The Spatial Nature of Mental Models

The idea that mental representations are “small-scaled models of external reality” can be traced back to Craik (1943). This means, that a mental representation shares its structure with what it represents. Therefore, it is possible to model reality in thoughts and to learn about it without affecting it. It enables us to reason, to know what will or would be the case if this and that were to happen. Johnson-Laird (1983) developed a theory of mental models that can explain a wide variety of phenomena in reasoning. The power of mental models is grounded in the way the world is represented. It is not the logical structure (such as in propositions) or some artificial constructions (such as circles standing for sets) that are represented, but rather single objects taking part in a situation and the relations among them. Of course, this very basic idea must be extended, for sets must be representable as such, for example. However, the advantage of this account is that the world is represented in a simple and natural way. Johnson-Laird shows that reasoning with mental models leads to logically valid conclusions when no limit of capacity is assumed.