A theory based on Kelson's adaptation level formulation was presented that was capable of integrating the data on the problem of transposition of intermediate size. These data were analyzed and new research reported that revealed several phenomena of importance for any theory of intermediate size discrimination. The modes of response either replicated or reported were transposition, systematic preference for the absolute stimulus, neither absolute nor relational choice, equal preference for 1 test stimuli, and random response. The ratio theory was able to deal with the experimental results and to predict the points of transition from one mode of response to another. The limitations and the implications of the theory were discussed.
THE INTERMEDIATE SIZE PROBLEMTransposition has been a central concern for theories designed to explain the essential nature of the stimulus in discrimination learning. However, evidence opposed to all theories that have attempted to account for transposition has been revealed by research on the intermediate size problem. The first study that employed three stimuli differing in size and simultaneously presented was that of Spence (1942) which showed that response in test was predominantly to the specific stimulus on which the subjects (chimpanzees) had originally been trained. Spence considered these data to be strong evidence against both a relational (Kohler, 1955) and a configurational (Gulliksen & Wolfle, 1938) theory and extended his earlier formulation (Spence, 1937) to account for the intermediate size 1 This paper is based on a dissertation submitted to the Graduate Faculty of the New School for Social Research in partial fulfill ment of the requirements for the PhD degree. The writer wishes to express his gratitude to the chairman of his dissertation committee, Jerome Wodinsky, for his interest and invaluable advice during all stages of the development of this study.1 Now at Wellesley College.