IntroductionIn the literature on cognitive psychology, the issues raised by Gibson's ecological theory of perception (Gibson, 1979) have been examined and taken on board. Gibson's theory was formulated primarily in order to overturn theories laden with subjective and objective knowledge, and to replace them with a model in which the agent and its environment are conjoined by a set of affordances so the agent perceives the contents of the environment directly and uses the affordances within it to guide its action without reference to superior representational models. Today, Gibson's work has been contextualised and broken into further models in which recognition and representation do play a part (Neisser, 1994). However, in the domain of agent-based modelling we still appear to ignore the original concerns he voiced öin particular`the model' is always preceded by a theoretical framework, rather than simply being a perceptual model in its own right (see, for example, Casti, 1998;Epstein and Axtell, 1996). This paper is not an attempt to overturn the body of literature which already exists in agentbased modelling, although it does constitute a plea for the use of direct perception where the approach is available, and to try to regard the environment as the provider of possibilities rather than as a place to be rationalised. As an example, consider human movement around an art gallery. There might be any number of causal factors for the routes people take. People might, for example, follow a map, or signage, take into account the direction other people are taking, a glimpse of a familiar painting, reject a route on the grounds of personal prejudice against a style, and so on. On the other hand, the possibility of exploring the walkable surface of the layout ahead (the rooms of the gallery) may simply be enough for a human to do exactly that. Abstract. Gibson's ecological theory of perception has received considerable attention within psychology literature, as well as in computer vision and robotics. However, few have applied Gibson's approach to agent-based models of human movement, because the ecological theory requires that individuals have a vision-based mental model of the world, and for large numbers of agents this becomes extremely expensive computationally. Thus, within current pedestrian models, path evaluation is based on calibration from observed data or on sophisticated but deterministic route-choice mechanisms;there is little open-ended behavioural modelling of human-movement patterns. One solution which allows individuals rapid concurrent access to the visual information within an environment is an`exosomatic visual architecture', where the connections between mutually visible locations within a configuration are prestored in a lookup table. Here we demonstrate that, with the aid of an exosomatic visual architecture, it is possible to develop behavioural models in which movement rules originating from Gibson's principle of affordance are utilised. We apply large numbers of agents programmed with these r...