Human spatial memory

点击翻译按钮获取中文摘要

The general objective of my research program has been to increase our understanding of how people represent spatial relations in memory and how people use spatial memory for orientation and reorientation. I propose three lines of research to investigate (1) how people select spatial reference directions for spatial structures experienced from multiple viewpoints, (2) what is the role of spatial reference direction in orienting to environments during locomotion, and (3) what is the role of environment features and spatial reference direction in reorienting to environment after disorientation. My previous research showed that when people learned a layout of objects from a single viewpoint, they tended to represent objects' locations with respect to an allocentric reference direction inside the layout. I intend to ask whether that will still happen when people sequentially learn the location of each object by walking on several segmented paths with changing viewpoints. Several studies in the literature showed that recognizing novel views in scenes was more accurate when the novel test view was caused by observer locomotion than when it was caused by scene rotation and indicated that people update egocentric spatial representations during locomotion. In the proposed research, I will demonstrate that people update with respect to an allocentric reference direction during locomotion. Previous research showed that when participants learned a layout of irregularly placed objects, their pointing consistency across objects decreased after disorientation. In the proposed research, I will examine what environmental features (e.g., connection, regularity) and reorientation cues (e.g., indication of the facing object or original learning direction) can remove the disorientation effect. The proposed research will elaborate and constrain theoretical models of spatial memory, both my own and those of others. It should also help our understanding the roles of the reference direction, viewpoint change, and locomotion in recognizing scenes, shapes, and objects. Hence, the findings should be informative in forming a universal model of visual processing in 3D environments.