Egocentric cues influence the spatial memory of landmark configurations for memory-guided actions

Abstract

Allocentric and egocentric reference frames are used to determine the spatial position of action targets in reference to objects in the environment, i.e., landmarks (allocentric), or the observer (egocentric). Previous research investigated reference frames in isolation, for example, by shifting landmarks relative to the target and asking participants to reach to the remembered target location. Systematic reaching errors were found in the direction of the landmark shift and used as a proxy for allocentric spatial coding. Here we examined the interaction of both allocentric and egocentric reference frames by shifting the landmarks as well as the observer. We asked participants to encode a 3D configuration of balls, and to reproduce this configuration from memory after a short delay followed by a landmark or observer shift. We also manipulated the number of landmarks to test its effect on the use of allocentric and egocentric reference frames. Shifting the observer resulted in larger configurational errors. In addition, an increase in the number of landmarks led to a stronger reliance on allocentric cues and a weaker contribution of egocentric cues. In sum, our results highlight the important role of egocentric cues for allocentric spatial coding in the context of memory-guided actions.