An evolutionary ecological approach to the study of learning behaviour using a robot based model

We are interested in the construction of ecological models of the evolution of learning behaviour using methodological tools developed in the field of evolutionary robotics. In this paper, we explore the applicability of integrated (i.e., non-modular) neural networks with fixed connection weights and simple "leaky-integrator" neurons as controllers for autonomous learning robots. In contrast to Yamauchi and Beer (1994a), we show that such a control system is capable of integrating reactive and learned behaviour without needing explicitly hand-designed modules, dedicated to particular behaviour, or an externally introduced reinforcement signal. In our model, evolutionary and ecological contingencies structure the controller and the behavioural responses of the robot. This allows us to concentrate on examining the conditions under which learning behaviour evolves.