Systematic Distortions in Urban Cognitive Maps

Aggregate cognitive maps of urban areas differ from cartographic maps for reasons other than differences in the mobility and idiosyncratic experiences of individuals. Systematic distortions in aggregate urban cognitive maps may be caused by the cognitive processes used to code spatial information into memory or to retrieve it from memory and by the way these processes relate to a particular urban area. A purpose of this study was to determine the extent to which systematic distortions are present in aggregate urban cognitive maps and to investigate the causes of such distortions. Subjects from three neighborhoods were asked to provide estimates of distances and directions between 105 pairs of landmarks. We analyzed differences between these estimates and true distances and directions to determine if the patterns of distortions were significantly different among the three neighborhoods. Differences for the three samples appeared to be related to the scale and orientation of the aggregate cognitive maps. Regressions with aggregate data for the three neighborhoods using cognitive distance as the dependent variable and actual distance as the independent variable indicated a tendency to overestimate shorter distances more than longer distances. Using multidimensional scaling and Euclidean regression, we mapped subjects'cognitive locations for landmarks and the actual locations in the same space. The aggregate cognitive maps appeared to be rotated to align major transportation axes with canonical direction axes. We argue that systematic distortions are related to a rotation heuristic and to key reference points used by the subjects to code and access spatial information.