On the use of ”Active Breaks” to perform Eye Exercises for more Comfortable VR Experiences

Abstract

Prolonged use of virtual reality (VR) head-mounted display (HMD) systems without removing the HMD remains a challenge, due to many factors. One is the fact that the eyes of the user suffer visual fatigue caused by stereoscopic HMDs. In clinics, people with eye-fatigue symptoms or who have difficulty seeing in daily life, are recommended to perform specific exercises, such as ”Thumbmoving” for focal changes and making a ”Figure-eight” by rolling the eyeballs. These practices efficiently help the eyes rapidly recover back to a normal state. In VR, existing solutions for visual comfort include adaptation practices, or taking breaks at different times: at the beginning (prior to exposure), in between by simply removing the HMD, and post-experiment with adaptation practices or natural decay. In this work, we mainly focus on the introduction of a recovery form called ”Active Breaks”, which briefly take place in between VR sessions, and are very different from passive methods. Our first motivation for this work is to bring eye exercises into VR experiences to enhance well-being. The second objective is to virtualize and deploy those exercises into the virtual environment for the user to practice without having to remove their HMD as in other approaches. These correspond to real-world exercises (RWE) after removing the HMD, and VR exercises (VRE) while still wearing the HMD, along with a Baseline (no exercises) condition. After an evaluation of user preference on the helpfulness of the introduced exercises, a majority of subjects rated both approaches as ”Helpful” or ”Very much helpful”, and rated RWE as more preferred over VRE. In terms of support for the superiority of VRE over RWE, we did not find significant differences in terms of eye-blink rate, eye tiredness, sense of presence, or VR sickness. However, we found one significant difference for a sub-component of IPQ: the RWE condition showed significantly higher levels of General Presence compared to VRE.