Cockroaches, drunkards, and climbers: Modeling the evolution of simple movement strategies using digital organisms

摘要

Even the simplest of organisms may exhibit low-level intelligent behaviors in their directed movements, such as in foraging. We used the Avida digital evolution research platform to explore the evolution of movement strategies in a model environment with a single local resource that diffuses to produce a gradient, which organisms have the ability to follow. Three common strategies that evolved, Cockroach, Drunkard, and Climber, exhibit how both environmental constraints and historical contingency play a role in the emergence of intelligent behaviors. The evolved programs are also suitable for use in controllers on robots.