Academic journal article Perception and Psychophysics

Perceiving Depth in Point-Light Actions

Academic journal article Perception and Psychophysics

Perceiving Depth in Point-Light Actions

Article excerpt

The present study investigates how observers assign depth in point-light figures, by manipulating spatiotemporal characteristics of the stimuli. Previous research on the perception of point-light walkers revealed bistability (i.e., that a point-light walker is perceived as either facing the viewer or facing away from the viewer) and the presence of a perceptual bias (i.e., a tendency to perceive the figure as facing the viewer). Here, we study the generality of these phenomena by having observers indicate the global depth orientation of different ambiguous point-light actions. Results demonstrate bistability for all actions, but the presence of a preferred interpretation depends strongly on the performed action, showing that the process of depth assignment takes into account the movements the point-light figure performs. Two additional experiments, using unfamiliar movement patterns without strong semantic correlates, show that purely kinematic aspects of an action also strongly affect depth assignment. Together, the results reveal the perception of depth in point-light figures to be a flexible process involving both bottom-up and top-down components.

Ever since Johansson (1973) introduced the point-light figure in biological-motion research, it has provided one of the most fascinating demonstrations of the perceptual system's sensitivity to actions performed by other animate beings. Observing nothing more than a swarm of bright dots indicating the major joints of a person suffices to convey the impression of that person engaged in a specific action (see, e.g., Ahlstrom, Blake, & Ahlstrom, 1997; Cutting & Kozlowski, 1977; Dittrich, 1993). Although much progress has been made in the study of computational and neural mechanisms underlying the perception of biological motion in general, and of point-light figures in particular (see Giese & Poggio, 2003, for an overview), one issue that has been largely ignored is the perception of depth in point-light stimuli and of the overall orientation of the figure (for two exceptions, see Proffitt, Bertenthal, & Roberts, 1984, and Verfaillie, De Troy, & Van Rensbergen, 1994). Consider a 2-D point-light walker, as in Figure 1: Although the animated version of such a 2-D stimulus could have been the projected image of an infinite number of possible 3-D configurations, it is typically interpreted as the projection of a 3-D human walker. However, due to the unspecified depth order, resulting from the lack of ordinal depth cues, two such "human walker" percepts are in fact possible: one that is oriented toward the viewer and one away from the viewer. An everyday analogue of this phenomenon can sometimes be experienced when one is observing people from a distance that is sufficient so that the people's internal features (such as facial elements or texture of clothing) are rendered uninformative. Although such an impoverished stimulus is typically highly recognizable ("Somebody's walking there... "), the orientation in depth often remains ambiguous (" ... but is he or she approaching or going away?").

Vanrie, Dekeyser, and Verfaillie (2004) investigated this issue and showed that point-light walkers seen from different viewpoints are perceived as 3-D human bodies in particular depth orientations and that a single point-light walker can indeed yield two different percepts, making it perceptually ambiguous or bistable (cf. Attneave, 1971 ). In addition, it was shown that despite the fact that both percepts are cquiplausible, they are clearly not equiprobable: Observers show a strong tendency to interpret a pointlight walker as facing them. The difference in response frequencies for the two veridical interpretations was taken as evidence for the existence of a perceptual bias, and suggested that the mechanism underlying the interpretation of ambiguous point-light walkers is not simply a random process.

The present study asks how observers perceive the depth orientation of point-light figures in general. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.