Combining Thinking Aloud and Eye Tracking

Abstract

Thinking Aloud and eye tracking are common in evaluations of human-computer interaction (HCI), but rarely applied simultaneously – presumably because speaking interferes with eye tracking. We tested to what extent this interference can be overcome with a head-mounted eye-tracking device that enables movements. Participants (N=12) read a text silently, read it aloud, and verbalized their thoughts, while their gaze was tracked. As usual, eye-tracking accuracy slightly decreased with time since calibration, but this was not affected by the act of speaking. Accordingly, we explored the combination of Thinking Aloud and eye tracking in a browser-based software usability test. In further participants (N=6), we found converging evidence for alternative software usage strategies in eye tracking and Thinking Aloud, while each method provided additional insights that were inaccessible to the other. We conclude that combining eye tracking and Thinking Aloud is feasible without compromising data quality and meaningful for comprehensive HCI evaluation.