An investigation of the mechanisms supporting navigation using virtual environments

2012-05-02How do people navigate in an enclosed environment? Previous research suggests either an image-matching system that stores environmental information in 2D snapshots or a domain-specific geometric system that records surface layout in encapsulated representations. These mechanisms have largely been studied, across species and human development, in full scale settings where precise control of environmental information is not possible. The current study seeks to establish a computer-based testing methodology utilizing virtual environments in which all visual information can be controlled and systematically varied in order to determine the precise contents of the mental representations used to facilitate navigation. Four experiments investigate whether people can form mental representations of virtual environments and whether the characteristics of these representations support the predictions of an image-matching system or a geometric module. Results indicate that surface layout geometry is encoded, but the possibility of a snapshot-based system functioning simultaneously is not entirely ruled out