A novel contextual dimension for use with an operant chamber: From simple to hierarchical forms of learning.

Abstract

A novel procedure is described in which the floor temperatures (warm and cool) in an operant chamber are used as contextual cues in 2 experiments with rats. Experiment 1 demonstrated that rats learn the relationship between these thermal contexts and auditory stimuli that have been paired with them. Experiment 2 showed that thermal contexts can serve a conditional function that (a) reflects the operation of a mechanism that is common to conventional, visual contexts and (b) is bound to these contexts' ability to retrieve the nature of the relationship between the auditory stimuli and food. The experimental study of rodent behavior within psychology and neuroscience has been dominated by the use of the operant chamber. Even in its basic form, this apparatus provides the opportunity to present animals with a range of stimuli, manipulanda, and events of motivational significance. Moreover, the ambient or contextual cues that accompany being placed in such a chamber (its visual characteristics, odor, and so on) are also cues that animals readily learn about (see Bouton, 1991). As versatile as this apparatus is, it remains the case that the number of stimulus dimensions that are both available and readily controlled (principally, the auditory and visual dimensions) serves as a constraint on the experiments that can be conducted. For example, when an additional stimulus dimension is required, different odor cues have been used that require procedures designed to minimize contamination between them. These procedures can be either expensive or inconvenient (e.g., Hall & Honey, 1989). At a practical level, therefore, the study of rodent behavior in the laboratory would be facilitated by the development of a system that allows another dimension to be presented. Moreover, many of our conceptual models of animal learning and memory (e.g., Gluck & Myers, 1993; McLaren, Kaye, & Mackintosh, 1989; Pearce, 1994; Rescoda & Wagner, 1972) assume that the principles that underlie them will be as likely to operate in one (sensory) domain as another. Use of an additional dimension would provide a way of assessing the generality of those effects on which our understanding of learning and memory has developed. With these practical and theoretical considerations in mind, we describe the development and use of a novel addition to an operant