is another. In each case, a stimulus that started as neutral has become a predictor of something significant, and the animal’s response shifts accordingly.

Classical conditioning was systematically described by Ivan Pavlov in the early twentieth century through his work with dogs, food, and bells, and is one of the foundational frameworks within learning theory. Unlike operant conditioning, where the animal’s behaviour is changing what happens, classical conditioning is about predictive associations forming whether the animal acts or not.

Classical conditioning runs continuously in the background of every training interaction, and is the mechanism underlying both positive associations (the feed bin, the click of a clicker that has been paired with food) and negative ones (the float that has predicted unpleasant journeys, the vet’s car park that has predicted unpleasant procedures). Modern training practice across species attends carefully to classical conditioning alongside operant conditioning, recognising that the two systems run simultaneously.

The technique of counterconditioning, used to change established negative associations to positive ones, is a direct application of classical conditioning principles. So is the use of bridging signals in modern training, where a previously neutral signal is paired with primary reinforcement until it acquires reinforcing value of its own.