A theoretical framework for the study of spatial cognition

The ultimate goal of this research is to contribute to the understanding of spatial cognition and its consequences on locomotion (that is, on the organization of the whole organism's movements in space). An organism's locomotion depends on what structure it superimposes on space itself and can therefore be understood as a form of interaction with a subjective environment, understandable in turn in terms of the organism's cognitive architecture. As a first step, we will propose here a large-scale classification of the cognitive architectures possible, outlining the subjective structure that each of them superimposes on space and the relevant consequences on locomotion. Our classification differs from others that have been proposed for spatial cognition (e.g., Papi 1990) in that it does not build on the idea of spatial behaviors. We will argue, first, that cognition is better understood as interaction rather than behavior; and, second, that an organism's interactions can only be understood as generated and controlled by its cognitive architecture. This corresponds, in a sense, to the adoption of the organism's (rather than the observer's) subjective point of view. The main division we will draw is between those architectures whose internal dynamics are entirely coupled to the dynamics in the world, and those that have at least some capability of decoupling. The latter correspond to representational architectures. Each class will be further decomposed according to a criterion of complexity. Our view of representation will not build upon a computationalist account of cognition.