Abstract
Spatial predictions of uncertain trajectories are challenging, but are often associated with overconfidence. This study explored how a visualization influenced prediction of uncertain spatial trajectories (e.g., unknown path of a downed aircraft or future path of a hurricane). Mean and variance estimates were compared for participants provided with a gradient-shaded “cone of uncertainty” visualization and those who were not provided with a visualization. Participants exhibited less error in mean estimations when a visualization was present, but performed worse than controls once the visualization was removed. For variance estimations, participants provided with a visualization did not retain any advantage in their estimations once the visualization was removed. Combined these findings suggest that visualizations may support some aspects of spatial predictions under uncertainty, but they can be associated with costs for the underlying knowledge being developed.
Get full access to this article
View all access options for this article.
