User evaluations of interactive and dynamic applications face various challenges related to the active nature of these displays. For example, users can often zoom and pan on digital products, and these interactions cause changes in the extent and/or level of detail of the stimulus. Therefore, in eye tracking studies, when a user's gaze is at a particular screen position (gaze position) over a period of time, the information contained in this particular position may have changed. Such digital activities are commonplace in modern life, yet it has been difficult to automatically compare the changing information at the viewed position, especially across many participants. Existing solutions typically involve tedious and time-consuming manual work. In this article, we propose a methodology that can overcome this problem. By combining eye tracking with user logging (mouse and keyboard actions) with cartographic products, we are able to accurately reference screen coordinates to geographic coordinates. This referencing approach allows researchers to know which geographic object (location or attribute) corresponds to the gaze coordinates at all times. We tested the proposed approach through two case studies, and discuss the advantages and disadvantages of the applied methodology. Furthermore, the applicability of the proposed approach is discussed with respect to other fields of research that use eye tracking-namely, marketing, sports and movement sciences, and experimental psychology. From these case studies and discussions, we conclude that combining eye tracking and user-logging data is an essential step forward in efficiently studying user behavior with interactive and static stimuli in multiple research fields.
Keywords Eye tracking . Interactivity . GeoreferencingEye tracking has proven to be a helpful technique in user research, especially when a visual element needs to be evaluated. By using eye tracking data, researchers can discover how long and how often a user looks at a particular area of interest, as well as the length and speed of the eye movements (Duchowski, 2007;Holmqvist et al., 2011). The position of the gaze (also termed the point of regard, or POR) is typically expressed using screen coordinates in pixels. From these basic screen coordinate measurements, various gaze metrics are derived in relation to what is displayed, such as the fixation duration (how long), fixation count (how often), and various scan-path characteristics (e.g., the length and speed of eye movements). The technique has been applied in a multitude of research fields, including software engineering, industrial engineering (e.g