Content uploaded by Brion Mario
Author content
All content in this area was uploaded by Brion Mario on Feb 18, 2019
Content may be subject to copyright.
Early Prediction of Cybersickness in Virtual,
Augmented & Mixed Reality Applications: A
Review
Brion M. Silva
Department of Computing
Informatics Institute of Technology
No 57, Ramakrishna Road,
Colombo 06, Sri Lanka.
brion.2015283@iit.ac.lk
Pumudu Fernando
Department of Computing
Informatics Institute of Technology
No 57, Ramakrishna Road,
Colombo 06, Sri Lanka.
pumudu.f@iit.ac.lk
Abstract - The field of Virtual, Augmented & Mixed Reality
(VAMR) has already taken up the consumer market by storm
and are demanding among users owing to their practicality and
aesthetics. Despite the massive popularity and hype, VAMR
carry a loathsome side-effect called “Cybersickness”, which is
often times severe enough to cause significant discomfort to the
users causing the discontinuation of use. Research have been
going on for years with the intentions of finding the theories and
factors which contribute to this unpleasant experience but we
are in no way shape to determine the underlying contrivances of
cybersickness. There are systems to quantify and measure
cybersickness but a symptom prediction tool which can be used
by application developers to evaluate their products for
cybersickness susceptibility is non-existent. This paper presents
a summary of literature and provides a review and a discussion
with the intentions of deriving a predictive algorithm which
would ultimately address the research gap and missing links.
Keywords – cybersickness, virtual reality, computer-vision,
human-computer-interaction
I. INTRODUCTION
Over the past few decades, the way people interact with
computers has changed drastically [1]. There had been
numerous attempts over the years to bridge the human-
computer barrier with the ultimate goal of making the human-
computer interactions as natural as possible. The most novel
idea was the thought of using hands as a device to connect and
communicate with computers mimicking human to human
interactions [2]. It is safe to say that this has been achieved
with the invention of digital reality systems also known as
Virtual, Augmented & Mixed Reality (VAMR) systems.
The field of VAMR is rapidly blooming as a result of the
massive popularity of new generation smart devices such as
displays, hand-held devices and wearables. VAMR is working
towards the reduction of cognitive load minimizing the
human-computer interaction efforts. These systems offer
immersion to the users diminishing attention shifts [3].
Despite of the popularity, there is an abundance of issues
when it comes to all three technologies. Most of these issues
are common across the platforms and research has been going
on for a long time trying to tackle them. The most prominent
problem that stands out has to be the concerns regarding the
user experience (UX). It has been discovered that VAMR
systems cause a nasty side effect called “Cybersickness”
which directly obstructs the UX forcing the user to terminate
the session instantly [4].
Cybersickness has persisted throughout the existence of
VAMR systems [5] and as a result, a large population are
unable to take advantage of this technology. Researchers are
busy finding the causes and the user response to cybersickness
but only a handful of people are working on to prevent or
predict the occurrence of cybersickness.
II. METHODOLOGY
A wide variety of literature was analysed to gather
knowledge and insight on the symptoms, theories and
contributing factors of cybersickness. The already existing
approaches of measuring and predicting cybersickness were
critically appraised giving more priority to recently published
research. The discovered evidence will be presented in a
narrative manner through the upcoming sections.
III. CYBERSICKNESS
The term “Cybersickness” was proposed by McCauley
and Sharkey (1992) [6] and it was described as the interim
side-effects caused by virtual reality immersion.
Cybersickness exhibits similarities to motion sickness and
simulator sickness and it’s often categorised as a form of
motion sickness [7]. But the major discrepancy when
comparing these phenomenon is the user’s sate of motion. In
cybersickness, the user is stationary for the most part but
experiences self-motion sensations as a result of the moving
imagery [8]. This sensation is commonly referred to as
Vection.
From a physiological standpoint, human body is not in a
position to cope cybersickness and according to researchers at
Coventry University, up to 80% of the population experience
some form of cybersickness [9], [10].
A. Symptoms
VAMR experiences can induce an array of symptoms
associated with cybersickness. Following are some commonly
identified side-effects and symptoms [8], [11], [12].
1. Eye strain
2. Nausea
3. Headache
4. Excessive Sweating
5. Dryness of mouth
6. Fullness of stomach
7. Vertigo
8. Ataxia
9. Vomiting
10. Dizziness
11. Disorientation
12. Fatigue
13. Strange hand eye coordination
The above symptoms can occur during and post exposure
and the majority of these side-effects are temporary whereas
few can linger for hours [13] or days [14]. As a result, many
virtual experience providers recommend users to limit
themselves from attempting risky activities such as driving for
at least an hour after exposure to VAMR.
B. Theories
Over the years, several research groups have identified
three major explanatory theories for the existence of
cybersickness.
Sensory Conflict Theory is the oldest and the most
accepted theory among the three [15]. It claims that VAMR is
likely to cause mismatches between the vestibular and visual
inputs of the human body and these discrepancies can lead to
cybersickness related symptoms such as nausea, headaches
and other oculomotor disturbances [16], [17]. However the
theory falls short in explaining the reason for the users to be
sick due to these cue conflicts [8].
Poison/ Intoxication Theory tries to base the
explanations for the occurrence of cybersickness from an
evolutionary standpoint. According to Treisman (1977) [18],
when the human body experiences abnormal coordination of
the visual, vestibular, and other sensory inputs, the nervous
system mistakes these circumstances with poison digestion
and responds by emptying the stomach. This corporal
response of the nervous system can cause symptoms like
nausea, vomiting and discomfort.
Postural Instability Theory was developed by Riccio
and Stoffregen and it states that the human body always tries
to maintain stable posture in the environment and the
unexpected changes to the environment could result in the loss
of postural control [19].
All these theories provide compelling arguments but
further research needs to be conducted to confirm the validity
of these claims.
C. Contributing Factors
Apart from the theories discussed above, there are three
main categories of factors that influence the intensification of
cybersickness. Namely, Hardware & Software factors,
Individual Differences and Environmental Factors [8].
Hardware & Software Factors
Issues related to hardware & software had been a huge
bottleneck at the dawn of the VAMR industry but fortunately,
they have improved quite a until then [20]. But there are few
outstanding issues that are left to be addressed.
Oftentimes virtual reality (VR) tech suffers from position
tracking errors. Inconsistencies in tracking could register
unstable information resulting in jitter. Jitter has a tendency to
cause dizziness and lack of concentration [21].
Head mounted displays (HMD) usually takes a bit of time to
update users visual feed in response to the head movements.
This delay is commonly referred to as “Lag” and has a
tendency to cause unsettling sensations which result in
cybersickness symptoms [22].
Flicker is another issue and the perception of flicker will differ
from person to person and the likelihood increases with the
increase of field of view (FOV). Flicker will cause eye fatigue
[23] and could be reduced by increasing the framerate of the
visual feed.
Individual Differences
Susceptibility to cybersickness hugely depend on individual
factors such as health, behavioural conditions, age, race,
ethnicity etc. [24]. Oftentimes individual differences are
overlooked when considering contributing factors of
cybersickness but research suggests that these shouldn’t be
disregarded [25]–[27].
Environmental Factors
Environmental conditions such as temperature, relative
humidity, light, air quality, ambient noise etc. can aid to
cybersickness during a VAMR experience. These conditions
aren’t so easy to control due to their unpredictable nature [24].
IV. EXISTING WORK
Cybersickness is not a new phenomenon [24] and the
concerns about it had paved the way for a rich body of
literature over the years. Its studies can be dated back decades
since VAMR has a broad spectrum of applications ranging
from home entertainment to military [28]. As stated in the
above section, cybersickness has a diverse range of symptoms.
Researchers have been using a wide verity of approaches and
measures to detect and rate the severity of these cybersickness
symptoms [29]. These approaches can be mainly categorised
in to four sections.
1. Questionnaire based approaches
2. Postural Instability based approaches
3. Physiological/ Biometric approaches
4. Predictive Model based approaches
Questionnaires were the first attempts made to rate and
quantify symptoms in simulators and these approaches tend to
lean towards simulator sickness and motion sickness rather
than cybersickness. On the contrary, postural instability
related approaches were inspired by the postural instability
theory in the recent years. Some have attempted to measure
physiological signals and biometrics in order to predict the
onset of cybersickness where as others have taken a predictive
model based approach. All these four categories are explained
in detail below.
A. Questionnaire based approaches
Kennedy et al. (1993) [30] utilized a 28 symptom version
of the Pensacola Motion Sickness Questionnaire (MSQ)
which was originally developed by Kellogg et al. (1964) [31]
and conducted an extensive research by analyzing more than
1000 sets of previous data and came up with a 16 term
measurement method to quantify cybersickness (see Figure 1).
Their main objective was to demystify the similarity concept
of motion sickness and simulator sickness and to point out the
need for a distinguished index for measuring simulator
sickness severity. Potential research subjects were subjected
to the “Simulator Sickness Questionnaire” (SSQ) pre and post
exposure to simulations. The questionnaire identified issues
such as disorientation, oculomotor effects, and nausea and the
total score at the end reflected the severity of simulator
sickness experienced by the users. SSQ is widely applied in
the research studying the propensity of inducing
cybersickness and simulator sickness.
Although SSQ is the primary measure of cybersickness
symptoms [11], it was never intended to be used in the context
of virtual reality. And to make matters even worse, it was
devised 25 years ago. As a result of the above mentioned
issues, SSQ has received quite a bit of criticism over the years.
There have been many attempts during the past to revise and
refactor the SSQ.
Figure 1. Symptoms in MSQ & SSQ [30]
Kim et al. (2004) [32] came up with a revised version of
the SSQ claiming that the original SSQ was missing several
key symptoms and lacked interpretability. The new scale was
named Revised-SSQ (RSSQ) by the team and consisted
significant changes to the measuring criteria and the scale.
They ditched the original 4 option ordinal response scale and
opted for an 11 option scale with two labels at the two
extremes. Despite the efforts RSSQ didn’t replace the original
SSQ mainly because of its intimidating scale and the under-
representative study group.
Bouchard et al. (2007) [33] attempted to refactor the SSQ
with the intentions of making it specific to cybersickness.
They were convinced that the SSQ had multiple subfactors
which resulted in a biased total score and expressed their
concerns regarding the use of military pilots in the SSQ
development study arguing the applicability to general adult
population. To address these problems, they used a non-
military study group to avoid cross-loading and devised a new
questionnaire with two main categories: nausea and
oculomotor. Although a higher portion of females were used
in the study, many researchers argued that the study sample
was still unrepresentative. As a result of the above mentioned
concern, the usage of this refactored version has been limited
[29].
Bruck and Watters (2011) [34] tried to incorporate
physiological responses to the factor structure of the SSQ by
using respiration and heart rate as variables alongside the
original 16 symptoms of the SSQ. The sample size used in the
study was better in terms of gender representation but had
received some negative remarks due to its limited size. And
the use of a principal components analysis (PCA) rather than
a common factor model has been criticised by some
researchers [11]. Same as the other refactoring attempts,
there’s no evidence to prove that this version is significantly
better than the original.
Ames et al. (2005) [35] designed a refactored version of
the SSQ called Virtual Reality Symptom Questionnaire
(VRSQ) to be distinctively used in the context of virtual
reality. A seven option response scale with four descriptive
labels was utilized in the VRSQ and unlike SSQ, VRSQ was
intended to be specifically used for the measure of
cybersickness. However, VRSQ has not been validated in the
context of virtual reality and the methodology followed during
the study is questionable mainly due to the use of a non-
interactive video feed on a HMD [11]. Hence the measure has
received very little attention.
Stone III (2017) [11] did a psychometric evaluation of the
SSQ and pointed out the need for an empirical measure for
cybersickness. He further argued that the use of the SSQ in the
context of virtual reality to be problematic and stated that the
refactored versions failed to provide an obvious advantage
over the original version. The study resulted in a new measure
called Cybersickness Questionnaire (CSQ), a two factor
exploratory model with a different factor structure and scoring
mechanism. What is noteworthy about this new measure is its
ability to be collaboratively used with the SSQ. According to
Stone III (2017) [11], researchers have the opportunity to
administer the original SSQ if they wish and use the CSQ’s
scoring criteria afterwards.
The use of single-item measures instead of the SSQ have
been a common practice lately since the longer questionnaires
usually take a lot of time to carry out requiring for frequent
breaks which might result in diminished cybersickness
symptoms [29]. Majority of these scales required the user to
rate their feelings of either nausea, motion sickness, or
discomfort throughout the course of the study at regular
intervals. Nalivaiko et al. (2015) [36] used a single-item, 11
option nausea rating scale in their study with labels “0-no
signs of nausea” and “10-ready to vomit” at corresponding
extremes. Rebenitsch and Owen (2014) [37] utilised an
immersion rating in their study and continuously inquired the
participants to rate the experience on a scale of 0 - 10. This
method ensured the safety of the participants and they
reportedly discontinued the exposure if anyone responded
“10”. Although these single-item scales are adequate to
monitor cybersickness symptoms, they are not as sensitive or
informative as the SSQ and its refactored versions [29].
B. Postural Instability based approaches
The core concept behind the postural instability theory is
that the maintenance of body posture is crucial and prolonged
instability of posture may lead to visually induced motion
sickness (VIMS) [19]. Some of the VIMS symptoms are
drowsiness, dizziness, fatigue, pallor, cold sweat, oculomotor
disturbances, nausea, and vomiting [38]. The aforementioned
symptoms are very similar to that of Vection and the
relationship between Vection and VIMS had been questioned
and debated over the years in the literature [39]. This approach
sounds convincing on paper since it doesn’t require to bother
the user during the simulation and most importantly due to its
objective nature. But in practice this approach seemed to
violate the above advantages simply because of the
standardized stance requirement [29].
Researchers at the University of Waterloo are
reportedly working on defining a predictive model to identify
the likelihood of cybersickness in individuals by analysing
their postural swaying patterns as a response to moving visual
field. According to the research team, knowing who might be
susceptible to cybersickness beforehand would help in
minimising the dawn of symptoms [40].
Dennison & Zmura (2017) [39] devised an experiment to
test out the link between postural sway and cybersickness.
They exposed a group of standing and seated HMD wearing
test subjects to a visual feed of a virtual tunnel that rotated
along their line of sight. It was discovered that only a minority
of the group exhibited postural sway in response to the
commencement of the tunnel rotation. The increase of the
rotational speed of the tunnel directly resulted in increase of
cybersickness. The duo claimed that the subjects who
exhibited greater levels of cybersickness, showed less
variations in postural sway.
C. Physiological/ Biometric approaches
Cybersickness symptom prediction using physiological
signals and biometrics have received a noticeable interest in
the past mainly due to the availability of affordable medical
technology. Several research groups have pointed out
prospective physiological consequences of cybersickness
from time to time [5], [41], [42]. And by following this
approach, researchers are able to gather evidence of symptoms
without intruding the user. Heart rate and blood pressure
measurements are commonly used for analysis [29] but some
research groups have managed to expand their horizons by
integrating different physiological signals such as brain
activity, skin reaction etc.
Kiryu et al. (2007) [43] used electrocardiogram (ECG),
blood pressure and respiration in their study to figure out the
trigger factors and accumulation factors in cybersickness. The
power frequencies of the physiological signals were used in
estimating the sensation intervals and the onset of
cybersickness.
Kim et al. (2005) [42] examined 16 electrophysiological
signals of 61 test subjects over a period of 9.5 minutes. Three
susceptibility questionnaires were obtained pre-exposure and
the physiological signals were recorded before, during and
after the immersion. The SSQ was administered at the end of
the exposure for the test subjects to report their cybersickness
severity. They were able to figure out a strong positive
correlation between cybersickness and gastric
tachyarrhythmia, eye blinking rate, heart rate and EEG.
Dennison & Zmura (2017) [5] examined ECG,
electrogastrogram (EGG), electrooculogram (EOG),
photoplethysmogram (PPG), breathing rate and galvanic skin
response (GSR) in their study to figure out the feasibility of
using physiological signals to predict cybersickness. Apart
from the physiological measures, they administered the SSQ
and asked the users to verbally rate the cybersickness severity
every two minutes. The main purpose of using the SSQ in the
study was to figure out the possibility of predicting sickness
scores on the SSQ by using the physiological changes. And
they were able to find significant correlations in the SSQ
scores and physiological signals.
D. Predictive Model based approaches
Use of a predictive model to predict cybersickness seems
highly constructive but upon close examination of the
literature, it was found that only a handful have even
attempted the said approach in their studies.
Design Interactive is a tech company based in the United
States and they have been studying cybersickness for over two
decades. They have been able to develop a predictive
algorithm which uses both the questionnaire & physiological
approaches in unison [44]. Unfortunately, Design Interactive
only provides services to their exclusive clientele. Therefore
the algorithm is not publicly available.
Bockelman and Lingum (2013) [24] have pointed out the
need for a cybersickness prediction tool & indicated their
interest in working towards an open source tool to predict the
possibility of cybersickness. They further reported that they
have been working on a predictive model to provide guidance
to virtual reality users. But the success rate or the progress of
the research is unknown.
V. ANALYSIS OF EXISTING WORK
By conducting a comprehensive investigation, we were
able to distinguish the pros and cons of each study and most
importantly we were able to eliminate the unlikely factors that
could potentially cause cybersickness to avoid any over fitting
issues.
A. Analysis of the existing questionnaires
The original version of the SSQ exhibits substantial
differences in comparison to the revised and refactored
versions with regards to the context it was intended to be used,
number of items in the questionnaire and the factor model.
Table 1 depicts a comparison between the original SSQ and
its refactored versions.
Study
Kennedy et al.
(1993)
Kim et al.
(2004)
Bouchard et al.
(2007)
Bruck and
Watters (2011)
Ames et al.
(2005)
Stone III (2017)
Intended
Context
SS
SS
CS
CS
CS
CS
Number
of Items
16
24
16
18
13
9
Factor
Model
3
4
2
4
2
2
* SS – Simulator Sickness, CS - Cybersickness
Table 1. Comparison between the exisiting questionnaires
B. Analysis of the physilogical signals
Table 2 displays a brief analysis of some important
physiological measures and their correlation to cybersickness.
Physiological Signal
Definition
Correlation to
Cybersickness
Blood Pressure
The pressure between
heart beats
High
Electrocardiogram (ECG)
Monitors the heart
High
Electroencephalogram (EEG)
Monitors electrical
activity due to current
flow in the brain
Medium
Electrogastrogram (EGG)
Detects stomach
movement and strength
High
Electrooculogram (EOG)
Detects the movement
of the eye
Non-existent
Galvanic Skin Response
(GSR)
Monitors changes in
electrical resistance of
the skin
Non-existent
Respiration (RSP)
Monitors inhalation of
oxygen
High
Table 2. Physiological signals and their correlation to cybersickness
C. Analysis of the Individual Differences affecting
cybersickness
Table 3 depicts an analysis of the individual factors taken
from cybersickness specific researches.
Factor
Correlation to
Cybersickness
Reasoning
Age
High
Young subjects proved to be more
susceptible
Gender
Medium
Females appeared to be more
susceptible
Ethnicity
Low
Not tested for cybersickness
Experience
High
Repeated immersion decreases
cybersickness susceptibility
Concentration Level
Medium
Diminished concentration likely
reduces cybersickness
Postural Instability
High
Unstable posture leads to
cybersickness
History of Illnesses
(Headaches, Motion
Sickness, Migraines,
Dizziness)
High
History of illness would likely
increase the likelihood of
cybersickness
BMI
Medium
BMI has proven to have a minor
effect on cybersickness
* BMI – Body mass index
Table 3. Analysis of individual differences
D. Analysis of Software and Hardware factors
Table 4 presents a summary of the prospective software
and hardware factors which affects cybersickness.
Hardware
Software
Lag
Flicker
Frame rate
Colour
Field of view (FOV)
Scene complexity
Head Tracking
Realism
Table 4. Summary of Hardware & Software factors affecting
cybersickness
VI. DISCUSSION
The ultimate goal of this study was to review the current
status of cybersickness research with the intentions of deriving
a predictive algorithm for cybersickness susceptibility. During
the study we realised that the performance of the already
existing approaches on their own were subpar and the use of a
hybrid approach seemed to be a highly versatile option.
In many cases, individual differences and software &
hardware factors had not been taken in to consideration. Some
of these factors have proven to be highly correlated with
cybersickness, hence ditching these would not be a smart idea.
Any new approach should take leverage of the existing work
and should carefully consider all the possible contributing
factors. On the other hand, machine learning and deep learning
is on the rise and the use of the said technologies would
enhance the results of the solution significantly.
VII. CONCLUSIONS & FUTURE DIRECTIONS
Overall, Regardless of the realism or interactive abilities,
current VAMR systems are not usable for a prolonged period
of time and they are a risk to the users because of
cybersickness.
The primary focus has been given to finding the
underlying reasons of cybersickness and VAMR’s
predisposition of inducing cybersickness but cybersickness
prediction has received very little to no attention. However,
upon close inspection of literature, it has been identified that
only subjective analysis is often used in cybersickness studies
and the current primary measure of cybersickness (SSQ) is
massively outdated. Whereas, other than subjective factors,
there are a numbers of possible contributing factors impacting
cybersickness which are often overlooked.
While exploring the literature, we strongly felt the need for
an all-round predictive model and through the analysis of the
existing work, we were able to identify the key factors and
parameters necessary for deriving such an algorithm. With the
gathered knowledge, we hope work towards a predictive
algorithm which can be used by the application developers to
check their products for cybersickness susceptibility prior to
release.
REFERENCES
[1] B. A. Myers, “A Brief History of Human Computer Interaction
Technology,” no. September, 2016.
[2] S. S. Rautaray and A. Agrawal, “Vision based hand gesture
recognition for human computer interaction: a survey,” AI Rev.,
vol. 43, no. 1, pp. 1–54, 2015.
[3] J. Chen, “11TH INTERNATIONAL CONFERENCE ON
VIRTUAL, AUGMENTED AND MIXED REALITY,” 2018.
[Online]. Available: http://2019.hci.international/vamr.
[Accessed: 18-Oct-2018].
[4] A. Tiiro, “Effect of Visual Realism on Cybersickness in Virtual
Reality,” University of Oulu, 2018.
[5] M. S. Dennison, A. Z. Wisti, and M. D’Zmura, “Use of
physiological signals to predict cybersickness,” Displays, vol. 44,
no. October, pp. 42–52, 2016.
[6] M. E. McCauley and T. J. Sharkey, “Cybersickness: Perception
of Self-Motion in Virtual Environments,” Presence Teleoperators
Virtual Environ., vol. 1, no. 3, pp. 311–318, 1992.
[7] B. Keshavarz, B. E. Riecke, L. J. Hettinger, and J. L. Campos,
“Vection and visually induced motion sickness: how are they
related? Behrang,” Front. Psychol., vol. 6:472, 2015.
[8] J. J. LaViola, “A discussion of cybersickness in virtual
environments,” ACM SIGCHI Bull., vol. 32, no. 1, pp. 47–56,
2000.
[9] Telegraph.co.uk, “Cybersickness: The new ‘illness’ sweeping the
nation,” 2015. [Online]. Available:
https://www.telegraph.co.uk/news/health/news/12001743/Cybers
ickness-The-new-illness-sweeping-the-nation.html. [Accessed:
19-Oct-2018].
[10] Dr. Brent Winslow, “Design Guidelines for Preventing VR
Sickness,” 2017. [Online]. Available:
https://www.slideshare.net/AugmentedWorldExpo/dr-brent-
winslow-design-interactive-design-guidelines-for-preventing-vr-
sickness. [Accessed: 20-Oct-2018].
[11] W. B. Stone III, “Psychometric evaluation of the Simulator
Sickness Questionnaire as a measure of cybersickness,” Iowa
State University, 2017.
[12] B. D. Lawson, D. . Graeber, and Mead A.M, Signs and symptoms
of human syndromes associated with synthetic experience. 2002.
[13] R. S. Kellogg, C. Castore, and R. Coward, “Psychophysiological
Effects of Training in a Full Vision Simulator,” Annu. Sci. Meet.
Aerosp. Med. Assoc., 1980.
[14] R. J. Blok, “Simulator sickness in the U.S. Army UH-60A
Blackhawk flight simulator.,” Mil. Med., vol. 157, no. 3, pp. 109–
111, Mar. 1992.
[15] J. Reason and Brand J. J, Motion sickness. Academic Press
London ; New York, 1975.
[16] B. Muthuraj, S. Mukherjee, C. R. Patra, and P. K. Iyer, “Effects
of visual flow direction on signs and symptoms of cybersickness
Alireza,” ACS Appl. Mater. Interfaces, vol. 8, no. 47, pp. 32220–
32229, 2016.
[17] S. Sharples, S. Cobb, A. Moody, and J. R. Wilson, “Virtual
reality induced symptoms and effects ( VRISE ): Comparison of
head mounted display ( HMD ), desktop and projection display
systems,” vol. 29, pp. 58–69, 2008.
[18] M. Treisman, “Motion sickness: An evolutionary hypothesis,”
Science (80-. )., vol. 197, no. 4302, pp. 493–495, 1977.
[19] G. E. Riccio and T. A. Stoffregen, “An Ecological Theory of
Motion Sickness and Postural Instability,” Ecol. Psychol., vol. 3,
no. 3, pp. 195–240, 1991.
[20] A. Vovk, F. Wild, W. Guest, and T. Kuula, “Simulator Sickness
in Augmented Reality Training Using the Microsoft HoloLens,”
Proc. CHI, pp. 1–9, 2018.
[21] F. Biocca, “Will Simulation Sickness Slow Down the Diffusion
of Virtual Environment Technology?,” Presence: Teleoper.
Virtual Environ., vol. 1, no. 3, pp. 334–343, Jan. 1992.
[22] R. Pausch, T. Crea, and M. Conway, “A Literature Survey for
Virtual Environments: Military Flight Simulator Visual Systems
and Simulator Sickness,” Presence: Teleoper. Virtual Environ.,
vol. 1, no. 3, pp. 344–363, Jan. 1992.
[23] K. Harwood and P. Foley, “Temporal Resolution: An Insight into
the Video Display Terminal (VDT) ‘Problem,’” Hum. Factors,
vol. 29, no. 4, pp. 447–452, 1987.
[24] P. Bockelman and D. Lingum, “Factors of Cybersickness,” vol.
374, no. October, 2013.
[25] E. M. Kolasinski, “Simulator Sickness in Virtual Environments
(Technical Report 1027),” U.S. Army Research Institute for the
Behavioral and Social Sciences, Alexandria, VA., 1995.
[26] L. Frank, R. S. Kennedy, M. E. McCauley, R. W. Root, R. S.
Kellogg, and A. C. Bittner., “Simulator Sickness: Sensorimotor
Disturbances Induced in Flight Simulators,” Image II Conf., pp.
417–426, 1983.
[27] J. Lackner, “Human Orientation, Adaptation, and Move- ment
Control,” Motion Sick. Vis. Displays, Armored Veh. Des., pp. 28–
50, 1990.
[28] D. M. Johnson, “Introduction to and Review of Simulator
Sickness Research,” 2005.
[29] L. Rebenitsch and C. Owen, “Review on cybersickness in
applications and visual displays,” Virtual Real., vol. 20, no. 2, pp.
101–125, Jun. 2016.
[30] R. S. Kennedy, N. E. Lane, K. S. Berbaum, and M. G. Lilienthal,
“Simulator_Sickness_Questionnaire,” The International Journal
of Aviation Psychology, vol. 3, no. 3. pp. 203–220, 1993.
[31] R. S. Kellogg, R. S. Kennedy, and A. Graybiel, “Motion Sickness
Symptomatology of Labyrinthine Defective and Normal Subjects
During Zero Gravity Maneuvers. Techn Docum Rep No. Amrl-
tdr-64-47.,” AMRL-TR. Aerosp. Med. Res. Lab., pp. 1–11, Jun.
1964.
[32] D. Kim, D. Parker, and M. Park, “A New Procedure for
Measuring Simulator Sickness–the RSSQ,” Univ. Washingt.
Hum. Interface Technol. …, no. 1998, pp. 1–14, 2004.
[33] S. Bouchard, G. Robillard, and P. Renaud, “Revising the factor
structure of the Simulator Sickness Questionnaire,” Annu. Rev.
Cybetherapy Telemed., vol. 5, no. January 2007, pp. 1–10, 2007.
[34] S. Bruck and P. Watters, “The factor structure of cybersickness,”
Displays, vol. 32, 2011.
[35] S. L Ames, J. Wolffsohn, and N. A McBrien, “The Development
of a Symptom Questionnaire for Assessing Virtual Reality
Viewing Using a Head-Mounted Display,” Optom. Vis. Sci., vol.
82, pp. 168–176, 2005.
[36] E. Nalivaiko, S. L. Davis, K. L. Blackmore, A. Vakulin, and K.
V. Nesbitt, “Cybersickness provoked by head-mounted display
affects cutaneous vascular tone, heart rate and reaction time,”
Physiol. Behav., vol. 151, no. October 2017, pp. 583–590, 2015.
[37] L. Rebenitsch and C. Owen, “Individual variation in
susceptibility to cybersickness,” UIST 2014 - Proc. 27th Annu.
ACM Symp. User Interface Softw. Technol., pp. 309–318, 2014.
[38] E. F Miller and A. Graybiel, “Experiment M-131. Human
vestibular function,” Aerosp. Med., vol. 44, pp. 593–608, 1973.
[39] M. S. Dennison and M. D. Zmura, “Cybersickness without the
wobble : Experimental results speak against postural instability
theory,” Appl. Ergon., vol. 58, pp. 215–223, 2017.
[40] S. Weech, J. P. Varghese, and M. Barnett-Cowan, “Estimating
the sensorimotor components of cybersickness,” J.
Neurophysiol., vol. 120, no. 5, pp. 2201–2217, 2018.
[41] L. R. Rebenitsch, “Cybersickness prioritization and modeling,” p.
177, 2015.
[42] Y. Youn Kim, H. Ju Kim, E. Nam Kim, H. Dong Ko, and H.-T.
Kim, “Characteristic changes in the physiological components of
cybersickness,” Psychophysiology, vol. 42, pp. 616–625, 2005.
[43] T. Kiryu, E. Uchiyama, M. Jimbo, and A. Iijima, “Time-varying
Factors Model with Different Time-scales for Studying
Cybersickness,” in Proceedings of the 2Nd International
Conference on Virtual Reality, 2007, pp. 262–269.
[44] Design Interactive, “Predicting Cybersickness,” 2017. [Online].
Available: http://designinteractive.net/predicting-cybersickness/.
