How do agent types represent human behaviour in land use change modelling?

Land use change models are frequently used as decision support systems, tools to explore possible future developments of a city, region or country. Given that spatial developments are to a large extent driven by human action and interaction, agent based approaches are a logical choice for simulating these processes, as agents are a natural representation of actors that make decisions. Over the past decade, several prototypes and more advanced simulation models have been reported that claim to include agents, or at least include agent behaviour. When analysing these different models, it quickly becomes clear that the definition of agent based in land use change modelling is rather broad and varies in the way decision-making bodies are perceived and represented in the model.