Sometimes now used as a synonym for LEARNING itself, the term CONDITIONING referred originally to the procedure introduced by I.P.Pavlov (1849–1936) in the early years of the twentieth century for the study of nervous function in animals. In PAVLOVIAN CONDITIONING (also called classical conditioning to distinguish it from other variants of the procedure developed subsequently) animal subjects are given paired presentations of two stimuli, one of which unconditionally evokes a response. As a result of this experience the other stimulus acquires the power to evoke a new response, often (although not invariably) similar in form to that evoked by the stimulus with which it has been paired. Since this newly acquired capacity is conditional upon the animal’s having undergone this training, the response has come to be called a conditional (conditioned) response (or reflex) (see CONDITIONED REFLEX) and the entire process gets referred to as conditioning. Pavlov’s own view, that the many other forms of learning might be explicable in terms of the basic processes at work in his classical conditioning paradigm, probably contributed to the more liberal use of the term conditioning.
Evidence to show that Pavlov’s version of conditioning could not explain all forms even of animal learning came from work conducted in the 1920s by the Polish neurophysiologist J. Konorski (1903–1973). He demonstrated that a dog induced to perform a particular response (for instance, to lift a paw) would come to perform this response with increasing frequency if the response was followed by access to food. In this procedure the critical contingency is between the response and the stimulus (food) that follows it, as distinct from Pavlov’s procedure in which the critical CONTINGENCY is between two stimuli.
Konorski still used the term conditioning to describe the learning process engaged by his experiments, but called it Type II conditioning to distinguish it from classical Pavlovian conditioning, which he referred to as Type I. The distinction was developed in the United States during the 1930s by B.F.Skinner who similarly pointed out that classical conditioning principles could not explain the increased readiness with which a rat would come to press a lever when that action resulted in the delivery of a food pellet. Skinner christened this form of learning (which is formally identical to that studied by Konorski) OPERANT CONDITIONING, the new term ‘operant’ being coined to emphasize the fact that the animal must operate upon its environment in order to obtain an effect. (The Pavlovian version was renamed RESPONDENT CONDITIONING.) Operant conditioning is also sometimes referred to as INSTRUMENTAL CONDITIONING, so called because the behaviour of the animal is instrumental in producing an effect.
Since the two types of conditioning procedure are operationally quite distinct (in classical the animal is exposed to a contingency between two stimuli; in operant the contingency is between an action and an outcome) the question arose as to whether they differed in more fundamental ways. One widely canvassed suggestion was that classical conditioning might be effective only in modifying the occurrence of simple REFLEX responses and that the modification of voluntary behaviour depended on operant conditioning processes. The distinction has proved to be untenable, however; there have been demonstrations of the operant conditioning of involuntary responses and that supposedly voluntary behaviour such as an animal shows in moving around its environment, can be susceptible to classical conditioning. The modern consensus is that both types of conditioning depend on a common ASSOCIATIVE LEARNING process, differing only in the events that enter into association: two stimuli in the case of classical conditioning, a response and a stimulus in the operant case.
Conditioning processes undoubtedly operate outside the animal learning laboratory. It seems likely, for instance that some instances of human PHOBIA are a product of the classical conditioning, of the fortuitous pairing, perhaps early in life, of the feared event with some trauma. And the way in which our everyday behaviour can be shaped by REWARD and PUNISHMENT may be taken to reflect operant conditioning in action. More generally, it has been suggested that the basic associative processes that are revealed in laboratory studies of animal conditioning constitute the underlying mechanism of all (or almost all) instances of individual learning and adaptation in both animals and humans. There is, at present, no convincing evidence to require us either to accept or reject this suggestion.
GEOFFREY HALL
This is the complete article, containing 718 words
(approx. 2 pages at 300 words per page).
