No abstract

Rats were trained in Tolman's "sunburst" apparatus modified to eliminate visual cues. Results consistent with his hypothesis of "cognitive maps" were obtained. There was a tendency to take the right sided test pathways pointing in the general direction of the food box. When the number of left and right turns in the training procedure were made equal and the ultimate turn was to the left, no tendency to turn rigllt was found during testing. Twenty years ago Tolman and his co-workers (Tolman, 1948j Tolman, Ritchie, & Kalish, 1946) suggested that "place" learning might be more important than "response" learning, and that animals built what they called "cognitive maps" of their environment. Although at the time other investigators criticized these and other experiments of Tolman (see Hilgard, 1956, for summary), these studies were in themselves not definitive, and Tolman's concepts have continued to occupy an important position in contemporary psychological theory (e.g., Altman, 1966, p. 390). We wished to repeat one of his original experiments with three major modifications: First, by using blind rats as well as normal rats to eliminate all visual cueSj second, by using primates so that if cognitive maps were a "higher" function they might be more readily demonstrated in a higher TRAINING RUNWAY