Effects of vection type and postural instability on cybersickness

Abstract

This study directly compared the novel unexpected vection hypothesis and postural instability-based explanations of cybersickness in virtual reality (VR) using head-mounted displays (HMD) for the first time within a commercial VR game. A total of 40 participants (19 males and 21 females) played an HMD-VR game (Aircar) for up to 14 min, or until their first experience of cybersickness. Based on their self-reports, 24 of these participants were classified as being ‘sick’ during the experiment, with the remainder being classified as ‘well’. Consistent with the unexpected vection hypothesis, we found that: (1) ‘sick’ participants were significantly more likely to report unexpected vection (i.e., an experience of self-motion that was different to what they had been expecting), and (2) sickness severity increased (exponentially) with the strength of any unexpected (but not expected) vection. Our results also supported the predictions of postural instability theory, finding that the onset of cybersickness was typically preceded by an increase in participants’ postural instability. However, when both sway and vection measures were combined, only unexpected vection was found to significantly predict the occurrence of sickness. These findings highlight the importance of unusual vection experiences and postural instability in understanding cybersickness. However, they suggest that developers should be able to make use of expected experiences of vection to safely enhance HMD-VR.