Using gaze data in evaluating interactive visualizations

Abstract

Evaluations have long been missing or imperfect in a publication presenting a new visualization technique, but proper evaluations are now becoming a standard. There are many reasons for the reluctance of evaluating visualization techniques, including the complexity of the task and the amount of work required. We propose a simple evaluation approach that consists of a set of tasks carried out in an experimental setting coupled with eye tracking to approximate the focus of the user's attention. In addition, we discuss three methods to visualize the gaze data to gain insight into the user's attention distribution, and show examples from a study where a parallel coordinate browser was evaluated. © 2011 Springer-Verlag.