A Study of Physical Fitness and Enjoyment on Virtual Running for Exergames

Abstract

Virtual Reality (VR) technology has advanced forward in everyday life where virtual fitness is possible through physically moving around in the real world. Exergame is a video game for exercise aimed at making exercise more fun. VR exergame applies these trends together for virtual fitness with immersive game play. The VR locomotion is traveling in VR, which is commonly used in adventure role-playing games (RPG). Virtual running can be applied as a locomotion technique for VR exergames. The design of virtual running in VR exergames should be considered as an exercise for fitness and also for enjoyment. This paper proposed two motion-based locomotion techniques: ArmSwing and Squat for virtual running, which are considered as aerobic and strength exercise. These two postures were used to study how physical exertion affected players while interacting in the test scene. Usability, motion sickness, and enjoyment were assessed to analyze the differences of each posture. The results showed that motion sickness and enjoyment of ArmSwing and Squat were not different, while usability was different where ArmSwing was rated higher than Squat. The results from the interviews suggest that most players preferred aerobic exercise (ArmSwing) more than strength exercise (Squat) for a long period of exercise. However, for a short period of exercise, players preferred strength exercise more than aerobic exercise. The adventure-based RPG for exercise needs a solution design appropriate for virtual running in VR, and our results can be a guideline for developers in order to handle motion-based locomotion for VR exergames.

Full text

Research Article
A Study of Physical Fitness and Enjoyment on Virtual
Running for Exergames
Chaowanan Khundam
1
and Frédéric Nöel
2
1
IICE, School of Informatics, Walailak University, 222 Thai-Bury, Thasala, Nakhon Si Thammarat, Thailand 80160
2
Institute of Engineering Univ. Grenoble Alpes, CNRS, G-SCOP, 38000 Grenoble, France
Correspondence should be addressed to Chaowanan Khundam; chaowanan.kh@gmail.com
Received 30 December 2020; Revised 10 March 2021; Accepted 9 April 2021; Published 30 April 2021
Academic Editor: Cristian A. Rusu
Copyright © 2021 Chaowanan Khundam and Frédéric Nöel. This is an open access article distributed under the Creative Commons
Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is
properly cited.
Virtual Reality (VR) technology has advanced forward in everyday life where virtual fitness is possible through physically moving
around in the real world. Exergame is a video game for exercise aimed at making exercise more fun. VR exergame applies these
trends together for virtual fitness with immersive game play. The VR locomotion is traveling in VR, which is commonly used in
adventure role-playing games (RPG). Virtual running can be applied as a locomotion technique for VR exergames. The design
of virtual running in VR exergames should be considered as an exercise for fitness and also for enjoyment. This paper proposed
two motion-based locomotion techniques: ArmSwing and Squat for virtual running, which are considered as aerobic and
strength exercise. These two postures were used to study how physical exertion affected players while interacting in the test
scene. Usability, motion sickness, and enjoyment were assessed to analyze the differences of each posture. The results showed
that motion sickness and enjoyment of ArmSwing and Squat were not different, while usability was different where ArmSwing
was rated higher than Squat. The results from the interviews suggest that most players preferred aerobic exercise (ArmSwing)
more than strength exercise (Squat) for a long period of exercise. However, for a short period of exercise, players preferred
strength exercise more than aerobic exercise. The adventure-based RPG for exercise needs a solution design appropriate for
virtual running in VR, and our results can be a guideline for developers in order to handle motion-based locomotion for VR
exergames.
1. Introduction
Virtual Reality (VR) technology has stepped up to have more
roles in our daily lives, including exercise and playing games.
Fitness game or exergame is a term referring to using video
games for exercise, which relies on technology that can track
body movement [1]. VR immersion is a new trend for exer-
games because body movements can be detected by VR sen-
sors and also immerse a user’s feelings [2]. Therefore, the
movement in the virtual environment corresponds to the
physical movement, which makes the VR application suitable
for exergames.
Especially, the use of exergames encourages users to
interact with the content, which makes VR fitness a perfect
integration of technology and exercise (Yoo, Carter, and
Kay [3]). Recently, the VR industry is growing at a fast pace,
and the market size of consumer VR hardware and software
is projected to increase from 6.2 billion U.S. dollars in 2019 to
more than 16 billion U.S. dollars by 2022 [4]. There is a high
possibility that VR exergames will become popular in the
future. VR systems have many possible advantages for sports
training that can precisely control the environment and
standardize the situation. Additional information can be
integrated to provide operational guidelines, and the envi-
ronment can be changed dynamically to create different
competitive situations [5]. Although this technology is new,
it shows promising results in high participation rates, more
enjoyment, more motivation [6–8], and more fun than other
forms of exercise. The fact is that most people are bored and
do not have the discipline to traditional exercise, which is one
of the main reasons for a high level of inactivity [9]. There-
fore, immersive VR exergaming is a new opportunity for
Hindawi
International Journal of Computer Games Technology
Volume 2021, Article ID 6668280, 16 pages
https://doi.org/10.1155/2021/6668280

engaging more players than standard exercise and may dis-
tract participants from exertion required and the feeling of
fatigue. We can play exergames from a variety of consoles.
The following examples can display about the tools to play
exergames.
There are many game consoles such as Xbox, PlayStation,
and Nintendo. However, for exergaming, additional accesso-
ries may be needed to physically interact with the console
instead of using just buttons on the gamepad. Although there
are other consoles available, we focused on these three con-
soles because they offer the most common exergames on
the market. The Xbox is a gaming console from Microsoft.
In order to play exergames with the Xbox, it requires the
Kinect system, which is a motion capture system that recog-
nizes player body movements and integrates them into the
game play. The Xbox system with Kinect does not require
any accessories, making it the first gaming system where free
body movement is possible. Exergames using the Xbox
Kinect have the potential to enhance balance training and
broader rehabilitation compared to standard exercise [10].
The PlayStation (PS) is a console from Sony. It includes the
PlayStation Camera, a motion sensor akin to the Xbox’s
Kinect peripheral. Moreover, there is the move controller,
which is conceptually similar to Nintendo’s Wii Remote.
These combinations allow for more free body movement
than the Nintendo but less than the Xbox. However, exer-
games on the PS are more preferable to players than the Xbox
[11]. The Nintendo Wii gaming console is popular for sport
games, aiming to develop new game franchises that target fit-
ness gaming. In particular, the Wii Remote controller can be
used as a handheld pointing device and can detect movement
in three dimensions. Since its release, the Wii has delivered
many peripheral devices for exergames such as the Wii Bal-
ance Board and Motion Plus. The Wii Fit is a popular exer-
game for Wii with several activities using the Balance
Board. It has been used for physiotherapy rehabilitation
[12] and adopted for fitness applications focused on the
elderly [13]. A new version called the Nintendo Switch
(NX) is a hybrid console, which can be used as a home con-
sole or a portable device. Its wireless Joy-Con controllers
come with standard buttons and directional analog sticks, a
motion sensor, tactile feedback, and a heart rate sensor, all
of which are useful for exergames. The Ring Fit Adventure
is a popular exergame for the NX, which is a turn-based
RPG, where player movements and battle actions are based
on performing certain physical activities.
These consoles are non-VR platforms and successfully
stimulate exercise for players. However, exergames can be
further integrated with VR immersion to go beyond in terms
of VR exergames. We found that exergames in the VR market
can be classified under several types as follows: fitness, danc-
ing, boxing, arm workout, leg workout, sport simulation, and
competition. We can see that most of them are for casual fit-
ness, and virtual running is still lacking due to low user inter-
est, and the VR device may not appropriate for running in
place. However, virtual running can be applied for
adventure-based role-playing games (RPG), which can moti-
vate players to play exergames for a longer time. There is also
a gap in posture design for virtual running corresponding to
the hardware. It is related to the locomotion technique in VR,
where motion sickness is concerned. There are comments
from VR exergame reviews [14] that should be considered.
Many players state that the quality of the graphics had a par-
ticularly strong impact on perceived enjoyment, which may
invoke motion sickness if the frame rate is not high enough.
Furthermore, players disliked the games when the controls
were too complex.
In this research, we proposed exergames to the next level
with VR immersion with the design and development of vir-
tual running for VR exergames to entertain players and stim-
ulate exercise. The next section about VR headset trend was
introduced to demonstrate the future and trend of VR
devices. Then, Section 3 described the virtual running and
related works. Section 4 presented the design and develop-
ment of virtual running for the experiment. The study was
related to the criteria of usability, enjoyment, and motion
sickness. The virtual running techniques used in the experi-
ment were designed differently according to low exertion
and high exertion. Section 5 showed our experiment with
the results and discussion in Section 6. The experimental
results can support VR exergame developers in understand-
ing the usage and the effects of virtual running for players,
facilitating RPG exergame design. Finally, Section 7
described the conclusion and contribution of this work.
2. VR Headsets Comparison and Trend
The design of the VR exergame is mainly dependent on the
characteristics of the head mounted display (HMD) device
and controllers. Considering the various commercial VR
headsets with controllers that have been developed since
2017 (Tables 1 and 2), we found that Oculus, HTC, and Win-
dows Mixed Reality (WMR) are the advanced products cre-
ated by technology vendors this market. The Steam
Hardware Survey results for November 2020 [15] showed
the following as the most popular active devices for VR: Ocu-
lus 53.34%, HTC 21.89%, WMR 5.82%, and others 18.95%.
The Oculus Go is the first standalone VR headset of the
Oculus dedicated display and mobile computing hardware
including a handheld controller using relative motion track-
ing. However, it is a nonpositional 3DoF tracking. The Ocu-
lus Rift S and Oculus Quest were released in May, 2019. The
Oculus Quest is proposed to be a standalone device, while the
Rift S is still PC-tethered with more performance. The Ocu-
lus Insight technology is used for motion tracking bundled
with an iteration of the controllers, and the inside-out tech-
nology is the new standard for positional tracking of VR
headsets [16, 17]. Hand tracking is the new feature of the
Oculus Quest, using the on-board cameras to identify the
movements of the fingers and hands without handheld
controllers [18].
The HTC Vive has been launched since 2016, and it is the
achieved product of HTC with a large working area, high per-
formance display, and tracking accuracy [19]. The control-
lers provide a new approach for interacting in VR [20]. The
HTC Vive Pro is a PC-tethered with full feature set for hard-
core gamers upgraded their resolution display and Light-
house 2.0 base stations, allowing more space to 10 m 10 m.
2 International Journal of Computer Games Technology

Table 1: Hardware and technology comparison of Oculus and HTC VR headsets.
Device Oculus Rift CV1 Oculus Go Oculus Rift S Oculus Quest HTC Vive HTC Vive pro HTC Vive Focus+ HTC Vive Cosmos
Release
date 2016-03-28 2018-05-01 2019-05-21 2019-05-21 2016-04-05 2018-04-05 2018-11 2019-02-19
Type PC-tethered Standalone PC-tethered Standalone PC-tethered PC-tethered Standalone Standalone
Price 599$ 199$ 399$ 399$ 599$ 1299$ 599$ 699$
Display
and
resolution
OLED 1080 ×
1200,94
°
FOV
Dual fast-switch
LCD 1280 × 1440,
100
°
FOV
Dual fast-switch
LCD 1280 × 1440,
90
°
FOV
OLED 1440 ×
1600, 100
°
FOV
OLED 1080 × 1200,
110
°
FOV
AMOLED 1440 ×
1600, 110
°
FOV
AMOLED 1440 ×
1600, 110
°
FOV
LCD 1440 × 1700, 110
°
FOV
Refresh
rate 90 Hz 60-72 Hz 80 Hz 72 Hz 90 Hz 90 Hz 75 Hz 90 Hz
Tracking
technology
Accelerometer,
gyroscope,
magnetometer,
camera-based
3DoF, gyroscope,
accelerometer,
magnetometer,
proximity sensors
for detecting
5x cameras-based,
insight tracking,
accelerometer,
gyroscope,
magnetometer
4x cameras-based,
insight tracking,
accelerometer,
gyroscope,
magnetometer
Accelerometer,
gyroscope, 2
lighthouse base
stations with IR laser
emitters
Accelerometer, G-
sensor, gyroscope,
proximity, IPD
sensor, 2 base stations
2.0 with SteamVR
tracking
2x cameras-based,
G-sensor,
gyroscope, VIVE
wave, hand
tracking SDK
6x cameras-based, G-
sensor, gyroscope, IPD
sensor, eye tracking
Positional
tracking
Constellation
outside-in
through USB-
connected IR
LED sensor
Non-positional
3DOF tracking
Inside-out
(Oculus Insight)
Inside-out
(Oculus Insight)
Outside-in through
USB-connected IR
LED sensor
Outside-in through
SteamVR tracking Inside-out tracking Inside-out tracking
Controller
Oculus touch
motion tracked
controllers
Orientation-tracked
remote 3DOF
controller with
pointer capabilities,
touchpad with 3
buttons
2nd generation
Oculus touch
motion tracked
controllers with
6DOF
Inside-out with a
system button, 2
app buttons, a
trigger, a grip
button, and a
joystick
2nd generation
Oculus touch
motion tracked
controllers with
6DOF
Inside-out with a
system button, 2
app buttons, a
trigger, a grip
button, and a
joystick
SteamVR wireless
motion tracked
controllers with a
track pad, grip
buttons, and a dual-
stage trigger
SteamVR wireless
motiontracked
controllers with a
track pad, grip
buttons, and a dual-
stage trigger
Dual 6DoF
controllers with
ultrasonic tracking,
2 trigger buttons, a
trackpad, 2 face
buttons
Dual 6DoF controllers
with system buttons, 2
app buttons, a trigger,
a bumper, a grip
button, and a joystick
Note that only headset products with the controllers included.
3International Journal of Computer Games Technology

Table 2: Comparison of HMDs compatible with Windows Mixed Reality (WMR) and SteamVR platform.
Device Lenovo Explorer Dell Visor Acer AH101 Asus HC102 Samsung
Odyssey+ Valve index HP Reverb Pimax 8 K PLUS
bundle
Release
date 2017-10-4 2017-10-17 2017-10-17 2018-02-20 2018-10-22 2019-05-01 2019-05-6 2019-12-16
Type PC-tethered PC-tethered PC-tethered PC-tethered PC-tethered PC-tethered PC-tethered PC-tethered
Price 449$ 450$ 399$ 399$ 500$ 999$ 599$/649$ 1399$
Display
and
resolution
LCD 1440 × 1440,
110
°
FOV
LCD (RGB
subpixel) 1440 ×
1440, 110
°
FOV
LCD 1440 × 1440,
95
°
FOV
LCD 1440 × 1440,
95
°
FOV
AMOLED
1440 × 1600,
110
°
FOV
LCD 1440 × 1600,
130
°
FOV
LCD 2160 × 2160, 114
°
FOV
LCD 3840 × 2160,
200
°
FOV
Refresh
rate 90 Hz 90 Hz 90 Hz 90 Hz 60 Hz/90 Hz 90 Hz, 120 Hz,
144 Hz 90 Hz 60-110 Hz
Tracking
technology
2x cameras-based,
accelerometer, gyro
sensor,
magnetometer,
proximity
2x cameras-based,
accelerometer, gyro
sensor,
magnetometer,
proximity
2x cameras-based,
accelerometer, gyro
sensor,
magnetometer,
proximity
2x cameras-based,
accelerometer, gyro
sensor,
magnetometer,
proximity
Gyroscope, 3-
axis compass,
proximity
sensor, IPD
sensor
SteamVR tracking,
2.0 lighthouse based
stations
2x cameras-based,
gyroscope,
accelerometer,
magnetometer
SteamVR tracking,
2.0 lighthouse based
stations
Positional
tracking Inside-out tracking Inside-out tracking Inside-out tracking Inside-out tracking Inside-out
tracking Outside-in tracking Inside-out tracking Outside-in tracking
Controller
6DOF controller
with haptic
feedback, thumb
stick, touchpad,
analog trigger, grap
button, Windows,
and menu button,
compatible with
Xbox One
Controller
6DOF controller
with haptic
feedback, thumb
stick, touchpad,
analog trigger, grap
button, Windows,
and menu button,
compatible with
Xbox One
Controller
6DOF controller
with haptic
feedback, thumb
stick, touchpad,
analog trigger, grap
button, Windows,
and menu button,
compatible with
Xbox One
Controller
6DOF controller
with haptic
feedback, thumb
stick, touchpad,
analog trigger, grap
button, Windows,
and menu button,
compatible with
Xbox One
Controller
6DoF dual
controllers
tracked by
HMD and
also
compatible
with Xbox
One
Controller
Knuckle controllers
with all fingers
detection through
capacitive sensors,
each controller
features a system
button, thumbstick,
touchpad, analog
index trigger, grip
button
6DoF dual controllers
tracked by HMD with
multifunction
touchpad, menu
button, Windows start
button, grab button,
thumbstick, trigger,
and also compatible
with Xbox One
Controller
Knuckle controllers
with all fingers
detection through
capacitive sensors,
each controller
features a system
button, thumbstick,
touchpad, analog
index trigger, grip
button
4 International Journal of Computer Games Technology

The Vive Trackers are integrated to support more tracking
and interactions. The HTC Vive Focus+ is the first standa-
lone VR headset of HTC with inside-out tracking technology.
The controller is redesigned to use 6DoF ultrasonic tracking.
The VIVE wave and hand tracking SDK offer an open inter-
face enabling interoperability between numerous mobile VR
headsets and accessories, supporting mainstream game
engines with a cross platform tool to track hand position
and gesture recognition. The HTC Vive Cosmos refined
inside-out tracking with six camera sensors. The controller
is upgraded from the HTC Vive Focus with precision joy-
sticks. Although all HTC Vive headsets are PC-tethered, the
wireless adapter with the WiGig technology can transform
HTC Vive to a standalone headset [21, 22].
WMR is a mixed reality platform based on the Windows
10 operating system, providing mixed reality experiences
with compatible HMDs from Acer, Dell, HP, Lenovo, and
Asus [23, 24]. Most WMR devices have the same specifica-
tion with PC-tethered and inside-out motion tracking fea-
tures integrated. The handheld motion controllers are
included. In addition, all WMR headsets are compatible with
the Xbox controller. The Samsung Odyssey+ come up with
the Anti-SDE (Screen-Door Effect) filter applied to the dis-
play and also has a slightly different controller design which
makes it different from the other WMR headsets [25] but also
compatible with the SteamVR platform. The Valve Index is a
high-end HMD, which is also compatible with the SteamVR
platform. The headset uses an improved version of the Light-
house tracking system. It has a 120 Hz display, with an option
up to 144 Hz, producing realistic VE even moving head. The
knuckle controllers come with all fingers trackable, which is
one of the most effective VR controllers [26]. The Pimax is
a HMD device with outside-in technology that focuses on
high resolution and a wide range of display up to 8K. The
new Pimax 8K PLUS comes with a bundle set, including
the Valve Index Controllers and SteamVR Base Stations.
The inside-out technology for tracking will come with expan-
sion modules in the next generation [27].
From the survey of the VR headsets available in the mar-
ket and the development of new devices, we found that there
is a trend for more standalone wireless devices that come
with inside-out tracking technology. The controller is more
effective in finger detection, which supports the development
of applications and more user interaction. However, applying
this technology to virtual running may not be suitable for sta-
tionary running because the headsets are still heavy. We have
to wait for new technology on HMD glasses to become com-
pact in size. Therefore, the virtual running design for VR
exergames should be physically consistent with movement,
which is a motion-based locomotion technique designed by
focusing on exertion by various parts of the body.
3. Virtual Running and Related Works
Virtual running is directly related to VR locomotion, which
comes in many forms. The motion-based locomotion is the
most suitable form that can be applied to exercise because it
must be activated by user movement. Related research
includes the application of transforming user movement into
locomotion in VR. However, motion-based locomotion was
previously developed with a focus on UI and UX [28] rather
than for exercise.
3.1. Motion-Based Locomotion. Motion-based locomotion is
a continuous locomotion through physical interaction of
the user’s movement in the VR space [28] and relies on the
guardian-based or room-scale-based locomotion. The
guardian-based locomotion focuses on stationary interaction
instead of moving around as with the room-scale-based
locomotion.
Cherni et al. [29] explored locomotion techniques in VR
between 2012 and 2019. Several studies applied body motion
for locomotion to enhance spatial perception and user expe-
rience, which increases the sensation of one’s own self-
motion. The techniques related to exercise are classified as
the following user-body centered techniques: leaning based
and walk simulation. Leaning-based simulation is separated
as arm-based, head-based, and trunk-based simulation. All
techniques rely on a part of the body, which is referred to
in its name. Head-based and truck-based simulations use
physical leaning or tilting in a standing or seating position,
focusing on control through posture.
The arm-based simulation uses the human arm to con-
trol movement, which is called the arm-swinging method.
Arm-Swinging is one of the techniques that lets the user
move forward when they move their arms. This technique
has many different approaches. Some research use the Myo
Armband, a forearm tracking device with various features
[30–32]. The direction of the armband is tracked to check
whether the user moves their arm or not, which is used for
movement tracking. The results [30, 32] showed that this
method outperformed the simple joystick in spatial orienta-
tion and that it is comparable to physically walking on foot.
Walking-in-place (WIP) is a technique to simulate as
similar as possible the manner of walking through the real
world, where feet movement is tracked and then translated
to movement in the virtual environment with the room-
scale-based locomotion. However, the WIP locomotion tech-
nique requires using additional hardware, and the interaction
area is limited by the room size [32, 33]. There were research
results showing that physical locomotion outperformed both
walking-in-place and arm-swinging in terms of spatial
awareness.
Gesture-based technique is one of motion-based tech-
niques that uses hand gestures to control movements in
VR. Some VR headsets do not have a hand tracking sys-
tem and require additional accessories. The Leap Motion
device has been modified allowing users to use hand ges-
tures, which are already defined [34, 35]. However, nowa-
days, HMD devices have built-in cameras with inside-out
technology to enable hand tracking allowing users to inter-
act with their hands.
3.2. Applications of Motion-Based Locomotion. From the lit-
erature review, there are many studies that have applied vir-
tual running or VR gaming for exercise. In general, these
studies focused on common issues, including usability,
motion sickness, and enjoyment of playing game. Following
5International Journal of Computer Games Technology

research are related to see their studies including the results
and can be used to be a basis for our work.
Yoo and Kay [36] presented VRun, a virtual running
exergame allowing players to jog in place through a virtual
world. The experiment has been tested on three systems: a
laptop, a large display, and a HMD using Google Cardboard.
The HMD received the best usability score from the SUS
(System Usability Scale), while the most immersive system
was the large display. The result showed that HMD is useful
for virtual running. However, the large display may be more
acceptable for longer exercise periods.
Next, VRmove [3] was proposed as a design framework
for VR games. This framework analyzed data on the basis
of exertion and enjoyment. They demonstrated the use of
the VRmove framework to inform the design of a new game.
The results showed that enjoyment was linked to light exer-
tion and multiple movements. The Goldilocks effect was
the highlight with the design of VR exergames. The balance
of exergames should take into account the level of exertion
while being engaging enough to distract players from actually
feeling fatigued.
The evaluation of the actual and perceived exertion from
VR games [37] was presented to compare gaming with exer-
cise. The contribution was in the insights about the exertion
in gaming, which highlighted the potential value for improv-
ing exercise levels. The Borg score was used to compare VR
games to be considered as exercise. Fruit Ninja was compara-
ble to walking, while Hot Squat to running, and Holopoint to
dancing. The results from this study pointed towards VR as
being able to deliver enough exertion to be considered as
exercise in terms of both the cardiovascular level and partic-
ular muscle groups.
There was research that applied the WIP method using
position and orientation tracking called jog-in-place [38].
This research proposed a recognition method to distinguish
a motion and transform it to virtual velocity. Their method
did not recognize the squatting motion but included WIP
steps. This locomotion provided users with a natural naviga-
tion experience, which can be used to walk in an infinite vir-
tual environment in VR applications. The results showed that
virtual velocity worked with an accuracy of 99.32%. There
were some participants that complained about dizziness,
but none experienced motion sickness.
RabbitRun [39] is a serious game with an immersive
experience to engage low back pain (LBP) patients in a
virtual environment and distract them from the pain while
performing LBP exercises. The usability evaluation results
showed that the RabbitRun game was enjoyable and
acceptable. Moreover, it was easy to learn and play, and
most of the participants were willing to play the game at
home. This game enhanced the rehabilitation outcome of
patients and can be used to motivate patients to move
and train for longer periods.
There was a research presenting Rift-a-bike [40], an
immersive VR game that enhances the exercise bike experi-
ence with challenges, levels, points, badges, and prizes to sup-
port physical exercise. Their results showed that gamification
increased the user’s enjoyment during the physical activity.
They also provided guidelines for applying gamification fea-
tures for fitmersive games. According to relevant research, we
found that VR exergames have many positive effects on exer-
cise [41, 42], such as motivating players to exercise, attracting
players to exercise for longer periods, and enhancing player
enjoyment. The results showed that playing VR exergames
can be compared to real exercise [43]. From previous
research, the HMD device is useful for virtual running
and increases immersion, while the large display is better
for longer periods of exercise. Virtual running is an inter-
esting point for VR exergames using HMD. The solution
to make users comfortable and acceptable is a concern in
our study. Therefore, it is interesting to study how exer-
tion should be for short and long exercise periods. The
design of different postures for virtual running is proposed
with light and heavy workouts to investigate usability,
motion sickness, and enjoyment of users.
4. Design and Development
The exercise posture consists of four exercise types: aero-
bics, strength, flexibility, and balance [44]. The aerobic
exercise such as walking, running, swimming, biking, and
dancing is the key to build and maintain cardio endur-
ance. The strength conditioning such as squat is the exer-
cise to strengthen the bones, muscles, and connective
tissues. The balance training such as yoga poses can
improve stability to keep body upright, including legs
and core. The body balance makes moving easier and pre-
vent injury. The flexibility is the activities that lengthen
and stretch muscles or functional abilities, such as reach-
ing, bending, or stooping. Initially, this research focused
on physical exercise postures to study the effects of exer-
cise postures on virtual running. The aerobic and strength
are the exercise using more exertion than balance and flex-
ibility. Therefore, we can use these two postures to inves-
tigate appropriate duration time and relationship between
physical fitness and user preference.
As for the VR headset trend review, we can see that the
inside-out tracking technology used with tracking controllers
is the trend of VR devices, and the Oculus Rift S also supports
this technology. Therefore, the design of the interaction tech-
nique was based on the Oculus Rift S with Touch controllers.
We separated virtual running techniques into two cases: low
exertion and high exertion. The ArmSwing virtual running
was a representation of less exertion using arm swinging for
locomotion, while the Squat virtual running was a represen-
tation of high exertion using squat posture for locomotion.
The implementation using OVRinput on Unity3D
(Figure 1) of our virtual running locomotion techniques is
defined with the following details:
4.1. ArmSwing Virtual Running. In the design of virtual run-
ning with arm swinging, the biomechanics of real walking
[45, 46] was considered using gait cycle to describe the
human walking pattern, which can be applied to the design.
However, gait cycle [47] consists of two main phases: (1)
stance and (2) swing, which is the posture of the legs and
arms, respectively. Recently, VR headset devices as reviewed
can detect only the position of the head and hands from the
6 International Journal of Computer Games Technology

HMD and controllers, but the leg position cannot be detected
without the use of accessories. In this research, we were
restricted to the VR headset without using other accessories.
We designed the use of arm swinging to move the arm only
with the HMD and controllers. Therefore, we designed the
locomotion movement by swinging the arms, which was cal-
culated by the speed from the distance of the controllers
being swung, while the locomotion direction was calculated
from the position and direction of the controllers. Biome-
chanics indicate that walking speed (v) is related to step fre-
quency (f) and step length (l) [47] in the following equation:
v=f×l:ð1Þ
The continuous walking or running movement has arms
and legs alternating with the similar frequency and pattern
of swinging at a particular speed value [48]. Then, we
claimed that the ratio of arm swinging and leg striding dis-
tance has the constant value and transformed the (1) equa-
tion to use with ArmSwing virtual running. The step
frequency (f) can be changed to step frame-by-frame, and
step length (l) can be changed to the distance of arm swing-
ing for each frame. The walking speed (v) was transformed
to the virtual running speed by calculation of the arm swing-
ing distance for each frame.
The arm swinging distance is calculated from the dis-
tance of the controller positions frame-by-frame, and both
distances are averaged together. If the value is greater than
the specified swing threshold, the user avatar will be moved
from the original position of H
!of the previous frame. The
direction of movement is calculated from the unit vector of
L
!(left controller direction) and R
!(right controller direction)
as in Figure 2.
Movement speed is calculated from the average control-
ler distance multiplied by the acceleration threshold. Since
A
RawAxis2D.Lumbstick
RawButton.Lumbstick
(le stick press)
RawButton.Y
RawButton.X
RawButton.Start
RawAxis1D.
RHandTrigger
RawAxis1D.
LIndexTrigger
RawAxis1D.LHandTrigger
RawAxis1D.
RIndexTrigger
RawAxis2D.Rumbstic
k
RawButton.Rumbstick
(right stick press)
RawButton.B
RawButton.A
Reserved
Y
X
B
Figure 1: OVRInput API for unity uses to query controller state.
Data:LHandT rigger, RHandT rigger, L
!,R
!, and H
!as trigger controllers and
position of left hand, right hand, and head, respectively.
Result: Locomotion by arm swinging applied on H
!.
begin
swingthreshold ,accthreshold ,inert hreshold ←swing range, acceleration, inertia
forframe i =1toNdo
ifLHandTrigger is TrueandRHandTrigger is Truethen
lL←kL
!
i−L
!
i−1k
lR←kR
!
i−R
!
i−1k
lswing ←lL+lR/2
if
fswing >swingthreshold
then
H
!
←H
!+ðlswing accthreshold Þð L
!+R
!/kL
!
i+R
!
ikÞ
else
H
!
←inerthr eshold H
!
else
H
!
←0
!
Algorithm 1: ArmSwing virtual running.
7International Journal of Computer Games Technology

the motion is a frame-by-frame calculation, the speed result
respects to the continuous movement. However, if the user
stops swinging their arms but still presses the LHandTrigger
and RHandTrigger buttons, the movement result will gradu-
ally slow down according to the inertia threshold until the
motion is stopped. If the user release the LHandTrigger and
RHandTrigger buttons, the movement will stop immediately.
4.2. Squat Virtual Running. The squat is an excellent lower
body workout, and it is a popular exercise that can be done
without using any equipment. Since the purpose of this
research needed a design with a high exertion posture greater
than swinging arms, the squat posture was applied to virtual
running. People have different physiology, including differ-
ences in the length of the thigh bone, body, and flexibility
[49]. Each person has a different squat posture depending
on their physiology. The design, therefore, emphasized the
continuous squat up and down without having to rush, but
the weight must always be put on both feet [50].
From equation (1), we can also apply this equation to use
with Squat virtual running. Step frequency (f) can be changed
to step frame-by-frame, and step length (l) can be changed to
the distance of squatting up and down. Then, the walking
speed (v) can be transformed to the virtual running speed by
calculation of the squatting distance for each frame.
The squat distance is calculated from the distance of
moving the HMD position up and down following the squat
posture. Therefore, only the Y-axis distance (Hyj
!) of the
HMD is calculated as detailed in Figure 3. If the value is
greater than the specified squat threshold, the user avatar will
be moved from the original position of H
!of the previous
frame. We added the gaze angle threshold to avoid a wrong
movement when the user bent or lifted his face more than
the specified degree. Since the motion is a frame-by-frame
calculation, the result is a continuous movement with the
same direction of the HMD. Similar to ArmSwing virtual
running, if the user stops swinging their arms but still presses
H
L
R
Figure 2: Virtual running with the ArmSwing posture.
Data:LHandT rigger, RHandT rigger, H
!, and γas trigger controllers, head
position and angle between H
!and Z-axis, respectively.
Result: Locomotion by squat applied on H
!.
begin
squatthr eshold ,gazethr eshold ,accthreshold ,inerthreshol d ←squat range, gaze, angle, acceleration, inertia
forframe i=1toNdo
ifLHandTrigger is TrueandRHandTrigger is Trueand
0<γ<gaze
threshold
then
lsquat ←Hyi−Hyi−1
if
fsquat >squatthr eshold
then
H
!
←H
!+ðlsquat accthreshold ÞðH
!/kH
!kÞ
else
H
!
←inerthr eshold H
!
else
H
!
←0
!
Algorithm 2: Squat virtual running.
8 International Journal of Computer Games Technology

the LHandTrigger and RHandTrigger buttons, the move-
ment result will gradually slow down according to the inertia
threshold until the motion is stopped. If the user releases the
LHandTrigger and RHandTrigger buttons, the movement
will stop immediately.
5. Experiment
The objective of this research was to evaluate the perfor-
mance of virtual running for exercise. Our experiments were
held on the university campus, announcing 30 volunteers via
social media. The method started with the introduction,
warm-up, and testing VR headset. Then, each participant
was required to test both virtual runnings. We designed dif-
ferent postures of virtual running to represent different exer-
tions, which included the ArmSwing and Squat. The
ArmSwing posture was a representation of aerobic exercise
for virtual running, while the Squat posture was a representa-
tion of strength exercise. Usability and motion sickness of
both the ArmSwing and Squat postures were evaluated to
see the different in terms of usage. Then, the performance
of each posture in terms of time and heart rate was assessed.
In addition, the enjoyment during usage was assessed to see
the differences of each posture. Both results were used to ana-
lyze the advantages and disadvantages of each posture. The
final step was an interview to confirm the results of the exper-
iment. This experiment was focused on VR exercises that
caused tiredness or fatigue, where participants could stop or
continue whenever they desired.
5.1. Virtual Environments. In our experiment, the scene was
designed to assess the virtual running techniques, which
included ArmSwing and Squat. The scene was a racetrack dec-
orated with the surrounding atmosphere. The virtual environ-
ments used in the experiment can be downloaded for free
from the asset store in the Unity Game Engine. The assets
and virtual environments were adapted and resized for our
virtual running test. Participants were required to exercise by
virtual running, which took around 1-2 minutes from the start
to finish point. All participants used the same route in the des-
ignated area so that everyone ran the same distance.
5.2. Evaluation. After testing, participants were measured for
their heart rates. Then, they were allowed to take some rest
and answered three questionnaires. The heart rate was mea-
sured to check the user’s fatigue of each posture in order to
confirm the exertion of the virtual running design. The Arm-
Swing posture was designed as low exertion virtual running. It
should result in a lower heart rate than the Squat posture,
which was designed as high exertion virtual running. While
the participants were taking their rest, they were given three
questionnaires about motion sickness, system usability, and
user enjoyment of the physical activity. After completing the
experiment, there was an interview about their preferences
for comparison of the two different virtual running postures.
5.3. Questionnaires and Interview. There were three question-
naires in our experiment. The first questionnaire was about
motion sickness, and simulator sickness was used to assess
symptoms of users while using the HMD (Oculus Rift S).
The second questionnaire was about system usability, which
was used to assess the design of the virtual running postures.
The last questionnaire was about physical activity enjoyment
to assess user enjoyment while using the virtual running pos-
tures. Finally, at the end of experiment, there was an interview
to compare user preferences for both virtual running postures.
The Simulator Sickness Questionnaire (SSQ) is related to
a type of motion sickness that is experienced in training. The
awareness of the differences between simulated motions in a
virtual environment and user movements can occur and lead
to simulator sickness [51, 52]. SSQ symptoms indicate three
constructs of simulator sickness, which are nausea, oculomo-
tor, and disorientation. These symptoms can reduce the effi-
ciency of virtual running and result in systematic effects such
as decreased simulator use and safety, which can affect user
preference. The users gave a score between 1 and 5 points
to tell how much they felt with each symptom. “1 point”is
strongly severe, “2 points”is severe, “3 points”is moderate,
H
yj
H
Xi
H
Zk
Figure 3: Virtual running with the Squat posture.
9International Journal of Computer Games Technology

“4 points”is slight, and “5 points”is no symptom. The SSQ
questionnaire was divided into different subscales as follows:
(i) Comfortable. The overall sensations of the user dur-
ing usage
(ii) Nausea. Symptoms such as increased salivation,
sweating, stomach awareness, and burping
(iii) Oculomotor. Symptoms such as fatigue, headache,
eyestrain, and difficulty focusing
(iv) Disorientation. Symptoms such as vertigo, dizziness,
and blurred vision
The System Usability Scale (SUS) was developed as a tool
to be used in usability engineering [53, 54] to provide a rough
estimate of a system’s ease of use. This questionnaire is a
Likert scale questionnaire, and the results obtained from this
test were quantitative. This SUS questionnaire contained 10
items for users to rate. The users gave a score between 1
and 5 points to tell how much they agreed with each item.
“1 point”is strongly disagreed, and “5 points”is strongly
agree. The questionnaire of SUS is described as follows:
(i) I think that I would like to use this system
frequently
(ii) I found the system unnecessarily complex
(iii) I thought the system was easy to use
(iv) I think that I would need the support of a technical
person to be able to use this system
(v) I found the various functions in this system were
well integrated
(vi) I thought there was too much inconsistency in this
system
(vii) I would imagine that most people would learn to
use this system very quickly
(viii) I found the system very cumbersome to use
(ix) I felt very confident using the system
(x) I needed to learn a lot of things before I could get
going with this system
The SUS score of both virtual running postures were evalu-
ated to see the performance in terms of efficiency, effectiveness,
and overall ease of use. Each participant rating had a scale of 0-
100 by calculating the answers with the formula as follows:
(i) X = Sum of the points for all odd-numbered ques-
tions –5
(ii) Y = 25 –Sum of the points for all even-numbered
questions
(iii) SUS Score = ðX+YÞ×2:5
The average SUS score was 68, which meant that system
was an okay system above 50th percentile. The SUS scores
can be divided into the following levels: >80.3 excellent, 68-
80.3 good, 68 okay, 51-68 poor, and <51 awful.
The Physical Activity Enjoyment Scale (PACES) is a
Likert Scale questionnaire as a tool to evaluate user enjoy-
ment when doing physical activity such as rehabilitation or
exercise. The 8-items PACES version [55] was used instead
of the original 18 items, and the revised scale is invariant
and a valid measurement to assess enjoyment of physical
activity. We adapted the questions to fit our experiment. Par-
ticipants were asked to rate the question, “How do you feel at
the moment about the virtual running you have been doing?”
using a5-point rating scale similar to the SUS questionnaire.
Higher PACES scores reflected greater levels of enjoyment.
The 8-item PACES questionnaire is described as follows:
(i) I find it pleasurable
(ii) It is a lot of fun
(iii) It is very pleasant
(iv) It is very invigorating
(v) It is very gratifying
(vi) It is very exhilarating
(vii) It is very stimulating
(viii) It is very refreshing
The interview at the end of experiment consisted of sev-
eral questions about user preference on low exertion and high
exertion of virtual running. We wanted to know user’s feel-
ings about the use of the virtual running exercises and to
compare their preferences in terms of design, exertion, and
playing time. The answer may be one of the postures they
preferred or may include both of them equally the same.
(i) Which virtual running posture do you prefer to use
(ii) Which virtual running posture do you prefer if play-
ing for a long period
(iii) If playing for a short period, which virtual running
posture do you prefer to use
5.4. Research Protocol. In our experiment, 30 participants
who volunteered were given step-by-step explanations as
shown in Figure 4. The introduction and importance of this
research are explained. The operation of virtual running
was described, which included the Arm-Swing and Squat
posture. Before testing, the participants were encouraged to
warm up because the experiment involved the use of muscles.
The participants were notified that after completing each vir-
tual running they had to answer the questionnaires and inter-
view at the end. Furthermore, the participants were notified
that using the Oculus Rift S with the controllers could cause
symptoms during the test, and they could stop the experi-
ment at any time. When the participants understood every-
thing clearly, then the experiment began.
Since each participant was required to assess two virtual
running postures, the sequence in the experiment affected
10 International Journal of Computer Games Technology

their fatigue and evaluation. Therefore, all participants were
separated into two groups with fifteen participants in each
group to switch starting postures and avoid bias from exer-
tion. All groups were tested two times with the ArmSwing
and the Squat posture. Figure 5 shows the selection menu
before starting and participants during the experiment as
shown in Figure 6. After each experiment, the participants
must remove the HMD device and complete the posttest
evaluations by answering the SSQ, SUS, and PACES ques-
tionnaires before moving on to test the next virtual running
posture. Finally, participants had to complete the experiment
with an interview to compare their preferences on the two
different postures.
6. Results and Discussion
The results of the study were collected from 30 participants,
consisting of 18 men and 12 women. The average age of all
participants was 27.40 years from 19 to 45 years old. The
average weight was 70.23 kilograms from 42 to 120 kilo-
grams. The average height was 165.40 centimeters from 152
to 187 centimeters. The Shapiro-Wilk test was used to check
the normal distribution of the observations and the results of
all questionnaires. The SSQ, SUS, and PACES results were
not normal distributed, which were considered as nonpara-
metric data. Then, a nonparametric Wilcoxon test was used
to evaluate the differences between ArmSwing and Squat,
where the samples of the participants were the same and con-
sidered as dependent samples. The results of the experiment
were divided into the following areas.
6.1. Heart Rate and Running Time. The results of the virtual
running testing showed that after using the ArmSwing pos-
ture, the participants had an average heart rate of 88.53
BPM, with an average time of 94.77 seconds. For virtual run-
ning with the Squat posture, the participants had an average
heart rate of 114.97 BPM with an average time of 112.83 sec-
onds. We can see that our testing in a short period of time,
the aerobic exercise of the virtual running design with Arm-
Swing had a lower heart rate than strength exercise with
Squat. In addition, participants used less time to run with
ArmSwing than Squat because the strength exercise required
more exertion, which made them more tired.
6.2. Motion Sickness Result. The results from the SSQ ques-
tionnaire (Table 3) showed that virtual running with Arm-
Swing and Squat had no differences in motion sickness (p
value =0.34722>0.05). In addition, 79.17% of the participants
had no abnormal symptoms between running with both pos-
tures. That meant both virtual running designs can be
adapted for locomotion in VR. The participants had very lit-
tle motion sickness during the virtual running experience.
Moreover, both designs can be applied to exergames, where
this movement is necessary in RPG game styles with VR
and locomotion.
6.3. Usability Results. The SUS results from the questionnaire
were converted to 0-100 points to evaluate the usability per-
formance of virtual running. The results showed that the
converted SUS scores from participants for the virtual run-
ning experience with ArmSwing and Squat were 86.5 and
82.67, respectively. The results implied that both virtual run-
ning postures’usability performance was excellent, and the
score of the ArmSwing was slightly more than the Squat.
In the aspects of effectiveness and efficiency by learnabil-
ity and usability, the learnability dimension can be evaluated
from the subscales at items 4 and 10. The average scores of
items 4 of ArmSwing and Squat were 2.07 and 2, respectively.
That meant a few of both virtual running participants
thought they needed a technical person to support using this
system. Similarly, the average score of items 10 of ArmSwing
and Squat equaled 1.5 and 1.7, respectively. That meant a few
of both virtual running participants thought they needed to
learn a lot to use this system. The usability dimension can
be evaluated from the other eight items, and the results were
in the same direction as the SUS score. It suggested that both
virtual running postures were effective and efficient to use.
However, when testing the differences with the Wilcoxon
signed-rank test (Table 3), it was found that the SUS scores of
both postures were significantly different (pvalue =0.03318<
0.05). Therefore, the design of virtual running with the Arm-
Swing posture was better than the Squat posture in terms of
usability.
6.4. User Enjoyment Results. The user enjoyment of virtual
running was measured using the PACES questionnaire.
The results (Table 3) showed that the enjoyment of virtual
running with ArmSwing and Squat had no differences (p
value =0.29372>0.05). The enjoyment scores from the
PACES questionnaire with a total score of 40 for the vir-
tual running experience with ArmSwing and Squat were
34 and 34.47, respectively. The highest average scores from
the PACES questionnaire for the ArmSwing and Squat
postures were “It’s very invig- orating”with scores of 4.4
and 4.67, respectively. It suggested that the participants felt
that using both postures in the virtual running experience
stimulated exercise, feeling strong, feeling healthy, and
feeling full of energy.
1.1 ArmSwing
1.2 Squat
10 mins
5–10 mins 5–10 mins
HR
SSQ
2
SUS
PACES
HR
SSQ
SUS
PACES
3 mins
Interview
User preference
2.1 Squat
2.2 ArmSwing
Virtual runningVirtual running
Introduction
- Explanation
- Consent form
- Warm up
Groups
Figure 4: Research protocol with two groups of participants in the experiment.
11International Journal of Computer Games Technology

6.5. Interview Results. Our first question was about which vir-
tual running posture the participant preferred in general
from overall usage. The results of the interview (Figure 7)
showed that 17 participants preferred the ArmSwing posture,
9 participants preferred the Squat posture, and 4 participants
liked both postures equally. Considering the overview of
results from usability, motion sickness symptoms, and enjoy-
ment, it can be seen that the participants preferred virtual
running with ArmSwing more than Squat.
In the second question, we asked about the duration of
the virtual running exercise. If the participant had to exercise
with these two postures for a long period of time, which pos-
ture would they prefer? The results of the interview showed
that 27 participants preferred the ArmSwing posture, with
only 2 participants who preferred the Squat posture more
and only 1 participant who voted both postures the same. It
can be seen that 90% of the participants preferred the aerobic
exercise (low exertion) more than the strength exercise (high
exertion) of virtual running when having to play for a long
period.
In the third question, we asked the same as the second
question but modified it to ask which posture they would
prefer for a short period of time? The results showed that
12 participants preferred the ArmSwing posture, whereas
17 participants preferred the Squat posture more and only
1 participant voted both postures the same. It can be seen
that 56.7% of the participants preferred strength exercise
(high exertion) when playing only for a short time. However,
even if just exercising for a short period of time, 40% of the
participants still chose the aerobic exercise (low exertion).
6.6. Association Results. The Pearson correlation coefficient
[56, 57] was used to measures linear correlation between
two variables. The value is between +1 and -1, where 1 is total
positive linear correlation, 0 is none correlation, and -1 is
total negative linear correlation. We used Pearson correlation
Figure 5: User interface of virtual running selection.
Figure 6: Participants during the experiment testing.
Table 3: The results of the Wilcoxon signed-rank test for motion
sickness, usability, and enjoyment (∗significant p<0:05,∗∗highly
significant p<0:01).
Motion sickness Usability Enjoyment
ArmSwing 19 86.5 34
Squat 18.87 82.67 34.47
W27 61 103.5
z-0.9414 -2.1265 -1.0493
pvalue 0.34722 0.03318 0.29372
Significant ∗
12 International Journal of Computer Games Technology

coefficient to measure linear correlation between gender, age,
and BMI with the SSQ, SUS, and PACES scores of both
postures. It was found that these values were not correlated
to each other.
However, when we studied the relationship between the
ArmSwing posture and the Squat posture, we found that
the scores from questionnaires of both postures were corre-
lated with highly significant (Table 4). In addition, when
the players had less motion sickness in the ArmSwing pos-
ture, they also have less motion sickness in the Squat posture.
When the players felt that usability of the ArmSwing posture
was good, the Squat posture was also good usability. And
when the players enjoyed with the ArmSwing posture, they
also enjoyed with the Squat posture. When we studied the
relationships between SSQ, SUS, and PACES scores on each
posture, we found that the scores between SUS and PACES
were correlated (Table 5); the ArmSwing posture was highly
significant (pvalue =0.0029<0.01), and the Squat was signif-
icant (pvalue =0.0386<0.05).
It implied that gender, age, and BMI have no effect on vir-
tual running of both aerobic exercise and strength exercise.
However, it was found that both ArmSwing and Squat pos-
tures were correlated to motion sickness, usability, and user
enjoyment on the same way. And in any posture, when the
players felt that the posture was good usability, the players
were also enjoyed to that posture (the usability is direct var-
iation to the user enjoyment).
6.7. Discussion. The inside-out tracking technology is the
trend of VR headsets with useful controllers and hand track-
ing. Exergames by this VR tracking will be the future with
immersive gameplay while doing exercise. Motion-based
locomotion in VR is related to physical interaction, which
walking or running simulation can apply for the exercise.
This is interesting for designing VR exergames that should
be physically consistent with movement focusing on exertion
by various parts of the body.
This research studied the criteria of usability, enjoyment,
and motion sickness. Our experiment and interview results
showed that virtual running with ArmSwing is preferred for
an extended period than Squat but preferred Squat in a short
period. Most research [30–32] have better results of WIP
than ArmSwing in terms of usability and motion sickness.
However, the trend of VR headsets is going to use inside-
out tracking. ArmSwing and Squat are interesting to promote
as a virtual running posture. Our heart rate result of Arm-
Swing was 88.53 BPM (99.77 seconds), which was not much
different from [31] with 95 BPM (mission completed). In
comparison, the heart rate result of Squat was 114.97 BPM
User_preference Long_period Short_period
6.7% 3.3%
40.0%
56.7%
3.3%
90.0%
56.7%
30.0%
13.3%
VirtualRunning
ArmSwing
Squat
Both
20
10
0
Count
Figure 7: Results from the interview with questions about preference and playing time.
Table 4: The association by the Pearson correlation coefficient
between ArmSwing and Squat in terms of motion sickness,
usability, and enjoyment (∗significant p<0:05,∗∗highly significant
p<0:01).
Motion sickness Usability Enjoyment
R0.8595 0.8286 0.8434
Tstatistic 8.8973 7.8315 8.3075
pvalue 1.1882E-09 1.5699E-08 4.8671E-09
Significant ∗∗ ∗∗ ∗∗
Table 5: The association by the Pearson correlation coefficient
between SUS and PACES scores of the ArmSwing and the Squat
(∗significant p<0:05,∗∗ highly significant p<0:01).
SUS-PACES (ArmSwing) SUS-PACES (Squat)
R0.5249 0.3794
Tstatistic 3.2634 2.1698
pvalue 0.0029 0.0387
Significant ∗∗ ∗
13International Journal of Computer Games Technology

(112.83 seconds) related to WIP [31] with 120 BPM (mission
completed).
In the interview, participants voted Squat because they
want more exertion if they were in the short period of
gameplay. However, they relaxed when using ArmSwing
by a little exertion. They felt the mission have no challenge,
which can make the exergame boring. This suggestion
implied that we should design VR exergames with appropri-
ate exertion and challenge. Moreover, intensive exertion
during gameplay can make players discouraged and do
not want to continue playing even though the enjoyment
for exergames did not mention too much in VR previously.
Our PACES result suggested that their experience in Arm-
Swing and Squat stimulated exercise with enjoyment feeling.
Design and development of RPG exergame in VR should
adjust effort for varying exertion difficulty to avoid boring
and encourage challenge.
7. Conclusion
This work presented exercise through virtual running. The
two postures were proposed with ArmSwing and Squat as
the representation of aerobic and strength exercise, respec-
tively. Virtual running with ArmSwing required low exer-
tion, whereas the Squat required high exertion based on the
posture and heart rate. The results showed that motion sick-
ness and user enjoyment of both postures were not different,
showing that both virtual running designs worked the same
way in terms of comfort and fun. In contrast, the usability
of the ArmSwing posture was better than the Squat posture.
However, both postures had SUS scores more than 80.3,
which were excellent usability. The interview results showed
that users preferred aerobic exercise (ArmSwing) when play-
ing for an extended period of time. The results suggested that
if playing for a short period, 56.7% of users chose strength
exercise (Squat) because they felt high exertion. However,
there were still 40% of users preferred aerobic exercise (Arm-
Swing) as it required less exertion and was more comfortable.
Moreover, if users preferred aerobic exercise, they preferred
strength exercise also.
The virtual running design can further develop in RPG
exergames. The results suggested that alternating postures
should be implemented in the design to avoid feeling bored
while also reducing injuries from the same pose for a long
time and supporting the challenge to exercise. The design
and development of VR exergames should entertain players
with acceptable exertion and timing to encourage exercise
by considering posture while playing. Future work should
focus on redesign virtual running extended to whole postures
(including aerobic, strength, flexibility, and balance) to
design virtual running postures for RPG exergames with
the appropriate time and fit for the physiology of each user.
Data Availability
The data used to support the findings of this study are avail-
able from the corresponding author upon request.
Conflicts of Interest
The authors declare that there are no conflicts of interest
regarding the publication of this paper.
Acknowledgments
This research was financially supported by the Walailak Uni-
versity Research Fund according to the contract number
WU64217.
References
[1] V. Benzing and M. Schmidt, “Exergaming for children and
adolescents: strengths, weaknesses, opportunities and threats,”
Journal of Clinical Medicine, vol. 7, no. 11, p. 422, 2018.
[2] W. Xu, H.-N. Liang, Y. Yu, D. Monteiro, K. Hasan, and
C. Fleming, “Assessing the effects of a full-body motion-
based exergame in virtual reality,”in Proceedings of the Seventh
International Symposium of Chinese CHI on - Chinese CHI '19,
pp. 1–6, Xiamen China, 2019.
[3] S. Yoo, M. Carter, and J. Kay, “Vrmove: design framework for
balancing enjoyment, movement and exertion in VR games,”
in Proceedings of the 2018 Annual Symposium on Computer-
Human Interaction in Play Companion Extended Abstracts,
pp. 295–307, 2018.
[4] K. Gordon, Virtual reality (vr) - statistics & facts, Statista, 2020,
https://www.statista.com/topics/2532/virtual-reality-vr/.
[5] J. Shepherd, L. Carter, G.-J. Pepping, and L.-E. Potter,
“Towards an operational framework for designing training
based sports virtual reality performance simulators,”Proceed-
ings, vol. 2, no. 6, p. 214, 2018.
[6] A. Christison and H. A. Khan, “Exergaming for health: a
community-based pediatric weight management program
using active video gaming,”Clinical Pediatrics, vol. 51, no. 4,
pp. 382–388, 2012.
[7] W. Peng and J. Crouse, “Playing in parallel: the effects of mul-
tiplayer modes in active video game on motivation and physi-
cal exertion,”Cyberpsychology, Behavior and Social
Networking, vol. 16, no. 6, pp. 423–427, 2013.
[8] A. E. Staiano, A. A. Abraham, and S. L. Calvert, “Adolescent
exergame play for weight loss and psychosocial improvement:
a controlled physical activity intervention,”Obesity, vol. 21,
no. 3, pp. 598–601, 2013.
[9] A. C. King, C. Castro, S. Wilcox, A. A. Eyler, J. F. Sallis, and
R. C. Brownson, “Personal and environmental factors associ-
ated with physical inactivity among different racial–ethnic
groups of U.S. middle-aged and older-aged women,”Health
Psychology, vol. 19, no. 4, pp. 354–364, 2000.
[10] G. Barry, P. Van Schaik, A. MacSween, J. Dixon, and
D. Martin, “Exergaming (xbox kinect™) versus traditional
gym-based exercise for postural control, flow and technology
acceptance in healthy adults: a randomised controlled trial,”
BMC sports science, medicine and rehabilitation, vol. 8, no. 1,
p. 25, 2016.
[11] I. Parry, C. Carbullido, J. Kawada et al., “Keeping up with video
game technology: objective analysis of xbox kinect™and plays-
tation 3 move™for use in burn rehabilitation,”Burns, vol. 40,
no. 5, pp. 852–859, 2014.
[12] B. Lange, S. Flynn, R. Proffitt, C.-Y. Chang, and A. Rizzo,
“Development of an interactive game-based rehabilitation tool
14 International Journal of Computer Games Technology

for dynamic balance training,”Topics in Stroke Rehabilitation,
vol. 17, no. 5, pp. 345–352, 2010.
[13] J. R. Franco, K. Jacobs, C. Inzerillo, and J. Kluzik, “The effect of
the nintendo wii fit and exercise in improving balance and
quality of life in community dwelling elders,”Technology and
Health Care, vol. 20, no. 2, pp. 95–115, 2012.
[14] N. Farǐc, H. W. Potts, A. Hon et al., “What players of virtual
reality exercise games want: thematic analysis of web-based
reviews,”Journal of Medical Internet Research, vol. 21, no. 9,
article e13833, 2019.
[15] V. R. Steam, “Hardware software survey november 2020,”
2020, https://store.steampowered.com/hwsurvey/Steam-
Hardware-Software-Survey-Welcome-to-Steam.
[16] C. Hillmann, “Comparing the gear vr, oculus go, and oculus
quest,”in Unreal for Mobile and Standalone VR, Apress,
Berkeley, CA, 2019.
[17] T. A. Jost, B. Nelson, and J. Rylander, “Quantitative analysis of
the Oculus Rift S in controlled movement,”Disability and
Rehabilitation: Assistive Technology, pp. 1–5, 2019.
[18] D. Heaney and J. Feltham, “Oculus quest built-in apps get
controller-free hand tracking, sdk out next week,”2019,
https://uploadvr.com/oculus-quest-finger-hand-tracking/.
[19] A. Borrego, J. Latorre, M. Alcaniz, and R. Llorens, “Comparison
of oculus rift and htc vive: feasibility for virtual reality-based
exploration, navigation, exergaming, and rehabilitation,”Games
for health journal,vol.7,no.3,pp.151–156, 2018.
[20] M. Suznjevic, M. Mandurov, and M. Matijasevic, “Perfor-
mance and QoE assessment of HTC Vive and Oculus Rift for
pick-and-place tasks in VR,”in 2017 Ninth International Con-
ference on Quality of Multimedia Experience (QoMEX), pp. 1–
3, Erfurt, Germany, 2017.
[21] Y. Huang, S. Shakya, and T. Odeleye, “Comparing the func-
tionality between virtual reality and mixed reality for architec-
ture and construction uses,”Journal of Civil Engineering and
Architecture, vol. 13, pp. 409–414, 2019.
[22] B. Lang, “Vive cosmos to support steamvr tracking with
optional faceplate add-on,”2019, https://www.roadtovr.com/
vive-cosmos-steamvr-tracking-lighthouse-base-stations-
faceplate-mod/.
[23] M. Hachman, Try vr for cheap: Windows mixed reality headsets
areonsalefornearlyhalfofftoday, PCWorld, 2017, https://
www.pcworld.com/article/3241949/virtual-reality/windows-
mixed-reality-headsets-microsoft-12-days-of-deals.html.
[24] C. Hunt, Get into windows mixed reality,WindowsCentral,
2020, https://www.windowscentral.com/best-windows-
mixed-reality-headsets.
[25] T. Warren, Samsung updates its windows VR headset with bet-
ter display tech, The Verge, 2018https://www.theverge.com/
circuitbreaker/2018/10/22/18008630/samsung-odyssey-plus-
windows-mixed-reality-headset-vr.
[26] A. Robertson, Valve index review: high-powered vr at a high-
end price, The Verge, 2019, https://www.theverge.com/2019/
6/28/19102584/valve-index-steamvr-headset-review-
shipping-today.
[27] R. Lai, “Chinese startup’s’8k’vr headset is surprisingly
advanced,”2020, https://www.engadget.com/2017-10-12-
pimax-8k-vr-headset.html.
[28] C. Boletsis, “The new era of virtual reality locomotion: a sys-
tematic literature review of techniques and a proposed typol-
ogy,”Multimodal Technologies and Interaction, vol. 1, no. 4,
p. 24, 2017.
[29] H. Cherni, N. Ḿetayer, and N. Souliman, “Literature review of
locomotion techniques in virtual reality,”International Jour-
nal of Virtual Reality, vol. 20, no. 1, pp. 1–20, 2020.
[30] M. McCullough, H. Xu, J. Michelson et al., “Myo arm: swing-
ing to explore a VE,”in Proceedings of the ACM SIGGRAPH
Symposium on Applied Perception, pp. 107–113, Tübingen
Germany, 2015.
[31] Y. S. Pai and K. Kunze, “Armswing: using arm swings for
accessible and immersive navigation in ar/vr spaces,”in Pro-
ceedings of the 16th International Conference on Mobile and
Ubiquitous Multimedia, pp. 189–198, Stuttgart Germany,
2017.
[32] P. T. Wilson, W. Kalescky, A. MacLaughlin, and B. Williams,
“VR locomotion: walking >walking in place >arm swinging,”
in Proceedings of the 15th ACM SIGGRAPH Conference on
Virtual-Reality Continuum and Its Applications in Industry -
Volume 1, pp. 243–249, Zhuhai China, 2016.
[33] J. N. Templeman, P. S. Denbrook, and L. E. Sibert, “Virtual
locomotion: walking in place through virtual environments,”
Presence, vol. 8, no. 6, pp. 598–617, 1999.
[34] J. C. Cardoso, “Gesture-based locomotion in immersive vr
worlds with the leap motion controller: comparison with
gamepad and gaze-directed locomotion,”2016.
[35] C. Khundam, “First person movement control with palm nor-
mal and hand gesture interaction in virtual reality,”in 2015
12th International Joint Conference on Computer Science and
Software Engineering (JCSSE), pp. 325–330, Songkhla, Thai-
land, 2015.
[36] S. Yoo and J. Kay, “Vrun: running-in-place virtual reality exer-
game,”in Proceedings of the 28th Australian Conference on
Computer-Human Interaction - OzCHI '16, pp. 562–566, 2016.
[37] S. Yoo, C. Ackad, T. Heywood, and J. Kay, “Evaluating the
actual and perceived exertion provided by virtual reality
games,”in Proceedings of the 2017 CHI Conference Extended
Abstracts on Human Factors in Computing Systems,
pp. 3050–3057, 2017.
[38] J. Lee, S. C. Ahn, and J.-I. Hwang, “A walking-in-place method
for virtual reality using position and orientation tracking,”
Sensors, vol. 18, no. 9, p. 2832, 2018.
[39] A. Alazba, H. Al-Khalifa, and H. AlSobayel, “Rabbitrun: an
immersive virtual reality game for promoting physical activi-
ties among people with low back pain,”Technologies, vol. 7,
no. 1, p. 2, 2019.
[40] E. Tuveri, L. Macis, F. Sorrentino, L. D. Spano, and R. Scateni,
“Fitmersive games: fitness gamification through immersive
VR,”in Proceedings of the International Working Conference
on Advanced Visual Interfaces, pp. 212–215, 2016.
[41] F. Laamarti, M. Eid, and A. El Saddik, “An overview of serious
games,”International Journal of Computer Games Technology,
vol. 2014, 15 pages, 2014.
[42] A. G. Thin, C. Brown, and P. Meenan, “User experiences while
playing dance-based exergames and the influence of different
body motion sensing technologies,”International Journal of
Computer Games Technology, vol. 2013, Article ID 603604, 7
pages, 2013.
[43] W. Peng, J.-H. Lin, and J. Crouse, “Is playing exergames really
exercising? A meta-analysis of energy expenditure in active
video games,”Cyberpsychology, Behavior and Social Network-
ing, vol. 14, no. 11, pp. 681–688, 2011.
[44] G. R. Oviedo, M. Guerra-Balic, T. Baynard, and C. Javierre,
“Effects of aerobic, resistance and balance training in adults
15International Journal of Computer Games Technology

with intellectual disabilities,”Research in Developmental Dis-
abilities, vol. 35, no. 11, pp. 2624–2634, 2014.
[45] F. C. Anderson and M. G. Pandy, “Dynamic optimization of
human walking,”Journal of Biomechanical Engineering,
vol. 123, no. 5, pp. 381–390, 2001.
[46] G. Cappellini, Y. P. Ivanenko, R. E. Poppele, and F. Lacquaniti,
“Motor patterns in human walking and running,”Journal of
Neurophysiology, vol. 95, no. 6, pp. 3426–3437, 2006.
[47] J. D. Wendt, M. C. Whitton, and F. P. Brooks, “GUD WIP:
Gait-Understanding-Driven Walking-In-Place,”in 2010 IEEE
Virtual Reality Conference (VR), pp. 51–58, Boston, MA,
USA, 2010.
[48] S. H. Collins, P. G. Adamczyk, and A. D. Kuo, “Dynamic arm
swinging in human walking,”Proceedings of the Royal Society
B: Biological Sciences, vol. 276, no. 1673, pp. 3679–3688, 2009.
[49] B. J. Schoenfeld, “Squatting kinematics and kinetics and their
application to exercise performance,”The Journal of Strength
& Conditioning Research, vol. 24, no. 12, pp. 3497–3506, 2010.
[50] R. F. Escamilla, “Knee biomechanics of the dynamic squat
exercise,”Medicine & Science in Sports & Exercise, vol. 33,
no. 1, pp. 127–141, 2001.
[51] S. Davis, K. Nesbitt, and E. Nalivaiko, “A systematic review of
cybersickness,”in Proceedings of the 2014 Conference on Inter-
active Entertainment, pp. 1–9, 2014.
[52] R. S. Kennedy, N. E. Lane, K. S. Berbaum, and M. G. Lilienthal,
“Simulator sickness questionnaire: an enhanced method for
quantifying simulator sickness,”The International Journal of
Aviation Psychology, vol. 3, no. 3, pp. 203–220, 1993.
[53] A. Bangor, P. T. Kortum, and J. T. Miller, “An empirical eval-
uation of the system usability scale,”International Journal of
Human-Computer Interaction, vol. 24, no. 6, pp. 574–594,
2008.
[54] J. Brooke, “SUS: a retrospective,”Journal of Usability Studies,
vol. 8, no. 2, pp. 29–40, 2013.
[55] S. P. Mullen, E. A. Olson, S. M. Phillips et al., “Measuring
enjoyment of physical activity in older adults: invariance of
the physical activity enjoyment scale (paces) across groups
and time,”International Journal of Behavioral Nutrition and
Physical Activity, vol. 8, no. 1, p. 103, 2011.
[56] J. Hauke and T. Kossowski, “Comparison of values of Pear-
son’s and Spearman’s correlation coefficients on the same sets
of data,”Quaestiones geographicae, vol. 30, no. 2, pp. 87–93,
2011.
[57] J. L. Myers, A. D. Well, and R. F. Lorch Jr., Research Design and
Statistical Analysis, Routledge, 2013.
16 International Journal of Computer Games Technology