The Representational Foundations of Updating Object Locations

Abstract

The goal of the experiments presented in this paper was to contribute to an ongoing debate in human spatial cognition research concerning the relative importance of dynamic egocentric and enduring allocentric representations for updating the locations of multiple objects. Several studies have demonstrated an increase in configuration error, which is a measure of the quality of the angular configurational knowledge of object locations, after disorientation. Based on the assumption that the fidelity of allocentric representations could not be affected by disorientation, those studies concluded that the observed increase in configuration error was evidence for the usage of dynamic egocentric representations in spatial updating. The experiments discussed in this paper challenge this conclusion and present evidence that supports the hypothesis that allocentric representations are the primary foundation of spatial updating. Two pilot experiments and Experiment 1 showed that performance on a tested heading was not determined by the amount of rotation needed to reach that heading, as predicted by the hypothesis that spatial updating is based on dynamic egocentric representations. Instead, performance was determined by the relationship between the tested heading and the walls of the surrounding room, as predicted by the hypothesis that spatial updating is based on allocentric representations that are specified with respect to reference directions that are intrinsic to the represented environment. Experiment 2 attempted to identify temporal and capacity limitations of egocentric updating that would have explained why no evidence for the usage of dynamic egocentric representations was observed in the earlier experiments. However, even when tracking a single object over the course of only a few seconds, no evidence in support of dynamic egocentric representations was apparent. The final experiment ruled out the possibility that the results of the previous experiments were caused by differences in the disparity between tested headings and the learning orientation.