Abstract
Disoriented humans and nonhuman animals reorient by the overall shape of the environment(1). For example, when subjects see an object being hidden in one corner of a rectangular room and then are disoriented, they restrict their search to the two geometrically correct corners (the correct corner and its diagonal). Geometry-based reorientation occurs only with extended surfaces: for example, 3-year-olds reorient by a rectangular array of wall panels but not by a rectangle drawn on the floor(2). In seven experiments, we asked what defines an extended surface by testing the reorientation of 4-year-old children in a circular enclosure containing a rectangular area defined by taped lines on the floor, by surfaces of various heights, or by extended columns, both separate and connected by borders on or off the ground.
Children reoriented by the shape of extended surfaces of all heights, including <1 inch, but not by lines on the floor or unconnected columns. When columns were connected by borders, children's use of geometry was more variable. These findings begin to shed light on the properties of the visual layout that guide navigation. They contribute to the evidence that visual representations of surfaces are crucial for the geometric encoding that underlies reorientation.
1. K. Cheng, N.S. Newcombe. (2005). Is there a geometric module for spatial reorientation? Squaring theory and evidence, Psychonomic Bulletin and Review, 12, 1-23.
2. S. Gouteux, E. S. Spelke. (2001). Children's use of geometry and landmarks to reorient in an open space. Cognition, 81, 119-148.