Academic journal article Perception and Psychophysics

Two Kinds of Visual Perspective Taking

Academic journal article Perception and Psychophysics

Two Kinds of Visual Perspective Taking

Article excerpt

Visual perspective taking can be used to determine where objects are located relative to another agent, or whether the agent can see a particular object. Four experiments indicated that different processes provide these different kinds of information. When participants were asked to report whether an object was to the left or to the right of another agent, response times (RTs) increased with increasing angular distance between the participant and the agent, suggesting that participants mentally transformed their perspective to align it with that of the agent. For visibility judgments, RTs were independent of the angle between the participant and the agent but increased with the distance between the agent and the object, suggesting that participants traced the agent's line of sight. Together, these data suggest that perspective taking encompasses at least two qualitatively different computational processes: one that updates the viewer's imagined perspective, and one that traces a line of sight.

Visual perspective taking (VPT) is the ability to predict the visual experience of another agent. This ability is valuable in contexts as diverse as avoiding predators, reasoning about what others know, navigation, and spatial problem solving. Depending on the situation and goals at hand, VPT allows one to predict qualitatively different kinds of information. In particular, one can predict (1) whether another person can see an object at all and (2) where objects are located relative to another person's egocentric reference frame. One possibility is that a general process of visual perspective taking is involved in both situations. However, an alternative hypothesis is that different operations are performed to calculate these two different types of information.

Studies of visual perspective taking in children suggest that more than one type of knowledge is necessary to achieve successful imaginary perspective changes. Specifically, a distinction has been made between knowledge about which objects are visible from another viewpoint (Level 1 knowledge) and knowledge about the visual aspects of a scene relative to an imagined viewpoint (Level 2 knowledge-Flavell, Everett, Croft, & Flavell, 1981; Salatas & Flavell, 1976). Level 1 knowledge is reflected by performance in hiding or occlusion tasks in which the child is asked either to position an object so that another person cannot see it or to decide whether or not another person can see a target object. Level 2 knowledge is usually tested using tasks in which children are asked to predict how an object or scene would look from another position. These two types of information are clearly different, but both require representing the fact that the other person has a perspective and calculating information about the difference between that person's perspective and one's own (Salatas & Flavell, 1976; Yaniv & Shatz, 1990). Level 2 knowledge typically appears later in development than Level 1 knowledge. However, this developmental progression does not reveal what processes support these two kinds of VPT.

Research on spatial transformations in adults provides one possibility. In these studies, paradigms have been used in which participants predict what a scene would look like if they were at a specified position (possibly different from their actual position). Typically, response times (RTs) are longer if the to-be-imagined position is misaligned with the participant's actual position (Amorim & Stucchi, 1997; Creem, Downs, Wraga, Proffitt, & Downs, 2001; Presson, 1982; Presson & Montello, 1994; Simons & Wang, 1998; Wang & Simons, 1999; Wraga, Creem, & Proffitt, 2000; Zacks, Vettel, & Michelon, 2003). This has been interpreted as reflecting the use of analogue perspective transformations, which update the location and/or orientation of one's egocentric perspective. The few studies of adult cognition that have been conducted to directly examine the degree to which participants take the perspective of another person suggest that perspective transformations are also used in this situation (Amorim, 2003). …

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.