Abstract

When normally sighted people observe a natural scene, their perceptions include seeing the supporting ground surface and the arrangements of the objects on that surface: where each object is in relation to the others, to the ground surface, and to the observer. This is defined as the perceived layout of a scene. Most current perceptual theories do not consider perceived layout, preferring to focus on the much narrower concern of perceived radial distance of objects from the observer, ignoring concern for the relationships among the objects of the scene, and among the observer in relation to the objects. In contrast, current cognitive theories have focused on cognitive maps as a spatial representation of scenes previously seen, maps which include explicit representation of spatial layout. However, cognitive theories have paid little theoretical attention to the spatial relationships in scenes currently on view, so they can tell us little about perceived layout, only remembered layout. Perceived layout has been missing from perceptual theories, and needs to be studied in its own right. In addition, to understand spatial cognition, cognitive maps, or any kind of representation of knowledge about the layout of space, we must study the properties of perceived layout—the perception of the arrangements of objects in a scene currently on view. To provide an example of the assessment of perceived layout, experimental data are presented on the psychophysical measurement of the perceived layout of a natural scene containing 13 objects arranged on a large flat lawn. Subjects are asked to draw a map of the scene, to estimate the absolute interobjects distances between every pair of objects in feet, and to compare the relative magnitudes of every possible triplet of interobject distances. The matrices of the interobject distances derived from each of these three measures produce highly reliable and consistent 2-dimensional constructions of the scene through a multi-dimensional scaling analysis. These constructions resemble the physical layout of the actual scene quite closely. However, subjects tend to underestimate the distances between objects when the direction of the distances are parallel to their line of sight, compared to directions perpendicular to their line of sight. Not only does this tend to produce perceived layouts of the scene that are slightly more ellipical than in actuality, but more important, it means that subjects' perceived layouts change when they move to new viewing positions. These analyses and others are used to examine the usefulness and validity of these measures as indices of perceived layout. Following the psychophysical descriptions of the measures, as theoretical analysis of some of the properties of the perceived layout of space is provided, along with a consideration of some of the variables that are expected to affect these properties. The paper concludes with a return to the theoretical issues concerning the contents of a proper theory of space perception.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call