Abstract
We recently reported that human participants rely on ordinal rather than metric spatial knowledge when navigating to known locations in a virtual hedge maze (Harrison et al, VSS 2001). Ordinal structure refers to the sequential relationship among junctions and paths in the maze, whereas metric structure refers to Euclidean distances and angles between them. In other experiments, we found that participants depend heavily on visual landmarks rather than spatial knowledge acquired from path integration when taking a shortcut between two known locations (Foo et al, JEP: LMC, in press; VSS 2004). Here we compare the contributions of ordinal structure and visual landmarks when navigating in the maze. Participants actively walk in an immersive virtual environment (12m × 12m) with a head-mounted display (60 deg H × 40 deg V) and a sonic/inertial tracking system (latency 50 ms). In the learning phase, participants freely explore a hedge maze with visual landmarks at some junctions, in order to learn 10 places. In the testing phase, they are instructed to walk to these places from a home location. On control trials, the environment is the same as during learning. On probe trials, the environment is changed in one of three ways. In the first condition, the maze layout remains constant while the landmarks are shifted by one junction; thus, metric and ordinal structure are preserved but landmark positions are changed. In the second condition, the maze is stretched along its major axis, preserving its ordinal structure, while landmark positions remain constant. In the third condition, a radial arm section of the maze is rotated by 45 deg, while landmark positions remain constant. If participants follow the landmarks, they turn onto incorrect paths in the probe trials. If they rely on ordinal knowledge of the maze, they ignore landmarks and take the correct paths. The results allow us to compare the relative contributions of ordinal spatial knowledge and visual landmarks in active navigation.
Acknowledgment: Funded by NSF BCS-0214383