Abstract
Throughout the history of research on animal learning, it has been widely acknowledged that the temporal relationship between events exerts a critical influence on the acquisition of a conditioned response. Until more recently, however, no explicit and systematic studies had investigated how animals learn the temporal relationship itself. In this article, we first review some basic functions of temporal information in classical conditioning. We then focus on one influential learning theory, temporal coding hypothesis, which posits that animals can automatically encode temporal relationships between events and express learned behaviour through integration of multiple temporal relationships acquired across contexts. After reviewing basic results supporting the temporal coding hypothesis, we present an alternative explanation of some temporal coding-like phenomena on the basis of AESOP model combined with the potentially different contributions of motivational and sensory US representations in higher-order conditioning. In a second article (Fujimaki, Shimbo, Matsui, Shi, & Kosaki, 2015), we will discuss interval timing in operant conditioning and neural substrates of timing behaviour.
