Comparing Eye Tracking and Head Tracking During a Visual Attention Task in Immersive Virtual Reality

Abstract

AbstractThe use of eye tracking (ET) and head tracking (HT) in head-mounted displays allows for the study of a subject’s attention in virtual reality environments, expanding the possibility to develop experiments in areas such as health or consumer behavior research. ET is a more precise technique than HT, but many commercial devices do not include ET systems. One way to study visual attention is to segment the space in areas of interest (AoI). However, the ET and HT responses could be similar depending on the size of the studied area in the virtual environment. Therefore, understanding the differences between ET and HT based on AoI size is critical in order to enable the use of HT to assess human attention. The purpose of this study was to perform a comparison between ET and HT technologies through the study of multiple sets of AoI in an immersive virtual environment. To do that, statistical techniques were developed with the objective of measuring the differences between the two technologies. This study found that with HT, an accuracy of 75.37% was obtained when the horizontal and vertical angular size of the AoIs was 25°. Moreover, the results suggest that horizontal movements of the head are much more similar to eye movements than vertical movements. Finally, this work presents a guide for future researchers to measure the precision of HT against ET, considering the dimensions of the AoI defined in a virtual scenario.KeywordsHead-mounted displayEye-trackingHead-trackingVirtual-realityArea of interest