An Investigation of Simulator Sickness In Virtual Reality Environments

Student: Suzan Szollar
Faculty: A. Richard Newton
(DARPA) DABT63-95-C-0074-NEWTON-06/96


Summary

An investigation of simulator sickness was conducted to determine the level of discomfort experienced by individuals immersed in virtual environments depending on the quality of the tracker and the lag, frame rate, weight, and resolution of the head mounted display (HMD). Furthermore, issues of permanent brain damage caused by low-end virtual reality (VR) displays were considered in the research.

The effect of VR depends primarily on the amount of time and frequency that a user is exposed to it, as well as the quality and structure of the VR device. Sensory conflict theory proposes that the vestibulo-ocular reflex (VOR) is not in agreement. The theory argues that the eyes are receiving information which would indicate to the brain that there should be a corresponding vestibular response, which is absent, and therefore there is a conflict between the data being processed by the brain. Other discrepancies the brain must compensate for are latency and drag in the field being viewed. The argument is that VOR adaptation will occur, while in the short term the flocculus contributes most significantly to the correction, more long term effects of motor learning indicate changes made in the synaptic weighing of synapses on vestibular neurons. Some researchers believe there is new axon growth that incurs permanent brain damage in extreme cases of prolonged exposure.

Learning occurs only when there is simultaneous head movement and retinal image motion. It is yet to be determined whether this is restricted to eye movement or also includes images changing in front of the eyes. If it is the former, it can be argued that minimum learning occurs with exposure to VR, where the eye remains relatively focused in a direction as the head turns.

Tests have done with various standard HMDs and trackers. Prior tests done by other groups indicate minimum cases of simulator sickness for bi-ocular versus binocular HMDs that have sufficient frame rates and meet other high-end parameters. Evidence of simulator sickness clearly decreases with improved quality VR devices. An alternative proposal is to investigate the result of distracting, rather than stimulating the vestibular apparatus. Stimulation has been attempted but the results did not indicate a reduction in the incidence of simulator sickness.

The research will attempt to clarify and suggest resolutions to issues regarding simulator sickness.


Suzan Szollar <szollar@ic.eecs.berkeley.edu>