The identification of concepts as a function of amounts of relevant and irrelevant information.

Several recent studies have shown that the efficiency of concept-identification depends upon the complexity of the stimulus-patterns to be categorized. Specifically, mean errors to attainment of the concept increased linearly with complexity in these studies.1 Complexity was quantified in terms of the number of binary dimensions, e.g. color (red-green), within which the patterns could vary. Any such dimension was either relevant to solution, i.e. necessarily used in classifying the patterns, or irrelevant, hence useless for correct identification of the patterns. As the number of dimensions increased, the number of alternative patterns increased along with the information contained in each alternative. The amount of infor-