A Study on Text Entry Methods Based on Eye Gestures

Purpose – In this paper we consider the two main existing text input techniques based on “eye gestures” -- namely EyeWrite and Eye-S -- and compare them to each other and to the traditional “virtual keyboard” approach. Design/methodology/approach – The study primarily aims to assess user performance at the very beginning of the learning process. However, a partial longitudinal evaluation is also provided. Two kinds of experiments have been implemented involving 14 testers. Findings – Results show that while the virtual keyboard is faster, EyeWrite and Eye-S are also appreciated and can be viable alternatives (after a proper training period). Practical implications – Writing methods based on eye gestures deserve special attention, as they require less screen space and need limited tracking precision. This study highlights the fact that gesture-based techniques imply a greater initial effort, and require proper training not only to gain knowledge of eye interaction per se, but also for learning the gesture alphabet. We think that our investigation can drive the designers of gaze-controlled writing techniques based on gestures to put more consideration on the intuitiveness of gestures themselves, as they may greatly influence user performance in the first stages of the learning process. Originality/value – This is the first study comparing EyeWrite and Eye-S. Moreover, unlike other analyses, our investigation is mainly aimed at assessing user performance with the three text entry methods at the inception of the learning procedure.