Gaze Graphs in spatial navigation: visual behavior during free exploration explains individual differences in navigation performance

Vision is an important sense for spatial navigation, yet little is known about the role of visual behavior during spatial knowledge acquisition and its link to spatial navigation performance. In this study, participants freely explored the immersive virtual reality city of Westbrook while eye-tracking data were collected. Subsequently, participants spatial knowledge was tested in an immersive pointing-to-building task. To address the freedom of movement during the spatial exploration, we transformed the eye-tracking data into gaze graphs and applied a graph-theoretical analysis. We assessed gaze graph architecture and identified gaze-graph-defined landmarks, finding that their properties were consistent with key results from our earlier work. Furthermore, we investigated which factors contribute to the observed individual differences in spatial navigation performance. We found that global gaze graph measures, specifically gaze graph diameter, could explain 40% of the observed individual differences in participants mean task performance. In comparison, a model based on participants responses to a self-report questionnaire was not predictive of such differences. Consequently, we suggest that visual behavior, as reflected in global gaze graph measures, is a strong predictor of individual performance differences. Our results not only contribute to a better understanding of landmarks and individual differences in spatial navigation but also outline the potential of vision and gaze-graph-based spatial navigation research.