The Viability of Web-based Eye Tracking

Abstract

The recent COVID-19 pandemic has forced vision scientists to develop innovative, online solutions to their research questions. After the introduction of new open-source software and improvements in personal laptop hardware, researchers have been able to recreate increasingly sophisticated measures online historically only measured in-person. In this paper, we explore the boundaries of online research to describe the development and testing of a new web-based eye tracking system, “Gazer”. Gazer is an accessible eye tracking system for vision scientists that harnesses the open-source Webgazer to record screen-based gaze locations, using the cameras present in participant’s personal laptops. We directly compare Gazer to the established Eyelink 1000 in two separate experiments, and determine it has comparable temporal and spatial precision for recording fast exogenous eye movements, and endogenous movements. In a final experiment, we record the development of gaze strategies in a repeated remote visual search task to indicate Gazer’s potential to address sophisticated attentional research questions. Overall, we present Gazer as a viable method for researchers to conduct gaze-based research online.