Virtual Reality in Higher Education: A Systematic Review Aligned with the Sustainable Development Goals

Abstract

Virtual Reality (VR) has emerged as a transformative tool in higher education, enabling immersive and interactive learning environments that support the assimilation of complex concepts, hands-on training, and innovative pedagogical practices. This systematic literature review analyzes studies published between 2020 and 2025 that examined the integration of VR in higher education and its connection with the United Nations Sustainable Development Goals (SDGs). Following the PRISMA guidelines, twelve studies were selected from the Web of Science and Scopus databases and assessed using predefined quality criteria. The findings highlight the predominance of mixed-methods approaches, with applications spanning diverse disciplines such as engineering, medical sciences, architecture, teacher training, and sustainability. The results emphasize VR’s potential to enhance student motivation, engagement, and digital competencies, while also contributing to Quality Education (SDG 4), along with other SDGs such as Good Health and Well-Being (SDG 3), Affordable and Clean Energy (SDG 7), Decent Work and Economic Growth (SDG 8), Reducing Inequalities (SDG 10), Sustainable Cities and Communities (SDG 11), and Climate Action (SDG 13). However, persistent challenges include high implementation costs, limited accessibility and teacher training, lack of standardization, and small short-term study designs. This review underscores the need for broader, longitudinal, and interdisciplinary research that integrates underrepresented SDGs and addresses inclusivity, equity, and long-term effectiveness, consolidating VR as a catalyst for innovation and sustainable development in higher education.

1. Introduction

Emerging technologies have gained increasing relevance across different sectors of society in recent years and have been progressively incorporated into the educational field [1]. They provide strategic support tools that facilitate teaching and learning processes and foster knowledge construction across disciplines [2]. These technologies promote student autonomy, collaboration, and the development of 21st-century skills while enhancing engagement and reducing barriers to access quality education [3,4,5,6,7,8,9,10].

The educational applications of emerging technologies have been widely studied, particularly in relation to design approaches, integration challenges, and their influence on creativity [1,11]. These contributions reflect a growing academic interest in understanding the impact of technological innovation and its role in transforming digital ecosystems [12]. Their adaptability, evolution, and constant development reinforce their central role in reshaping education, enabling innovative and effective learning environments that respond to societal changes [13]. Nevertheless, despite students’ positive perceptions of technology-mediated learning, further research is needed to assess its long-term impact on learning outcomes and acceptance [6,14,15].

Evidence has consistently shown that the structured integration of digital tools enhances students’ technological and digital competencies, which are essential for modern workforce demands [16,17,18,19] and adapting to technology-driven processes [20]. However, successful implementation depends largely on educator acceptance, digital competence, and ongoing professional development [9,21,22,23,24]. Supportive and creative learning environments are essential for maximizing the benefits of technology for creativity [25,26,27]. Technological innovation enhances pedagogical processes and creative skills in diverse contexts [11], with creativity support tools and collaborative design platforms enabling experiential learning and problem solving [28,29,30].

Within this context, Extended Reality (XR) has emerged as a response to educational demands, encompassing VR, Augmented Reality (AR), and Mixed Reality (MR) [31]. VR immerses users in fully computer-generated environments, AR overlays digital elements onto the real world, and MR merges both realms to allow interactions between them [11,32,33,34,35,36,37]. These technologies, especially STEAM education, require further infrastructure and research [6,7,38]. The metaverse, an XR-based learning environment, has been recognized for its potential to reduce the digital divide and foster social inclusion [12].

VR is an educational tool that generates immersive and interactive experiences. Through specific hardware and software, VR replaces real-life objects with virtual simulations [39], allowing the exploration of 3D environments that go beyond traditional teaching media [1,40]. Typically implemented via head-mounted displays, VR offers learners opportunities for dynamic simulations, historical reconstructions, and scientific visualizations [32,33,34,37,41,42]. This capacity enables the replication of experiential learning activities, such as academic excursions, within virtual environments [43], thereby promoting engagement, comprehension, critical thinking, collaboration, and practical skills across disciplines [44,45,46,47,48,49,50,51,52].

Student-centered learning has positioned VR as an effective tool for addressing abstract concepts, problem solving, and skill development across multiple disciplines [2,53,54,55]. Its immersive character strengthens teaching and learning in both classroom and professional training contexts and is more effective than traditional technologies [2]. Furthermore, accessibility promotes educational inclusion and reduces inequalities [55].

At the policy level, the United Nations [56] and the European Union have advanced the 2030 Agenda, a global framework comprising 17 Sustainable Development Goals (SDGs) designed to address equity, sustainability, and inclusion across different social domains [57]. Although multiple studies have documented VR’s educational benefits, there is limited consolidated evidence connecting its implementation to specific SDG targets and indicators. Aligning VR with sustainability principles not only enriches the learning experience but also enhances awareness of global challenges, particularly in relation to SDG 4 (Quality Education) [58,59,60].

Effective integration of VR into higher education requires alignment with curricular and sustainable objectives, pedagogical strategies, and continuous collaboration among teachers, researchers, and developers [52,61,62,63]. This necessitates further research to consolidate theoretical and practical frameworks that ensure sustainable adoption and long-term impact, while also addressing inclusivity and equity.

In this context, a systematic review of the literature is essential to synthesize the existing evidence, evaluate VR’s impact in higher education, and assess its contribution to the SDGs [64]. Focusing on studies published between 2020 and 2025, this review provides an updated analysis that reflects current pedagogical and technological developments, avoiding the use of outdated frameworks. This approach provides a robust foundation for guiding future VR research, informing practices, and strengthening the role of higher education in advancing the global objectives of social development, equity, and sustainability.

Thus, the main objective of this systematic review was to analyze the implementation of VR in the context of higher education, study its practical implications, and link it to the SDGs. Accordingly, the specific objectives of this study were as follows:

• To identify the context of VR in higher education and define the research methodology used in this study.
• To examine the main characteristics of VR and its implementation in a higher education context.
• To identify the skills and competencies developed by students using VR in different educational areas and disciplines in higher education.
• To determine and analyze the educational benefits achieved by implementing VR in the higher education context in terms of meaningful learning, motivation, and student participation/engagement.
• To study the intrinsic relationship between the use of VR in higher education and SDGs.
• To identify and define the SDGs pursued and implemented in the classroom during VR in the context of higher education.

2. Methodology

A systematic review of the relevant literature was conducted. The development phases were based on those referenced in the Marin-Juarros (2022) [65] and Garcia-Peñalvo (2022) [66] studies. The review adopted a predominantly descriptive approach as an initial analytical layer, cataloging the characteristics, contexts, and methodological features of the included studies to ensure transparency and facilitate reproducibility of the synthesis process. This descriptive mapping is essential in systematic reviews, particularly when dealing with a heterogeneous evidence base, as it enables the identification of patterns, gaps, and relevant contextual factors before engaging in a deeper critical synthesis. While this structure prioritizes a clear classification of evidence, it is complemented by integrative comparisons and quality assessments that highlight methodological strengths, limitations, and thematic trends across studies, thereby ensuring that the analysis remains both comprehensive and interpretative.

2.1. Guidelines and Research Process

A systematic literature search was conducted from November 2024 to June 2025 across two multidisciplinary databases, Web of Science (n = 57 records retrieved) and Scopus (n = 40 records retrieved), yielding a total of 97 records prior to deduplication.

The first phase involved developing a theoretical framework to justify the topic’s importance and relevance to the educational context. Subsequently, the main objective was defined, and exploration or mapping questions were established together with the research questions to be analyzed based on the results generated in the literature search. To ensure the quality, objectivity, and transparency of the study, the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) method of Moher et al. (2015) [67] was used. This method established guidelines, defining the databases used in the study (Web of Science and Scopus) and delimiting the results to the last six years (2020–2025). Sample collection was based on a search for keywords extracted from the UNESCO and ERIC thesauri. Box 1 defines the terms and Boolean operators used in the databases.

Box 1. Keywords used in the search and combinations of the Boolean operators.

(“Virtual Reality”) AND (“Higher Education” OR “Education”) AND (“Sustainability” OR “Sustainable Development Goals”)

Publications were limited to English and Spanish to ensure the accurate interpretation of the methodological and pedagogical nuances described in the studies. This restriction was justified by the research team’s linguistic competencies and the predominance of scholarly output in these languages in the targeted field from 2020 to 2025. No restrictions were imposed on the country or region of origin.

The search was restricted to the 2020–2025 period to ensure relevance and timeliness of the evidence. This decision responds to the rapid evolution of VR technologies and their pedagogical applications, which have advanced significantly in recent years. Limiting the timeframe makes it possible to capture contemporary methodological approaches and implementations closely aligned with current higher education practices and the SDGs while avoiding the inclusion of outdated studies that do not reflect the present state of the field.

The results were recorded and organized using the Zotero (Personal Research Assistant bibliographic reference manager v. 6.0.36) [68], which was used to register and organize all records of the results. In the first search, 97 bibliographic records were identified, stored, cataloged, and downloaded in RIS format using Rayyan software (AI-Powered Systematic Review Management Platform, 2025) [69]. Two independent reviewers performed duplicate removal and eligibility screening using this software. Discrepancies in the inclusion decisions were resolved through discussion; in cases of persistent disagreement, a third reviewer was consulted.

2.2. Selection and Eligibility Criteria

At the beginning of the screening, the title, keywords, and abstract of each article were considered to check the study topic and its linkage with the research objective (direct agreement with the keywords) in the higher education context, considering the following inclusion criteria (IC) (

Table 1. Inclusion criteria (IC) for the systematic literature review.

N	Inclusion Criteria (IC)
1	Language: Written in English or Spanish, justified by the research team’s linguistic competence and the predominance of scholarly output in these languages within the topic area.
2	Access: Open-access availability to ensure full-text retrieval for in-depth analysis and methodological evaluation.
3	Stage of publication: Fully published articles (excluding in-press, preprints, or early-access versions without final peer review).
4	Study design: Quantitative, qualitative, or mixed-methods research providing sufficient methodological detail to assess quality using the predefined quality criteria (QC) checklist.
5	Study orientation: Empirical studies with a clear pedagogical dimension (teaching strategies, learning outcomes, skills development, assessment practices) rather than purely technical or engineering-focused research.
6	Educational context: Studies conducted specifically in the higher education/university context, involving undergraduate, postgraduate, or doctoral students, and/or higher education faculty.
7	Focus on VR: Research focuses on immersive and non-immersive VR technologies in higher education classrooms and blended learning contexts.
8	Link to SDGs: Explicit mention of SDGs in the title, abstract, or keywords, and/or integration of SDG-related content, indicators, or targets into the pedagogical intervention.
9	Type of publication: Research articles excluding systematic review of the literature.

): The inclusion of studies in this systematic review was guided by explicit and predefined criteria to ensure methodological rigor and relevance to the research objectives. Each criterion was directly linked to the scope of the review, which focused on the implementation of VR in higher education and its explicit connection to the SDGs.

During the review process, a second screening was performed to review the quality and methodological rigor of the twelve documents selected for the subsequent literature review. To this end, a more exhaustive reading was carried out, defining the minimum quality requirements defined by the methodological quality criteria (QC) shown in

Table 2. Quality criteria (QC) were applied for record selection during the second screening phase of the systematic literature review (Kitchenham, 2007) [70].

N	Quality Criterion (QC)	Question/Response (Yes = 1.0, Partially = 0.5, No = 0.0)
1	Sample typology	Is the sample adequate in quantity and quality according to the study objectives and research questions?
2	Methodological design	Is the design well-founded and consistent with the objectives?
3	Clarity of objectives	Are the research objectives/questions clearly defined?
4	Techniques and instruments	Are the techniques or instruments used adequately described?
5	Measurement of variables	Have the variables been measured and assessed rigorously?
6	Research methods	Are the methods clearly defined and justified?
7	Study timeframe	Is the timeframe appropriate for the study typology?
8	Response to objectives/hypotheses	Is there an adequate response to the stated objectives or hypotheses?
9	Publication quality	Is the publication indexed in JCR and peer-reviewed?

. This phase was performed according to the guidelines proposed by Kitchenham (2007) [70]. For this purpose, the level of acceptance by each of the documents with respect to the criteria and questions raised was evaluated: “Yes”, confirming the criterion, with a score of 1.0 points; “partially confirming”, with a score of 0.5 points; as well as “not confirming” the criterion raised, with a score of 0.0 points, thus assessing and confirming with greater precision the quality of the research of the documents analyzed. Finally, a minimum cutoff score of 6.5 out of 9 was established for inclusion in this systematic review. Despite this systematic approach, the final corpus contained a notable concentration of studies (n = 6) from the same journal. This overrepresentation can be attributed to the journal’s high publication output at the intersection of VR, higher education, and SDGs, combined with its open-access policy and high-indexing standards. While this reflects the thematic suitability of the journal’s scope, it also underscores the influence of publication patterns on the composition of evidence-based research. To address potential concerns regarding editorial bias, all selected studies from this journal underwent the same rigorous quality appraisal and scoring process as publications from other sources, ensuring that their inclusion was based on methodological robustness and direct relevance to the review’s research questions rather than solely on their origin.

After screening and applying the inclusion, exclusion, and quality criteria outlined above, twelve records (n = 12) were obtained for the subsequent review and analysis. Figure 1 shows the flowchart developed using the PRISMA method [67,71].

2.3. Mapping and Research Question Selection

Once the parameters that make up the object of study have been selected, mapping questions are defined, which stage the main characteristics of the documents:

• In which years have there been a greater production of research related to VR in higher education?
• Which journals have published research articles, and what is their scientific impact according to their Journal Impact Factor (JIF) or other indicators?
• What is the geographical distribution of VR studies (region, country, etc.) in higher education?

Based on the research objective, the following research questions (RQ) were proposed for further analysis and representation of the results (

Table 3. Research questions (RQ) formulated with respect to the documents selected for review.

N	Research Question
RQ1	What research methodology (quantitative, qualitative, or mixed) is being used and what types of data collection and analysis instruments/tools are used in higher education VR research?
RQ2	What is the subject matter of study for the VR implementation as a learning methodology and what sample has been analyzed in higher education VR implementation research?
RQ3	What are the main areas of application of VR and which academic disciplines contribute to/influence VR research in higher education?
RQ4	What are the latest innovations in VR devices, software, and applications, and what complementary technologies/methodologies are integrated in conjunction with VR for application in higher education classrooms?
RQ5	Which SDGs are addressed and what methodology has been carried out by the application of VR in higher education classrooms? What are the educational benefits?
RQ6	What are perceived limitations and drawbacks regarding the implementation of SDG-linked classroom VR in the higher education context and what research limitations are mentioned/detected in the studies conducted?
RQ7	What is the direction of research on VR in higher education and what recommendations are expressed for future research linked to the SDGs?

):

3. Analysis

Data Extraction and Coding

Based on the theoretical framework and research questions, a structured coding system was designed to extract the most relevant information from each record. The analysis aimed to define and answer the research questions based on the categorization of the variables listed in

Table 4. Research questions and variables from the documents selected for the systematic review.

N		Registration Indicator
RQ	V	Description	Categories/Values
RQ1	V1	Research methodology	Quantitative methodology
			Qualitative methodology
			Mixed methodology
	V2	Data collection instrument	Quantitative instruments
			Qualitative instruments
	V3	Analysis instrument both quantitative and qualitative data
RQ2	V4	Themes addressed in higher education context for the implementation of VR	Description of the research for the implementation of VR
			Discipline and subject area where VR-based learning is implemented in the classroom
	V5	Research sample (students and/or teachers of higher education)	Age range/average age/gender
			Source
			Academic/professional type and/or university degree/master’s/doctorate course.
RQ3	V6	Academic areas related to the implementation of VR in higher education	Areas of application of VR in higher education
			Academic disciplines that have contributed to/influenced research on VR in higher education
RQ4	V7	Recent innovations in VR devices, software, and applications in higher education
	V8	Complementary technologies and/or methodologies that integrate with VR in higher education
RQ5	V9	SDGs that are put into practice when implementing VR in the higher education context	SDG 1: End Poverty; SDG 2: Zero Hunger; SDG 3: Health and Well-Being; SDG 4: Quality Education; SDG 5: Gender Equality; SDG 6: Clean Water and Sanitation; SDG 7: Affordable and Clean Energy; SDG 8: Decent Work and Economic Growth; SDG 9: Industry, Innovation, and Infrastructure; SDG 10: Reducing Inequalities; SDG 11: Sustainable Cities and Communities; SDG 12: Responsible Production and Consumption; SDG 13: Climate Action; SDG 14: Life Below Water; SDG 15: Life on Land; SDG 16: Peace, Justice, and Strong Institutions, and/or SDG 17: Partnerships for Goals
	V10	Educational benefits of using VR in the higher education context and in achieving the SDGs	Favorable educational effects of VR implementation in higher education: motivation, satisfaction, and learning generated (needs covered by students)
			SDGs put into practice during the implementation of VR in the classroom: SDG implementation methodology in higher education
RQ6	V11	Limitations and drawbacks of VR in the context of higher education and/or its linkage to the SDGs	Limitations of the research study when implementing VR in the higher education classroom
			Drawbacks and/or limitations of practical implementation of VR in higher education classrooms and its linkage to the SDGs
RQ7	V12	Future direction of VR research and linkage to SDGs.	Recommendations for future research on VR and SDGs

, as follows: A coding framework comprising variables V1 to V12 was developed to ensure a systematic and comprehensive analysis of the selected studies, directly addressing the seven research questions (RQs) established for this review. The variables were chosen following a deductive–inductive approach: they were deductively derived from the theoretical framework and the explicit dimensions embedded in the RQs, while also incorporating recurrent themes and methodological descriptors identified during the initial scoping of the literature. Specifically, variables V1 to V3 (research methodology, data collection instruments, and analysis instruments) provided the background necessary to answer RQ1, which examines the methodological approaches and analytical strategies employed in VR research in higher education. Variables V4 to V6 (themes addressed in higher education, research sample, and academic areas) corresponded to RQ2 and RQ3, focusing on the subject matter of VR implementation, the type of participants involved, and the disciplinary contributions influencing VR research. Variables V7 and V8 (innovations in VR devices/software and complementary technologies/methodologies) aligned with RQ4 and provided insights into technological advances and pedagogical integration in higher education classrooms. Variables V9 (SDGs put into practice) and V10 (educational benefits) were linked to RQ5, which explored both the SDGs addressed and the pedagogical benefits of VR. Variable V11 (limitations and drawbacks) corresponded to RQ6, identifying methodological and contextual challenges in VR adoption, particularly in SDG-linked classrooms. Finally, Variable V12 (future directions of VR research and linkage to SDGs) supported RQ7, addressing recommendations and perspectives for advancing the field. This explicit alignment between the coding variables and the research questions ensured that the analysis was both methodologically coherent and directly relevant to the aims of the review, while also facilitating the comparability and synthesis of results across studies.

Considering the twelve bibliographic documents registered, a review of tables designed using Microsoft Office Excel 365 software was conducted, structured by columns to organize and extract information from each study, considering the established research questions. The tables present concrete and synthesized information to provide an updated view of the current state of the field of VR studies and its link with the SDGs. The Google Notebook LM tool (Note-Taking and Research Assistant Powered by AI, 2025) [72] was used to consult complementary information in the documents/articles analyzed. To ensure rigor and mitigate the risk of selection bias, the processes of data extraction, coding, and quality appraisal were independently conducted and cross-checked by the three authors of this review, reaching a consensus through discussion in cases of discrepancy, thereby reinforcing the robustness and reliability of the evidence presented.

4. Results

A bibliometric study (n = 12) was conducted to answer the mapping and research questions.

Table 5. Research articles selected for the study and analysis.

N	Article Title	Authors	Year	Authors Institution/Country	Publication Journal
1	The continuum of simulator-based maritime training and education [73]	Kim, T., Sharma, A., Bustgaard, M., Gyldensten, W.C., Nymoen, O.K., Tusher, H.M., & Nazir, S.	2021	Faculty of Technology, Natural and Marine Sciences/Norway.	WMU Journal of Maritime Affairs, 20(2). SPRINGER NATURE.
2	Technological Solutions for Sustainable Development: Effects of a Visual Prompt Scaffolding-Based Virtual Reality Approach on EFL Learners’ Reading Comprehension, Learning Attitude, Motivation, and Anxiety [60]	Wang, Z., Guo, Y., Wang, Y., Tu, Y.-F., & Liu, C.	2021	University of Wenzhou/China & Fu Jen Catholic University/Taiwan.	Sustainability, 13(24). MDPI
3	How Prior Knowledge Affects Visual Attention of Japanese Mimicry and Onomatopoeia and Learning Outcomes: Evidence from Virtual Reality Eye Tracking [74]	Wang, C.-C., Hung, J.C., & Chen, H.-C.	2021	Christian University and National Chang Jung University of Science and Technology/ Taiwan.	Sustainability, 13(19). MDPI
4	Virtual Field Trips in Binational Collaborative Teacher Training: Opportunities and Challenges in the Context of Education for Sustainable Development [75]	Leininger-Frézal, C. & Sprenger, S.	2022	Université Paris Cité/France & University of Hamburg/ Germany.	Sustainability, 14(19). MDPI
5	Innovative Practice of Sustainable Landscape Architecture Education—Parametric-Aided Design and Application [76]	Hsu, C.-Y., & Ou, S.-J.	2022	Chaoyang University of Technology/ Taiwan.	Sustainability, 14(8). MDPI
6	The virtual reality: a sustainable technology [77]	Vergara-Rodríguez, D., Fernández-Arias, P., Santos-Iglesia, C., & Antón-Sancho, A.	2022	Catholic University of Avila/Spain.	DYNA. Ingeniería e Industria, 97(5)
7	Boosting urban community resilience to multi-hazard scenarios in open spaces: A virtual reality-serious game training prototype for heat wave protection and earthquake response [78]	De Fino, M., Tavolare, R., Bernardini, G., Quagliarini, E. & Fatiguso, F.	2023	Politecnico di Bari, Bari/Italy & Università Politecnica delle Marche, Ancona/Italy.	Sustainable Cities and Society, 99. ELSEVIER
8	Exploring the Use of Avatars in the Sustainable Edu-Metaverse for an Alternative Assessment: Impact on Tolerance [79]	Zaky, Y.A.M. & Gameil, A.A.	2024	Ain Shams University/Egypt. King Faisal University/Saudi Arabia.	Sustainability, 16(15). MDPI
9	Horizons architecture with virtual reality for complexity: Mixed methods [80]	Montoya, M.S.R., Perez, S.M., & Orantes, L.P.Z.	2024	Tecnológico de Monterrey/Mexico. & University of Seville/Spain.	Journal of Technology and Science Education, 14(1). JOTSE
10	Comparing cognitive load in learning spatial ability: immersive learning environment vs. digital learning media [11]	Jian, Y., & Abu Bakar, J.A.	2024	University of Utara/Malaysia & Faculty of Electronic Engineering, Chongqing/China	Discover Sustainability, 5(1). SPRINGER
11	Empathy, Education, and Awareness: A VR Hackathon’s Approach to Tackling Climate Change [58]	AlQallaf, N., Elnagar, D.W., Aly, S.G., Elkhodary, K.I., & Ghannam, R.	2024	University of Glasgow/United Kingdom & American University of Cairo/Egypt.	Sustainability, 16(6). MDPI
12	Harnessing the Power of Virtual Reality Experiences as Social Situation of Development to Enrich the Professional Experiences of Early Childhood Pre-Service Teachers [81]	Lee, Q., Devi, A., & Cutri, J.	2025	Swinburne University of Technology, Hawthorn/Australia	Education Sciences, 15(5). MDPI

shows the selected research studies (publication title, authors, year of publication, authors’ country/institution, and journals in which they were published) in chronological order.

A critical assessment is provided regarding the quality of the scientific documents/records selected for the systematic literature review, following the quality criteria and methodological rigor assessment framework outlined by Kitchenham (2007) [70]. Subsequently, the scores obtained from each selected study are presented. The minimum acceptance score (cut-off point) was set at 6.5 out of 9, resulting in the inclusion of the twelve documents listed in Table 5 and

Table 6. Quality scores for each question were obtained from the research papers included in the systematic literature review, following the quality criteria and methodological rigor assessment framework outlined by Kitchenham (2007) [70]. (* The minimum cut-off score established for the selection of documents is 6.5 out of 9 points).

N	Quality Criterion (QC)	Formulated Question	Research Papers Reviewed
			1 [73]	2 [60]	3 [74]	4 [75]	5 [76]	6 [77]	7 [78]	8 [79]	9 [80]	10 [11]	11 [58]	12 [81]
1	Sample typology	Is the sample adequate in quantity and quality according to the study objectives and research questions?	0.5	1	0.5	0.5	0.5	0.5	0.5	1	0.5	1	0.5	0.5
2	Methodological design	Is the design well and consistent with the objectives?	1	1	1	1	1	0.5	1	1	1	1	1	1
3	Clarity of objectives	Are the research objectives/questions clearly defined?	1	1	1	1	1	1	1	1	1	1	1	1
4	Techniques and instruments	Are the techniques or instruments used adequately described?	1	1	1	1	1	1	1	1	1	1	1	1
5	Measurement of variables	Have the variables been measured and assessed rigorously?	0.5	1	1	0.5	1	1	0.5	1	1	1	0.5	0.5
6	Research methods	Are the methods clearly defined and justified?	1	1	1	1	1	1	1	1	1	1	1	1
7	Study timeframe	Is the timeframe appropriate for the study typology?	1	1	1	0.5	1	0.5	0.5	1	1	0.5	1	1
8	Response to objectives/hypotheses	Is there an adequate response to the stated objectives or hypotheses?	1	1	1	1	1	1	1	1	1	1	1	1
9	Publication quality	Is the publication indexed in a JCR quality publication and peer-reviewed?	1	1	1	1	1	1	1	1	1	1	1	1
Total score of the article out of 9 *			8	9	8.5	7.5	8.5	7.5	7.5	9	8.5	8.5	8	8

.

4.1. Mapping Questions Results

All the papers (100%) were published in English. Regarding the temporal distribution of publications, three research papers were published in 2021 (25%) [60,73,74], three in 2022 (25%) [75,76,77], one in 2023 (8.3%) [78], four in 2024 (33.3%) [11,58,79,80], and one in 2025 (8.3%) [81]. These studies indicate an upward trend in VR research in higher education over the past few years, suggesting that this field is steadily growing. The studies analyzed were published in journals with recognized impact factors. Specifically, 50% of the selected papers have been published in Sustainability (MDPI) [58,60,74,75,76,79], with a Journal Impact Factor (JIF) of 3.3 (2023). The rest of the papers reviewed came from diverse journals: the journal Discover Sustainability (SPRINGER) [11], with a JIF of 2.4 (2023); Journal of Technology and Science Education (JOTSE) [80], which presented a Scimago Journal Ranking (SJR) of 0.396; Sustainable Cities and Society (ELSEVIER) [78], with a JIF of 12.0; Dyna-Engineering and Industry [77], with a JIF of 1.0; WMU Journal of Maritime Affairs (SPRINGER NATURE) [73], with a JIF of 2.4; and Education Sciences (MDPI), with a JIF of 2.6 (8.3%, respectively). The predominance of six articles (50%) published in the Sustainability Journal (MDPI) within the final selection can be explained by their thematic alignment with the research focus of this systematic review. Its editorial scope explicitly covers topics such as SDGs, educational innovation, and technology-enhanced learning, thereby directly matching the search keywords and inclusion criteria used in this review. Furthermore, the journal’s broad international authorship, high publication frequency, and commitment to multidisciplinary approaches have facilitated the dissemination of empirical VR studies in higher education across various contexts. This alignment between the journal’s aims and the objectives of the present review naturally resulted in a higher number of eligible studies meeting the quality and relevance thresholds established during the screening process.

Research has been conducted in a wide variety of countries, including Taiwan (25%), China, Spain and Egypt (16.7% respectively), Germany, Saudi Arabia, Mexico, the United Kingdom, France, Norway, Italy and Australia. This confirms the international research nature of the topic addressed in this study: VR in higher education and SDGs.

4.2. Reasearch Questions Results

From the sample research articles and data collected, the findings focused specifically on the application of VR in the context of higher education, representing the implementation of the SDGs in an international educational setting.

4.2.1. Research Methodology and Data Collection/Analysis Instrument Used

According to the research methodology (RQ1/V1), most studies adopted a mixed-methods approach (83.3%), whereas only two studies relied exclusively on a qualitative design (16.7%) [73,81] (one contribution corresponds to a design/implementation article without an empirical sample [78]). This trend reflects a clear preference for methodological triangulation, which combines quantitative rigor with qualitative insights to better capture the complexities of VR learning experiences (

Table 7. Research methodology and data collection/analysis instruments used.

N	Authors and Year	Methodology (Research Method)	Data Collection Instrument Used		Analysis Data Instrument Used
			Quantitative	Qualitative
1	Kim et al. (2021) [73]	Qualitative	Quantitative instruments are not used.	SWOT analysis (identification and evaluation of strengths, weaknesses, opportunities, and threats). Group discussion workshops with experts.	Focus group discussions and review of documents and related research results.
2	Z. Wang. et al. (2021) [60]	Mixed	Pre- and post-tests of reading comprehension in English. Motivation, attitude, and learning anxiety questionnaires (5-point Likert scale).	Semi-structured interviews (nine questions to collect student perceptions: experience, performance, and acceptance of technology). The interviews were recorded audio.	Analysis of covariance (ANOVA) was used to compare the group results and control pre-test scores. Post hoc analysis was used to identify specific differences between the groups.
3	C. Wang et al. (2021) [74]	Mixed	Tests to measure participants’ understanding of onomatopoeias before and after the VR experience: pre-test (cognition, comprehension, and application of onomatopoeia types (15 min) and post-test (similar to pre-test), but conducted after the experience. Eye movement tracking with “HTC VIVE Pro Eye”.	Observation and analysis of the distribution of visual attention.	One-way ANOVA was used to examine correlations. Modified analysis tools from EyeNTNU-120p, software group analysis of eye movement data.
4	Leininger-Frézal & Sprenger (2022) [75]	Mixed	Online questionnaire (four closed-ended questions: 5-point Likert satisfaction scale from “strongly disagree” to “strongly agree”).	Online questionnaire with three free-response questions.	Descriptive analysis of the data obtained from the closed questions.
5	Hsu & Ou (2022) [76]	Mixed	Pre-test and post-test to assess students’ knowledge through mobile learning platforms and flipped classrooms (learning capability by reviewing learning history and providing feedback). Questionnaires: 13 Likert scale questions (5 points) to assess awareness/cognition of learning (7 questions) and satisfaction with learning (6 questions).	Interviews at the end of the semester to collect feedback from students about their experiences on the course (open-ended questions about learning and their recommendations)	-Mean pre-test and post-test scores for each thematic unit. Analysis of the difference between these scores: a paired samples t-test was used to determine the statistical significance of the difference between pre-test and post-test scores. Qualitative analysis of interview responses to identify relevant themes and patterns: teacher observation to assess the effectiveness of teaching strategies.
6	Vergara-Rodríguez et al. (2022) [77]	Mixed	Likert scale questionn-aires from 1 to 10 were used to evaluate variables: usability, ease of use, interaction experience, and motivation.	Teachers’ opinions and direct observation: Phase I and II (areas of influence of VR in the three dimensions of sustainable development (economic, social, and environmental through the analysis of other research); Phase III (perspective of teachers responsible for the activity through direct observation to explain students’ opinions).	Comparative analysis of two technical variants of Virtual Reality (IVR and NIVR). Descriptive statistical analysis of means and standard deviations and Student’s t-test were used to compare the means of the variable.
7	De Fino et al., 2023 [78]	Mixed	Likert-scale questionnaires (engagement, usefulness, ease of use, simplicity/effectiveness, realism); in-game data (play/response time, wrong answers, completion rate).	Open-ended questions (mitigating elements for heat waves, hazards during earthquakes, safe areas); demographic information (age, gender, occupation, education level, VR/training experience).	Mean and standard deviation (Likert items); content analysis of correct responses (open-ended questions); automatic in-game data storage and analysis (time, errors, completion rate); knowledge retention evaluation through repeated post-tests.
8	Zaky & Gameil (2024) [79]	Mixed (Quasi-experimental design)	Pre-test and post-test questionnaires with modified IPTS scale (4-point Likert scale): interpersonal tolerance and respect for others survey (34 closed-ended items). Product design quality sheet: 20-item rubric (4-point Likert scale) assessing interaction and participation, educational suitability, ease of use and interface, sustainability, and scalability (novice = 1; emerging = 2; developing = 3; proficient = 4).	Discussion tools on the metaverse platform: frame VR—compilation of student comments and opinions.	Mann–Whitney U test (significant differences between groups at the beginning of the study). Wilcoxon test: differences between students and teachers’ evaluations of product quality. Calculation of arithmetic means and standard deviations to compare group results. Peer evaluation of the design of sustainable educational environments: students evaluated their peers’ work using avatars. Quality analysis of the product design: evaluation of card criteria.
9	Montoya et al. (2024) [80]	Mixed (Concurrent triangulation design)	Questionnaire: Likert scale to assess student perceptions: 28 Likert scale indicators (5 points). It is structured in two parts: innovation (change/novelty and added value) and types of educational innovation (incremental, systematic, disruptive, and open).	Semi-structured questionnaires: open-ended questions to collect demographic data, SDG interests, and perceptions of students (11 questions), and innovative projects that were generated. Mozilla Hubs platform for interaction and feedback (virtual interaction space “NISA”, where student teams collaborated and received feedback on their projects).	Quantitative analysis was performed using Microsoft Office Excel, pivot tables, filters, statistical functions, and graphs: frequencies and percentages, comparisons with the existing literature, and measurement of the reliability of the Likert scale instrument using Cronbach’s alpha coefficient. Qualitative content analysis: Voyant Tools automatically extract themes from the text, thematic categories, and the frequency of ideas in each category. Responses related to the VR environment were categorized (positive and negative comments) and triangulated with sociodemographic data of the participants.
10	Jian & Abu Bakar (2024) [11]	Mixed (Experimental Design)	Experimental design with two groups: a control group (using digital learning media (DLM) based on slides) and an experimental group (in an immersive learning environment (ILE) based on VR technology. Cognitive load scale. Expert-assessed three-view test (TVT): students’ ability to perform 3D to 2D model conversions (drawing front, top and side views from 3D model) and 2D to 3D conversions (drawing the 3D shape from front, top and side views).	Observation of student behavior: ILE (how they interact with 3D virtual objects, as opposed to the more passive experience in the DLM environment).	Learning performance using a three-view test (TVT): conversion from 3D to 2D models and vice versa. Paired samples t-tests to compare learning performance and cognitive load between DLM and ILE groups; one-way ANOVA to analyze the influence of gender on learning performance and cognitive load in different learning environments; Shapiro–Wilk test to verify normality of data.
11	AlQallaf et al. (2024) [58]	Mixed	Online survey with a 5-point Likert scale to assess understanding of the SDGs, awareness of climate change issues, and hackathon experience (23 questions): rating skills developed, evaluation, experience, and teamwork.	Students’ comments on how their understanding of SDGs improved. 10 min group presentation (with evaluation criteria: clarity and organization, understanding and application of the SDGs, and use of VR/AR technology).	Calculation of percentages of responses on Likert scales to measure participants’ level of agreement with each question; mean (µ) and standard deviation (σ) of responses: scale from 1 to 5 to measure the level of mastery of the VR skills acquired during the hackathon (quantitative) and student comments (qualitative).
12	Lee et al. (2025) [81]	Qualitative	Quantitative instruments are not used.	VR simulations via Mursion© (virtual EC classroom with avatar children); online collaborative teaching sessions (Zoom); online reflections via Padlet; open-ended guiding questions on strategies, challenges, and observations.	Dramatic Events–Social Situation of Development (D-SSD) framework; three levels of interpretation (common sense, situated practice, thematic analysis); color-coding of patterns; reflective vignettes and direct participant quotes.

). However, relevant distinctions exist within this broad pattern. For example, two studies [11,79] applied quasi-experimental designs to test cause-and-effect relationships in controlled settings, such as by comparing immersive learning environments with digital resources or by examining interactions in metaverse-based platforms. In contrast, qualitative contributions [73,81] prioritized qualitative SWOT analyses and open-ended questionnaires, placing greater emphasis on contextual and subjective dimensions.

The analysis of the data collection instruments (RQ1/V2) also revealed convergence across studies (Table 7). Online questionnaires dominate (66.7%), with Likert-type scales appearing in almost all cases (83.3%), often combined with pre-/post-test measures (41.7%) to assess changes in knowledge, skills, or attitudes [60,74,76,78,79]. However, complementary instruments provide additional nuances: Interviews and focus groups were employed in studies such as Wang et al. [60] and Hsu & Ou [76], while teacher-led observations were used by Jian & Abu Bakar [11] and Vergara-Rodríguez et al. [77] to capture students’ authentic interactions with VR technologies; Zaky & Gameil [79] used product-quality evaluation rubrics and platform discussion tools, and Lee et al. [81] implemented VR classroom simulations (Mursion), online collaborative sessions (Zoom/Padlet), and reflective coding using the D-SSD framework, without quantitative instruments.

The SWOT analysis by Kim et al. [73] highlights the institutional strengths and weaknesses of adopting immersive environments, an aspect that is absent from other experimental approaches. Taken together, these results show that while quantitative strategies remain predominant, qualitative methods play a crucial role in contextualizing the learning experience. The overlaps between studies (for instance, the combined use of Likert questionnaires and interviews in Wang et al. [60] and Hsu & Ou [76]) illustrate a shared concern for balancing measurement precision with contextual interpretation. Simultaneously, divergences in methodological emphasis reflect not only different research traditions but also the diversity of objectives pursued, from evaluating performance outcomes to capturing students’ perceptions and institutional challenges. This combination of overlaps and contrasts suggests that VR research in higher education is methodologically diverse but still uneven, with a strong inclination toward short-term evaluation designs, leaving a gap for longitudinal and cross-institutional studies (Table 7). Finally, De Fino et al. [78] did not collect user data at all (it details hazard modeling, scenario sequencing, and validation), and specified in-game telemetry variables (play/response time, errors, completion rate) and repeated post-tests for knowledge retention evaluation.

Regarding the type of data analysis instrument (RQ1/V3), most studies relied on quantitative techniques (66.7%), while qualitative tools accounted for 33.3%. Descriptive statistical analyses (means, standard deviations, and frequencies) appear in 66.7% of the studies [11,58,60,74,75,76,78,79]; inferential statistical tests, such as Student’s t-tests and paired-sample tests, were used in 33.3% of the studies [11,76,77,79]; and ANOVA/ANCOVA was used in 25% of the studies [11,60,74]. Non-parametric approaches (Mann–Whitney U, Wilcoxon, and Shapiro–Wilk tests), were less common, appearing in only 16.7% of the cases [11,79]. In line with its design demonstration nature, De Fino et al. [78] reported modeling and specification rather than statistical hypothesis testing. In contrast, Lee et al. [81] applied analytic procedures D-SSD-guided thematic coding (three interpretation levels), color-coded patterning, reflective vignettes, and direct participant quotations without using quantitative statistical tests. These patterns indicate a shared interest in validating VR’s effectiveness through comparative and quasi-experimental testing, although they are often constrained by small sample sizes and short time frames.

Notably, Likert scales combined with reliability measures (Cronbach’s alpha) were used in 16.7% of the studies [60,80], signaling concern for internal consistency when measuring attitudes and perceptions. However, reliance on standardized scales may limit the depth of understanding of the nuanced or contextualized aspects of the VR learning experience. In contrast, qualitative analysis tools provided richer contextual perspectives, although they were used less frequently. Focus groups have been employed in studies such as Kim et al. [73] and Hsu & Ou [76], allowing participants to articulate subjective experiences with VR, while AlQallaf et al. [58] and Hsu & Ou [76] triangulated teacher observations with student feedback to provide a more holistic interpretation of results. These overlaps reveal a convergence in the use of observational data to complement quantitative measures, illustrating that some studies have aimed to mitigate the reductionism of purely statistical results. These findings highlight a methodological tension: while quantitative analysis dominates, ensuring measurable and comparable outcomes, qualitative approaches remain secondary and complementary, often relegated to supporting roles rather than being integrated as central tools. This uneven balance mirrors the field’s broader trend toward prioritizing effectiveness testing over contextual understanding, pointing to a gap for future studies to incorporate more balanced mixed-methods analyses capable of capturing both measurable impacts and the lived experiences of students in VR learning environments.

4.2.2. Study Object and Research Sample

Several thematic clusters were identified regarding the subject matter addressed in the implementation of VR as a learning methodology in higher education (RQ2/V4) (

Table 8. Subject matter and sample used in the study.

N	Authors and Year	Subject/Study Object	Research Sample
1	Kim et al. (2021) [73]	Simulator-based maritime training and education (MET). Role of simulators in post-COVID-19 education and their contribution to SDG 4; highlights the importance of instructors in distance learning environments.	Seven maritime researchers with expertise in simulator-based maritime training and education
2	Z. Wang. et al. (2021) [60]	The paper examines technological solutions for sustainable development, focusing on the effects of a Virtual Reality (VR) approach based on visual prompt scaffolding (VPS) on the reading comprehension, learning attitude, motivation, and anxiety of English as a Foreign Language (EFL) learners.	98 university students from China, with an average age of 19 years. The participants were divided into three groups: Experimental group A (N = 31): learning with the VPS-VR2 approach; Experimental group B (N = 32): learning with the VR2 approach; Control group (N = 35). All students with previous digital reading experience; however, none of them used VR in their English courses.
3	C. Wang et al. (2021) [74]	Influence of prior knowledge on Japanese students’ visual attention when learning Japanese onomatopoeia and mimicry (MIO) using VR eye tracking. Exploring differences in the distribution of their visual attention and learning six mimicry and onomatopoeia topics, including weather change, speed, mood, rotation, animal sounds, and food temperature.	20 students from the Department of Applied Japanese at a university in Taiwan. The average age was 20.6 years (range, 20–22 years). They were divided into two groups according to their Japanese Language Proficiency Test (JLPT) certification level—high prior knowledge group (levels N1-N3): 7 participants; low prior knowledge group (level N4 or below): 13 participants.
4	Leininger-Frézal & Sprenger (2022) [75]	Virtual field trips (VFT) in binational collaborative teacher training in the context of education for sustainable development (ESD). The main objective is to develop a didactical approach for the use of virtual fieldwork in ESD with geography teachers in the initial and continuing education in two universities (Hamburg and Paris). Geography in the context of education for sustainability development (ESD).	22 university students (11 from Germany and 11 from France) of Master’s in Geographic Education and Master’s in Didactics (19 women and 3 men) participated in this study. Teacher training: initial training in Germany (Bachelor of Education program (Master’s in Geographical Education) and continuous training in France (Master’s program to improve their didactic skills, Master’s in Didactics).
5	Hsu & Ou (2022) [76]	Innovative practice of sustainable landscape architecture education using parameter-aided design and its application: innovate teaching design by combining BOPPPS teaching structure and a design-based learning model to build a knowledge chain of landscape architecture design modeling and inspired logical thinking.	24 students (11 males and 13 females) from the Taiwan University of Science and Technology (final year of undergraduate study) participated in this study. Students’ familiarity with 3D modeling was considered. The students were divided into high-, medium-, and low-performing groups according to their scores in the first semester of the initial course.
6	Vergara-Rodríguez et al. (2022) [77]	Use of VR as a sustainable technology, comparing immersive VR (IVR) and non-immersive VR (NIRVR): advantages and limitations in sustainable development (economic, social, and environmental).	11 students of the Materials Engineering course of the Mechanical Engineering degree at the Catholic University of Avila during the 2020–2021 academic year and two professors responsible for the training action, who provided their qualitative observations on the use of RVI and RVNI in higher education.
7	De Fino et al., 2023 [78]	Development of a VR-serious game (VR-SG) prototype for urban community resilience training in multi-hazard outdoor scenarios. Focus on heat wave protection and earthquake response, aiming to enhance adaptive human–urban–building interactions and promote safety and sustainability. The literature review highlights VR applications in Architecture, Engineering, and Construction (AEC) for design education, worker safety preparation, operational instruction, and emergency training.	No participants analyzed yet, as the study focuses on prototype development. Future validation is planned with a representative Italian population sample segmented by age groups (18–35, 35–50, 50–60 years). Participants will test different training modalities (non-interactive visualization, non-immersive game, immersive VR headset game).
8	Zaky & Gameil (2024) [79]	Use of avatars in the sustainable Edu-Metaverse (the cause) as an alternative assessment method and its impact on the development of tolerance and respect for others: cause and effect relationships in the metaverse as a virtual world that simulates the real environment using 3Deducational graphics.	36 (100% female) undergraduate students (22–35 years old) from the Master’s program in Educational Technology at King Faisal University (Saudi Arabia). The participants had no prior experience using the metaverse. They were randomly divided into two groups of 18 students each.
9	Montoya et al. (2024) [80]	Innovations that graduate students perceive in environments that use horizontal architecture to integrate VR. This study focuses on how this combination fosters complex reasoning and the search for new solutions in undergraduate education (through individual and collaborative activities).	99 graduate students from Spain, China, the United States, and Latin America: (20–55 years old, predominantly 35–39 years old). Enrolled in Humanities and Education programs: Educational Entrepreneurship (MTO), Digital Humanities (MHD), and Management for Educational Leadership and Innovation (EHE) at Tecnológico de Monterrey. Most of the students worked in the field of education (faculty members or administrative positions in educational institutions). The group included students from various graduate programs (a multidisciplinary approach).
10	Jian & Abu Bakar (2024) [11]	Comparison of cognitive load in spatial skill learning using an immersive learning environment (ILE) based on Virtual Reality (VR) vs. slide-based digital learning media (DLM). Learning performance and cognitive load were also assessed. (Theoretical knowledge by playing videos, handling 3D virtual objects with Virtual Reality devices, etc.	28 first-year undergraduate students (14 males and 14 females), average age 19.96 years old, from the art degree program at Chongqing University (China). Group control (DLM) and experimental group (ILE) based on their admission qualifications and gender, ensuring equal distribution in both groups. Almost 90% of the students had not previously used VR.
11	AlQallaf et al. (2024) [58]	Three-day VR hackathons represent a novel approach to climate change education. Objective: To foster empathy, education, and awareness regarding four SDGs: Quality Education, Affordable and Clean Energy, Sustainable Cities and Communities, and Climate Action. Use of VR technology to design immersive environments that will demonstrate their understanding of the SDGs: collaborative nature, development of teamwork skills, and problem solving.	14 students from interdisciplinary backgrounds were selected from 22 shortlisted disciplines (and from the 36 initially interested): computer science, computer engineering, information and computer systems, mechanical engineering, and renewable energy. They were selected based on their technical skills in VR development and commitment. They were divided into five practical groups.
12	Lee et al. (2025) [81]	Professional preparation and development of pre-service teachers (PSTs) in Early Childhood (EC) education through Virtual Reality (VR) experiences using the Mursion© program. Focus on placement preparation, classroom management, behavioral guidance, adaptation to diverse children, application of theoretical knowledge, and development of professional and procedural competencies.	66 post-graduate PSTs enrolled in an Australian Initial Teacher Education (ITE) program (EC specialization). Mainly international students with diverse cultural/linguistic backgrounds. Data collected during Quarters 2–4 of 2024.

). The first group of studies focused on the impact of VR on education and cognitive processes, such as language acquisition and comprehension [60,74], engineering, and spatial learning [11,77], and the motivational effects of immersive and non-immersive environments [77]. Although these studies vary in discipline, they converge on reporting improvements in comprehension, attention, and student motivation. In addition, teacher education research using avatar-based VR simulations [81] reinforces this cluster by demonstrating professional skill development among pre-service teachers. However, they also revealed differences in the scope of outcomes: while Wang et al. [60] and C. Wang et al. [74] emphasized perceptual and linguistic skills, Vergara-Rodríguez et al. [77] and Jian & Abu Bakar [11] addressed more technical and cognitive dimensions such as spatial ability and engineering performance.

The second group emphasizes VR and sustainability, exploring its role as a sustainable technology [76] and its potential to promote awareness of climate change and SDGs through innovative practices such as hackathons [58]. Hsu & Ou [76] employed parametric design and VR modeling to teach sustainable landscape architecture, while AlQallaf et al. [58] used collaborative VR activities to foster climate awareness. De Fino et al. [78] presented a VR-serious game prototype for multi-hazard training (heat wave protection and earthquake response) in urban open spaces, outlining instruments for engagement/realism and knowledge-retention evaluation. Despite the different strategies, both highlight the potential of VR as a cross-cutting tool linking education with the 2030 Agenda.

The third group addresses technological innovation and digital education, where VR is applied to assess tolerance and inclusion in the Edu-Metaverse [79] or virtual field trips (VFT) to foster collaborative geography teacher education [75]. Similarly, Kim et al. [73] analyzed simulators in maritime education post-COVID-19, highlighting the relevance of VR for training in the context of restricted physical mobility. A teacher education study [81] also fits this strand, using VR classroom simulations and online collaborative sessions to support reflective practice and professional preparation. These examples illustrate how VR can expand learning opportunities beyond traditional classrooms, although the studies differ in their focus; some emphasize ethical and affective dimensions [80], whereas others highlight professional and technical training [73,77].

Regarding the research samples (RQ2/V5), diversity was evident across geographic origins, ages, and educational levels (Table 8). Most participants were undergraduate and graduate students (83.3%), although experts and teachers were also included in some cases [73,77]. This combination broadens perspectives but also introduces heterogeneity, complicating cross-study comparisons. Geographically, studies from Asia, Europe, and Latin America reflect the global scope of VR research. However, certain regions, particularly Africa and North America, remain underrepresented. Gender balance is inconsistent, with some studies reporting a predominance of women [75,76,77] or equal participation [11]. Approximately 75% of the studies did not report disaggregated gender data, which represents a notable limitation in assessing inclusivity and aligns with broader critiques of gender underreporting in educational technology research. In summary, although the subject matter and samples analyzed demonstrate broad thematic and geographic diversity, patterns of convergence can be observed: a predominance of studies in STEM and sustainability-related fields, reliance on student populations, and repeated alignment with SDG 4 (Quality Education). At the same time, divergences such as whether the emphasis lies on cognitive, technical, or ethical dimensions underline the multiplicity of research approaches in this field. These overlaps and contrasts suggest that while VR is recognized as a versatile tool for higher education, important gaps remain in terms of inclusivity, disciplinary balance, and the systematic integration of underexplored SDGs.

4.2.3. Academic Areas and Disciplines of Application and Influence of VR in Higher Education

The application of VR in higher education spans a wide range of academic disciplines (RQ3/V6), reflecting its versatility and revealing specific disciplinary concentrations (

Table 9. Influence and application of VR in academic areas and disciplines in higher education.

N	Authors and Year	Academic Areas and Disciplines of Application and Influence of VR in Higher Education
1	Kim et al. (2021) [73]	Support learner motivation and engagement and promote higher-order learning. VR simulators can be used in different environments because of their compactness and high mobility, resulting in a higher return on training investment.
2	Z. Wang. et al. (2021) [60]	VR has been applied in a variety of educational contexts, including STEM-related disciplines, such as physics, chemistry, medical courses, art, and history.
3	C. Wang et al. (2021) [74]	VR has applications in a variety of educational contexts: STEM, medicine, computer engineering, industrial engineering design and environmental education.
4	Leininger-Frézal & Sprenger (2022) [75]	VR focuses on the use of virtual field trips (VFT) in teacher education, specifically in the context of education for sustainable development (ESD).
5	Hsu & Ou (2022) [76]	VR has an influence on areas such as landscape architecture and industrial design. VR has gained presence and influence on the entertainment and film industries.
6	Vergara-Rodríguez et al. (2022) [77]	VR impacts academic areas, such as engineering (materials and mechanical), education, mathematics, and experimental science. The sustainability of VR in economic (development costs, maintenance costs, facilities, transportation, economic activity, energy cost, and reusability), social (human safety, education and knowledge transfer, health and quality of life), and environmental (biodiversity conservation, pollution (noise, air, and environmental), reusability, and resource scarcity) areas.
7	De Fino et al., 2023 [78]	Main VR application: development of a VR-serious-game prototype for multi-hazard urban resilience training (heat wave and earthquake response). Literature review also highlights VR in AEC education, emergency training, and single-hazard simulations (fire, earthquake, flood, terrorist attack). Disciplines involved: Civil, Environmental, and Architectural Engineering (urban planning, risk reduction); Computer Science and Game Development (game engines, 3D modeling, serious games); Behavioral Science/Psychology (responsive behaviors, user experience); Urban Planning and Disaster Management (community resilience, sustainable cities); Data Science and Simulation (CFD, hydraulic and agent-based models); Ethics (autonomy, privacy, motion sickness, realism implications).
8	Zaky & Gameil (2024) [79]	Use of avatars in sustainable educational metaverse, but the academic areas and disciplines of VR applications were not explicitly stated.
9	Montoya et al. (2024) [80]	Explore the integration of architecture with VR in complex environments: distance learning modalities and project presentations.
10	Jian & Abu Bakar (2024) [11]	VR is used to design an immersive learning environment (ILE) in the context of art (art education, audiovisual communication design, and architectural design).
11	AlQallaf et al. (2024) [58]	Implementation of VR with four SDGs (Quality Education, Affordable and Clean Energy, Sustainable Cities and Communities, and Climate Action) in academic areas and disciplines such as computer engineering, mechanical engineering, renewable energy engineering, and computer science.
12	Lee et al. (2025) [81]	Teacher education (Initial Teacher Education—Early Childhood). VR bridges theory–practice gaps, supports professional (child development, assessment, professionalism) and procedural competencies (classroom management, health/safety, curriculum, reflective practice), and builds confidence/readiness, especially for international students. VR was also applied in elementary, special needs, and secondary education (though less common in EC), with established precedents in medicine and aviation as benchmarks. Disciplines contributing/influencing: Education Sciences and Teacher Education (esp. Early Childhood, Special Education); Psychology/Cultural–Historical Theory (Vygotsky’s SSD and Dramatic Events framework); Social Sciences (interpretation of social interactions); Computer Science and Technology (VR platforms like Mursion©, Padlet, Zoom, mixed-reality/immersive technologies).

). A significant proportion of studies (33.3%) situate VR within STEM (Science, Technology, Engineering, and Mathematics)-related fields, notably computer science, engineering, and mathematics [11,73,77,78]. In these contexts, VR is typically employed for simulation, visualization, and skill-based training, reinforcing its value in technical-learning environments. Architecture and artistic design represented 33.3% of the contributions [11,73,76,80], where VR was leveraged to enhance spatial skills, design processes, and immersive creativity. These overlaps underscore the predominance of VR in practice-oriented disciplines, where immersion directly supports task performance and experiential learning.

Beyond STEM and design, environmental education and sustainability appeared prominently in 41.7% of the studies [58,74,75,77,78]. VR is used to foster awareness of social, economic, and ecological challenges and link education to global sustainability goals. Hsu & Ou [76] integrated VR with parametric modeling for sustainable landscape architecture, whereas AlQallaf et al. [58] organized VR hackathons to promote climate change awareness, and De Fino et al. [78] added a resilience-training perspective for heatwaves and earthquake responses in urban open spaces. These approaches differ in strategy but converge in highlighting VR as a tool for addressing sustainability challenges in line with the 2030 Agenda.

Other disciplinary applications, although less frequent, expand VR’s potential impact. For instance, teacher training has been addressed in 25% of the studies [73,75,81], with virtual field trips, collaborative environments, and avatar-based classroom simulations supporting the development of digital and professional competencies. Likewise, medical education [60] and cinema/entertainment remain niche areas, and the strand on inclusion/tolerance is represented in the Edu-Metaverse study [79]. While these applications demonstrate VR adaptability across contexts, their limited presence indicates that research has been unevenly distributed, privileging technical disciplines over humanities and social sciences.

The main academic areas and disciplines influenced by the application of VR in higher education (RQ3/V6) are diverse, as shown in the analyzed documents (Table 9). In summary, the main academic areas influenced by VR in higher education remain diverse: STEM (33.3%), architecture/design (33.3%), environmental/sustainability (41.7%), and teacher education (25%), with smaller footprints in health and entertainment fields. This underscores a persistent gap: extending VR into underexplored domains so that its benefits are not restricted to technical fields but also support holistic objectives, such as inclusivity, critical citizenship, and cultural understanding.

4.2.4. Innovation (Device and Software) and Complementary VR Technologies

VR and innovation in the use of technological tools in the form of devices, hardware, and applications are used in higher education (RQ4/V7) (

Table 10. Devices, software, applications used in VR, and technologies or methodologies complementary to VR in the context of higher education.

N	Authors and Year	Devices, Software and/or Applications Used	Complementary VR Technologies/Methodologies
1	Kim et al. (2021) [73]	VR simulators: head-mounted displays (HMD) to immerse users in a realistic work environment. Cloud-based S-simulators, which allow instructors and learners to run the simulation online: web browser with their own devices (PCs, laptops, tablets or mobiles).	- Fullmission simulators can be used in combination with emerging technologies, such as VR. - Desktop simulators: for product/equipment familiarization. - Potential methods such as eye tracking can be used in conjunction with VR.
2	Z. Wang. et al. (2021) [60]	DPVR HMD M2 Pro (Shanghai Lexiang Information Technology Co. Ltd), M-Polaris compatible device that allows free movement within the VR scene and accurately captures the user’s position and movement: a portable and easy-to-operate device with a touch panel and physical buttons, including an object distance adjustment button for students with myopia. Software: Adobe Premiere Pro, as an editing tool for teachers to create learning environments with VPS- VR: to edit 3D videos and provide visual cues to students during VR reading activities.	- Visual Prompt Scaffolding (VPS): VPS is integrated with VR to improve reading comprehension in EFL. - Explanatory and complementary information while reading in VR environments: didactic cards with explanatory information that appear when a new word appears in the VR video. - Spherical videos: The learning content is based on videos edited by teachers. They were required to complete reading comprehension tests at the end of the video.
3	C. Wang et al. (2021) [74]	HTC VIVE Pro Eye (HTC Corporation City, Taoyuan Country, Taiwan): Head-Mounted Display (HMD) device used for eye tracking while viewing VR content. It has two AMOLED displays with a resolution of 2880 × 1600 pixels, a refresh rate of 90 Hz, and a field of view of 110°. EyeNTNU-120p: Integrated analysis software for collecting eye movement data with a sampling rate of 120 Hz and an accuracy of 0.5°. It is used to capture pupil movements and serves as a VR eye tracker. Unreal Engine 4 (UE4): Development tool used to design a virtual 3D theme park called “MIO Land” (Mimicry and Onomatopoeia).	- Integration of VR and eyetracking technology: Analyze dynamic visual behavior during Japanese onomatopoeia learning. Eye tracking technology: Tracking and recording eye movements of participants while interacting with VR content (allows collecting data on visual attention). Eye tracking was integrated with the HTC VIVE Pro Eye HMD and the EyeNTNU120p software. - VR Dynamic ROI Tool and Fixation Calculator Tool: Modified auxiliary tools of the EyeNTNU120p eye movement analysis software. - The VR Dynamic ROI Tool helps operators define regions of interest (ROIs), while the Fixation Calculator Tool automatically processes ROIs according to priority.
4	Leininger-Frézal & Sprenger (2022) [75]	The type of VR device used is not specified. Esri Story Maps: Offers a way to explore the world from home. It allows one to mark locations on a map and link them with text, images, or videos to create a tour. The National Aeronautics and Space Administration (NASA) website contains images and animations useful for exploring places.	- Padlet: Allows you to mark places digitally with pins on a map for collaborative work. - Google Earth: Allows you to search and zoom in to specific locations and virtually walk the streets using Street View. - Open Street Map: Similarly to Google Earth, it is useful for virtual exploration of places. - Geoportal: It is a graphical interface with a Geographic Information System (GIS) that allows space exploration through a wide range of layers, although it does not allow walking in 3D images. - uMap: It is opensource software that allows users to create maps quickly.
5	Hsu & Ou (2022) [76]	Use of VR visualization devices such as the following: Parametric VR modeling: Develop models and provide optimal designs under different conditions and constraints.	- Visual Programming Language (VPL): Used through the Rhino. - 3D Printing: Used to print the results of parametric modeling; allows students to evaluate the feasibility of their design ideas and reduce potential problems in actual production.
6	Vergara-Rodríguez et al. (2022) [77]	The type of VR device used was not specified, although an HMD was used. Use of educational VR applications, both IVR and NIVR. In this context, it is described that the IVR application used by the students is an adaptation of the NIVR application: Non-Immersive VR (NIVR): The user is not immersed in a virtual world, and his senses (except sight and hearing) receive sensations from the real world and Immersive VR (IVR). The user’s senses receive more virtual sensations, increasing the degree of immersion and interaction in a virtual world.	There has been no mention of complementary technologies/methodologies that integrate VR.
7	De Fino et al., 2023 [78]	Meta Quest 2 (Meta Platforms Technologies Ireland Limited, Merrion Road, Dublin 4, D04 X2K5, Irlanda) headset for immersive mode; gyroscopic sensors for visual orientation; Oculus Touch controllers for gesture interaction; mouse/keyboard for non-immersive mode. VR Software: Epic Unreal Engine 4.6® (simulation engine); MAXON Cinema4D® (3D modeling, NPC rigging, debris simulation); microclimate toolkit for UTCI mapping; damage matrix for falling debris; agent-based simulations for crowd behavior. VR Applications: Development of a modular VR-serious game prototype for multi-hazard urban training (heat wave, earthquake, post-earthquake). Broader VR applications in AEC.	VR is the central technology, applied through a VR-serious game (VR-SG) prototype for multi-hazard resilience training in urban open spaces. The system integrates phenomenological and behavioral analyses, Built Environment Typologies (BETs), and simulation-based data (UTCI maps, debris damage matrices, and agent-based crowd simulations). Developed with Unreal Engine 4.6® and MAXON Cinema4D®, the prototype features animated NPCs with AI-based pathfinding, modular learning units, and dual immersive/non-immersive modes.
8	Zaky & Gameil (2024) [79]	The type of VR device used is not specified, although HMD is used. Edu-Metaverse Platforms: Immersive virtual spaces (3D) and avatars—virtual representations of users collaborating to explore and implement sustainable practices. Frame VR: Metaverse platform used for practical applications in this study. It allows virtual classes to enhance the learning experience.	- Product quality evaluation cards: They allow the measurement of students’ performance in the design of sustainable EduMetaverse environments through specific criteria and rubric. - Use of tools within the metaverse platform linked to VR: virtual whiteboards, video files, PDF files, and the ability to display different websites.
9	Montoya et al. (2024) [80]	Does not specify the type of VR device but indicates that users are immersed in virtual environments using HMDs. The Mozilla Hubs platform is used to create a virtual interaction space called NISA Space Center, where student teams interact to learn about their projects, and provide and receive feedback.	- LMS courses with visual elements for space missions, including logos for each “spacecraft” (APOLLO, SATURN, ATLAS) representing different project approaches. - Horizons Architecture: This methodology is used to structure students’ entrepreneurship and innovation projects, helping them to visualize future scenarios and manage the resources needed to achieve their SDGs.
10	Jian & Abu Bakar (2024) [11]	Use of HMD allows students to enter the immersive 3D virtual space to view, explore, and participate in the interactive process as an active element of the environment. The type of VR device used is not specified. Teacher’s avatar: Students in the experimental group learned theoretical knowledge.	The focus is not on complementary technologies/methodologies, but on the impact of the VR learning environment.
11	AlQallaf et al. (2024) [58]	HMD: HTC Vive (HTC Corporation City, Taoyuan Country, Taiwan) and Meta Quest 2 (Meta Platforms Technologies Ireland Limited, Merrion Road, Dublin 4, D04 X2K5, Irlanda) VR headsets. VR/AR software and platforms for acquiring skills. Game graphics engines such as Unity and Unreal are used to develop and create VR applications. An app was developed to increase empathy, education, and awareness of SDGs.	- Game design software: Game design software to design immersive environments that will demonstrate their ODS understanding. - The hackathon integrated authentic assessments, which reflect realworld engineering tasks and provide a more practical and relevant learning experience. - Teambased learning (TBL): Incorporated teambased learning to respond to the demands of engineering accreditation bodies.
12	Lee et al. (2025) [81]	Mursion© simulation software creates realistic Australian Early Childhood classrooms with avatar children (3–5 years) to practice teaching skills, classroom management, behavioral guidance, and adaptation to diverse learners.	- Online platforms: Zoom for collaborative smallgroup teaching (20 min sessions to avatar children); Padlet for anonymous reflections and peer learning. - Pedagogical frameworks: Cultural–Historical Theory (Social Situation of Development—SSD, Dramatic Events—D); practicetotheory approach; iterative cycles of practice, feedback, and reflection; curriculumaligned lesson plans. - Analytical methods: Hedegaard’s (2008) three levels of interpretation (common sense, situated practice, thematic analysis).

).

The analyzed studies revealed clear patterns and fragmentation in the use of devices, software, and complementary technology. In terms of hardware, head-mounted displays (HMDs) predominated rather than being universal (75%), confirming their central role in VR immersion. Representative models, such as the DPVR M2 Pro [60], HTC VIVE Pro Eye [58,74], and Meta Quest 2 [58,78], reflect reliance on commercially available technologies. However, the diversity of equipment and the absence of standard reporting practices hinder comparability across studies. In parallel, other devices such as desktop/full-mission simulators [73], 3D printing technologies [60], and interactive/virtual whiteboards [79] illustrate VR’s integration with broader ecosystems of learning tools. Although less common, these complementary devices highlight efforts to enhance immersion through multimodal experiences, such as the dual immersive/non-immersive configuration [78], which combines HMD + motion controllers and mouse/keyboard modes. Along with HMDs, VR ecosystems integrate other devices and peripherals (object tracking and eye-tracking sensors [74] and physical prototyping/3D printing for instructional video production [60]). Screen-based VR implementations (Mursion) without dedicated HMDs have also been developed [81]. However, they also underscore inequalities in terms of accessibility and cost.

The most frequently employed Unreal Engine (UE4) and Unity [60,74,78,79] are used to develop immersive educational environments alongside specialized tools, cloud-based simulation software [73], and MAXON Cinema4D for asset creation, NPCs, AI pathfinding, and agent-based crowd simulation [78]. Digital applications for geospatial exploration (Google Earth, OpenStreetMap) and collaborative mapping/boards (uMap, Padlet, Geoportal) are common in geography education [75]. Metaverse platforms, such as FrameVR, support multi-user collaboration and artifact sharing [79]. Specialized editing tools, such as Adobe Premiere Pro for 3D/360° video, are used for educational media workflows [60]. This breadth illustrates VR flexibility but also exposes the absence of a unified technological reporting framework.

The integration of complementary methodologies (RQ4/V8) reflects an attempt to maximize VR’s educational potential by aligning technology with pedagogical approaches (Table 10). Examples include eye tracking to analyze visual attention [74], visual prompt scaffolding (VPS) to improve EFL reading comprehension [60], and team-based learning (TBL) in VR hackathons to promote collaboration and problem solving [58].

Hsu & Ou [76] incorporated visual programming languages (VPL) to enable parametric modeling, while horizon-based design approaches were employed for resource management in sustainable architecture. In teacher education, Lee et al. [81] combined screen-based VR classroom simulations (Mursion) with Padlet/Zoom activities, guided by the D-SSD framework (reflective vignettes, thematic coding).

De Fino et al. added in-game telemetry (play/response time, errors, completion) and repeated post-tests for knowledge-retention evaluation in a multi-hazard VR-serious game [78]; metaverse tasks used product-quality evaluation cards and platform discussion tools to structure collaborative design and reflection [79]. These combinations show that VR is increasingly embedded in hybrid multi-technology learning environments.

Taken together, this evidence reveals both convergence and disparities. On the one hand, the universal reliance on HMDs (75%) and the recurring use of Unity/Unreal highlight common ground in the technological foundations of VR research. On the other hand, heterogeneous device configurations (simulators without HMDs [73], screen-based avatar simulations [81], dual immersive/non-immersive modes [78]) and uneven reporting practices indicate fragmentation, which limit comparability and replicability across contexts. This duality suggests that while innovation in VR integration is advancing, the field still requires greater consistency in technical documentation and clearer pedagogical frameworks to ensure the scalability and sustainability of technological innovations in higher education.

4.2.5. Sustainable Development Goals Related to the Application of VR in the Higher Education Context, Methodology for Classroom Application and VR Benefits

The integration of VR into higher education is strongly connected to the Sustainable Development Goals (RQ5/V9), but in an uneven manner (

Table 11. SDGs related to VR applications in the context of higher education. (An “X” indicates that the respective article addresses or contributes to the achievement of the corresponding Sustainable Development Goal (SDG) in the context of higher education).

Sustainable Development Goals (SDGs)			Article
	1 [73]	2 [60]	3 [74]	4 [75]	5 [76]	6 [77]	7 [78]	8 [79]	9 [80]	10 [11]	11 [58]	12 [81]
SDG 1: No Poverty
SDG 2: Zero Hunger
SDG 3: Good Health and Well-Being							X	X
SDG 4: Quality Education	X	X	X	X	X	X	X	X	X	X	X	X
SDG 5: Gender Equality
SDG 6: Clean Water and Sanitation
SDG 7: Affordable and Clean Energy								X			X
SDG 8: Decent Work and Economic Growth	X					X		X
SDG 9: Industry, Innovation, and Infrastructure									X
SDG 10: Reducing Inequalities	X	X										X
SDG 11: Sustainable Cities and Communities				X	X	X	X		X		X
SDG 12: Responsible Production and Consumption
SDG 13: Climate Action					X					X	X
SDG 14: Life Below Water
SDG 15: Life on Land					X	X
SDG 16: Peace, Justice, and Strong Institutions
SDG 17: Partnerships for Goals

An “X” indicates that the respective article explicitly addresses or contributes to the achievement of the corresponding Sustainable Development Goal (SDG) in the context of higher education.

). SDG 4 (Quality Education) was consistently addressed in the revised articles (100%). This reflects the central role of VR as a tool for improving teaching and learning processes through experiential, immersive, and interactive methodologies. However, the fact that all studies focused on SDG 4 also suggests a potential bias: VR is frequently conceptualized as an educational innovation, but its broader contributions to sustainability agendas are less systematically developed.

The other SDGs displayed more fragmented coverage.

-

SDG 11 (Sustainable Cities and Communities) appears in 50% of the corpus [58,75,76,77,78,80] and is typically linked to architecture, landscape design/urban planning, and resilience training in open spaces.

-

SDG 8 (Decent Work and Economic Growth) reached 25% [73,77,79], reflecting competency development for employability and innovation/entrepreneurship training.

-

SDG 13 (Climate Action) was addressed by 25% [11,58,76], highlighting VR’s use for disaster risk reduction, climate-awareness hackathons, and sustainability-oriented projects.

-

SDG 10 (Reducing Inequalities) was referenced in 25% of studies [60,73,81] via access/equity considerations in training and inclusive competency frameworks.

-

SDG 3 (Good Health and Well-Being) is present in 16.7% [78,79] through reductions in learning anxiety and well-being-oriented instructional designs.

-

SDG 7 (Affordable and Clean Energy) with 16.7% [58,79] reported for climate-energy hackathon activities.

-

SDG 15 (Life on Land) appeared in 16.7% of the studies [76,77] through biodiversity/land use topics embedded in sustainable education projects.

A closer comparison reveals different ways of integrating the SDGs. Some studies embed them directly into curricular design and assessment, such as parametric/sustainable landscape modeling [76], climate-awareness hackathons [58], and resilience training with multi-hazard VR-serious games [78], while others refer to SDGs more tangentially, framing VR as a technology that indirectly supports sustainability goals: professional/technical training and simulator-based education [73,77] and teacher education programs with avatar-based classroom simulations [81]. This variability suggests a lack of standardized frameworks for linking VR with SDGs, which limits the comparability and reduces the possibility of cumulative evidence.

Overall, the findings indicate that while VR in higher education has been predominantly aligned with educational improvement (SDG 4), its role across other SDGs remains selective and fragmented. Expanding research on under-represented goals, such as SDG 1 (No Poverty), SDG 2 (Zero Hunger), SDG 5 (Gender Equality), SDG 6 (Clean Water and Sanitation), SDG 9 (Industry, Innovation, and Infrastructure), SDG 12 (Responsible Production and Consumption), SDG 14 (Life Below Water), SDG 16 (Peace, Justice, and Strong Institutions), and SDG 17 (Partnerships for Goals), and consolidating explicit curricular mappings and assessment rubrics would broaden the social impact and strengthen VR’s position in the 2030 Agenda as a genuinely cross-cutting educational technology.

The implementation of SDGs through VR in higher education is characterized by both breadth and variability in classroom methodologies and outcomes (RQ5/V10). SDG 4 (Quality Education) was consistently embedded across all cases but with diverse strategies: simulators used in maritime and vocational training [73], immersive environments designed to improve reading comprehension in foreign languages [60,74], experiential projects that require students to model and demonstrate their understanding of sustainability concepts [58], and metaverse-based collaborative activities with product-quality rubrics [79]. In resilience education, a VR-serious game integrates multi-hazard training and dual immersive/non-immersive modes, emphasizing preparedness and knowledge retention assessments [78]. These overlaps reveal a common orientation toward enhancing inclusivity and engagement in learning, although the specific learning objectives vary considerably (

Table 12. Methodology for the application and implementation of SDGs in the classroom and educational VR benefits in the context of higher education.

N	Authors and Year	Educational Benefits of the Use of Virtual Reality in the Higher Education Context	Methodology for Application and Implementation of the SDGs in the Classroom.
1	Kim et al. (2021) [73]	Accessibility to destinations and geographic work methods. Immersion and interactivity allow for the manipulation of tools and exploration of space. Facilitates international cooperation and develops digital skills. Safe environment for high-risk tasks. Improved feedback and adaptability. Improved understanding when interacting with virtual objects.	Maritime training through simulators to acquire vocational and technical skills. Digital and remote learning opportunities through cloud-based simulators and VR for post-COVID-19 maritime training. Related to SDGs 4, 8, and 10.
2	Z. Wang. et al. (2021) [60]	Learning with the VPS-VR approach: Their reading comprehension, information location, and text comprehension skills significantly improved. Students’ motivation and attitude towards learning is improved. Provides reading confidence. Reduces anxiety when learning English compared to traditional approaches. Students enjoy VR learning environments.	It is linked to SDG 4 and SDG 10: it is applied in practice through a VR approach based on visual cues to improve the reading comprehension skills of English as a Foreign Language (EFL) learners.
3	C. Wang et al. (2021) [74]	Extends the applications of visual behaviors immersed in real-time VR. Engages learners. Enables understanding of the fundamental principles of sustainable development skills through foreign language learning.	Integration of sustainable development into the sociocultural aspects of education through Japanese language learning and VR eye-movement analysis to develop students’ competence. Relates to SDG 4. The VR Dynamic ROI Tool helps operators define regions of interest (ROIs), while the Fixation Calculator Tool automatically processes ROIs according to priority.
4	Leininger-Frézal & Sprenger (2022) [75]	Accessibility to destinations that cannot be visited due to distance, economic, health, or time restrictions. Increases motivation. Encourages inclusion. Easily integrated into lessons. Promotes the development of competencies in the use of digital media. It is a more effective tool.	Transnational geography teacher training approach: Participants collaborate to explore space and sustainability issues using virtual field trips (VFT) with students. Focus on SDG 1 and 4: The project is based on the 4Is approach (Immersion, Interaction, Open Spaces).
5	Hsu & Ou (2022) [76]	Improving learning effectiveness by integrating thematic issues related to SDGs into the design-based teaching and learning model. Promotes cooperative group-learning experiences. Reduces the learning effectiveness gap between high-, middle-, and low-achieving students. Allow various useful teaching strategies. Learning is faster and clearer. Stimulates students’ interest and curiosity regarding sustainable land use. Better understanding of the needs of the landscape profession, development of practical skills for the workplace, and quick understanding of key points.	Design and application of parametric models by students based on different design requirements, providing optimal designs under various conditions and constraints, using 3D printing in the context of sustainable landscape architecture (SDG 4). SDG 11 “Make cities and human settlements inclusive, safe, resilient, and sustainable”, 13 “Take urgent action to combat climate change and its impacts” and 15 “Protect, restore, and promote sustainable use of terrestrial ecosystems, sustainably manage forests, combat desertification, and halt and reverse land degradation and halt biodiversity loss” are implemented.
6	Vergara-Rodríguez et al. (2022) [77]	Advantages for three areas of sustainable development: economy, society, and the environment. Promotes emerging technological activity with a high economic impact; Reduces pollution and resource consumption and ions, and favors the reuse of software and hardware, which contributes to the conservation of biodiversity; Allows the design of dynamic and interactive learning environments. It reduces energy costs, which promotes economic activity with a low environmental impact; it favors the design of dynamic and interactive learning environments, which facilitates the manipulation of certain objects.	It does not specify which SDGs are prioritized or directly addressed in the practice of the study, but it does mention how VR, in its FTI and NIVR variants, can be advantageous for three areas of sustainable development: economy, society, and environment. It promotes SDG 4, SDG 8, and SDG 11.
7	De Fino et al., 2023 [78]	Development of responsive behaviors; enhanced preparedness and risk awareness; improved cognitive learning, knowledge retention, and physiological engagement compared to traditional methods; higher engagement, satisfaction, immersion, and sense of presence; embodiment effects via first-person perspective; flexibility/scalability through dual immersive/non-immersive modes; reinforcement of learning through error-based feedback; promotion of participatory decision-making and communication of mitigation strategies; support for behavioral analysis and guideline improvement; and wide educational transferability across disciplines (AEC for design, safety training, operations, and emergency response).	Its educational design aligns with SDGs related to sustainable cities (SDG 11), climate action (SDG 13), and safety/resilience (SDG 3, SDG 4). It applies a multi-hazard approach, integrating phenomenological/behavioral analyses, Built Environment Typologies (BETs), and quantitative hazard simulations (UTCI maps, debris matrices, agent-based crowd models). Structured in modular training phases with dual immersive/non-immersive access, it aims to foster preparedness, awareness, and sustainable disaster risk reduction through experiential learning.
8	Zaky & Gameil (2024) [79]	This helps to overcome the limitations of traditional education. Develop an understanding of SDGs and climate change issues. Improve digital literacy and interpersonal skills, such as teamwork. They help interact with student avatars in an enjoyable manner, improve the learner’s experience, and increase engagement (which can lead to higher grades). Safe educational environment for students: impact on equal values of tolerance. Helps spread ideas about sustainability by overcoming spatial and temporal barriers, enabling the exchange of sustainable ideas and practices through Edu-Metaverse platforms.	The metaverse offers opportunities to achieve sustainability: students do not have to travel to school, thus reducing the consumption of fossil fuels. SDG 3 (preserving and protecting the natural environment and its resources), SDGs 7 and 8 (sustaining a strong and resilient economy over the long term, fostering economic growth, and creating employment opportunities), and SDG 11 (promoting the well-being and quality of life of people and communities) are implemented. The principles of computational science are applied to address climate change issues, and SDGs to enhance the ability to design practical solutions to complex environmental challenges.
9	Montoya et al. (2024) [80]	Provides different types of innovation in complex learning environments. Facilitates changes and solutions in the learning process. Provides opportunities to generate creative scenarios that promote educational innovation for improvement. Facilitator of changes and solutions in learning processes. Support for distance education and motivating scenarios. Provide knowledge about new tools. Develop a greater understanding of SDGs and climate change issues. Improve the ability to design practical solutions to complex environmental challenges. Encourages complex reasoning and invites the search for new solutions, training citizens with critical, scientific, systemic, innovative and entrepreneurial thinking (SDG 9), empathetic, cooperative, and committed to sustainable development (SDG 11).	LMS courses with visual elements for space missions, including logos for each “spacecraft” (APOLLO, SATURN, ATLAS) representing different project approaches. In the “Entrepreneurship and Innovation” course, students must select an SDG to focus on in their project and carry it out. Projects are classified by the element and type of innovation determined by the scope, objective, context, and expected innovation outcomes, seeking solutions to the challenges described by the SDGs. Related to SDG 4, SDG 9, and SDG 11.
10	Jian & Abu Bakar (2024) [11]	Increases intrinsic user motivation, which optimizes learning and performance by transferring skills from the virtual environment to the real world. Promotes a more stimulating learning experience (active participants, not mere spectators). Facilitates the visualization of information and future scenarios. Improves technical skills in coding, problem solving, and 3D model design with emphasis on user experience design (UX). Incentivizes participation in new experiences that are impossible, costly, or dangerous in other contexts. Creates a need for students to develop an understanding of the SDGs and climate change issues.	Supports efforts towards sustainable development by promoting innovative educational approaches aligned with SDG4 and SDG 13.
11	AlQallaf et al. (2024) [58]	Facilitate team collaboration by removing barriers. Students were motivated to use VR and build applications. Allows interaction with ODS in a more tangible and meaningful way. Satisfaction with aspects such as understanding VR hardware, game engine mastery, and collaborative testing/debugging. Ability to integrate into curricula to enhance learning outcomes by creating interactive environments and encouraging experiential learning and active participation. More training in VR development and more time to develop projects could further enhance the experience: use of technology to solve real problems.	Engineering students used VR technology and game design software to design immersive environments that demonstrated their understanding of these SDGs (preferably climate change, SDG 13). For example, a multiplayer VR game raised awareness of SDG 4 and 7: Affordable and Clean Energy: players were given a budget to invest in different types of energy (fossil fuels, solar energy, wind energy, etc.) and had to maximize energy production while minimizing environmental damage.
12	Lee et al. (2025) [81]	VR supports professional preparation of pre-service teachers (PSTs) in Early Childhood education by achieving the following: Bridging the gap between theory and practice before first placements. Developing professional competencies (child development, assessment, professionalism) and procedural competencies (classroom management, health and safety, curriculum application, reflective practice). Enhancing confidence, readiness, and cultural competence of PSTs (particularly international students). Allowing safe practice of teaching techniques, behavior guidance, and adaptation to diverse learners. Encouraging reflective practice through cycles of practice, feedback, and reflection.	Methodology aligns with SDG 4 (Quality Education) by improving teacher training and learning outcomes, and indirectly with SDG 10 (Reduced Inequalities) by supporting international students and diverse learning needs. VR sessions employed the Mursion© program with avatar children, structured around curriculum-aligned lesson plans. Complementary platforms (Zoom, Padlet) fostered collaboration and reflection. The analysis followed the Dramatic Events–Social Situation of Development (D-SSD) framework and Hedegaard’s levels of interpretation, enabling researchers to trace developmental moments linked to competence building and sustainable professional preparation.

).

Other SDGs have been operationalized through more discipline-specific applications. For example, SDG 7 (Affordable and Clean Energy) was implemented in climate-energy hackathon activities with budgeting and trade-off decisions [58]. Similarly, SDG 8 (Decent Work and Economic Growth) was linked to VR simulators for professional and technical skills [73], innovation/entrepreneurship, and teacher education ecosystems [80,81]. Meanwhile, SDG 11 (Sustainable Cities and Communities) appeared in transnational geography teacher training [75], metaverse-supported urban projects [79], and multi-hazard preparedness in open spaces [78], whereas SDG 13 (Climate Action) and SDG 15 (Life on Land) were advanced through parametric modeling and VR-integrated sustainable landscape design [76].

The educational benefits derived from these interventions (RQ5/V11) demonstrated convergence and fragmentation (Table 12). The most frequently reported advantages were experiential and hands-on learning (66.7%), motivation and engagement (58.3%), and immersion and digital-skill development (41.7%). These benefits align with the broader literature that emphasizes VR’s role in fostering active learning and competence-based education. Other gains, such as teamwork, accessibility, and critical thinking (41.7%), were supported by metaverse collaboration [79], and innovation/entrepreneurship and teacher education programs [80,81]. Benefits such as anxiety reduction and performance optimization were observed in 25.0% of the studies, with evidence from language-learning anxiety decreases [60], quasi-experimental performance gains [11], and error-based feedback/retention improvements in multi-hazard training [78].

Beyond registering the presence of SDGs, the analysis reveals specific pathways for operationalization. Contributions to SDG 3 (Good Health and Well-Being) are visible in reduced learning anxiety and improved well-being indicators [60], innovation/entrepreneurship ecosystems with reflective practice [80], and reflective/innovation-oriented courses and teacher education practice [81]. Links to SDG 9 (Industry, Innovation, and Infrastructure) emerge from VR-enabled innovation and entrepreneurship courses [80,81], while SDG 13 (Climate Action) is strengthened by virtual fieldwork/modeling [76] and multi-hazard risk-reduction training [78], reducing carbon-intensive or unsafe real-world tasks.

In summary, these findings show that while SDG 4 dominates (directly or indirectly reported in the articles), VR in higher education can translate sustainability goals into practice through targeted pedagogical strategies, disciplinary integration, and innovative technology. However, the uneven distribution across SDGs and the lack of systematic frameworks for their implementation suggest that VR’s potential to advance the 2030 Agenda remains partially unexplored, requiring broader and more coordinated efforts to maximize its transformative role in higher education.

4.2.6. Research and VR Limitations in the Higher Education Context

Despite the promising benefits of VR, the studies analyzed revealed several recurring limitations in its implementation in higher education (RQ6/V11), which can be grouped into technological barriers and methodological constraints (

Table 13. Limitations of the research study and the use of VR in higher education contexts.

N	Authors and Year	Limitations of the Research Study	Limitations in the Application of VR in Higher Education
1	Kim et al. (2021) [73]	No limitations of the research study are mentioned.	Initial learning curve, possible dizziness, visual fatigue, and lack of direct feedback. VR simulators in the early stage of technological maturity. Lack of fidelity, team cooperation, and immersion in desktop simulators. Costs and need for competent instructors in full-mission simulators. Depth perception issues, student–instructor interaction, limited cooperation, and questionable learning effects in VR simulators. Lack of social interaction, formative assessment, limited transfer of learning, and team-based training in cloud-based simulators.
2	Z. Wang. et al. (2021) [60]	The duration of the study was three weeks; this may be an insufficient period to significantly improve EFL reading comprehension. English as a Foreign Language (EFL) learning resources based on 3D VR videos were limited. The study adopted 3D videos to train students in EFL reading comprehension skills, which did not provide opportunities to interact with virtual learning environments. VPS strategies only improved the lower level of reading comprehension skills—information location and comprehension of unfamiliar vocabulary—rather than evaluation and reflection.	Students can easily feel dizziness or discomfort when using VR devices. It is costly to implement VR in terms of the preparation of learning materials and design of learning activities. Learning behaviors and interactions with others are difficult to record.
3	C. Wang et al. (2021) [74]	Small sample size: 20 participants. However, it is argued that the amount of eye movement data collected is sufficiently large for an exploratory study: each participant experienced six MIO expression situations with 24 ROIs.	There are no limitations to the use and implementation of VR in higher education.
4	Leininger-Frézal & Sprenger (2022) [75]	Scarcity of empirical findings regarding TFVs.	Lack of direct confrontation with reality: Students cannot fully experience space. Little self-determination: VFTs do not encourage student self-determination; inadequate technical equipment: There is concern about the lack of digital devices at school or at home, as well as insufficient internet connection. Information overload: Students may feel overwhelmed by it.
5	Hsu & Ou (2022) [76]	No limitations are mentioned.	No limitations are mentioned.
6	Vergara-Rodríguez et al. (2022) [77]	No limitations are mentioned.	Although students stated that the user experience was better when using IVR (Immersive VR), this perception does not imply that IVR has educational advantages over NIVR (Non-Immersive VR). According to the teachers’ perspective, IVR could generate a fascinating effect on students that divert their attention from the learning objective. From an economic point of view, the greater amount of infrastructure and support required by the FTI compared to the NIVR makes its use more expensive, and the negative impact on its implementation in different sectors could be potentially useful.
7	De Fino et al., 2023 [78]	Pending validation: The VR-SG prototype still requires comprehensive testing with diverse age groups beyond previous studies’ narrow participant samples. Ethical and social issues: Risks of reduced autonomy, motion sickness, privacy concerns, and desensitization to disasters require future management. Realism vs. effectiveness: Excessive realism may increase cognitive load and hinder knowledge transfer. Technical compromises: Simplified environments/avatars balance computational load but limit realism. Typological setting: Designed as a generic typological model rather than photorealistic replicas, reducing fidelity but avoiding superrealism.	The VR-serious game prototype highlights multiple ethical, social, and technical limitations: Ethical issues: Reduced learner autonomy, motion sickness/fatigue, privacy risks from data collection, potential “superrealism”, risk of desensitization or biased stereotypes, and cognitive overload affecting knowledge transfer. Technical/operational trade-offs: High computational load requiring simplified environments and avatars; disorientation and motion sickness mitigated by teleporting but still present; cost/benefit constraints making Non-Immersive VR preferable in some cases.
8	Zaky & Gameil (2024) [79]	Sample size: (n = 36). Pre-existing familiarity between participants could have influenced their interaction and tolerance-related outcomes. The teacher’s supervision and monitoring of the environment to provide a safe educational environment may have had an impact on the equality of tolerance values between the two groups in the intervention. The study was conducted in a specific virtual environment (Frame VR), and the results may vary in other metaverse platforms. The duration of the experiment was approximately four weeks, which may be a limited period to observe significant changes in tolerance and respect.	The open and unrestricted nature of the metaverse, with freedom of action without limitations: the likelihood that students will not be fully engaged in learning and will be prone to distraction and inattention. Competitiveness among student avatars and differences in their perspectives on the design of the environment, as well as the virtual interaction of student avatars, which was not observed naturally, but rather as a virtual representation of their personalities, could have affected tolerance levels. No other potential limitations of using VR in education are addressed.
9	Montoya et al. (2024) [80]	The VR environment was used at a specific point in the course to present the students’ final projects. The sample size and the Horizons Architecture methodology were limited, as the open-ended questions to collect comments on these aspects were not mandatory, and the results were based solely on the perceptions of the students who responded. The study did not include a control group; therefore, a causal relationship between the use of the VR environment and the results could not be established.	Students questioned the technical requirements and felt that they could limit their beneficial experiences. They felt that the time spent on their use was detracted from the tangible learning outcome and its potential added value to the learning experience. Usability problems, such as lack of interaction, language adaptations, and hardware and network problems.
10	Jian & Abu Bakar (2024) [11]	The study focused exclusively on first-year art students. Participants were selected using purposive sampling. The study examines the cognitive load and learning performance of students in different learning environments using the same didactic content.	Students in immersive learning environments may experience cognitive load because of the large amount of information they must process. Spatial ability involves transforming 2D and 3D relationships, which consume more cognitive resources. DLM operation is limited by time, place, and materials.
11	AlQallaf et al. (2024) [58]	Limitations related to the availability and accessibility of VR viewers, which affected and limited participation. Evaluation of the effectiveness of VR applications is subjective, and it is difficult to develop standardized criteria for various projects within a hackathon.	Implementing VR/AR technologies in education can require expensive devices and software and the need for ongoing maintenance. The development of high-quality VR applications for renewable energy systems requires specialized skills and resources. The acquisition of technical skills in “3D Modeling and Animation” was challenging within the timeframe of the event, demonstrating the need for prolonged exposure, like traditional educational environments.
12	Lee et al. (2025) [81]	Limited scope: Data from a single postgraduate course, restricting generalizability. Subjectivity/selectivity: Reliance on self-reported reflections, which may omit key experiences. Absence of video data: Lack of deeper analysis of interactions within VR sessions; suggested for future research. Need for longitudinal studies: Current findings show immediate gains but not long-term impact. Under-researched area: Few studies focus on VR in Early Childhood teacher education compared to medicine/aviation or other education levels.	Lack of adaptive learning content: Many VR tools cannot tailor to learner needs, limiting effectiveness. Technological constraints: Inability to use physical aids (e.g., cards) within VR; required workarounds seen as monotonous. Gap in theoretical frameworks: VR studies sometimes lack robust frameworks to maximize learning/development. Emerging use in Early Childhood education: Less explored than in elementary, special needs, secondary, or other sectors, making applications less mature.

).

From a technological perspective, the learning curve and usability challenges were the most frequently cited barriers (58.3%) [11,58,60,73,75,77,78]. Students and teachers often encounter difficulties owing to limited prior experience, compatibility issues, and technical requirements such as hardware calibration or network stability. These barriers are compounded by high implementation costs (33.3%) [58,60,73,77], which include equipment acquisition, maintenance, and the need for specialized software, making scalability difficult for many institutions. Issues of equity and accessibility are also evident in 33.3% of the studies [11,58,75,77] since not all learners have access to HMDs or robust technological infrastructure, which risks widening the digital divide in higher education.

Other limitations are related to cognitive and health factors. Approximately 25% of the studies reported attention deviation or information overload [75,77,79], where the immersive appeal of VR distracted students from learning objectives. Side effects (dizziness and visual fatigue) were noted in 16.7% of the participants [60,75], particularly during prolonged VR sessions. Concerns about the authenticity of social interaction appeared in 16.7% of the studies: limited interaction in simulator-based contexts [73] and avatar-mediated exchanges that may distort tolerance levels in metaverse settings [79].

From a research design perspective, methodological weaknesses limit the generalizability of these findings. Small sample sizes (25%) [74,79,80] (including narrow/absent participant cohorts in design demonstration work [78]) and short intervention durations (16.7%) [60,79] make it difficult to assess long-term effectiveness. Similarly, the absence of control groups [60,79,81] constrains causal inference, whereas reliance on specific technologies or environments [58,78] reduces replicability. Other methodological gaps include the lack of stage implementation frameworks [79] and omission of key variables [60], with a predominant focus on immediate outcomes rather than long-term impacts, such as retention, motivation, or the transfer of skills to real-world contexts. Study-specific notes [78] highlight ethical/privacy risks, motion sickness, and “superrealism” concerns; [79] reports distraction/competitiveness due to open-ended metaverse freedom and platform dependence; and [80] uses optional responses and no control group, weakening causal claims.

These findings indicate that VR research in higher education is constrained by significant structural and methodological limitations. Although technical barriers and costs affect scalability, the predominance of short-term, small-sample designs and platform-specific setups limits the robustness of the evidence base. Addressing these challenges requires more rigorous longitudinal and cross-institutional designs, together with policies that improve accessibility, standardize technical reporting, and clarify pedagogical frameworks for VR implementation.

4.2.7. Recommendations for Future VR Research

The recommendations emerging from the reviewed studies (RQ7/V12) converge on several key priorities for advancing VR research in higher education, although with different emphases depending on the disciplinary and methodological context (

Table 14. Recommendations for future research.

N	Authors and Year	Recommendations for Future Research
1	Kim et al. (2021) [73]	Conduct quantitative and qualitative research to evaluate and validate the effectiveness of VR and cloud-based (CB) simulators in different operational contexts, comparing their impact on the acquisition of technical and nontechnical skills. Design and analyze collaborative and self-directed training scenarios in virtual environments to optimize learning in maritime contexts.
2	Z. Wang. et al. (2021) [60]	Research learning strategies embedded in VR learning environments to enhance higher-order thinking skills. Longer duration of studies: A longer intervention is needed to improve students’ reflection and assessment of reading comprehension skills. Explore more English language learning resources. Design of interactive VR learning environments.
3	C. Wang et al. (2021) [74]	Use larger sample sizes. Acquire more empirical eye-movement data to reveal a meaningful index for learning effectiveness. To investigate how to raise awareness and positive perceptions of students in applying foreign language knowledge in quality education and discover the prevailing trends in the field of higher education for ESD (Education for Sustainable Development).
4	Leininger-Frézal & Sprenger (2022) [75]	Experimental or quasi-experimental studies examining individual aspects of virtual excursions. Comparisons at different stages of education or differences between real and virtual field trips. Efficacy analyses should ideally be conducted through group comparisons. To discuss the conditions for successful approaches in an educational environment characterized by differences between digital and real formats.
5	Hsu & Ou (2022) [76]	Implementation of educational models other than VR or 3D printing technologies for later comparison. How to more effectively integrate the SDGs into landscape architecture curricula and other design-related disciplines.
6	Vergara-Rodríguez et al. (2022) [77]	To perform a quantitative analysis of the impact of VR technologies on different dimensions of sustainability.
7	De Fino et al., 2023 [78]	To validate the VR-SG prototype with diverse age groups, address ethical/social issues (autonomy, motion sickness, privacy), and refine the balance between realism, fidelity, and cognitive load. Further work should apply the prototype to real urban contexts, support participatory decision-making, enable behavioral analyses for policy insights, and test photorealistic environments to enhance knowledge transfer, linked to SDG 11 and SDG 13.
8	Zaky & Gameil (2024) [79]	To investigate the impact of avatars on sustainable educational Edu-Metaverse platforms in terms of gender on educational assessment. To conduct studies examining the effects of different ethical values on Edu-Metaverses, such as fairness, privacy, and ethical artificial intelligence. To conduct studies on different metaverse platforms and educational stages to clarify their impact on students: quality of life, educational flexibility, and digital well-being.
9	Montoya et al. (2024) [80]	Expand the sample and profile of participants, specifically for students’ perceptions of the VR environment. Analyze the use of VR environments predominantly throughout the learning experience. Consider an experimental design that includes a control group to establish a causal relationship between the use of the VR environment and the results generated.
10	Jian & Abu Bakar (2024) [11]	Consider gender variables in cognitive load theory, visualization research, and learning outcomes between static and dynamic learning resources. Build relevant immersive learning environments to reduce the cognitive load of students or those with low spatial ability and provide them with a good learning environment. Efficacy analyses should be conducted, ideally with group comparisons.
11	AlQallaf et al. (2024) [58]	Explore long-term retention of knowledge and skills. Examine the scalability and accessibility of this approach in various educational environments. Examine the adaptability of VR-based learning to other renewable energy educational subjects.
12	Lee et al. (2025) [81]	Expand data collection to include PSTs from multiple institutions to improve generalizability. Include diverse stakeholder perspectives, especially placement mentors, to enrich analysis of professional preparation. Incorporate video data to deepen analysis of interactions and enhance application of the D-SSD framework. Conduct longitudinal studies to assess lasting impacts of VR on professional competencies and teaching skills. Address technological limitations, such as lack of adaptive learning content in VR applications. Strengthen theoretical frameworks to guide VR studies with greater rigor and depth. Analyze VR teaching data beyond reflections for a more complete picture of PSTs’ development. Focus further on Early Childhood education, an under-researched area compared to other fields.

).

First, increasing sample size and diversity is a recurring priority (41.7%) [73,74,78,80,81]. Many studies have relied on small, relatively homogeneous groups, limiting the generalizability of their findings. Expanding participant pools across disciplines, levels of study, and geographic contexts would strengthen the robustness of the conclusions and enable cross-comparability.

Second, the need for longitudinal research designs was highlighted (33.3%) [60,75,80,81] to assess the medium- and long-term effects of VR on learning. Short interventions dominate the current evidence base, offering insights into immediate impacts but not retention, transfer, or sustained motivation.

Methodologically, the inclusion of control groups and experimental/quasi-experimental designs was underscored in 25% of the studies [75,78,80]. This responds to a clear gap in causality analysis; while most research demonstrates correlations between VR use and learning gains, few studies establish causal relationships. Equity, accessibility, and sustainability also appear as lines for deeper analysis (16.7%) [76,77], acknowledging that the benefits of VR will only be transformative if they can be extended, regardless of socioeconomic status or institutional resources.

Other recommendations emphasize the importance of connecting VR more explicitly with the SDGs and sustainability dimensions (25%) [76,77,78]. This includes not only embedding sustainability topics into VR activities but also investigating how VR technology can contribute to institutional sustainability agendas. Relatedly, several authors have suggested broadening research across different metaverse platforms and educational stages [79] and exploring outcomes such as digital well-being, flexibility, and quality of life. Furthermore, studies encourage the examination of long-term retention of knowledge and skills [58], an area largely neglected in short-term interventions, and cross-cutting recommendations such as addressing technological limitations (lack of adaptive learning content), strengthening theoretical frameworks beyond reflective reports, expanding multi-site data collection), and devoting greater attention to Early Childhood teacher education [81].

These recommendations highlight the need to shift from small-scale, short-term, and context-specific VR applications to more rigorous, inclusive, and future-oriented research design. Addressing these gaps would not only strengthen the evidence base but also ensure that VR’s integration into higher education is sustainable, equitable, and aligned with global educational priorities such as the 2030 Agenda.

4.2.8. Overall Results

In summary, the analysis of the twelve studies reviewed revealed both convergence and divergence in the implementation of VR within higher education, highlighting methodological tendencies, areas of application, and educational benefits, while also exposing limitations and gaps. Methodologically, most studies relied on mixed designs (83.3%) combining quantitative and qualitative approaches, with a strong reliance on Likert-type questionnaires (83.3%), pre/post-tests (41.7%), and statistical analyses (t-tests, ANOVA/ANCOVA) (25%). Complementary tools, such as interviews, focus groups, and direct observation, were used to provide contextual insights, although their use was comparatively less frequent than other tools.

In terms of academic areas and disciplinary integration, VR applications have been concentrated in STEM-related disciplines (41.7%), industrial/civil/engineering, and architectural design (33.3%), with additional weight in environmental/sustainability education and fewer examples in the humanities and social sciences.

The connection to the Sustainable Development Goals (SDGs) was evident but uneven. SDG 4 (Quality Education) appeared directly or indirectly in 100% of the studies, SDG 11 (Sustainable Cities and Communities) in 33.3%, and SDG 13 (Climate Action) and SDG 3 (Good Health and Well-Being) in 25.0%. SDG 7 (Affordable and Clean Energy), SDG 8 (Decent Work and Economic Growth), SDG 10 (Reducing Inequalities), and SDG 15 (Life on Land) were reported in 16.7% of the studies. SDG 1, 2, 5, 6, 12, 14, 16, and 17 remained unaddressed.

This suggests that VR’s potential as a cross-cutting technology for advancing sustainability agendas has yet to be fully realized. Despite these imbalances, the results consistently demonstrate significant educational benefits, including enhanced student motivation, engagement, experiential learning, digital competence, and collaborative skills. Moreover, VR applications contribute to inclusive and equitable learning experiences, offering opportunities for accessibility and skill development in contexts where real-world practice is limited or risky. Finally, recommendations for future research converge on expanding sample diversity, adopting longitudinal and experimental designs, integrating equity and sustainability considerations, and explicitly linking VR implementation to broader educational and institutional agendas. Taken together, these findings indicate that while VR consolidates as a powerful educational innovation in higher education, its integration remains fragmented, discipline-specific, and methodologically limited in scope. Addressing these challenges is essential for maximizing VR’s transformative potential and ensuring its alignment with the 2030 Agenda for Sustainable Development.

To complement the results presented, a summary table (

Table 15. Summary of the studies reviewed, including author, year, context, type of VR employed, SDGs addressed, methodological approach, and key findings.

N	Author(s) and Year	Context/Discipline (Country)	Type of VR	SDGs Addressed	Method	Key Findings
1	Kim et al. (2021) [73]	Maritime training (Norway)	Immersive and cloud-based simulators	SDG 4, 8, 10	Qualitative (SWOT, focus groups)	Improved technical skills in safe environments; international cooperation; remote learning post-COVID.
2	Z. Wang. et al. (2021) [60]	English language teaching (China/Taiwan)	VR with visual scaffolding (VPS-VR)	SDG 4, 10	Mixed (pre-/post-test, Likert, interviews)	Enhanced reading comprehension, motivation, and attitude; reduced anxiety in EFL learning.
3	C. Wang et al. (2021) [74]	Japanese language and onomatopoeia (Taiwan)	VR with eye-tracking	SDG 4	Mixed (eye-tracking, ANOVA)	Improved visual attention; strengthened sociocultural understanding linked to sustainable development.
4	Leininger-Frézal & Sprenger (2022) [75]	Teacher training in geography (France/Germany)	Virtual field trips (VFT)	SDG 1, 4	Mixed (questionnaires, descriptive analysis)	Increased motivation and inclusion; accessibility to remote destinations; development of digital competencies.
5	Hsu & Ou (2022) [76]	Landscape architecture (Taiwan)	Parametric modeling and 3D VR	SDG 4, 11, 13, 15	Mixed (t-test, interviews, observation)	Enhanced sustainable design learning; curricular integration of sustainability.
6	Vergara-Rodríguez et al. (2022) [77]	Engineering (Spain)	Immersive vs. non-immersive VR (IVR/NIVR)	SDG 4, 8, 11	Mixed. Comparative (t-test, descriptive analysis)	IVR improved performance and comprehension; VR viewed as a sustainable technology.
7	De Fino et al., 2023 [78]	Civil, Environmental and Architectural Engineering; Urban Planning; Disaster Risk Reduction (Italy)	Immersive and Non-immersive VR-serious game (heatwave and earthquake multi-hazard training)	SDG 3, 4, 11	Mixed. VR-SG prototype integrating simulation data (UTCI maps, damage matrix, agent-based models); modular storyline training; tested with Meta Quest 2 and desktop mode	VR-SG improves disaster preparedness, awareness, engagement, embodiment, and knowledge retention; scalable and adaptable for multi-hazard urban resilience training.
8	Zaky & Gameil (2024) [79]	Edu-Metaverse, alternative assessment (Egypt)	Avatars and immersive environments	SDG 3, 4, 7, 8, 11	Mixed (Mann–Whitney U, Wilcoxon, peer review)	Avatars foster tolerance, respect, and inclusion; strong ethical and social potential.
9	Montoya et al. (2024) [80]	Higher education (Mexico, Latin America)	Immersive VR environments	SDG 4, 9, 11	Mixed (Likert surveys, qualitative analysis)	High student acceptance; VR improved attitudes and performance; need for larger samples.
10	Jian & Abu Bakar (2024) [11]	Arts education (China)	Immersive VR (ILE) vs. digital learning materials (DLM)	SDG 4, 13	Mixed. Quasi-experimental (control/experimental, t-test)	ILE reduced cognitive load and enhanced spatial performance, especially for students with low spatial ability.
11	AlQallaf et al. (2024) [58]	Interdisciplinary hackathon (Saudi Arabia)	Collaborative VR and AR	SDG 4, 7, 13	Mixed (surveys, presentations, focus groups)	VR hackathons fostered climate awareness, teamwork, and digital competencies linked to SDGs.
12	Lee et al. (2025) [81]	Initial Teacher Education (Early Childhood Education), Australia	Mixed-reality VR simulations using Mursion© software with avatar children (3–5 years); non-immersive collaborative sessions supported by Zoom and Padlet	SDG 4, 10	Qualitative study. Reflections collected via Padlet; analyzed through Hedegaard’s (2008) three levels of interpretation and the Dramatic Events–Social Situation of Development (D-SSD) framework	Fosters reflective practice, cultural competence, and confidence; effective in preparing “classroom-ready” graduates, particularly benefiting international PSTs (child development).

) synthesizing the main characteristics of the studies reviewed, including author, year, context, type of VR employed, SDGs addressed, methodological approach, and key findings, was included.

5. Discussion

5.1. Expansion and Multidisciplinary Integration of VR in Higher Education

Research on VR in higher education shows a clear upward trend between 2020 and 2025, with widespread international dissemination and growing consolidation in high-impact journals. This increase reflects not only academic interest in immersive technologies but also their positioning as strategic pedagogical resources. However, although the studies originated from diverse countries and disciplines, there is a thematic concentration in health sciences and engineering; this indicates that VR adoption remains strongly associated with contexts in which simulation offers immediate added value, improving engagement, confidence, and knowledge retention compared to conventional or 2D digital learning, especially for 3D structures, safety, and procedural skills [49,82,83,84,85,86]. This concentration, which is beneficial for skill-based domains, limits the potential to explore VR’s broader transformative capacity in underrepresented fields, such as social sciences, humanities, and teacher education, where immersive environments could foster ethical reasoning, intercultural dialog, and civic competence. Thus, a key implication is the need for deliberate diversification strategies to avoid reinforcing a narrow disciplinary scope.

5.2. Methodological Approaches

The predominance of mixed and quasi-experimental methodologies in the analyzed studies reflects the scientific community’s interest in measuring the effects of VR on learning through cause-and-effect comparisons (control vs. experimental groups). There is increasing advocacy for mixed-methods research that combines the strengths of both approaches to provide more comprehensive and actionable insights [87,88,89,90].

While this emphasis adds empirical rigor to the field, it also reveals a bias toward short-term impact evaluation designs, in which learning outcomes are primarily assessed in experimental or laboratory contexts. The scarcity of longitudinal and follow-up studies limits our understanding of VR’s sustained impact on competence development, restricting insights into its medium- and long-term potential. This gap hampers the formulation of evidence-based policies in higher education, as isolated short-term gains cannot confirm the scalability or sustainability of VR adoption.

The widespread use of online questionnaires, Likert-type scales, and pre-and post-tests contributes to comparability across studies but also raises the risk of overreliance on self-reported measures, which may overestimate positive perceptions without fully capturing the complexity of immersive experiences. In this regard, complementary qualitative methods (interviews, observations, and SWOT analyses) represent a step toward a more contextualized evaluation; however, they remain secondary or marginal compared to the dominance of quantitative approaches. In practice, qualitative research is often used to inform or supplement quantitative studies rather than as a stand-alone approach. This status is evident in fields such as medicine, social sciences, and policy evaluation, where qualitative findings are sometimes undervalued or considered less rigorous [87,91,92]. Calls for more balanced appraisal frameworks and reporting standards for qualitative research are growing, but challenges remain in shifting entrenched quantitative biases [93,94,95]. A stronger interpretative use of qualitative evidence would allow VR studies to illuminate not only measurable outcomes but also the emotional, ethical, and institutional challenges of immersive learning environments.

The choice of test and post hoc procedures should be guided by data characteristics, such as sample size, variance equality, and distribution shape, to ensure valid inference [96,97,98]. Regarding data analysis, the recurrent use of classical statistical tests ensures methodological rigor and enables the identification of significant differences between experimental conditions. Parametric tests (t-tests, ANOVA, ANCOVA) are powerful under correct assumptions, whereas nonparametric tests (Mann–Whitney U, Wilcoxon) are robust to violations of normality or variance homogeneity [98,99]. Newer methods, such as permutation tests, further enhance flexibility and rigor, particularly in complex or small sample settings [99,100].

However, a strong reliance on these quantitative approaches reinforces a fragmented view of the learning experience, focusing on immediate performance indicators without integrating affective, ethical, or social dimensions that are equally salient in immersive environments. While qualitative approaches provide interpretative value, they are often employed to corroborate numerical findings rather than to generate critical insights in their own right. Consequently, although methodological approaches in this field demonstrate maturity in terms of internal validity, they show limitations in ecological validity and transferability, as most studies are conducted in highly controlled environments that do not always reflect the dynamics of real higher education classrooms. This raises a major challenge: how to design studies that capture authentic learning processes, including cultural, emotional, and sustainability-related dimensions, without sacrificing rigor. To consolidate VR and SDG research, future work should advance toward integrated multimodal designs capable of capturing not only academic performance but also students’ holistic experiences in authentic contexts, including emotional, social, and sustainability-related variables.

5.3. Educational Impact and Contribution to Competence Development

Research on VR highlights the breadth of its applications in higher education, emphasizing its impact on teaching, sustainability, and innovation across various disciplines. VR supports active, experiential, and collaborative learning, which is aligned with educational theories such as constructivism and experiential learning. It fosters a deeper understanding, engagement, and motivation, and can improve learning outcomes, especially when thoughtfully integrated into pedagogical frameworks [44,45,47,49,50,51,52,101,102,103,104]. VR also facilitates the development of critical competencies, including spatial reasoning, procedural skills, creativity, and self-directed learning, and democratizes access to high-quality educational resources [45,50,101,102].

The thematic and geographic diversity of these studies demonstrates the global appeal of immersive technology. However, despite this broad scope, evidence has revealed persistent gaps in gender representation, with many studies failing to adequately report or balance women’s participation. This omission constrains the analysis of inclusivity and limits the potential of VR research to contribute meaningfully to equity-oriented educational agendas. In practical terms, the lack of gender-sensitive reporting undermines the possibility of designing inclusive pedagogical interventions and perpetuating systemic blind spots in digital-educational research.

The predominance of VR applications in STEAM disciplines [105,106,107] (computer engineering, computer science, materials engineering, medicine, and industrial design) reflects the suitability of the technology for simulation-based and practice-oriented learning environments. While this concentration underscores VR’s strength in fields requiring visualization and procedural training (surgical simulations, maritime operations, and 3D urban modeling), it also highlights the underrepresentation of the social sciences and humanities [108,109], where the integration of immersive technologies could enrich critical thinking, ethical reasoning, and sociocultural awareness. This disciplinary imbalance reduces the scope of VR’s transformative potential in fostering competencies beyond technical skill. Therefore, future research must deliberately extend VR applications to disciplines in which creativity, empathy, and civic responsibility are central, ensuring a more balanced contribution to higher education goals.

VR enhances real-time educational communication and management and improves communication skills in students and teachers [110]. Current evidence suggests that VR supports competency-based education (development frameworks such as collaboration, digital literacy, intercultural communication, and problem solving skills) by enabling experiential, interactive, and cross-contextual learning. However, much of the literature still emphasizes technological novelty rather than systematically evaluating its contribution to long-term competency acquisition in academic and professional contexts [12]. Some studies have reported that traditional or video-based methods outperform VR for hands-on psychomotor skill acquisition, suggesting that VR is best used as a complement rather than a replacement for existing strategies [111,112,113].

In summary, while the findings validate VR’s relevance in creating immersive and interactive learning scenarios, unequal disciplinary distribution, insufficient attention to gender inclusivity, and limited alignment with comprehensive competence models represent ongoing challenges. These limitations imply that, without a stronger focus on inclusivity and transversal skills, VR risks are consolidated as niche innovations rather than truly transformative educational tools.

5.4. VR Technology Innovation (Device and Software) and Complementary VR Technologies

VR adoption in higher education is driven by advances in hardware, software, and complementary technologies that enhance interaction and immersive learning. The predominance of Head-Mounted Displays (HMDs) and the use of simulators and collaborative platforms, such as Mozilla Hubs, Frame VR, and portable VR devices (HTC VIVE Pro Eye, Meta Quest 2, and DPVR M2 Pro), demonstrate the rapid diversification of devices available to educational institutions. These VR platforms enable geographically distributed learners to interact in shared virtual spaces, thereby supporting teamwork, communication, and collaborative problem solving [114,115,116,117]. HMDs foster physical presence, body ownership, and agency, which can enhance cognitive and socio-emotional interactions, leading to improved learning outcomes in distributed settings [116,117] and engagement in vocational and laboratory settings, allowing exposure to complex equipment and scenarios that are otherwise inaccessible to students [118,119,120].

However, the lack of standardization in reporting hardware and software specifications across studies limits replicability and weakens the comparability of the results. This inconsistency suggests a pressing need for uniform guidelines for documenting VR-based educational interventions, particularly given the pace of technological evolution. The identification of major development tools (including Unreal Engine (UE4), Unity, and Adobe Premiere Pro) confirms the integration of professional-grade software in academic contexts. While this enhances the realism and interactivity of educational environments, it also raises issues of accessibility and scalability, as high technical and financial requirements may restrict adoption by institutions with fewer resources. Similarly, reliance on specialized modeling and programming applications for architecture and design, though innovative, risks reinforcing disciplinary inequalities by privileging fields with existing technical infrastructure. As reiterated in numerous studies, the financial and technical requirements of VR (such as hardware, software, and infrastructure) are major obstacles, particularly in resource-limited settings and developing countries [121,122,123,124,125]. These inequalities create the risk of deepening the digital divide, whereby institutions with more resources benefit disproportionately from immersive technologies, while others remain excluded.

The integration of VR with advanced methodologies and complementary technologies, such as eye tracking for visual attention analysis, virtual exploration platforms (Google Earth and Open Street Map) for contextual learning, and visual programming for design, demonstrates the potential of VR to extend beyond immersion and contribute to multimodal learning experiences. Nevertheless, most studies focus on proof-of-concept rather than long-term institutional strategies, raising concerns about the sustainability and pedagogical scalability of these innovations. Approaches such as VR-by-proxy (in which a single instructor broadcasts a VR session) and mobile VR platforms offer more scalable and cost-effective alternatives, thereby increasing accessibility without requiring every student to have specialized equipment [121,126]. The challenge lies in transforming these isolated innovations into systemic, long-term strategies that can be replicated on a scale without losing pedagogical depth.

AI-driven VR systems tailor learning paths, resources, and assessments to individual needs, supporting self-directed and collaborative learning [9]. However, most current evidence remains exploratory, and systematic evaluations of AI–VR synergies in authentic higher education contexts are still scarce. Without rigorous longitudinal studies, transformative claims regarding AI- and robotics-enhanced VR risks are prematurely overstated. A more critical stance is required, as overstating these synergies without robust evidence could lead to unrealistic expectations and wasted institutional investments.

In summary, although the technological ecosystem surrounding VR adoption in higher education is rapidly expanding and diversifying, its fragmented reporting practices, uneven accessibility, and lack of longitudinal validation represent persistent challenges to its effective implementation. Future work should move beyond technical feasibility to critically assess scalability, equity, and ethical implications, ensuring that technological innovation contributes to systemic educational transformation rather than isolated novelty.

5.5. Alignment with the United Nations’ Sustainable Development Goals

Research has increasingly positioned VR as a pedagogical tool aligned with the SDGs, with notable applications in engineering, renewable energy, and urban planning. The emphasis on VR’s role in distance education, virtual field trips, and immersive metaverse environments underscores VR’s capacity to reshape interactions and digital learning practices, consolidating its status as a driver of educational innovation. VR-based hackathons and immersive courses have been shown to significantly improve students’ understanding of the SDGs. These approaches enhance empathy, technical skills, teamwork, and problem solving abilities, while providing authentic real-world experiences [58,127,128]. However, much of the evidence remains fragmented across disciplines, with limited attempts to establish a systematic framework for aligning VR adoption with sustainability agendas. This fragmentation reduces comparability across studies and weakens the capacity to accumulate evidence to inform policy and curricular frameworks.

The predominance of SDG 4 (Quality Education) in the analyzed studies reflects VR’s immediate relevance to teaching and learning contexts. Simulators, digital classrooms, and language-learning applications [129,130] illustrate how VR can strengthen students’ engagement, comprehension, and motivation. These findings are consistent with prior evidence highlighting VR’s ability to deliver multisensory and dynamic learning experiences [53]. However, this focus on SDG 4 creates a disciplinary and thematic imbalance, as the contribution of VR to other SDGs, such as equity, social inclusion, and ethical citizenship, remains underexplored. Moreover, many studies have emphasized students’ enthusiasm and short-term motivational outcomes [2], but few have examined the sustained impact of VR on long-term learning or the development of transversal competencies. This indicates a gap between the rhetoric of sustainability and the actual evidence of VR’s contribution to broad societal goals.

The integration of VR with other SDGs, such as SDG 3 (Good Health and Well-Being), SDG 7 (Affordable and Clean Energy), SDG 8 (Decent Work and Economic Growth), SDG 10 (Reducing Inequalities), SDG 11 (Sustainable Cities and Communities), SDG 13 (Climate Action), and SDG 15 (Life on Land), highlights its potential as a cross-cutting technology capable of addressing global challenges through immersive simulations, urban design, and sustainability-focused learning environments. These applications are promising and help visualize and optimize green spaces, energy efficiency, and noise control in city planning [131]. However, they often remain isolated case studies or pilot projects, raising questions regarding their scalability and institutional adoption in higher education. The risk here is that VR is perceived as a showcase technology rather than a sustainable educational tool unless frameworks for institutionalization are developed to support its use. This research also underscores VR’s potential to deliver educational, psychological, and social benefits, particularly for students with specific learning needs or disabilities, aligning with the objectives of SDG 10 (Reducing Inequalities) [59].

Simultaneously, there are research gaps in relation to SDG 1 (No Poverty), SDG 2 (Zero Hunger), SDG 5 (Gender Equality), SDG 6 (Clean Water and Sanitation), SDG 9 (Industry, Innovation, and Infrastructure), SDG 12 (Responsible Production and Consumption), SDG 14 (Life Below Water), SDG 16 (Peace, Justice, and Strong institutions), and SDG 17 (Partnerships for Goals). These omissions are particularly notable, given the transformative potential of VR in promoting inclusive education, reducing gender disparities, and fostering sustainable production. There is a lack of evidence in these areas, such as robust evidence on VR’s role in directly reducing gender disparities or systematically promoting inclusive education [132]. The absence of systematic evidence in these areas suggests that VR research risks reinforcing narrow interpretations of sustainability and overlooks its holistic agenda.

5.6. Benefits and VR Implementation Challenges in Higher Education

VR has been presented as a tool for promoting accessibility and inclusion, enabling students with physical, economic, or geographic barriers to participate in enriched learning experiences. The reported benefits include increased student motivation and engagement in immersive interactive environments [44,49,59,133]. Its capacity to reduce anxiety and improve academic performance by offering safe spaces for practice without the risks associated with real-world mistakes further reinforces its potential to transform educational practice. As stated in numerous studies, VR supports the development of academic, cognitive, and practical skills such as problem solving, metacognition, and daily living tasks [44,134,135,136]. Transformative roles in higher education and adaptive, personalized, and inclusive VR systems can tailor experiences to individual needs, promoting equity and accessibility in education [59,125,133].

Despite these advantages, several challenges hinder effective integration of these technologies. Teacher preparation and professional development remain critical issues; many instructors remain insufficiently trained in both general Information and Communications Technology (ICT) integration and the specialized pedagogical use of advanced technologies such as VR [1,137]. This gap is often due to insufficient opportunities for targeted training and a lack of ongoing practice-oriented Professional Development (PD) [137,138,139]. This training deficit means that even when institutions acquire VR infrastructure, its pedagogical use may remain superficial, perpetuating the view of VR as a technological “add-on” rather than an integrated learning tool.

The use of VR in teacher education is emerging. When VR is used, it shows promise in improving teacher motivation, self-efficacy, and specific teaching skills; however, research and adoption are limited [140,141,142,143]. Most PD programs focus on basic ICT tools (e.g., LMS and discussion forums), with VR being less commonly implemented [140,144,145]. Without targeted training programs, there is a risk of superficial or inefficient implementation, undermining VR’s benefits and perpetuating inequalities rather than reducing them. However, resource and institutional constraints complicate this adoption. High hardware and infrastructure costs [39,58], limited technical support, and uneven institutional capacity create access inequality. These challenges imply that equity in VR adoption is a question of teacher readiness and institutional vision, as well as financial investment.

Cognitive and physiological factors are also barriers. VR has, on the one hand, been associated with reductions in anxiety, internalizing symptoms such as depression and fear, and improvements in academic performance [146,147,148,149,150,151]. However, studies have emphasized that VR may cause visual fatigue, cognitive overload, or reduced social interaction [39,152,153,154], particularly when it is poorly designed or misaligned with user skills and curricular objectives. The literature also documents instances where VR may act as a distraction, particularly for neurodivergent learners who may be more sensitive to ambient noise and irrelevant stimuli [152]. This duality of positive outcomes vs. potential risks highlights the urgent need for careful instructional design and ethical considerations when adopting VR technology in the classroom.

In this sense, while VR demonstrates strong potential to act as a catalyst for inclusive and equitable education, its successful adoption requires systematic instructional design, institutional support, and capacity building among educators. Addressing these gaps is essential to ensure that VR contributes effectively to both academic success and the broader social goals outlined in the 2030 Agenda. Otherwise, VR risks reinforce educational inequalities, benefiting those already advantaged by resources and training.

5.7. Literature Limitations and Future Research in VR and SDGs

As discussed above, limited teacher training and readiness further hinder the effective integration of VR, especially when support systems are lacking [124,125]. Several studies have confirmed that teacher training in technological competencies is often inadequate [31,155,156], creating a gap that directly affects the effectiveness of VR adoption. Addressing this deficit through sustained professional development is essential, as teachers’ digital skills and ICT proficiency are the cornerstones of sustainable education in a rapidly evolving technological landscape [157,158].

The texts analyzed in this review refer to the technical and financial barriers associated with VR implementation, particularly the high costs of hardware and immersive equipment [39,58,159]. These constraints are significant and highlight persistent inequalities in the institutional capacity to adopt innovative technologies. Beyond cost, pedagogical training and institutional support are necessary to ensure the meaningful integration of VR, which requires the involvement of a broader educational community [159]. Nonetheless, the literature lacks comprehensive cost–benefit analyses, leaving open questions regarding the scalability and sustainability of VR initiatives. This absence weakens the ability of universities and policymakers to make informed decisions regarding long-term investments in immersive technologies.

As prior research suggests [136], the effective use of VR learning objects requires careful instructional design to minimize extraneous cognitive load while enhancing the germane cognitive load. Although some studies highlight VR’s ability to overcome the limitations of traditional teaching by fostering engagement and motivation, the evidence remains inconsistent and heavily context dependent. This reinforces the need for caution when generalizing results, as the success of VR appears to be highly contingent on context-specific variables, such as subject matter, institutional support, and learner profile.

At the methodological level, this section accurately identifies limitations, such as small sample sizes and reduced generalizability. This is consistent with prior studies that applied VR to specific learning needs, including dyslexia interventions [160,161]. Although these case-based contributions are valuable, they often restrict conclusions to narrow populations and short-term results. The need for longitudinal and large-scale experimental designs is evident, particularly for assessing the durability of learning outcomes, retention, and skill transfer [53]. Expanding the sample to include students from diverse universities and disciplines would significantly strengthen external validity and provide a more comprehensive understanding of VR’s educational impact of VR. This expansion is also critical to ensure that VR adoption is not concentrated in elite institutions with greater resource availability.

The proposal to design VR learning scenarios that integrate collaborative and self-directed activities represents an important step toward competency-based education, with specific relevance for disciplines such as maritime training, language learning, and architecture [60,73,76]. Equally important is the need for studies assessing long-term retention, scalability, and adaptability, especially in fields related to sustainability and renewable energy [10]. However, further efforts should focus on the underexplored contributions of VR to broader SDGs, including gender equality, inclusion, and equitable, quality education [76,77]. Neglecting these aspects risks reducing VR to a predominantly technical innovation, overlooking its potential as a driver of social justice and sustainability (the social and ethical objectives of the 2030 Agenda).

The design of VR-embedded strategies that foster higher-order thinking skills, such as critical reflection, advanced reading comprehension, and complex problem solving, is a valuable perspective, as most existing studies focus on short-term learning gains or surface-level outcomes. The integration of cognitive and metacognitive dimensions would allow for a deeper understanding of VR’s pedagogical potential. Likewise, the suggestion to incorporate complementary technologies, such as eye-tracking, to measure attentional processes and learning effectiveness [60,74] provides opportunities for a more precise evaluation of how students engage in immersive environments. However, current evidence remains limited to exploratory studies, and there is still a lack of systematic frameworks that explain how these tools can be scaled and integrated into diverse educational contexts. Developing such frameworks would be decisive for moving from experimental novelty to a structured educational innovation.

The design of VR environments that minimize cognitive load, particularly for learners with low spatial abilities, is a crucial concern. Empirical studies have shown that inappropriate instructional design can lead to cognitive overload, thereby reducing the effectiveness of immersive experiences, particularly among novice learners [154,162,163]. For example, adding unnecessary multisensory elements or complex navigation without clear guidance can distract learners and hinder their ability to recall information [154,162]. Therefore, future research must combine VR innovation with evidence-based instructional design principles to ensure that immersive learning environments enhance understanding rather than hinder it.

In addition, the recognition of gender differences and the influence of resource types in VR-mediated learning [11] are relevant, yet underexplored; for example, immersive vs. desktop learning [164]. This gap underscores the need for a stronger equity perspective in VR research, particularly regarding access, inclusivity, and differentiated learning outcomes. Despite the immersive nature of VR, significant male advantages in visuospatial mental rotation tasks persist, with males outperforming females in terms of accuracy, even in 3D VR settings. This suggests that VR does not universally eliminate gender gaps in cognitive domains [165]. Other studies have shown that female students report higher motivation and engagement across several dimensions (attention, satisfaction, and confidence) when using VR applications, indicating that gender-specific impacts may depend on the measured learning outcomes [166]. This inconsistency highlights the importance of adopting a gender-sensitive approach in VR research and pedagogy to ensure that interventions do not inadvertently perpetuate inequality.

Finally, reference is made to the ethical dimensions of VR adoption, including privacy, equity, and the influence of artificial intelligence on Edu-Metaverse platforms [29]. VR and Edu-Metaverse platforms collect extensive biometric, behavioral, and cognitive data to personalize learning, raising the risk of identity theft, unauthorized surveillance, and data misuse. The integration of AI amplifies these risks by enabling more pervasive data collection and analysis, robust data governance, transparency, and user control over personal information. Current security measures are often inadequate, and breaches can have severe consequences, especially in sensitive educational or healthcare contexts [129,166,167,168,169,170,171]. These aspects remain largely absent from current empirical studies, yet they are essential for safeguarding students’ digital well-being. Incorporating ethical and social considerations into VR research would reinforce its alignment not only with pedagogical goals but also with the broader sustainability and equity frameworks underpinning the SDGs.

In conclusion, the evidence suggests that VR should complement, rather than replace, traditional and collaborative dynamics, ensuring that immersive learning environments contribute to inclusivity and human connection. Aligning VR adoption with the vision of the United Nations 2030 Agenda underscores its potential as a tool for training globally competent citizens who are aware, engaged, and prepared to address the challenges of the 21st century. In summary, future research should consider the following aspects:

-

Longitudinal and large-scale designs should be adopted to assess durability, skill transfer, and scalability [10,53].

-

Increasing the sample size and improving generalizability are recommended.

-

Explore diverse educational stages (primary and secondary) and disciplines beyond STEM to broaden inclusivity and equity.

-

Integrate evidence-based instructional design to minimize cognitive load [136,154,162,163] and incorporate complementary tools (eye-tracking) to study attention and learning processes [60,74].

-

Address gender differences and resource disparities [11,164,165,166] to ensure accessible and inclusive practices.

-

Strengthening ethical and data governance frameworks for VR- and AI-based platforms [29,129,166,167,168,169,170,171].

6. Conclusions

This systematic review confirms that VR in higher education has consolidated itself as a rapidly expanding field of educational technology research between 2020 and 2025. The evidence highlights VR’s global reach, disciplinary diversity, and capacity for innovation, with applications extending beyond technical domains such as engineering and medicine to include the arts, design, sustainability education, and social sciences. Its versatility underscores VR’s potential to foster technical, cognitive, and socio-emotional competencies while supporting active, experiential, and interdisciplinary learning approaches.

The results demonstrate that VR contributes to enhancing conceptual understanding, knowledge transfer, student motivation, and engagement, as well as promoting inclusion and accessibility by reducing geographical, economic, or physical barriers. Furthermore, its alignment with the SDGs, particularly SDG 3 (Good Health and Well-Being), SDG 4 (Quality Education), SDG 7 (Affordable and Clean Energy), SDG 8 (Decent Work and Economic Growth), SDG 10 (Reducing Inequalities), SDG 11 (Sustainable Cities and Communities), SDG 13 (Climate Action), and SDG 15 (Life on Land), reinforces its role as a transformative tool for advancing quality, inclusive, and sustainable education in higher education institutions. However, other SDGs, such as SDG 1 (No Poverty), SDG 2 (Zero Hunger), SDG 5 (Gender Equality), SDG 6 (Clean Water and Sanitation), SDG9 (Industry and Innovation), SDG 12 (Responsible Production and Consumption), SDG 16 (Peace, Justice, and Strong Institutions), and SDG 17 (Partnerships for Goals), remain underexplored and represent an important agenda for future research. These omissions constrain VR’s potential to contribute comprehensively to the 2030 Agenda and highlight the need for more systematic integration of immersive technologies into sustainability-oriented educational frameworks.

This review presents several significant challenges in this field. Methodologically, the predominance of small-scale, short-term quasi-experimental studies constrains generalizability, highlighting the need for longitudinal research with large sample sizes. The lack of standardized reporting of hardware and software limits comparability across studies, whereas high implementation costs and uneven institutional access raise concerns regarding equity and scalability. At the pedagogical level, teacher training emerges as a decisive factor: insufficient preparation in ICT and advanced tools hinders effective integration, risking superficial adoption and reducing the transformative potential of VR. Looking ahead, maximizing the impact of VR in higher education requires systematic instructional design to minimize cognitive overload, especially for students with lower spatial abilities, and to integrate collaborative and self-directed activities that promote higher-order thinking skills such as problem solving, critical reflection, and advanced reading comprehension. Attention to cognitive and physiological risks (visual fatigue, overload, and distraction in neurodivergent learners) must be integrated into design principles to ensure accessibility and well-being.

The convergence of VR with artificial intelligence, robotics, and eye-tracking technologies opens promising opportunities for adaptive and data-informed teaching but also intensifies ethical challenges regarding privacy, surveillance, and data governance, especially within Edu-Metaverse platforms.

Ultimately, this review positions VR as a key educational technology for the 21st century, capable of transforming higher education if its implementation is conceived not as a replacement for traditional teaching but as a complement to hybrid learning models that preserve social interactions as a central element. However, to avoid reinforcing inequalities and disciplinary silos, VR adoption must move beyond isolated pilot projects toward systemic, ethically grounded, and pedagogically integrated strategies. Strengthening inclusivity, gender-sensitive approaches, sustainability alignment, and robust data governance is essential to bridge technological innovation with the imperatives of equity, social justice, and the 2030 Agenda for Sustainable Development.

7. Limitations of Study and Future Research

Although this review offers valuable insights, certain limitations should be acknowledged to contextualize these findings. Despite the methodological rigor applied in the study selection and quality assessment, the final sample (n = 12) may have limited the generalizability of the results and increased the influence of individual study context, design, and methodological choices. In particular, the relatively small number of studies constrains the breadth of perspectives captured and restricts the possibility of identifying more robust cross-contextual patterns. Expanding the evidence base through broader searches, including additional peer-reviewed journals, conference proceedings, and gray literature, would enrich the diversity of perspectives and reduce potential biases.

The search strategy was designed to cover multiple high-impact databases (Web of Science and Scopus) and to apply standardized inclusion and quality criteria. However, another notable limitation is the overrepresentation of studies published in a single journal/editorial (Sustainability, MDPI). While this reflects the journal’s thematic alignment with VR, higher education, and SDGs, it also raises concerns regarding editorial concentration and potential regional or thematic biases in the literature. Consequently, the findings of this review should be interpreted considering that editorial and thematic focus may have influenced the nature and emphasis of the evidence analyzed.

The scope of this review is restricted to higher education- and sustainability-related VR applications only. Future studies should explore other educational stages, such as primary and secondary education, to examine differences in perceptions, implementation strategies, and learning outcomes across a wider spectrum. Such heterogeneity also means that the observed effects may be context-dependent and influenced by specific institutional settings, disciplinary focus or participant profiles. Recognizing these potential biases is crucial, as they underscore the need for cautious generalization of the results. Similarly, research should diversify the SDGs addressed, as most studies focus on SDG 4 (Quality Education), with limited integration of other goals such as gender equality or responsible consumption. Such an expansion would contribute to a more holistic understanding of VR’s role in promoting sustainability-oriented education.

Potential publication bias must also be considered; studies reporting positive or innovative outcomes are more likely to be published, whereas null or negative results remain underrepresented. Furthermore, methodological heterogeneity (differences in research design, sample size, intervention duration, and types of VR technologies) complicates direct comparisons and limits meta-analytic synthesis. The sample of articles analyzed could be enlarged to capture a broader range of works, geographic regions, and publication sources relevant to the field, particularly given the growing number of scientific publications on VR in education. The selective availability of evidence can lead to an overestimation of the perceived effectiveness of VR in advancing pedagogical goals and SDG-related outcomes.

Future research should therefore prioritize the promotion of methodologically sound studies regardless of outcome direction, adopt standardized reporting frameworks to enhance comparability and replicability, conduct longitudinal and cross-institutional studies to strengthen external validity, and integrate VR with broader educational and cultural contexts to ensure inclusivity and sustainability. Addressing these points will contribute to a more comprehensive and balanced understanding of VR’s role in advancing education and the SDGs.