Understanding cognitive maps from landmark and configurational representations

Abstract

Previous studies on spatial cognition have rarely adopted a behaviorally objective approach for survey representations that may capture the Euclidean qualities of configurations. The current study integrates subjectively conventional coding and a recently developed objective Gardony Map Drawing Analyzer (GMDA) technique to link landmark and survey representations. The combined approach is able to present a more complete story. Participants were instructed to tour a real aquatic environment, which provided more ecologically valid results. Conventional analysis results revealed that the top two most commonly recalled and located landscape categories were lakes and a cascade due to their embedded sociopsychological meanings and physical salience. GMDA results provided quantitative measures regarding configurational outputs, angle accuracy, and distance accuracy. Based on these results, relative contributions of different measures to spatial knowledge were identified using the k-means method. From the two approaches, the results indicated that participants who were better at landmark recall led to a better global perspective of the research site. The study also points out that missing landmarks, which involve participants’ loss of components in response to the target environment map, play critical roles in understanding configurations. In addition, directional information is more important than distance information in constructing spatial representations.