Abstract
Research on spatial perspective taking has suggested that including an agent in the display benefits performance. However, little research has examined the mechanisms underlying this benefit. Here, we examine how an agent benefits performance by examining its effects on three mental steps in a perspective-taking task: (1) imagining oneself at a location (station point) within in the array, (2) adopting a different perspective (heading), and (3) pointing to an object from that perspective. We also examine whether a non-agentive directional cue (an arrow) is sufficient to improve performance in an abstract map-like display. We compared a non-directional cue to two cues for position and orientation: a human figure (agentive, directional) and an arrow (non-agentive, directional). To examine the effects of cues on steps 2 and 3 of the perspective-taking process, magnitude of the initial perspective shift and pointing direction were varied across trials. Response time and error increased with the magnitude of the imagined perspective shift and pointing to the front was more accurate than pointing to the side, or back, but these effects were independent of directional cue. A directional cue alone was sufficient to improve performance relative to control, and agency did not provide additional benefit. The results overall indicate that most people adopt an embodied cognition strategy to perform this task and directional cues facilitate the first step of the perspective-taking process, imagining oneself at a location within in the array.