Two views of the stimulus controlling learned behavior are contrasted. On the elemental view, the organism fractionates its environment into isolatable features or elements, and experience alters how it responds by strengthening or weakening associations among the elements. On the confrgural view, it is the representation of the pattern or conjunction of stimulus elements acting on the organism that acquires control over behavior. This paper addresses the issue of how to describe the stimulus from a comparatiuelorganismic perspective and argues that separate neural systems are available that support both elemental and configural associations. Normal adult rats and rats with hippocampalformation damage, and different-aged rats and children were compared on tasks that permit a solution based only on elemental associations and tasks that only can be solved if the organism can acquire configural associations, Subjects in all comparison conditions were able to solve tasks that permit an elemental solution, but animals with hippocampal-formation damage, preweanling rats, and children less than 55 months of age were significantly impaired on tasks that required solving configural associations. Some implications of this data pattern are discussed.Learning theorists have long debated how to best describe the stimulus that controls a learned behavior. Two basic approaches to this issue can be identified (see Rudy & Wagner, 1975 An example will help clarify the two positions. Suppose an organism experiences a compound stimulus composed of two elements, a and b, followed by a third element, c . How should one think about the nature of the associative linkages that are formed between the ab compound and c? Elemental theorists assume the organism fractures its world into independent stimulus elements, features, or attributes and that experience alters how it responds to its environment by strengthening or weakening associations among the elements. In the example, the components a and b independently associate with c. Elemental theorists also assume that Reprint requests should be sent to J. W. Rudy, Department of Psychology, CB-345, University of Colorado, Boulder, CO 80309.