To be presented at International Workshop on Motion Sickness , Marbella Spain, May 26-28, 1997.
INTRODUCTION
Current virtual interfaces imperfectly simulate the
motion dynamics of the real world. These imperfections have a
range of consequences which this research explores. Conflicting
visual and vestibular cues for self-motion are believed to drive
physiological adaptations and simulator sickness, which raises
significant health and safety issues regarding virtual environment
exposure. Our research investigates the nature of human physiological
adaptation to virtual interfaces through a detailed study of the
vestibulo-ocular reflex (VOR). The VOR is a compensatory eye movement
response that functions to keep the visual scene stabilized on
the retina during head movements. We hypothesize that simulator
sickness susceptibility can be partially predicted by speed of
VOR adaptation to visual-vestibular mismatches. Those prone to
simulator sickness are thought to be slow adapters to stimulus
rearrangements. Figure 1 provides an overview of the theoretical
relationship between VOR adaptation and simulator sickness.
BACKGROUND
The above hypothesis assumes the existence of an
individual 'adaptability' trait. Reason and Brand (1975) argue
that this stable adaptability trait reflects the rate at which
a person typically adjusts to sensory rearrangements. In terms
of sensory rearrangement theory, adaptability is the time it takes
for the 'internal model' of expected combinations of motion signals
to be updated. A person with high adaptability would rapidly adjust
to sensory rearrangements and would therefore avoid motion (or
simulator) sickness. A person who had low adaptability would be
prone to more sickness symptoms due to the increased duration
of signal mismatch before the neural stores were updated. There
is experimental evidence for the existence of this adaptability
trait (Reason & Graybiel, 1972). In addition, many investigators
consider differences in adaptability to be a major determinant
of inter-subject difference in susceptibility to motion sickness
(Reason & Graybiel, 1972; Griffin, 1990; Kennedy, Dunlap &
Fowlkes, 1990; Guedry, 1991). If individuals exhibit a fixed trait
which governs relative adaptive ability to altered sensory rearrangements,
then an objective measure of this adaptability trait would likely
predict individual susceptibility to simulator sickness.
There are data supporting the existence of a correlation
between VOR adaptation response and simulator sickness. First,
a functioning vestibular organ is a fundamental requirement for
both VOR adaptation and simulator sickness processes (Reason &
Brand, 1975). Second, both processes are thought to be driven
by visual-vestibular sensory rearrangements found in virtual interfaces
(Peli, 1995). Third, VOR adaptation is often accompanied by symptoms
(e.g., headaches, dizziness, nausea, eye strain) that are similar
to those identified with simulator sickness. However, if VOR
adaptation and simulator sickness are correlated, the association
is neither perfect nor causal. VOR adaptation can occur without
the onset of simulator sickness symptoms and simulator sickness
can occur without obvious concurrent VOR adaptation. Therefore,
one must look deeper to uncover an objective measure of adaptability.
Speed of VOR adaptation response is suggested as an objective measure of adaptive ability to altered visual-vestibular motion cues. This metric leverages the close association between VOR adaptation with sickness with the concept of an individual adaptability trait in an attempt to predict sickness susceptibility. Both time course of adaptation and level of adaptation achieved after a fixed exposure period may provide meaningful correlations with sickness likelihood. The relationship is hypothesized to be along the lines of:
where the likelihood of experiencing sickness symptoms
following sensory rearrangement is related to the speed of VOR
recalibration to a sensory rearrangement and the maximum level
of adaptation achieved.
APPROACH
Experiments are currently underway to determine the
magnitude of VOR gain and phase adaptation to specific parameters
of virtual interfaces. Parameters being evaluated include time
delays between head movement and visual scene update and scene
scale changes due to geometric field-of-view manipulations. These
preliminary experiments provide the foundation of support for
the main experiment.
The VOR-sickness experiment involves the formation
of two susceptibility groups (LOW, HIGH) of 15-20 subjects each,
based upon their scores on a motion/simulator sickness history
questionnaire. VOR gain and phase data will be collected before,
during and after a 45 minute exposure to a cross-axis visual-inertial
stimulation, with horizontal inertial oscillations being paired
with vertical oscillation of the visual scene as per Khater, et
al., (1990). VOR gain and phase changes from baseline will be
averaged and compared across groups. Readaptation time course
will also be examined for both groups. It is anticipated that
VOR adaptation speed will partially predict sickness susceptibility
and that, when combined with other partial predictors, a majority
of individual variance in sickness susceptibility can be accounted
for.
ACKNOWLEDGMENTS
Supported by Grant F49620-93-0339 from the Air Force
Office of Scientific Research and Grant NAS 0-703 from the National
Aeronautics and Space Administration to the University of Washington.
REFERENCES