Virtual Reality (2024) 28:62

[Link]

ORIGINAL ARTICLE

A Multi-user Cross-platform hands-on virtual lab within the Metaverse

– the case of machining training
Tung-Jui Chuang1 · Shana Smith1

Received: 26 June 2023 / Accepted: 20 February 2024 / Published online: 4 March 2024
© The Author(s) 2024

Abstract
Distance learning has become a popular learning channel today. However, while various distance learning tools are avail-
able, most of them only support a single platform, offer only the trainer’s perspective, and do not facilitate student-instruc-
tor interaction. As a result, distance learning systems tend to be inflexible and less effective. To address the limitations of
existing distance learning systems, this study developed a cross-platform hands-on virtual lab within the Metaverse that
enables multi-user participation and interaction for distance education. Four platforms, HTC VIVE Pro, Microsoft Holo-
Lens 2, PC, and Android smartphone, are supported. The virtual lab allows trainers to demonstrate operation steps and
engage with multiple trainees simultaneously. Meanwhile, trainees have the opportunity to practice their operational skills
on their virtual machines within the Metaverse, utilizing their preferred platforms. Additionally, participants can explore
the virtual environment and interact with each other by moving around within the virtual space, similar to a physical lab
setting. The user test compares the levels of presence and usability in the hands-on virtual lab across different platforms,
providing insights into the challenges associated with each platform within the Metaverse for training purposes. Further-
more, the results of the user test highlight the promising potential of the architecture due to its flexibility and adaptability.

Keywords Metaverse · Multi-user · Distance learning · Cross-platform · Hands-on

1 Introduction virtual reality (VR), augmented reality (AR), and mixed

reality (MR), becoming more accessible and affordable,
Distance learning offers the advantages of overcoming time it has been increasingly utilized in job training and indus-
and space constraints. However, it also has some drawbacks, trial applications. To achieve a high level of presence and
such as the lack of face-to-face interaction and hands-on immersion, XR technology has been increasingly used
experimentation (Dickson-Karn 2020; Faulconer et al. in distance learning. For instance, Monahan et al. (2008)
2018). Studies have reported that students are less engaged developed a web-based e-learning system that employed
in distance learning (Chen et al. 2021), which can nega- VR and multimedia to create a 3D virtual classroom where
tively impact their academic performance (Jaggars 2013). students can attend group meetings and discussions. McFaul
While some instructors use screen-sharing to simulate face- and FitzGerald (2020), as well as Kondratiuk et al. (2022),
to-face interaction, communication is often delayed and created 3D virtual environments for users to practice pre-
unclear (May et al. 2022). sentation skills. Similarly, Harfouche and Nakhle (2020)
Previous studies have found that virtual presence with designed a virtual lecture hall for bioethics education. Chan
a high level of immersion plays a critical role in the suc- et al. (2023) found that VR games significantly enhanced
cess and satisfaction of online learners (Annetta et al. 2008, learner motivation and engagement in chemical safety train-
2009). With extended reality (XR) technology, including ing compared to a video lecture-only approach. In science,
technology, engineering, and mathematics (STEM) educa-
tion, Mystakidis et al. (2022) indicated that students who
utilized AR achieved better outcomes compared to their
Shana Smith
ssmith@[Link] counterparts in the traditional training formats.
While the XR applications mentioned above have proven
1
Department of Mechanical Engineering, National Taiwan effective in enhancing students’ distance learning, their
University, Taipei, Taiwan

13
62 Page 2 of 16 Virtual Reality (2024) 28:62

capabilities have been constrained to presentation or view- (2018) demonstrated that various types of VR displays can
ing functions, lacking online hands-on capabilities or multi- influence users’ engagement and sense of presence.
user interactions. The significance of hands-on practices Modern virtual labs should be compatible with multiple
and multi-user interactions in STEM disciplines is crucial. platforms. This ensures that students can access the labs
However, providing virtual laboratories or training in a dis- from various devices and places, enhancing flexibility and
tance or non-traditional format has posed a challenge (Faul- usability. To enhance collaboration and engagement, virtual
coner et al. 2018). labs should also support multi-user interactions. This allows
students to work together in real-time, fostering teamwork
1.1 XR-based hands-on distance learning and communication, similar to a physical lab setting.

Some researchers have made attempts to incorporate hands- 1.2 Metaverse education
on functions into XR-based distance learning. For example,
Potkonjak et al. (2010) created a mechatronics and robotics With the advancements in computing technology, the con-
virtual lab for distance learning. Students could operate and cept of the Metaverse has emerged as an innovative social
observe the virtual robots. Aziz et al. (2014) developed a and interactive tool built upon XR technology. Metaverse is
game-based hands-on virtual lab for gear design. Xu and a 3D online multi-user environment that combines physi-
Wang (2022) proposed a virtual lab for wafer preparation cal reality with digital virtuality (Mystakidis 2022). The
training. Hurtado-Bermúdez and Romero-Abrio (2023) advent of the Metaverse opens up new possibilities for dis-
used an online virtual lab suite to teach students how to tance learning (Inceoglu and Ciloglugil 2022). Based on
operate a scanning electron microscope. Chan et al. (2021) the concept of the Metaverse, some of the aforementioned
demonstrated that virtual labs have the capability to deliver XR-based distance learning systems indeed fall into the cat-
training effects comparable to those of real hands-on labs. egory of the Metaverse.
Therefore, virtual labs have been used as a preparatory tool For example, Siyaev and Jo (2021) used a Metaverse in
to enhance students’ learning in real laboratories (Dalgarno HoloLens 2 for MR aircraft maintenance training. Trainees
et al. 2009; Elfakki et al. 2023; Kiourt et al. 2020; Manyilizu without HoloLens 2 could use mobile devices to view the
2023; Potkonjak et al. 2016; Sung and Chin 2023). maintenance lessons simultaneously. Khan et al. (2022)
Conventional virtual labs are typically regarded as a created a Metaverse in Quest VR goggles to deliver engi-
teaching approach, lacking any human-to-human interac- neering technology courses. Two presentation tools were
tion (Reeves and Crippen 2021). Therefore, although online available for the instructor, one was a whiteboard and the
hands-on lab provided a safe and time-efficient learning other was the ability to share computer screen into the VR
platform, research showed that students often considered classroom. Hwang (2023) implemented a Metaverse on PC
the online lab to be less intuitive than the physical lab, more to showcase artworks.
difficult to ask for help, more difficult to diagnose errors, The aforementioned research highlights that while numer-
and lacking in communication (Li et al. 2020). ous online virtual labs have been developed, to the best of
Many of the prior hands-on virtual labs were limited to our knowledge, none of them support both cross-platform
a single platform, thereby limiting their usability. In order and multi-user interactions. To address the research gap, in
to cater to the diverse needs of learners, there is a growing this study, a novel cross-platform multi-user hands-on vir-
demand for cross-platform distance learning solutions. For tual lab within the Metaverse was developed, supporting
example, Bao et al. (2022) used WebXR to build a multi-user HTV VIVE Pro, Microsoft HoloLens 2, PC, and Android
cross-platform system for construction safety training. The smartphone. The implementation of the Metaverse enables
system was designed to support desktop, mobile devices, trainers and trainees to immerse themselves in the same
and head-mounted displays (HMDs). Kambili-Mzembe and virtual lab using different platforms, facilitating interactive
Gordon (2022) used Unreal Engine to build a multi-user engagement with both each other and the virtual objects,
cross-platform system for secondary school teaching. Their replicating the experience of a physical lab environment.
system supported desktop, HMDs, and Android mobile This study states the following two research questions:
devices. Delamarre et al. (2020) developed a cross-platform
virtual classroom to simulate student’s emotional and social RQ 1. What is the level of presence of each platform
disruptive behaviors to facilitate teacher training. Their sys- in the multi-user cross-platform hands-on virtual lab
tem supported desktop, HMDs, and cave automatic virtual within the Metaverse?
environment (CAVE) setups. Delamarre et al. (2020) men- RQ 2. What is the usability of each platform in the multi-
tioned that the main difference between the platforms is the user cross-platform hands-on virtual lab within the
navigation techniques. Furthermore, Buttussi and Chittaro Metaverse?

13
Virtual Reality (2024) 28:62 Page 3 of 16 62

Since machining training not only requires extensive hands- lab information, such as the number of users and the names
on practice but also often involves close interaction between of users. The machine status includes the position of virtual
trainers and trainees, in this study, machining training was objects, operation process, and the user’s speech data. The
used as an example to demonstrate the feasibility and capa- Photon Network Module also sends non-local machine sta-
bility of the developed tool. This paper is organized as tus to the Operation Simulation module, sends lab status to
follows: Sect. 2 introduces the system architecture. Sec- the Lab Manager module, and receives local machine sta-
tion 3 describes the training environment. Section 4 dis- tus from the Operation Simulation module. The Operation
cusses the interaction designs. Section 5 presents the user Simulation module is the system’s core, containing all vir-
test. Section 6 provides discussions. Finally, Sect. 7 offers tual machines. The Operation Simulation module manages
conclusions. the local machine’s status, including the location of each
component, operation status, operation process, interactive
functions, and user’s speech data. Users communicate with
2 System architecture each other using audio devices. Speech data from a non-
local machine will be transmitted to virtual objects, which
In this study, Unity and Photon Unity Networking (PUN) in turn will play the speech. This process enables users to
SDK were selected for creating a cross-platform Metaverse. communicate with each other using a microphone.
PUN offers several advantages. The Photon cloud server The Lab Manager module oversees the lab status, includ-
facilitates a client-server-based connection, enabling com- ing the number of clients in the lab, and their names and
munication between clients within the same virtual environ- representative colors. When a trainee enters into the lab,
ment. The client’s system architecture, as depicted in Fig. 1, this module creates a new virtual machine in the Operation
comprises six modules: Photon Network Module, Opera- Simulation module. When a trainee leaves the lab, the Lab
tion Simulation Module, Lab Manager Module, Interaction Manager module deletes the corresponding virtual machine.
Module, Rendering Module, and Information Module. The Interaction module collects the user’s operation
The Photon Network Module communicates with Photon information and user’s speech data, and then sends them
cloud server by sending local machine status and receiving to the Operation Simulation module. User’s operation

Fig. 1 System architecture

13
62 Page 4 of 16 Virtual Reality (2024) 28:62

information is also sent to the Rendering module. The Ren- can occupy the same space, as shown in Fig. 3(b). Fig. 3(c)
dering module includes a virtual camera, which captures the depicts users operating their own machines.
images of the virtual environment and projects the virtual The training process is shown in Fig. 4. Once a user joins
object accurately on the display hardware. The Information a virtual lab, the system loads the lab scene, which includes
module retrieves information from the Operation Simula- both VR and AR versions. The HoloLens 2 device utilizes
tion module and creates visual feedback to users, such as the AR version, while other devices use the VR version.
message texts, operation status, warning signals, and indica- Then, the Lab Manager module is initialized for each cli-
tion pointers. ent. In the VR version, users can move around in the virtual
scene without physically moving their bodies. In contrast, in
the AR version, users are required to physically move their
3 Training environment bodies to maintain the relative position between the virtual
objects and real objects. Since rendering multiple objects
The system includes four platforms: HTC VIVE Pro, Micro- is a significant computing load for HoloLens 2, only one
soft HoloLens 2, PC, and Android smartphone. Due to the machine is displayed with 40 frames per second.
differences in input interfaces across the platforms, it is nec- Once all trainees are ready, the trainer can initiate
essary to utilize different software development kits (SDKs) the training. VR users will be transported to the front of
to manage the diverse input data. In this study, SteamVR the trainer’s machine, while AR users will have the local
was used to handle the VIVE Pro input, while the Mixed machine hidden and only the trainer’s machine displayed.
Reality Toolkit (MRTK) was used to manage the HoloLens The system then proceeds to the “Hands-on Demo” session,
2 input. The PC and smartphone platforms use the Unity where the trainer demonstrates how to operate the machine.
input utilities. All trainees have the ability to change their viewpoint to
The LC-1/2 series milling machines were used to demon- observe the hands-on demo operation from different angles.
strate the capabilities of the multi-user cross-platform online After the hands-on demo session, each trainee will be
hands-on virtual lab. The Metaverse consists of a virtual transported back to their respective local machines to prac-
factory and several virtual milling machines, as depicted in tice the operation process, as depicted in Fig. 5. Trainees will
Fig. 2. The trainer’s local machine is always in blue, while be required to replicate each operation step demonstrated by
other trainees’ local machines are in other colors. Since all the trainer. The system will only advance to the next step
machines are situated in the same scene, users can observe when trainees accurately perform the required actions. The
each other’s operations freely without any restrictions. A Operation Simulation module in each client also simulates
user list is used to display all users participating in the lab, the status of non-local machines, allowing the trainer and
along with the respective colors representing their local trainees to move to other non-local machines and observe
machines, as shown in Fig. 2. the operations of other users. Storing milling data requires a
Other than using a user list to represent the presence of minimum of 2.28 MB of memory per second. Due to limita-
other users, the system also uses virtual heads with differ- tions in the capacity of certain platforms that cannot accom-
ent colors to denotes the locations of different users. For modate this large data volume, the system was not designed
example, Fig. 3(a) displays trainees viewing the trainer’s to log or record practice milling data.
demonstration from different locations. Taking advantage
of the Metaverse, unlike the physical world, multiple users

Fig. 2 Training environment in the Metaverse

13
Virtual Reality (2024) 28:62 Page 5 of 16 62

Fig. 3 Multi-user in the Metaverse

13
62 Page 6 of 16 Virtual Reality (2024) 28:62

Fig. 4 Training process

Fig. 5 Trainees operate on different platforms

4 Interaction design 4.1 VIVE pro interaction

Due to the differences in user interface (UI) designs across VIVE Pro utilizes the SteamVR SDK to manage the inputs
the platforms, it is necessary to utilize different software of the platform. SteamVR enables the system to synchro-
development kits (SDKs) to manage the diverse input data. nize the position and rotation of the virtual camera with the
HMD. Additionally, SteamVR generates a virtual hand at
the controller position to assist users in locating their hands
in the virtual environment, as illustrated in Fig. 6. The red

13
Virtual Reality (2024) 28:62 Page 7 of 16 62

Fig. 6 Using the virtual hand to grab a virtual object

Fig. 7 Using the “pinch” gesture to grab a virtual

object on HoloLens 2

sphere with a radius of 30 mm in Fig. 6 is referred to as the This allows users to interact with virtual objects using their
trainer pointer. When the trainer uses the VIVE Pro platform, bare hands.
trainees on their respective platforms can only see the red The user’s viewpoint corresponds to the position and
trainer pointer instead of the virtual hand. This is due to the rotation of the HMD, enabling natural control of the view-
complexity of the virtual hand coordinates, which involve point by moving their heads. However, HoloLens 2 only
tracking over 20 points to record the hand status. Transmit- moves the viewpoint through head tracking, so users must
ting excessive data to synchronize the virtual hands’ status ensure they have enough physical space to move around and
can lead to an unstable connection and delays, which would operate the virtual machine.
undermine the effectiveness of the training.
4.3 PC interaction
4.2 HoloLens 2 interaction
PC and smartphone platforms utilize the Unity input utili-
HoloLens 2 utilizes MRTK to manage the inputs of the plat- ties to manage the interactions. PC uses a mouse and a
form, including the position of the HMD, hand tracking, and keyboard as the input interfaces. Users can grab a virtual
gesture recognition. Gesture recognition is limited to the object at the location of the mouse cursor by pressing the
pinch gesture, as shown in Fig. 7, where a pinch gesture and left mouse button. They can release the grabbed object by
the trainer pointer are used to grab an object. MRTK cre- releasing the left mouse button. When users click the left
ates virtual hands based on hand-tracking input, which are mouse button, the mouse cursor typically provides the 2D
superimposed on the real hands in the virtual environment. position in the image plane using pixel units. On the other
Since MRTK has the capability to detect objects touching hand, the position of virtual objects within the virtual envi-
the virtual index fingertip, users can interact with virtual ronment is defined in the 3D global coordinate system using
buttons by pressing them as they would with actual buttons.

13
62 Page 8 of 16 Virtual Reality (2024) 28:62

meter units. The 2D mouse position needs to be translated step when users accurately complete the required actions.
to the corresponding 3D position in the virtual environment. Consequently, the time taken in the second stage was uti-
lized to assess the trainees’ performance. After completing
4.4 Smartphone interaction the second stage, trainees were asked to fill out a presence
questionnaire (PQ) version 3 (Witmer et al. 2005), a system
The primary input interface of a smartphone is its touch- usability scale (SUS) questionnaire (Brooke 1996), a sub-
screen, which also serves as the image plane. Consequently, jective questionnaire, and to provide comments about the
the smartphone platform utilizes the same interaction algo- system.
rithm as the PC platform. However, users adjust the object The PQ evaluates each platform’s involvement, sensory
distance by using two fingers. The UI on smartphones is fidelity, adaptation/immersion, and interface quality. For
identical to the UI on PC since Unity UI elements are also enhanced reliability and validity, irrelevant questions were
compatible with smartphones. Moreover, both platforms excluded. In this study, the PQ comprised 25 questions
use the position on the image plane as the primary input, aimed at measuring the presence of the virtual lab, using a
enabling for a unified UI design. 7-point Likert scale (1 for strongly disagree, 7 for strongly
agree). Higher scores indicate better involvement, sensory
fidelity, and adaptation/immersion, whereas a lower score
5 User test is preferable for interface quality. As this system does not
simulate the environmental sounds of milling machines or
5.1 Method provide haptic feedback, the Presence Questionnaire (PQ)
questions related to audio and haptic aspects were removed.
The user test was conducted in two stages to evaluate the However, since participants can communicate with each
effectiveness of the cross-platform hands-on virtual lab other verbally, a subjective questionnaire was designed to
within the Metaverse. VIVE Pro, HoloLens 2, and smart- evaluate the verbal audio quality in the system.
phone utilized their built-in speakers and microphones for The purpose of the SUS is to evaluate the usability of
recording and playing users’ voice. In this study, the smart- each platform. The SUS questionnaire consists ten questions
phone used was the SAMSUNG Galaxy S8+, and the PC to evaluate the usability of a system (Brooke 1996), with a
was equipped with the HONG JIN HJ-MX3 multimedia 5-point Likert scale ranging from 1 (strongly disagree) to
microphone and the DIKE DSM230 speaker. 5 (strongly agree). The scores are then converted to a SUS
In the first stage, the trainer demonstrated how to set up score using a hundred-point system. A system is considered
a milling machine and machine a 60 × 60 × 60 mm3 work- usable if its SUS score is higher than the average score of 70
piece to the target shape, as shown in Fig. 8. In the second (Bangor et al. 2008).
stage, trainees were asked to repeat the operation in the first The subjective questionnaire was designed to assess par-
stage. The milling operation included several steps, such as ticipants’ communication and navigation capabilities within
installing a workpiece and a milling tool, moving an object the virtual lab across different platforms. The subjective
to a specific position, rotating a handwheel to a particular questionnaire comprised twelve questions. Questions 1 to
angle, tightening a nut, and machining the workpiece into 4 evaluated the clarity of the audio communication for spe-
a specific shape. The system will only advance to the next cific platform, while Questions 5 to 8 focused on partici-
pants’ perception of audio delays. Questions 9 to 11 aimed
to assess satisfaction with the navigation function on vari-
ous platforms, excluding HoloLens 2 due to its limited com-
puting power for rendering all machines in the virtual lab at
the same time.
To determine the minimum sample size, this study used
the G* power tool. The statistical power P, type I error α,
and effect size d were set to 0.8, 0.05, and 0.833, respec-
tively (Cohen 1988). The result showed that at least 11 par-
ticipants were needed for the user test in each platform. In
this study, a total of 52 participants aged between 20 and
30 were recruited, with 13 using VIVE Pro, 13 using Holo-
Lens 2, 13 using PC, and 13 using smartphones. None of
the participants were professional milling machine opera-
tors, but all had operation experience of less than one year.
Fig. 8 The target product of the user test

13
Virtual Reality (2024) 28:62 Page 9 of 16 62

Table 1 Results of the operation time

Task VIVE Pro HoloLens 2 PC Smartphone
Mean SD Mean SD Mean SD Mean SD
(sec) (sec) (sec) (sec) (sec) (sec) (sec) (sec)
Set up a milling machine 224.23 78.91 231.15 23.84 230.77 82.00 249.23 73.88
Machine a virtual workpiece 806.00 106.75 831.92 131.02 823.31 133.53 911.62 151.48
Total time 1050.23 87.78 1183.08 143.65 1174.08 139.98 1280.85 157.99

Fig. 9 The PQ factor scores of each platform

Participants using VIVE Pro and HoloLens 2 had no prior p = .882) and adaptation/immersion score (F(3,48) = 2.734,
experience with these devices, whereas those in the PC and p = .054) showed no significant difference among platforms.
smartphone groups had over two years of experience with The smartphone platform tends to exhibit lower scores in
their respective devices. involvement, sensory fidelity, and adaptation/immersion.
The Tukey’s test indicated that the interface quality of the
5.2 Results HoloLens 2 was significantly lower than other platforms.
Table 2 provides a breakdown of the detailed results for
5.2.1 Operation time each PQ factor. The bolded number indicates a significant
difference from the neutral point of 4. Questions 1 to 12
Table 1 presents the average operation time for participants assess the level of involvement, reflecting participants’ inter-
who used different platforms. The smartphone platform had actions and engagement with the virtual lab. Sensory fidel-
the longest average operation time, while the VIVE Pro ity questions (13 and 14) measure the realism of the sensory
platform had the shortest average operation time. However, experience in the virtual lab. Questions 15 to 22 evaluate
the one-way ANOVA analysis showed that operation time the level of adaptation and immersion, indicating whether
on different platforms did not significantly differ (p = .097). participants adapt to the virtual lab and feel immersed.
Questions 23 to 25 assess interface quality, emphasizing the
5.2.2 Presence questionnaire impact of device functionality and the quality of interface
design.
Figure 9 displays the PQ results for each factor category. The results of the PQ for VIVE Pro platform indicated
Regarding interface quality, for the sake of easy compari- that participants exhibited high levels of involvement in the
son, the scores were transposed, with higher values indicat- training environment and demonstrated effective interaction
ing better performance. The statistical analysis results show with virtual objects (Q 1–12). The system was found to be
that the VIVE Pro platform received the highest score for responsive (Q 1 and 2) and provided a natural interaction
the involvement factor, and it is significantly higher than mechanism for participants (Q 3, 5, 6, and 9), allowing them
the smartphone platform by the Tukey’s test. The ANOVA to engage effectively with the virtual lab (Q 11).
results of the sensory fidelity score (F(3,48) = 0.219,

13
62 Page 10 of 16 Virtual Reality (2024) 28:62

Table 2 The results of PQ

Q (Question) Overall VIVE Pro HoloLens 2 PC Smartphone
Involvement
1 How much were you able to control Mean = 5.40 Mean = 6.15 Mean = 4.62 Mean = 6.38 Mean = 4.46
events? SD = 1.485 SD = 0.801 SD = 1.710 SD = 0.650 SD = 1.450
p < .001 p < .001 p = .281 p < .001 p = .189
2 How responsive was the environment to Mean = 5.35 Mean = 6.23 Mean = 4.38 Mean = 6.00 Mean = 4.77
actions that you initiated (or performed)? SD = 1.385 SD = 0.927 SD = 1.660 SD = 0.707 SD = 1.166
p < .001 p < .001 p = .420 p < .001 p = .038
3 How natural did your interactions with the Mean = 5.23 Mean = 5.77 Mean = 4.85 Mean = 5.62 Mean = 4.69
environment seem? SD = 1.366 SD = 1.166 SD = 1.519 SD = 0.506 SD = 1.750
p < .001 p < .001 p = .068 p < .001 p = .179
4 How much did the visual aspects of the Mean = 5.17 Mean = 6.08 Mean = 5.38 Mean = 4.92 Mean = 4.31
environment involve you? SD = 1.630 SD = 1.115 SD = 1.261 SD = 1.754 SD = 1.888
p < .001 p < .001 p = .002 p = .095 p = .516
5 How natural was the mechanism which Mean = 5.33 Mean = 6.23 Mean = 5.23 Mean = 5.23 Mean = 4.62
controlled movement through the SD = 1.382 SD = 0.599 SD = 1.536 SD = 0.927 SD = 1.758
environment? p < .001 p < .001 p = .021 p < .001 p = .231
6 How compelling was your sense of objects Mean = 5.19 Mean = 5.92 Mean = 4.92 Mean = 5.31 Mean = 4.62
moving through space? SD = 1.253 SD = 0.760 SD = 1.320 SD = 0.630 SD = 1.710
p < .001 p < .001 p = .027 p < .001 p = .219
7 How much did your experiences in the Mean = 4.79 Mean = 5.31 Mean = 4.69 Mean = 5.00 Mean = 4.15
virtual environment seem consistent with SD = 1.226 SD = 0.751 SD = 1.437 SD = 0.707 SD = 1.573
your real world experiences? p < .001 p < .001 p = .108 p < .001 p = .778
8 How completely were you able to actively Mean = 5.44 Mean = 6.00 Mean = 5.00 Mean = 5.85 Mean = 4.92
survey or search the environment using SD = 1.364 SD = 0.913 SD = 1.472 SD = 0.899 SD = 1.754
vision? p < .001 p < .001 p = .031 p < .001 p = .082
9 How compelling was your sense of mov- Mean = 5.10 Mean = 5.54 Mean = 6.00 Mean = 4.62 Mean = 4.23
ing around inside the virtual environment? SD = 1.550 SD = 1.330 SD = 1.291 SD = 1.660 SD = 1.363
p < .001 p < .001 p < .001 p = .206 p = .553
10 How well could you move or manipulate Mean = 5.08 Mean = 5.77 Mean = 4.46 Mean = 5.83 Mean = 4.31
objects in the virtual environment? SD = 1.495 SD = 0.927 SD = 1.664 SD = 0.718 SD = 1.750
p < .001 p < .001 p = .337 p < .001 p = .538
11 How involved were you in the virtual Mean = 5.46 Mean = 6.08 Mean = 5.38 Mean = 5.31 Mean = 5.08
environment experience? SD = 1.514 SD = 0.862 SD = 1.660 SD = 1.601 SD = 1.754
p < .001 p < .001 p = .018 p = .012 p = .060
12 How easy was it to identify objects Mean = 5.27 Mean = 5.85 Mean = 4.85 Mean = 5.54 Mean = 4.85
through physical interaction, like touch- SD = 1.443 SD = 1.405 SD = 1.281 SD = 1.127 SD = 1.772
ing an object, walking over a surface, or p < .001 p < .001 p = .028 p < .001 p = .111
bumping into a wall or object?
Sensory Fidelity
13 How closely were you able to examine Mean = 5.52 Mean = 5.62 Mean = 5.54 Mean = 5.77 Mean = 5.15
objects? SD = 1.336 SD = 0.961 SD = 1.198 SD = 1.166 SD = 1.908
p < .001 p < .001 p < .001 p < .001 p = .051
14 How well could you examine objects from Mean = 5.79 Mean = 5.77 Mean = 5.85 Mean = 5.85 Mean = 5.69
multiple viewpoints? SD = 1.258 SD = 1.166 SD = 1.144 SD = 1.214 SD = 1.601
p < .001 p < .001 p < .001 p < .001 p < .001
Adaptation/Immersion
15 Were you able to anticipate what would Mean = 5.71 Mean = 6.23 Mean = 5.38 Mean = 6.08 Mean = 5.15
happen next in response to the actions that SD = 1.194 SD = 0.599 SD = 1.446 SD = 0.760 SD = 1.463
you performed? p < .001 p < .001 p = .012 p < .001 p = .021
16 How quickly did you adjust to the virtual Mean = 6.15 Mean = 6.31 Mean = 5.62 Mean = 6.46 Mean = 6.23
environment experience? SD = 1.092 SD = 0.947 SD = 1.502 SD = 0.776 SD = 0.927
p < .001 p < .001 p < .001 p < .001 p < .001
17 How proficient in moving and interacting Mean = 5.94 Mean = 6.08 Mean = 5.69 Mean = 6.08 Mean = 5.92
with the virtual environment did you feel SD = 1.259 SD = 1.038 SD = 1.437 SD = 1.115 SD = 1.498
at the end of the experience? p < .001 p < .001 p < .001 p < .001 p < .001
18 How well could you concentrate on the Mean = 5.33 Mean = 5.85 Mean = 4.77 Mean = 5.77 Mean = 4.92
assigned tasks or required activities rather SD = 1.133 SD = 0.801 SD = 1.092 SD = 1.092 SD = 1.188
than on the mechanisms used to perform p < .001 p < .001 p = .026 p < .001 p = .016
those tasks or activities?

13
Virtual Reality (2024) 28:62 Page 11 of 16 62

Table 2 (continued)
Q (Question) Overall VIVE Pro HoloLens 2 PC Smartphone
19 How completely were your senses Mean = 5.31 Mean = 6.15 Mean = 5.31 Mean = 5.23 Mean = 4.54
engaged in this experience? SD = 1.553 SD = 0.899 SD = 1.316 SD = 1.691 SD = 1.854
p < .001 p < .001 p < .001 p = .030 p = .316
20 Were there moments during the vir- Mean = 5.46 Mean = 6.08 Mean = 5.08 Mean = 5.69 Mean = 5.00
tual environment experience when you SD = 1.306 SD = 0.760 SD = 1.553 SD = 1.109 SD = 1.472
felt completely focused on the task or p < .001 p < .001 p = .028 p < .001 p = .031
environment?
21 How easily did you adjust to the control Mean = 5.44 Mean = 6.08 Mean = 4.77 Mean = 5.92 Mean = 5.00
devices used to interact with the virtual SD = 1.259 SD = 0.760 SD = 1.013 SD = 0.862 SD = 1.732
environment? p < .001 p < .001 p = .026 p < .001 p = .076
22 Was the information provided through Mean = 5.40 Mean = 6.08 Mean = 5.38 Mean = 5.00 Mean = 5.15
different senses in the virtual environment SD = 1.512 SD = 0.862 SD = 1.609 SD = 1.354 SD = 1.951
(e.g., vision, hearing, touch) consistent? p < .001 p < .001 p = .024 p = .035 p = .065
Interface Quality
23 How much delay did you experience Mean = 2.65 Mean = 2.46 Mean = 3.31 Mean = 2.46 Mean = 2.38
between your actions and expected SD = 1.736 SD = 1.664 SD = 1.494 SD = 2.106 SD = 1.660
outcomes? p < .001 p < .001 p = .090 p = .022 p < .001
24 How much did the visual display quality Mean = 2.94 Mean = 3.31 Mean = 3.46 Mean = 2.54 Mean = 2.46
interfere or distract you from performing SD = 1.638 SD = 1.437 SD = 1.941 SD = 1.506 SD = 1.561
assigned tasks or required activities? p < .001 p = .098 p = .304 p = .013 p = .011
25 How much did the control devices inter- Mean = 2.71 Mean = 2.23 Mean = 3.92 Mean = 2.15 Mean = 2.54
fere with the performance of assigned SD = 1.684 SD = 0.832 SD = 2.397 SD = 1.144 SD = 1.450
tasks or with other activities? p < .001 p < .001 p = .905 p < .001 p = .013

Participants also reported positive opinions on sensory users, indicating that participants generally reported a posi-
fidelity, indicating that they could examine virtual objects tive experience in terms of examining virtual objects using
closely from multiple viewpoints (Q 13, 14). Participants HoloLens 2 (Q 13–14).
demonstrated a high level of adaptability to the virtual envi- The results of the PQ show that the PC platform demon-
ronment and were able to operate virtual machines with a strated a comparable level of involvement, sensory fidelity,
strong sense of immersion (Q 15–22). Participants could adaption/immersion, and interface quality with the VIVE
quickly adjust to the control devices and the system (Q 16 Pro users, with the exception of visual and motion aspects
and 21). The mechanisms of the system allowed participants (Q 4 and 9). It might be attributed to the stationary 2D screen
to focus on their tasks and have an immersive experience display and constrained spatial movement limitations inher-
(Q 18 and 20). However, it is worth noting that some par- ent in the PC platform. However, it is surprising that the
ticipants experienced occasional issues with unclear images PC platform received the highest scores on sensory fidelity
when using the VIVE Pro headset even though other ques- among the four platforms (Q 13 and 14). This may be attrib-
tions did not reveal the impact of the display issue (Q 24). uted to participants’ familiarity with the PC platform.
Compared to the VIVE Pro, participants considered The PQ results for the smartphone platform suggested
the HoloLens 2 to have a lower level of involvement and that participants felt a low level of involvement compared
adaption/immersion. This suggests that participants felt to the VIVE Pro platform, likely attributed to difficulties in
challenged when interacting with virtual objects using interacting with virtual objects and controlling the virtual
the HoloLens 2 platform (Q 1–3 and 10). These interac- camera using a small hand-held device. These challenges
tion issues had an impact on the overall experience of the also make the smartphone platform have the lowest sen-
virtual lab compared to real-world scenarios (Q 7). Con- sory fidelity. However, due to participants’ familiarity with
sequently, some participants perceived that their ability to smartphone interfaces (Q 23–25), they felt they could pre-
fully engage with tasks and maintain a high level of focus dict the system’s response and adapt quickly to it (Q 15 and
was compromised. Users also indicated dissatisfaction with 16).
the interface quality of the HoloLens 2 platform (Q 23–25).
This might be attributed to occasional tracking issues and 5.2.3 System usability scale
gesture recognition failures of the HoloLens 2 platform,
which were perceived as distracting and disruptive to their The results of the SUS are presented in Table 3. On the
concentration. However, HoloLens 2 users demonstrated VIVE Pro platform, the average SUS score was 87.69, sig-
a comparable level of sensory fidelity with the VIVE Pro nificantly higher than the average score of 70 (t = 6.397,

13
62 Page 12 of 16 Virtual Reality (2024) 28:62

Table 3 System usability scale results

Question Overall VIVE Pro HoloLens 2 PC Smartphone
1 I think that I would like to use Mean = 3.52 Mean = 4.08 Mean = 3.23 Mean = 3.69 Mean = 3.08
this system frequently. SD = 0.980 SD = 0.862 SD = 0.725 SD = 0.751 SD = 1.256
p < .001 p < .001 p = .257 p = .014 p = .829
2 I found this system unneces- Mean = 1.60 Mean = 1.46 Mean = 1.62 Mean = 1.62 Mean = 1.69
sarily complex. SD = 0.846 SD = 0.660 SD = 0.870 SD = 0.961 SD = 0.947
p < .001 p < .001 p < .001 p < .001 p < .001
3 I thought this system was easy Mean = 3.96 Mean = 4.54 Mean = 3.46 Mean = 4.38 Mean = 3.46
to use. SD = 1.009 SD = 0.660 SD = 0.519 SD = 0.650 SD = 1.450
p < .001 p < .001 p = .014 p < .001 p = .273
4 I think that I would need the Mean = 1.96 Mean = 1.46 Mean = 2.15 Mean = 2.15 Mean = 2.08
support of a technical person SD = 1.137 SD = 0.877 SD = 0.987 SD = 1.214 SD = 1.382
to be able to use this system. p < .001 p < .001 p < .001 p = .029 p = .04
5 I found the various func- Mean = 4.06 Mean = 4.54 Mean = 4.08 Mean = 3.85 Mean = 3.77
tions in this system were well SD = 0.916 SD = 0.519 SD = 0.641 SD = 1.068 SD = 1.166
integrated. p < .001 p < .001 p < .001 p = .021 p = .038
6 I thought there was too much Mean = 1.62 Mean = 1.54 Mean = 1.46 Mean = 1.77 Mean = 1.69
inconsistency in this system. SD = 0.771 SD = 0.519 SD = 0.660 SD = 0.725 SD = 1.109
p < .001 p < .001 p < .001 p < .001 p < .001
7 I would imagine that most Mean = 4.13 Mean = 4.77 Mean = 3.69 Mean = 4.23 Mean = 3.85
people would learn to use this SD = 1.103 SD = 0.439 SD = 0.947 SD = 0.927 SD = 1.573
system very quickly. p < .001 p < .001 p = .22 p < .001 p = .083
8 I found this system very cum- Mean = 1.90 Mean = 1.23 Mean = 2.54 Mean = 1.69 Mean = 2.15
bersome (awkward) to use. SD = 1.125 SD = 0.599 SD = 1.198 SD = 0.947 SD = 1.281
p < .001 p < .001 p = .19 p < .001 p = .028
9 I felt very confident using this Mean = 3.98 Mean = 4.46 Mean = 3.54 Mean = 4.38 Mean = 3.54
system. SD = 1.019 SD = 0.660 SD = 1.127 SD = 0.768 SD = 1.127
p < .001 p < .001 p = .083 p < .001 p = .11
10 I needed to learn a lot of things Mean = 1.71 Mean = 1.62 Mean = 1.92 Mean = 1.54 Mean = 1.77
before I could get going with SD = 1.109 SD = 1.261 SD = 1.188 SD = 0.660 SD = 1.301
this system. p < .001 p < .001 p = .01 p < .001 p < .001

p < .001). Similarly, the SUS score of the PC platform 5.2.4 Subjective questionnaire
(79.42) was significantly greater than the average score
(t = 2.59, p = .024). All participants strongly agreed that the The results of the subjective questionnaire are presented in
system on the VIVE Pro and PC platforms was easy to use Table 4, utilizing a 7-point Likert scale (1 for strongly dis-
and that they would like to use it. agree, 7 for strongly agree). The Wilcoxon-signed rank test
However, the SUS score of the HoloLens 2 platform was applied to all questions due to the non-normally distrib-
(70.77) was not significantly different from the average uted data. The results of Questions 1 to 4 were significantly
score (t = 0.214, p = .834). It implied that participants could different from the neutral point of 4, suggesting that partici-
accept the system on HoloLens 2 but had some issues using pants found that the audio communication on all platforms
it. Questions 7 to 9 of the SUS indicated that participants to be clear, and the ANOVA result (F(3,204) = 0.8233,
found it difficult to operate HoloLens 2. The reported chal- p = .48) indicated that the clarity of the audio communica-
lenges with hand tracking and gesture recognition functions tion was not significantly different. This implies that all plat-
may have influenced the overall SUS scores, resulting in forms effectively transmitted clear audio data. The results
scores that are not significantly different from the average of Questions 5 to 8 showed that participants perceived no
score. delay in communication with others in the virtual lab, and
The SUS score of the smartphone platform (70.77) was the ANOVA result (F(3,48) = 0.426, p = .732) showed no
also not significant different from the average SUS score significantly difference among the platforms. Questions 9
(t = 0.385, p = .7). Questions 3, 7, and 9 revealed that some to 11 indicated that participants found the navigation func-
participants found the system challenging to use, leading to tion easy to use on all platforms, and the ANOVA result
a lack of confidence. The primary issue causing this prob- (F(2,36) = 0.142, p = .868) indicated that there is no signifi-
lem was the difficulty of placing objects to the designated cant difference.
location after grabbing them.

13
Virtual Reality (2024) 28:62 Page 13 of 16 62

Table 4 Results of the subjective questionnaire While prior research indicated that steering might lead to
Question Mean SD Z more cybersickness than teleportation (Chin et al. 1988), in
1 How clear is the audio communica- 5.79 1.21 5.757***
this study, the steering time is very short. Participants did
tion with the VIVE Pro users?
2 How clear is the audio communica- 5.77 1.15 5.810***
not even notice the existence of the middle path, and they
tion with the HoloLens 2 users? could easily navigate to different locations in the virtual lab
3 How clear is the audio communica- 6.08 1.14 6.056*** without any difficulty or experiencing cybersickness.
tion with the PC users? Prior research showed that VR head mounted displays
4 How clear is the audio communica- 5.94 1.09 5.996*** provide higher presence and engagement than a PC (But-
tion with the smartphone users?
tussi and Chittaro 2018). The success and satisfaction of
5 How much delay is there in audio 1.46 0.88 3.217**
communication when using the online learners are critically linked to virtual presence and a
VIVE Pro? high level of immersion. (Annetta et al. 2008, 2009). In this
6 How much delay is there in audio 1.69 0.95 3.140** study, the use of the VIVE Pro platform resulted in the high-
communication when using the est levels of involvement and adaptation/immersion, with
HoloLens 2?
the PC platform following as the second.
7 How much delay is there in audio 1.85 1.21 3.101**
communication when using a PC?
Although VIVE Pro controllers could not fully replicate
8 How much delay is there in audio 1.92 1.38 3.006** actual actions, users considered the platform intuitive and
communication when using a immersive. However, users encountered challenges with
smartphone? the VIVE Pro platform related to display issues. This was
9 How easy is the navigation between 6.62 0.75 3.276** attributed to the optical characteristics of the device, where
machines when using the VIVE
the edges of the display may appear blurred. Additionally, if
Pro?
10 How easy is the navigation between 6.54 0.66 3.275**
the VIVE Pro was not worn in the optimal position, it could
machines when using the PC? further contribute to image blurriness. The unclear images
11 How easy is the navigation 6.46 0.66 3.256** posed challenges for participants when operating the virtual
between machines when using the machine, hindering their ability to perceive and interact
smartphone? with the virtual environment effectively.
* p < .05, ** p < .01, *** p < .001
Using HoloLens 2 with bare hands allowed participants
to quickly learn the operations. The AR scene made par-
6 Discussion ticipants feel immersive, and they could perceive the actual
milling machine’s size, tools and feel like they were oper-
After the user test, participants were asked about their expe- ating a real milling machine. However, the scores for the
rience using the system. They found it exciting to operate SUS and PQ interface quality were relatively low. Users
virtual milling machines with other participants in the same encountered occasional challenges with hand tracking when
virtual factory. As the interaction designs vary across dif- interacting with virtual objects. Due to HoloLens 2’s limita-
ferent platforms, making a fair comparison becomes chal- tion of tracking hands within the user’s field of view (FOV),
lenging. However, specific platforms may be more suitable users needed to consistently look at their hands. This made
for certain applications (Delamarre et al. 2020). While the it challenging to read text messages while operating the
operation time did not significantly differ among the plat- machine. Some participants found the grabbing operation
forms, VIVE Pro users were the quickest, followed by PC confusing, as HoloLens 2 only supports the “pinch” gesture.
users, and HoloLens 2 and smartphone users took the third Moreover, the platform’s constrained computing power
and longest, respectively. This pattern might be associated restricted it to displaying only one machine at a time, and it
with the usability of the platforms, with VIVE Pro receiving lacked teleportation functionality, further constraining user
the highest SUS score, while smartphone and HoloLens 2 interaction.
received the lowest scores. The PC platform generally received higher scores on
Prior research has demonstrated that clear communi- both the PQ and SUS, despite a few participants express-
cation and verbal instruction have a significant impact on ing a lack of immersion due to the 2D display. The larger
online virtual lab instruction.(Li et al. 2020; May et al. image size on the PC provided participants with an easy and
2022). According to the subjective questionnaire results, clear view, greatly facilitating their virtual machine opera-
participants reported no audio communication issues on any tions. Participants found it effortless to grab objects with-
platform. The navigation technique used in the VIVE Pro out needing additional practice, and they highlighted the
is teleportation (from one location to another directly with- straightforward nature of the UI on the PC platform. How-
out a middle path), while PC and smartphone are steering ever, there were instances where the message panel would
(from one location to another rapidly with a middle path).

13
62 Page 14 of 16 Virtual Reality (2024) 28:62

occasionally cover virtual objects, causing participants 2 platform, due to occasional tracking issues and gesture
minor inconvenience. recognition failures. Additionally, due to computing power
Participants using the smartphone platform expressed limitations and the nature of AR displays, HoloLens 2 can
satisfaction with its mobility, appreciating the freedom from only display one machine at a time and lacks teleportation
space constraints. However, the SUS score, as well as the functionality. Despite the interface design of the smart-
involvement, sensory fidelity, and adaptation/immersion phone platform being similar to that of the PC platform,
factors in PQ, were relatively low, indicating participants participants were dissatisfied with the smartphone’s usabil-
were not entirely satisfied with the virtual labs on smart- ity. They also considered the smartphone platform to lack
phones. It might be attributed to the small smartphone involvement, sensory fidelity, and immersiveness due to its
screen, as participants reported that virtual objects appearing small display.
much smaller than actual objects, even when attempts were The developed architecture and user interfaces developed
made to enlarge them. Users’ fingers sometimes obstructed in this study serves as a roadmap for the implementation of
their view of the virtual objects, hindering complete engage- a multi-user cross-platform online hands-on interactive vir-
ment with the virtual lab experience. As a result, more time tual lab. The study is subject to limitations arising from the
was spent relocating the objects to their designated loca- diverse range of AR, VR, PC, and smartphone devices used.
tions, leading participants to find it challenging to interact The levels of presence and usability may vary among these
with virtual objects. This issue was not encountered on the devices. Due to differences in interaction designs across
PC platform, where virtual object images were larger than platforms, conducting a fair comparison poses challenges.
the mouse cursor. Certain platforms may prove more suitable for specific
applications. In the future, different devices and multi-user
hands-on training tasks will be assessed and tested. Com-
7 Conclusions parative studies of learning effectiveness across different
platforms will also be conducted. The interaction functions
Distance learning has become a global trend, as it allows of the platforms should be optimized to ensure better hands-
students to acquire knowledge at any time and from any on effectiveness. Finally, a more compact data structure for
location. Modern virtual labs must be compatible across recording hands-on working data will be developed.
various platforms to ensure students can access them from
different devices and locations, promoting flexibility and Acknowledgements The authors would like to thank the Ministry of
Science and Technology, Taiwan, Republic of China for financially
usability. Additionally, supporting multi-user interactions supporting this research under Contract MOST 110-2221-E-002-145.
enhances collaboration, mirroring the teamwork and com-
munication found in physical labs. This study developed a Author contributions All authors wrote the main manuscript text and
multi-user cross-platform hands-on virtual lab within the reviewed the manuscript.
Metaverse. This innovative approach aims to enhance the
ubiquity of applications in distance learning, providing stu- Data availability The datasets generated during and/or analyzed dur-
ing the current study are available from the corresponding author.
dents with more interactive and practical hands-on learning
opportunities.
Declarations
In this study, a virtual lab was developed specifically for
machining training, serving as a case study. To facilitate Ethical approval All procedures carried out in studies involving hu-
the participation of multiple trainees, the system employed man participants adhered to the ethical standards of the institutional
Photon Unity Networking, which enabled the connection and/or national research committee, with an IRB-approved consent
form.
of users across different platforms such as HTC VIVE Pro,
Microsoft HoloLens 2, PC, and smartphones. Each plat-
Competing interests The authors declare no competing interests.
form offered unique interactive interfaces, allowing mul-
tiple users to engage in the virtual lab and interact with each Consent for Publication Informed consent was obtained from all indi-
other. vidual participants included in the study.
In implementing multi-user cross-platform hands-on
virtual labs, the user test revealed that participants were Open Access This article is licensed under a Creative Commons
Attribution 4.0 International License, which permits use, sharing,
satisfied with the presence and usability of the VIVE Pro adaptation, distribution and reproduction in any medium or format,
platform within the virtual lab. The PC platform also pro- as long as you give appropriate credit to the original author(s) and the
vided comparable levels of presence and usability satis- source, provide a link to the Creative Commons licence, and indicate
faction for users. However, participants were dissatisfied if changes were made. The images or other third party material in this
article are included in the article’s Creative Commons licence, unless
with the interface quality and usability of the HoloLens indicated otherwise in a credit line to the material. If material is not

13
Virtual Reality (2024) 28:62 Page 15 of 16 62

included in the article’s Creative Commons licence and your intended Elfakki AO, Sghaier S, Alotaibi AA (2023) An efficient system based
use is not permitted by statutory regulation or exceeds the permitted on experimental laboratory in 3D virtual environment for stu-
use, you will need to obtain permission directly from the copyright dents with Learning Disabilities. Electronics 12:989
holder. To view a copy of this licence, visit [Link] Faulconer E, Gruss A, Distance Science Laboratory Experiences
org/licenses/by/4.0/. (2018) A review to weigh the pros and cons of Online, Remote,
and. Int Rev Res Open Distrib Learn 19:155–168. [Link]
org/10.19173/irrodl.v19i2.3386
Harfouche AL, Nakhle F (2020) Creating Bioethics Distance Learning
References through virtual reality. Trends Biotechnol 38:1187–1192. https://
[Link]/10.1016/[Link].2020.05.005
Annetta L, Klesath M, Holmes S (2008) V-Learning: how Gaming and Hurtado-Bermúdez S, Romero-Abrio A (2023) The effects of combin-
avatars are engaging online students. Innovate 4. ing virtual laboratory and advanced technology research labora-
Annetta L, Klesath M, Meyer J (2009) Taking Science Online: evaluat- tory on university students’ conceptual understanding of electron
ing Presence and Immersion through a Laboratory experience in microscopy. Interact Learn Environ 31:1126–1141. [Link]
a virtual learning environment for Entomology Students. J Coll org/10.1080/10494820.2020.1821716
Sci Teach 39:27–33 Hwang Y (2023) When makers meet the metaverse: effects of creat-
Aziz E-SS, Chang Y, Esche SK, Chassapis C (2014) A multi-user vir- ing NFT metaverse exhibition in maker education. Comput Educ
tual laboratory environment for gear train design. Comput Appl 194:104693. [Link]
Eng Educ 22:788–802. [Link] Inceoglu MM, Ciloglugil B (2022) Use of Metaverse in Education.
Bangor A, Kortum PT, Miller JT (2008) An empirical evaluation of the Springer International Publishing, Cham, pp 171–184. [Link]
system usability scale. Int J Human–Computer Interact 24:574– org/10.1007/978-3-031-10536-4_12
594. [Link] Jaggars SSE, Stacey N (2013) Georgia West Creating an Effective
Bao L, Tran SV-T, Nguyen TL, Pham HC, Lee D, Park C (2022) Cross- Online Instructor Presence. Columbia University: Community
platform virtual reality for real-time construction safety training College Research Center. (accessed 2022/11/7) Retrieved from
using immersive web and industry foundation classes. Autom Con- [Link]
str 143:104565. [Link] Kambili-Mzembe F, Gordon NA (2022) Synchronous Multi-User
Brooke J (1996) SUS-A quick and dirty usability scale. Usability eval- Cross-Platform Virtual Reality for School Teachers. 2022
uation in industry, 1st edn. CRC, London, pp 4–7 8th International Conference of the Immersive Learning
Buttussi F, Chittaro L (2018) Effects of different types of virtual real- Research Network (iLRN), Vienna, Austria, pp.1–5. [Link]
ity Display on Presence and Learning in a Safety Training sce- org/10.23919/iLRN55037.2022.9815966
nario. IEEE Trans Vis Comput Graph 24:1063–1076. [Link] Khan V, Basith II, Boyd M (2022) Application of Metaverse as an
org/10.1109/TVCG.2017.2653117 Immersive Teaching Tool in Engineering Technology Class-
Chan P, Van Gerven T, Dubois J-L, Bernaerts K (2021) Virtual chemi- rooms. 2022 ASEE Illinois-Indiana Section Conference. Ander-
cal laboratories: a systematic literature review of research, son, Indiana: ASEE Conferences, 36131
technologies and instructional design. Computers Educ Open Kiourt C, Kalles D, Lalos AS, Papastamatiou NP, Silitziris P, Paxinou
2:100053. [Link] E, Theodoropoulou H, Zafeiropoulos V, Papadopoulos A, Pavli-
Chan P, Van Gerven T, Dubois J-L, Bernaerts K (2023) Study of dis G (2020) XRLabs: Extended Reality Interactive Laboratories.
motivation and engagement for chemical laboratory safety train- International Conference on Computer Supported Education.
ing with VR serious game. Saf Sci 167:106278. [Link] 601–608. [Link]
org/10.1016/[Link].2023.106278 Kondratiuk L, Musiichuk S, Zuienko N, Sobkov Y, Trebyk O, Yefimov
Chen E, Kaczmarek K, Ohyama H (2021) Student perceptions of D (2022) Distance Learning of Foreign languages through virtual
distance learning strategies during COVID-19. J Dent Educ reality. BRAIN Broad Res Artif Intell Neurosci 13:22–38. https://
85:1190–1191. [Link] [Link]/10.18662/brain/13.2/329
Chin JP, Diehl VA, Norman KL (1988) Development of an instru- Li R, Morelock J, May D (2020) A comparative study of an online
ment measuring user satisfaction of the human-computer lab using Labsland and Zoom during COVID-19. Adv Eng Educ
interface. Proceedings of the SIGCHI Conference on Human 8:1–10
Factors in Computing Systems, Washington, D.C., USA, Manyilizu MC (2023) Effectiveness of virtual laboratory vs. paper-
pp.213–218. Association for Computing Machinery. [Link] based experiences to the hands-on chemistry practical in Tan-
org/10.1145/57167.57203 zanian secondary schools. Educ Inform Technol 28:4831–4848.
Cohen J (1988) Statistical power analysis for the behavioral sci- [Link]
ences. New York: Routledge. 2nd Edition [Link] May D, Morkos B, Jackson A, Hunsu NJ, Ingalls A, Beyette F (2022)
org/10.4324/9780203771587 Rapid transition of traditionally hands-on labs to online instruc-
Dalgarno B, Bishop AG, Adlong W, Bedgood DR (2009) Effective- tion in engineering courses. Eur J Eng Educ 48:842–860. https://
ness of a virtual laboratory as a preparatory resource for Distance [Link]/10.1080/03043797.2022.2046707
Education chemistry students. Comput Educ 53:853–865. https:// McFaul H, FitzGerald E (2020) A realist evaluation of student use of
[Link]/10.1016/[Link].2009.05.005 a virtual reality smartphone application in undergraduate legal
Delamarre A, Lisetti C, Buche C (2020) A Cross-Platform Class- education. Br J Edu Technol 51:572–589. [Link]
room Training Simulator: Interaction Design and EvaluationA bjet.12850
Cross-Platform Classroom Training Simulator: Interaction Monahan T, McArdle G, Bertolotto M (2008) Virtual reality for col-
Design and Evaluation. 2020 International Conference on Cyber- laborative e-learning. Comput Educ 50:1339–1353. [Link]
worlds (CW). Caen, France, 86–93. [Link] org/10.1016/[Link].2006.12.008
CW49994.2020.00020 Mystakidis S (2022) Metaverse Encyclopedia 2:486–497. [Link]
Dickson-Karn NM (2020) Student Feedback on Distance Learning org/10.3390/
in the Quantitative Chemical Analysis Laboratory. J Chem Educ Mystakidis S, Christopoulos A, Pellas N (2022) A systematic map-
97:2955–2959. [Link] ping review of augmented reality applications to support STEM

13
62 Page 16 of 16 Virtual Reality (2024) 28:62

learning in higher education. Educ Inform Technol 27:1883– Sung C, Chin S (2023) Contactless Multi-user virtual hair design
1927. [Link] synthesis. Electronics 12:3686. [Link]
Potkonjak V, Vukobratović M, Jovanović K, Medenica M (2010) Vir- electronics12173686
tual Mechatronic/Robotic laboratory – a step further in distance Witmer BG, Jerome CJ, Singer MJ (2005) The factor structure of the
learning. Comput Educ 55:465–475. [Link] Presence Questionnaire. Presence: Teleoperators Virtual Environ
compedu.2010.02.010 14:298–312
Potkonjak V, Gardner M, Callaghan V, Mattila P, Guetl C, Petrović Xu X, Wang F (2022) Engineering Lab in Immersive VR—An Embod-
VM, Jovanović K (2016) Virtual laboratories for education in ied Approach to Training Wafer Preparation. J Educational Comput
science, technology, and engineering: a review. Comput Educ Res 60:455–480. [Link]
95:309–327. [Link]
Reeves SM, Crippen KJ (2021) Virtual Laboratories in Undergradu- Publisher’s Note Springer Nature remains neutral with regard to juris-
ate Science and Engineering courses: a systematic review, dictional claims in published maps and institutional affiliations.
2009–2019. J Sci Edu Technol 30:16–30. [Link]
s10956-020-09866-0
Siyaev A, Jo G-S (2021) Towards aircraft maintenance metaverse
using speech interactions with virtual objects in mixed reality.
Sensors 21:2066. [Link]

13