Spatial mental models from descriptions

Abstract

Spatial language is widely used, both literally to describe space and figuratively to express a broad range of ideas. This article reviews two projects studying the nature of mental representations of space induced entirely by language. The first project investigates perspective in descriptions of large-scale (e.g., convention center, town) space. People typically describe environments using a route or survey perspective, or a mixture of both. Route perspectives take a view from within the environment and describe the locations of landmarks with respect to a moving observer in terms of the observer's left, right, front, and back. Survey perspectives take a view from above the environment and describe locations of landmarks with respect to each other in terms of north, south, east, and west. Features of the environment, such as having a single or multiple path, affect choice of perspective. In comprehension, readers seem to form the same perspective-free mental representation irrespective of description perspective. They respond with equal speed and accuracy to inference questions from either perspective regardless of read perspective. The second project investigates mental representations of the objects located immediately around the body. Readers seem to form mental spatial frameworks, extensions of the three body axes, associating objects to the frameworks. The accessibility of the three axes depends on characteristics of the body, characteristics of the perceptual world, and posture of the observer. For example, for an upright observer, times to access objects along the head/feet axis are fastest because it is an asymmetric axis of the body and is correlated with the only asymmetric axis of the world, that created by gravity. Times to access objects along the front/back axis are next fastest as it is an asymmetric axis of the body, and times to the left/right axis are slowest as it is an axis with few asymmetries. Evidence for the spatial framework hypothesis was obtained in a variety of situations, varying posture, perspective, number of observers, and cause of reorientations. © 1994 John Wiley & Sons, Inc.