Spatial cognition: systematic distortions in cognitive maps.

Stevens and Coupe demonstrated that people tend to regularize their cognitive maps by distorting the position of relatively small features (e.g., cities) to make them conform with the position of larger features (e.g., state boundaries). The present studies replicated those of Stevens and Coupe by asking college students to study prototypical maps with three kinds of boundaries: none, straight, and curved. They extended Stevens and Coupe's work by asking students to (a) physically reproduce the maps rather than simply answer questions about them, (b) recall the information not only immediately after study, but also one hour later and again two days later, and (c) reproduce the state boundaries as well as the city locations. Students who studied the curved boundary maps made more distortion errors than those who studied the straight boundary and the no-boundary maps. In addition, students' distortion errors were greatest when recall was delayed, and distortion errors in the recall of state boundaries were correlated with their distortion errors in the recall of city locations. In general, the findings provide additional evidence for the notion that memory for map information is schematic.