作者: Brian Dupuis , Michael R. W. Dawson
DOI: 10.3758/S13420-013-0111-0
关键词:
摘要: Spatial learning and navigation have frequently been investigated using a reorientation task paradigm (Cheng, Cognition, 23(2), 149-78, 1986). However, implementing this typically involves making tacit assumptions about the nature of spatial information. This has important theoretical consequences: Theories focus on angles at corners as geometric cues ignore information present noncorner locations. We neural network model that challenges these use to generate predictions in novel variant task. test against human behavior virtual environment. Networks humans alike exhibit even when goal locations are not corners. Our simulated our experimental results suggest processed manner more similar features, acting focal point for reorientation, mechanisms governing may be inhibitory rather than excitatory.