'Reinforcement' is the process which results when reinforcers are delivered. In deciding which pattern of behaviour is to be associated with a reinforcer, the experimenter is essentially establishing a teaching situation. A reinforcer is said to increase and maintain the frequency of behaviour with which it is associated; by specifying a form of behaviour upon which the reinforcer is to be dependent, its effect is demonstrated more unequivocally. Essentially, the 'trainer' builds up a new pattern of behaviour by making reinforcers dependent upon behaviour which gradually approximates to the desired act. The frequency of approach to that part of the cage is a function of the demonstrable fact that a reinforcer is dependent upon, and therefore associated with, that behaviour. The technical term 'reinforcer' was introduced during the discussion of 'superstitious' behaviour in animals. B. F. Skinner entitled 'How to teach animals', in which he explains how the method of successive approximation may be used develop tricks in household pets.