A Systematic Review of the Design of Serious Games for Innovative Learning: Augmented Reality, Virtual Reality, or Mixed Reality?

: The recent integration of educational technologies and emerging learning approaches into education systems has been driven largely by the pandemic. This paper conducts a systematic review and delves into the new wave of research on serious games designed for innovative learning using augmented reality (AR), virtual reality (VR), and mixed reality (MR). The review was referenced to the review protocol, PRISMA 2020. Using the Scopus Database with a time filter from 2007 to 2023 (27 July), we searched 329 articles and shortlisted 273 relevant studies. Notably, European countries contributed the most (62.9%) to this research area. Among the most frequent keywords, VR (90.9%) was commonly used in AR/VR/MR, while e-learning (95.3%) was among the popular innovative learning approaches. Further research studies are needed to employ AR and MR technologies, as well as other innovative learning approaches, to enable performance evaluation and comparison of various educational technologies and learning approaches. We conducted an in-depth analysis of the relevant studies and their basic characteristics. Additionally, we introduced 15 essential and recently published AR/VR/MR standards to ensure better reliability, quality, and safety of architectures, systems, products, services, and processes. To facilitate performance evaluation and analysis, we surveyed 15 recently published benchmark education datasets. This review suggested four future research directions, including multisensory experiences, generative artificial intelligence, personalization and customization, and real-time interaction.


Introduction
Education is a vital element of the sustainable development of individuals, nations, and the world [1].Many worldwide alliances and agendas have been established to promote education.One of the most prominent is the 2030 Agenda for Sustainable Development [2], which was agreed upon and adopted by 193 countries in the United Nations General Assembly in September 2015.The United Nations Educational, Scientific and Cultural Organization (UNESCO) coordinates communities across the globe to achieve Sustainable Development Goal 4 (SDG 4), namely "Quality Education", through different means such as partnerships and policy guidance [3].SDG 4 is one of the 17 SDGs outlined in the agenda and aims to "ensure inclusive and equitable quality education and promote lifelong learning opportunities for all" [2].Another initiative is UNESCO's Global Education Coalition [4], which aims to support countries in scaling up their best distance learning practices and reaching children and youth who are most at risk.
Innovative learning.According to the SDG Report 2022 [5] compiled by the United Nations Statistics Division, the COVID-19 pandemic has led to 147 million children missing over half of the in-person instruction in 2020-2021, and remote learning is offered to 3 million Ukrainian children in the chaos of war as of April 2022.All of these crises have accelerated the transition of traditional education toward innovative learning, such as e-learning, agile learning, and blended learning.E-learning has become one of the mainstream ways of learning new skills and knowledge; for example, a common form of e-learning is online courses such as those on MIT's Open Courseware (ocw.mit.edu)(accessed on 1 July 2023) and Coursera (www.coursera.org)(accessed on 1 July 2023).E-learning allows students to access learning materials and learn anywhere and anytime on any smart device, and its learning effectiveness has also been supported by a lot of research studies [6].On the other hand, agile learning refers to the ability to adapt learning strategies quickly in response to change and bridge the gap between various identified needs of stakeholders and the learning process being deployed [7], and it is a concept transferred from agile methods in software engineering to the learning design process [8].Noguera et al. [9] showed that agile strategies for teamwork organization can improve students' team regulation and project management in collaborative project-based learning.Another example of innovative learning is blended learning, which refers to the combination of face-to-face and computer-mediated instructions, such as online learning and self-paced study [10].Blended learning is an effective approach to creating a more engaging and effective learning experience for students [11,12].During the pandemic, it has become a common solution to blend online learning and remote face-to-face learning to keep the students' study progress on track and to ensure effective learning [11].
Serious games.Serious games are "digital games created with the intention to entertain and to achieve at least one additional goal (e.g., learning or health)" [13].They have been applied in different educational contexts because of their effectiveness in promoting engagement and learning in different modes of learning, see, e.g., ref. [14] for e-learning, ref. [15] for agile learning, and [16] for blended learning.Two closely related concepts on serious games in education are gamification and game-based learning (GBL).Gamification is the process of combining game theory and design, game elements, game aesthetics, and game mechanics into a learning experience, in which serious games are created as products, while GBL is a pedagogical approach to teaching in which students can explore relevant aspects of games in a learning context designed by the teachers [17].
VR, AR, and MR in education.Virtual Reality (VR) is a computer technology that creates a 3D virtual environment and simulates the user's physical presence in a virtual world [18].Augmented Reality (AR) also generates 3D virtual environments for users, but AR integrates the virtual world with reality by anchoring 3D models to specific locations or objects in the physical environment, such as those captured on a camera screen.Mixed Reality (MR) combines elements of both VR and AR to create an interactive and immersive experience that blends the digital and physical worlds, in which digital objects and information are also superimposed onto the real world like AR, yet MR creates a spatially aware environment that allows users using a see-through display or a headset, e.g., Microsoft's HoloLens 2 device, to interact with virtual objects in a 3D environment in a way different to AR [19].Over the past decade, the affordability and user-friendliness of the hardware and software supporting VR and AR (referred to collectively as immersive reality) have improved, leading to their growing use in education (e.g., refs.[20][21][22][23][24]; refer to surveys [25,26] for more examples).The utilization of VR and AR in education enables learners to practice skills and gain knowledge in a safe and controlled virtual environment.It also provides them with access to experiences and resources that may not be available in traditional classrooms [26].VR and AR serious games are particularly well-suited for skill-based education and attract students' interest in learning [27].
Existing reviews on innovative learning.Numerous reviews have been conducted on innovative learning across diverse subject domains, with each review emphasizing a specific technology within its respective field of study.In the medical domain, Gaballa et al. [28] conducted a study to explore the potential of augmented reality (AR), virtual reality (VR), and mixed reality (MR) systems in enhancing upper-limb prosthesis rehabilitation.Kleinsorgen et al. [29] performed a descriptive analysis of publication activities related to digital teaching and learning in the GMS Journal for Medical Education from 1984 to 2020.Shahid et al. [30] compared various information technology applications for the treatment of autism.On the other hand, the education domain also benefits from the integration of innovative learning technologies.For instance, Ongoro and Fanjiang [31] conducted a systematic review that examined the use of digital game-based learning technology for English language instruction in preschool and elementary schools.Baxter et al. [32] developed a framework through a comprehensive preliminary discursive review of the literature on innovative education and computer technology, specifically in the context of teaching history.Ullah et al. [33] conducted a systematic literature review focused on serious games for science education, highlighting research trends and patterns from 2011 to 2021.
In addition to industrial training, innovative learning technologies have also been examined in other domains, such as personal skills development.For example, Vigoroso et al. [34] conducted a synthesis of available works on game-based safety training across various occupational domains.Williams-Bell et al. [35] reviewed several serious games and virtual simulations that aid in fire service training.Wahyudin and Hasegawa [36] focused on the role of serious games in disaster and safety education.Feng et al. [37] explored the use of immersive virtual reality serious games for building evacuation training and research, specifically for indoor emergencies like fire and earthquakes.Some reviews cover multiple domains and investigate specific technologies or methodologies.For instance, Hare and Tang [38] categorized player modeling and game adaptation methods in serious games for higher education.Chavez and Bayona [39] selected articles published between 1999 and 2017 across various fields of study to identify the characteristics of virtual reality technology and its impact on the learning process.Checa and Bustillo [27] reviewed 135 proposals for immersive virtual reality serious games, examining their standards and differences between training and learning applications.Yomeldi et al. [40] conducted a systematic literature review of serious games for mobile learning, highlighting the potential influence of visualization, enjoyment, immersion, and interactivity on learning outcomes.
However, there is currently a lack of research specifically focused on the design of serious games for innovative learning.

Research Contributions
To address the above significant aspects and gaps, the objective of this systematic review using a review protocol (PRISMA 2020) [41] is to provide an in-depth analysis of the relevant studies on serious games designed for innovative learning using augmented reality (AR), virtual reality (VR), and mixed reality (MR):

•
We provide an in-depth analysis of the 273 relevant studies and their basic characteristics.

•
Our work introduces 15 essential AR/VR/MR standards to ensure better reliability, quality, and safety of architectures, systems, products, services, and processes.

•
Lastly, to facilitate performance evaluation and analysis, we survey 15 benchmark education datasets.

Organization of the Article
The rest of the article is organized as follows.Section 2 presents the methodology of article search and selection processes to shortlist relevant articles in AR/VR/MR research areas for innovative learning using serious games.It is followed by the discussion of results in Section 3, comprising basic characteristics of shortlisted articles and an in-depth analysis of journal articles.To support environments with good quality, repeatability, safety, interoperability, and compatibility, AR, VR, and MR standards (published and under development) will be summarized in Section 4. Lastly, a conclusion is drawn in Section 5. Figure 1 summarizes the structure of this article.
Electronics 2024, 13, x FOR PEER REVIEW 4 of 23 • Lastly, to facilitate performance evaluation and analysis, we survey 15 benchmark education datasets.

Organization of the Article
The rest of the article is organized as follows.Section 2 presents the methodology of article search and selection processes to shortlist relevant articles in AR/VR/MR research areas for innovative learning using serious games.It is followed by the discussion of results in Section 3, comprising basic characteristics of shortlisted articles and an in-depth analysis of journal articles.To support environments with good quality, repeatability, safety, interoperability, and compatibility, AR, VR, and MR standards (published and under development) will be summarized in Section 4. Lastly, a conclusion is drawn in Section 5. Figure 1 summarizes the structure of this article.

Methodology
As a review article, this section starts with an illustration of the search process, followed by a selection process to exclude and shortlist articles for an in-depth literature review in the following sections.

Search Process
Advanced document search of Scopus Database was used to search for relevant articles, which is the largest database in this domain.The search was based on article title, abstract, and keywords.The query string comprised: ("serious game") AND ("blended learning" OR "hybrid learning" OR "e-learning" OR "agile learning") AND ("virtual reality" OR "VR" OR "augmented reality" OR "AR" OR "mixed reality" OR "MR").The learning approaches covered the latest categories of innovative learning.The shortlisted articles must be serious games-related and fall into at least one of the learning and reality approaches.Three authors (L.-K.L., X.W., and K.T.C.) checked the query and confirmed the results.The initial search responses were 329 articles with a time filter from 2007 to 2023 (27 July, the first date on which the team prepared this article).

Methodology
As a review article, this section starts with an illustration of the search process, followed by a selection process to exclude and shortlist articles for an in-depth literature review in the following sections.

Search Process
Advanced document search of Scopus Database was used to search for relevant articles, which is the largest database in this domain.The search was based on article title, abstract, and keywords.The query string comprised: ("serious game") AND ("blended learning" OR "hybrid learning" OR "e-learning" OR "agile learning") AND ("virtual reality" OR "VR" OR "augmented reality" OR "AR" OR "mixed reality" OR "MR").The learning approaches covered the latest categories of innovative learning.The shortlisted articles must be serious games-related and fall into at least one of the learning and reality approaches.Three authors (L.-K.L., X.W., and K.T.C.) checked the query and confirmed the results.The initial search responses were 329 articles with a time filter from 2007 to 2023 (27 July, the first date on which the team prepared this article).

Articles Selection Process
The inclusion criteria for the review were that the articles must be related to education purposes and that details must be presented to describe the game setting and approach of innovative learning.Three authors independently and randomly reviewed all papers' titles and papers.Each author read about 74 papers.Discussions were made when there was any inconsistency.To shortlist appropriate articles, the following exclusion criteria were adopted.
• To ensure technical content is discussed, only non-review articles under the document type are included.Therefore, 30 articles are excluded, including conference review (n = 21), article review (n = 6), short survey (n = 1), note (n = 1), and book (n = 1); • To ensure the best communication in academia, only articles presented in English are included.Thus, four articles are excluded, including Spanish (n = 1), French (n = 1), German (n = 1), and Hungarian (n = 1); • The authors carefully read the titles and abstracts of the articles and further excluded 22 articles, including conference review (n = 11), article review (n = 10), and book chapter review (n = 1).
After applying the criteria, 273 articles were shortlisted.Among the shortlisted articles, an in-depth analysis will be conducted on journal articles (n = 44).To summarize the article selection process, a workflow is presented in Figure 2.

Articles Selection Process
The inclusion criteria for the review were that the articles must be related to education purposes and that details must be presented to describe the game setting and approach of innovative learning.Three authors independently and randomly reviewed all papers' titles and papers.Each author read about 74 papers.Discussions were made when there was any inconsistency.To shortlist appropriate articles, the following exclusion criteria were adopted.

•
To ensure technical content is discussed, only non-review articles under the document type are included.Therefore, 30 articles are excluded, including conference review (n = 21), article review (n = 6), short survey (n = 1), note (n = 1), and book (n = 1); • To ensure the best communication in academia, only articles presented in English are included.Thus, four articles are excluded, including Spanish (n = 1), French (n = 1), German (n = 1), and Hungarian (n = 1); The authors carefully read the titles and abstracts of the articles and further excluded 22 articles, including conference review (n = 11), article review (n = 10), and book chapter review (n = 1).After applying the criteria, 273 articles were shortlisted.Among the shortlisted articles, an in-depth analysis will be conducted on journal articles (n = 44).To summarize the article selection process, a workflow is presented in Figure 2.

Results
The shortlisted 273 articles are analyzed in various aspects: (i) basic characteristics include the number of publications, distribution of the document type, geographical distribution of country affiliations, distribution of subject areas, publication sources, keywords analysis, and Word cloud; and (ii) literature review of shortlisted journal articles.

Basic Characteristics
The number of articles published between 2007 and 2023 (up to 27 July 2023) is summarized in Figure 3.It can be seen from the trend that the research areas have generally received increasing intentions (although it is not monotonically increasing) from 2007

Results
The shortlisted 273 articles are analyzed in various aspects: (i) basic characteristics include the number of publications, distribution of the document type, geographical distribution of country affiliations, distribution of subject areas, publication sources, keywords analysis, and Word cloud; and (ii) literature review of shortlisted journal articles.

Basic Characteristics
The number of articles published between 2007 and 2023 (up to 27 July 2023) is summarized in Figure 3.It can be seen from the trend that the research areas have generally received increasing intentions (although it is not monotonically increasing) from 2007 (two articles) to 2022 (40 articles).Notably, more articles were published during the COVID-19 pandemic (2020-2022) because the face-to-face learning mode was usually not feasible [42,43].To further study the distribution of the document type of these articles, Figure 4 presents a pie chart showing 223 conference papers (82%), 44 articles (16%), and six book chapters (2%).The distribution indicates that many research studies published as conference papers are initial ideas, preliminary analyses, and case studies.It reflects that there is room for future research for innovative learning using VR and AR.
COVID-19 pandemic (2020-2022) because the face-to-face learning mode was usually not feasible [42,43].To further study the distribution of the document type of these articles, Figure 4 presents a pie chart showing 223 conference papers (82%), 44 articles (16%), and six book chapters (2%).The distribution indicates that many research studies published as conference papers are initial ideas, preliminary analyses, and case studies.It reflects that there is room for future research for innovative learning using VR and AR.Table 1 summarizes the geographical distribution of the authors' country affiliations (n = 160).Of the 273 articles, 57 countries/territories contributed to the research.The continents' contributions are ranked in descending order: Europe (62.9%),Asia (13.5%),North America (13.2%),South America (6.29%),Africa (2.10%), and Oceania (2.10%).The research studies of VR and AR for innovative learning are dominated by European institutions, which is attributable to the unique characteristics of their educational systems [44,45].However, this does not mean other continents far less value innovative learning, where they devote effort to other learning tools (less on AR/VR/MR).During a pandemic, the research and adoption of innovative learning have been accelerated because the changing mode of learning drives the mitigation (at least some trials) from face-to-face learning to any innovative learning.
Figure 5 shows the distribution of the articles based on 22 subject areas, where 7 out of 22 (31.8%)subject areas comprise at least ten articles encompassing AR/VR/MR for innovative learning.The top three dominant subject areas are computer science (82.1%), engineering (43.4%), and social sciences (28.8%).The first two subject areas can be explained by the fact that AR/VR/MR and educational technologies usually cover discussion in the domain of computer science and engineering.To further analyze the key contributions, the annual publications in the top three subject areas are shown in Figure 6.The (two articles) to 2022 (40 articles).Notably, more articles were published during the COVID-19 pandemic (2020-2022) because the face-to-face learning mode was usually not feasible [42,43].To further study the distribution of the document type of these articles, Figure 4 presents a pie chart showing 223 conference papers (82%), 44 articles (16%), and six book chapters (2%).The distribution indicates that many research studies published as conference papers are initial ideas, preliminary analyses, and case studies.It reflects that there is room for future research for innovative learning using VR and AR.Table 1 summarizes the geographical distribution of the authors' country affiliations (n = 160).Of the 273 articles, 57 countries/territories contributed to the research.The continents' contributions are ranked in descending order: Europe (62.9%),Asia (13.5%),North America (13.2%),South America (6.29%),Africa (2.10%), and Oceania (2.10%).The research studies of VR and AR for innovative learning are dominated by European institutions, which is attributable to the unique characteristics of their educational systems [44,45].However, this does not mean other continents far less value innovative learning, where they devote effort to other learning tools (less on AR/VR/MR).During a pandemic, the research and adoption of innovative learning have been accelerated because the changing mode of learning drives the mitigation (at least some trials) from face-to-face learning to any innovative learning.
Figure 5 shows the distribution of the articles based on 22 subject areas, where 7 out of 22 (31.8%)subject areas comprise at least ten articles encompassing AR/VR/MR for innovative learning.The top three dominant subject areas are computer science (82.1%), engineering (43.4%), and social sciences (28.8%).The first two subject areas can be explained by the fact that AR/VR/MR and educational technologies usually cover discussion in the domain of computer science and engineering.To further analyze the key contributions, the annual publications in the top three subject areas are shown in Figure 6.The Table 1 summarizes the geographical distribution of the authors' country affiliations (n = 160).Of the 273 articles, 57 countries/territories contributed to the research.The continents' contributions are ranked in descending order: Europe (62.9%),Asia (13.5%),North America (13.2%),South America (6.29%),Africa (2.10%), and Oceania (2.10%).The research studies of VR and AR for innovative learning are dominated by European institutions, which is attributable to the unique characteristics of their educational systems [44,45].However, this does not mean other continents far less value innovative learning, where they devote effort to other learning tools (less on AR/VR/MR).During a pandemic, the research and adoption of innovative learning have been accelerated because the changing mode of learning drives the mitigation (at least some trials) from face-to-face learning to any innovative learning.in social sciences remained steady (about 10 on average).7 provides the trends of the top three keywords.Some key observations are concluded: (i) The number of annual publications of all three keywords dropped in 2018; (ii) The number of annual publications in the top keyword, e-learning, increased monotonically from 2017 to 2021; (iii) The number of annual publications in the second top keyword, serious games fluctuated (up and down) from 2017 to 2021; and (iv) The number of annual publication in the third top keyword, virtual reality increased monotonically from 2017 to 2022.As a full picture, Figure 8 shows a word cloud of all keywords.

Literature Review of Shortlisted Journal Articles
To balance a concise and in-depth discussion of AR/VR/MR for innovative education, only shortlisted journal articles (n = 44) are fully investigated in this subsection.Table 4 summarizes the crucial elements of the existing studies .The publication years of the journal articles ranged from 2011 to 2023.Among 44 articles, the majority of studies

Conclusions
This paper explores the integration of serious games into education to enhance the quality of learning.The post-pandemic era has facilitated the adoption of innovative approaches to teaching and learning, including the use of gaming elements such as AR, VR, and MR.While VR dominates (81.8% among three reality approaches) in the literature review of 273 articles on developing serious games for innovative learning between 2007 and 27 July 2023, AR and MR possess unique characteristics that support different educational research.AR often provides extra information to users in real-world environments, whereas MR considers the interactions between digital and physical elements.Interactions between AR and MR enable the joint environment between computers, physical elements, and humans [124].However, this increases the computing load for realistic environment simulation and deployment [125].Our paper also provides an overview of typical guidelines and settings in software and hardware development for 15 AR/VR/MR standards and 15 benchmark datasets (six open-access datasets and nine datasets with reasonable requests to the corresponding authors) for conducting research studies.Although all authors devoted total effort to preparing this article, we observe limitations of the review processes, such as the search being restricted to the Scopus Database, only English articles being included, and journal articles being thoroughly analyzed in Table 4.The presented article is believed to provide a good systematic review of the latest developments in the design of serious games for innovative learning using AR/VR/MR.
Last but not least, the authors suggest several future research directions: (i) Apart from vision and hearing, consider the multisensory experiences (e.g., taste, smell, and touch) of the virtual environments, which enhances participant engagement.A review article of 105 articles revealed that 85% of them show a positive impact of multisensory VR experiences [126].Another work [127] compared the sense of presence between audiovisualand multisensory-based VR.The results found that multisensory experiences enhanced the sense of presence (from 4.37 to 4.44), in a participant size of 80.In the multisensory environment, the system complexity will be increased, and a larger space is required; (ii) employ generative artificial intelligence (e.g., data generation algorithms) to synthesize data and AR/VR/MR environments.This extends the data distribution from only ground truth information.The generative adversarial network has been a leading approach for data generation [128].An augmented flow network was generated to support the AR-and VR-based ceramic art exhibitions to enhance visitors' experience [129].It took advantage of updating the design easily and frequently.It is a common issue of huge computational power requirements to render AR and MR information.A deep learning-based 3D point cloud generation was proposed to reduce the requirements [130]; (iii) enhance the learning experience with personalization to customize the learning environments and systems for individual users.One can consider balancing various factors, such as comfortability, excitability, and durability.An immersive VR neuro-learning platform was proposed to customize learning scenarios for medical education [131].Both qualitative and quantitative revealed the effectiveness.In one paper [132], customized VR-based exercises (based on patients' performance) were designed to enhance gait rehabilitation.The ease of the system was confirmed by three experienced physiotherapists (15-33 years of experience); and (iv) real-time interaction, engagement, and feedback between teachers and students to support multi-user using the same AR/VR/MR environment.In one paper [133], researchers studied the real-time interaction between teachers and students in construction teaching using VR from a teaching perspective.Another work [134] also focused on teaching perspective to design VR-based lectures to enhance teacher-student interaction and after-class activities.Further work is required to analyze the interaction from a student perspective.

Figure 1 .
Figure 1.Structure of this article.

Figure 1 .
Figure 1.Structure of this article.

Figure 2 .
Figure 2. Workflow of the article selection process.The purple, blue, and green colors denote the information related to the initial search, screening, and analysis, respectively.

Figure 2 .
Figure 2. Workflow of the article selection process.The purple, blue, and green colors denote the information related to the initial search, screening, and analysis, respectively.

Figure 3 .
Figure 3. Number of articles published between 2007 and 2003 (up to 27 July 2023).

Figure 4 .
Figure 4. Distribution of the document type of the 273 shortlisted articles.

Figure 3 .
Figure 3. Number of articles published between 2007 and 2003 (up to 27 July 2023).

Figure 3 .
Figure 3. Number of articles published between 2007 and 2003 (up to 27 July 2023).

Figure 4 .
Figure 4. Distribution of the document type of the 273 shortlisted articles.

Figure 4 .
Figure 4. Distribution of the document type of the 273 shortlisted articles.

Figure 5
shows the distribution of the articles based on 22 subject areas, where 7 out of 22 (31.8%)subject areas comprise at least ten articles encompassing AR/VR/MR for innovative learning.The top three dominant subject areas are computer science (82.1%), engineering (43.4%), and social sciences (28.8%).The first two subject areas can be explained by the fact that AR/VR/MR and educational technologies usually cover discussion in the domain of computer science and engineering.To further analyze the key contributions, the annual publications in the top three subject areas are shown in Figure 6.The contributions started to be more obvious in 2017.Key observations are illustrated in each area: (i) The number of annual publications in computer science remained steady every two years (2017-2018, 2019-2020, and 2021-2022) and stepped up to the next level (about five articles increment); (ii) The number of annual publications in engineering fluctuated (up and down) four times from 2017 to 2021; and (iii) The number of annual publications in social sciences remained steady (about 10 on average).

Figure 5 .
Figure 5. Distribution of the articles based on 22 subject areas.Figure 5. Distribution of the articles based on 22 subject areas.

Figure 5 . 23 Figure 6 .
Figure 5. Distribution of the articles based on 22 subject areas.Figure 5. Distribution of the articles based on 22 subject areas.Electronics 2024, 13, x FOR PEER REVIEW 8 of 23

Figure 6 .
Figure 6.Number of articles in the top three subject areas between 2007 and 2003 (up to 27 July 2023).

Figure 7 .
Figure 7. Number of articles in the top three keywords between 2007 and 2003 (up to 27 July 2023).

Figure 7 .
Figure 7. Number of articles in the top three keywords between 2007 and 2003 (up to 27 July 2023).

Figure 7 .
Figure 7. Number of articles in the top three keywords between 2007 and 2003 (up to 27 July 2023).

Figure 8 .
Figure 8. Word cloud of all keywords.

Figure 8 .
Figure 8. Word cloud of all keywords.

Table 1 .
Geographical distribution of the country affiliations of all the authors.

Table 1 .
Geographical distribution of the country affiliations of all the authors.

Table 2
summarizes the selected list of AR/VR/MR publication sources for innovative learning, with at least two articles published between 2007 and 2023 (up to 27 July 2023).In total, there are 160 sources, 36 of which (22.5%) are listed in Table 2.The leading publication source Lecture Notes in Computer Science, Including Subseries Lecture Notes in

Table 2
summarizes the selected list of AR/VR/MR publication sources for innovative learning, with at least two articles published between 2007 and 2023 (up to 27 July 2023).In total, there are 160 sources, 36 of which (22.5%) are listed in Table 2.The leading publication source Lecture Notes in Computer Science, Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics, is mainly supported by collecting various conference proceedings since 2012.

Table 2 .
Selected AR/VR/MR publication sources for innovative learning (from 2007 to 27 July 2023).

Table 2 .
Cont.Table 3 summarizes the top 40 AR/VR/MR keywords for innovative learning.Figure

Table 4 .
Summary of journal articles in AR/VR/MR for innovative learning (from 2007 to 27 July 2023).

Table 6 .
VR and AR Datasets in Education.