Individual Differences in Spatial Knowledge Acquisition from a Virtual Environment: An Eye-Tracking Study

Previous studies have shown that spatial knowledge acquisition differs across individuals in both real and virtual environments. For example, in a real environment, Ishikawa & Montello (2006) showed that some participants had almost perfect configural knowledge of the environment after one or two learning trials, whereas others performed at chance even after repeated learning trials. Using a virtual version of Ishikawa & Montello's layouts, we measured eye movements while participants were learning a layout of a route, as eye movements are shown to be closely linked to performance in spatial navigation tasks. We prepared three different layouts of a route depicted in a desktop virtual environment, along with the locations of four landmarks on that route. After learning each of the routes, we administered three different measures of spatial knowledge: numbering the landmark order, estimation of direction, and map sketching. Self-reported sense-of-direction (SDQ-S) was also measured. Behavioral analyses showed positive correlations across the routes in the estimation of direction. However, consistent correlations were not observed between eye movements and performance of the estimation of direction in each route. Those results suggest that eye movements do not predict individual differences in spatial knowledge acquisition