Abstract
Classical conditioning can be characterized as an experimenter-defined set of operations, involving a temporal contingency between two stimuli, that are independent of the subject’s behavior. Viewed from this perspective, classical conditioning is unique among the various approaches to the study of behavior because it allows for a level of experimental control that is rarely encountered in the behavioral sciences. Furthermore, if the experimenter chooses the appropriate stimulus parameters, an analysis of the temporal relationships among the responses elicited and the stimuli used permit the externalization of otherwise internal events that may not be directly observed. The information gleaned from this type of analysis can be used to study the adaptive functions of learned associations (Hollis, 1984; Shettleworth, 1983; Culler, 1938; Palmerino et al., 1980; Dworkin, 1993) or the processes that underlie the associations formed during learning (Gormezano, 1972; Schneiderman, 1972, 1974; Moore, 1986; Thompson et al., 1983). In recent years, the “internal events” described by (1972) have been investigated directly using electrophysiological techniques to monitor neural activity during conditioning. In addition, modern neuroanatomic tracing techniques have allowed investigators to chart the central nervous system (CNS) pathways that provide the links between the stimuli and responses involved in conditioning (Weinberger & Diamond, 1987; Kapp et al., 1990; Thompson & Steinmetz, 1992; Davis, 1992; McCabe et al, 1992; Powell, 1994; Ledoux, 1995).