Learning is frequently proposed as a mechanism that enables an individual to adapt to changes in the environment during its lifetime (Levins, 1968; Stephens, 1991). These studies generally take as their point of departure the Òabsolute fixity argument.Ó That is, in an absolutely fixed environment, an individual should develop a genetically fixed pattern of behavior (assuming some cost is associated with learning). On the other hand, in an absolutely unpredictable environment, where the past and present states of the environment offer no information about the future, there is nothing to learn and, assuming some cost to computation, there is again no driving force for a learning capability to evolve. Stephens (1991) proposed an approach to the evolution of learning that takes into account the individualÕs life history. Stephens argues that the pattern of predictability in relation to an individualÕs life history could determine the evolutionary advantage of learning. He concludes that the evolutionary advantage of learning is in dealing with sequences of environmental events that change between generations and are regular within a generation. Another approach is to view learning as the ability of an individual to construct a representation of its environment, and, by proper use of the representation, to predict future states of its environment. This requires some regularity in the environmental signals and the individualÕs capacity to capture this regularity.
