The theory outlined in this chapter is what has sometimes been called the ‘radical’ form of the simulation theory. It conceives simulation as more than just a heuristic for finding out the mental states of others and predicting their behaviour quickly and economically, using a minimum of specially dedicated brain power. Beyond this, the theory holds that even our ability to grasp the concepts of mind and the various mental states depends on our having the capacity to simulate others.

Before discussing the radical position, I present a broad characterisation of the simulation theory in general and a new argument that clearly favours it over competitors.

Hot and cold methodologies contrasted

Concerning the ‘theory of mind’, or more broadly the methodology, by which people anticipate and predict one another's actions, there are basically two kinds of theory. One kind holds that we use what I call a cold methodology: a methodology that chiefly engages our intellectual processes, moving by inference from one set of beliefs to another, and makes no essential use of our own capacities for emotion, motivation, and practical reasoning. So-called ‘theory’ theories are of this kind, both those that hold we develop or somehow acquire something comparable to a scientific theory and those that say the work is done by a ‘theory’ in an extended sense, a set of rules of symbol manipulation embodied, like a Chomskian universal grammar, in an innate module.