Peter Todd

Minding the environment in simulations of cognition

Research in artificial life and the simulation of adaptive behavior constantly deals with the importance of the environment, for shaping an agent's moment-to-moment behavior and sculpting the evolution of the agent's behavioral programs over time. But psychology and cognitive science have too often lost sight of the outside world, instead focusing on the nature of an idealized disembodied rationality and how far humans and other creatures fall short of that ideal. Artificial-life-inspired models are now coming to the rescue of cognitive research by demonstrating how effective behavior can be generated via simple information-processing mechanisms that take advantage of the structure of the environment, rather than ignoring it. By letting the world do some of the work in producing adaptive behavior, agents can be ecologically, isntead of logically, rational. In this talk, I will show how some of our recent agent-based modeling of search behavior in the domains of mating, eating, and even parking have provided evidence for understanding how humans and other agents can be ecologically rational by taking into account the structured environments they inhabit.