Purchase this article with an account.
George S. W. Chan, Patrick Byrne, Suzanna Becker, Hong-Jin Sun; Differential encoding of environmental features in spatial representation. Journal of Vision 2006;6(6):457. doi: 10.1167/6.6.457.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Originally demonstrated by Wang and Spelke (2000), it has since been shown that different features of an environment may be encoded either allocentrically or egocentrically. Following this line of research, we studied subject's directional judgments of environmental features (objects and corners) from an imagined viewpoint that was either aligned or misaligned with the originally learned viewpoint. Subjects were first brought to a fixed learning position in a four-sided, irregularly shaped room and learned the locations of four corners and four different objects relative to two testing viewpoints (aligned or misaligned). They were then blindfolded, brought out of the room, and required to point in the directions of the corners and objects while imagining themselves at one of the two testing viewpoints. Three experiments were conducted: different objects placed in the middle of the room (Exp 1); different objects placed against the wall (Exp 2); and identical objects placed against the wall (Exp 3). The results showed that absolute error for both corners and objects and configurational error for corners (Exp 1, 2, and 3) and objects with the same identity (Exp 3) was higher from the misaligned viewpoint compared to the aligned viewpoint. However, the configurational error for objects with different identities was similar between viewpoints (Exp 1 and 2). The fact that configuration errors were different relative to different viewpoint under particular circumstances (e.g., environmental features) argues against a definitive allocentric representation. Thus, disregarding important variables such as viewpoint can potentially misrepresent the true nature of spatial representations.
This PDF is available to Subscribers Only