In cognitive science, mental representations of spatial knowledge are metaphorically referred to as cognitive maps. However, investigations in cognitive psychology reveal that the cognitive map metaphor is inadequate and that more suitable conceptions of human spatial knowledge processing are needed.

This book addresses mental processing of knowledge about geographic space from an AI point of view by presenting an experimental computational modeling approach. Results about human memory and visual mental imagery from cognitive psychology are combined with AI techniques of spatial and diagrammatic knowledge processing. The author develops the diagrammatic reasoning architecture MIRAGE as a comprehensive conception of human geographic knowledge processing.

Contents

• Mental Processing of Geographic Knowledge
• Theses and Assumptions
• Research Questions and Goals
• Approach: Experimental Computational Modeling

• Spatial Knowledge Conceptions: Cognitive Maps and Other Metaphors
• Human Memory
• Mental Imagery
• Spatial Reasoning
• Diagrammatic Reasoning
• Summary

• Characteristics of the Model
• MIRAGE - Outline of the Model
• Types of Entities and Spatial Relations in MIRAGE
• Subsystems, Structures, and Processes

• A More Demanding Scenario
• Diagrammatic Representations of Lean Knowledge
• Consequences for Image Construction
• Image Revision Strategies in MIRAGE
• Summary

• Computational Tools for Modeling: SIMSIS
• Realization of the Model
• Operation and Behavior of MIRAGE

• Summary
• Results and Discussion
• Future Work