Graph-based models of space in architecture and cognitive science - a comparative analysis∗

Graph-based operationalizations of space are used in architecture as well as in cognitive science. In such models, environments are usually described by means of nodes and edges, roughly corresponding to places and their spatial relations. In the field of cognitive science, view and place graphs are models of mental representations of environments and used for the explanation of wayfinding behavior such as exploration and route planning. In architecture, space syntax and visibility graph analysis aim at identifying and describing structural properties of built environments that determine their usage and experience. In cognitive science, mental representations of space cannot be seen independently from the formal and configurational properties of the corresponding environments that are well captured by architectural description systems. Vice versa, formal descriptions of space as used in architecture gain plausibility and relevance by incorporating results from cognitive research that allow the prediction and explanation of actual human behavior. In this paper approaches from the two different disciplines are therefore reviewed and compared. Special interest concerns their scope, structure, and representational content. Parallels, differences, and specific strengths are discussed. Furthermore, based on recent empiric work, strategies to integrate aspects from both disciplines are outlined.