ABSTRACT
An animal (or person) trained to respond in a given way to one stimulus will tend to show the same behavior (although usually with reduced vigor) when confronted with another similar stimulus. The standard explanation for this phenomenon of generalization is, in principle, very simple. It is acknowledged that the event or object that the experimenter refers to, for convenience, as a stimulus will always be, in fact, a complex made up of many features or elements. The simplest of tones, such as might be used in the animal conditioning laboratory, has a definable intensity, duration, frequency, site of origin, and so on. All of these features may be presumed to be capable of entering into associations and thus of contributing to a conditioned response (CR) that the tone can come to elicit as a result of training. Another stimulus such as a clicker may appear to be quite different from the tone but it will hold some features in common with the latter. The situation is shown in Fig. 9.1. Each circle represents the set of features that constitutes a stimulus. Each of the stimuli, A and B, possesses certain unique and defining features (represented by the areas marked a and b). However, A and B also have some features in common (the overlapping area marked c). Establishing a CR to A will mean conditioning both a and c elements. Generalization to B will occur because presentation of this stimulus will activate some elements (the c elements) that have been conditioned when presented as parts of Stimulus A.
