The dynamics of the earth and its inhabitants have become a core topic and focus in research in the spatial sciences. The spatio-temporal data avalanche challenges researchers to provide efficient and effective means to process spatio-temporal data. It is of vital importance to develop mechanisms that allow for the transition of data not only into information but also into knowledge. Knowledge representation techniques from artificial intelligence play an important role in laying the foundations for theories dealing with spatio-temporal data. Specifically, the advances in the area of qualitative spatial representation and reasoning (QSTR) have led to promising results. Categorical distinctions of spatio-temporal information identified by QSTR calculi potentially correspond to those relevant to humans. This article presents the first behavioral evaluation of qualitative calculi modeling geographic events associated with scaling deformations of entities, that is, changes in size by either expansion or contraction. Examples of such dynamics include a lake flooding its surroundings or an expanding oil spill in the ocean. We compare four experiments using four different semantic domains. Each domain consists of two spatially extended entities: one entity is undergoing scaling deformations while the other is static. We kept the formal QSTR characterization, which are paths through a topologically defined conceptual neighborhood graph, identical across all semantic domains. Our results show that for geographic events associated with scaling deformations (a) topological relations are not equally salient cognitively; (b) domain semantics has an influence on the conceptual salience of topological relations.
All Science Journal Classification (ASJC) codes
- Civil and Structural Engineering
- Geography, Planning and Development
- Management of Technology and Innovation