A neural network model for the evolution of reconstructive social learning

Learning from others is an important adaptation. However, the evolution of social learning and its role in the spread of socially transmitted information are not well understood. Few models of social learning account for the fact that socially transmitted information must be reconstructed by the learner, based on the learner's previous knowledge and cognition. To represent the reconstructive nature of social learning, we present a modelling framework that incorporates the evolution of a neural network and a simple yet biologically realistic learning mechanism. The framework encompasses various forms of individual and social learning and allows the investigation of their interplay. Individual-based simulations reveal that an effective neural network structure rapidly evolves, leading to adaptive inborn behaviour in static environments, pure individual learning in highly variable environments, and a combination of individual and social learning in environments of intermediate stability. However, the evolutionary outcome depends strongly on the type of social learning (social guidance versus social instruction) and the order of individual and social learning. Moreover, the evolutionary dynamics of social learning can be surprisingly complex, with replicate simulations converging to alternative outcomes. We discuss the relevance of our modelling framework for cultural evolution and suggest future avenues of research.

On the Evolution of Behaviors through Embodied Imitation

This article describes research in which embodied imitation and behavioral adaptation are investigated in collective robotics. We model social learning in artificial agents with real robots. The robots are able to observe and learn each others' movement patterns using their on-board sensors only, so that imitation is embodied. We show that the variations that arise from embodiment allow certain behaviors that are better adapted to the process of imitation to emerge and evolve during multiple cycles of imitation. As these behaviors are more robust to uncertainties in the real robots' sensors and actuators, they can be learned by other members of the collective with higher fidelity. Three different types of learned-behavior memory have been experimentally tested to investigate the effect of memory capacity on the evolution of movement patterns, and results show that as the movement patterns evolve through multiple cycles of imitation, selection, and variation, the robots are able to, in a sense, agree on the structure of the behaviors that are imitated.