ISSN: 1130-3743 - e-ISSN: 2386-5660
DOI: https://doi.org/10.14201/teri.31921

A SYSTEMATIC REVIEW OF COLLABORATIVE LEARNING THROUGH VIRTUAL REALITY, AUGMENTED REALITY AND MIXED REALITY

Revisión sistemática sobre aprendizaje colaborativo mediante realidad virtual, realidad aumentada y realidad mixta

Joel Manuel PRIETO ANDREU
Universidad Internacional de La Rioja. Spain.
joelmanuel.prieto@unir.net
https://orcid.org/0000-0002-2981-0782

Date received: 26/03/2024
Date accepted: 16/05/2024
Online publication date: 01/01/2025

ABSTRACT

The integration of augmented reality (AR), virtual reality (VR), and mixed reality (MR) in education holds great promise for advancing teaching and learning models that cater to the needs of 21st-century students. AR- and VR-based learning offer efficiency and total immersion, while MR merges the real world with the virtual world for overlapping experiences. However, there is a lack of studies that examine the pedagogical factors and potential benefits of collaborative learning in these environments. This work aims to conduct a systematic literature review to explore the impact of collaborative experiences facilitated by AR, VR, and MR. The rapid systematic review was prepared following the protocols of the PRISMA 2020 declaration in the selection of studies. This review analysed 62 studies centred on collaborative learning through AR, VR or MR, comprising 7 theoretical reviews, 21 experimental studies, and 34 observational studies. The criteria set out in the Guidance for Quality Assessment Tool of the National Heart, Lung, and Blood Institute (NHLBI, 2020) were followed. Technical limitations and a lack of instructional design hinder the usability of virtual reality and augmented reality in education. Real-world sessions and adequate interaction are suggested for monitoring and manipulating virtual content. Integrating MR technologies into higher education can enhance educational practice, but challenges remain. Further research is needed to assess their effectiveness in collaborative learning. By understanding the challenges and opportunities presented by these technologies in teaching scientific subjects, policymakers and educators can develop effective strategies to leverage current trends and foster meaningful educational change.

Keywords: augmented reality; virtual reality; mixed reality; collaboration; education; learning.

RESUMEN

La integración de la realidad aumentada (RA), la realidad virtual (RV) y la realidad mixta (RM) en la educación es muy prometedora para el avance de modelos de enseñanza y aprendizaje que satisfagan las necesidades de los estudiantes del siglo XXI. El aprendizaje basado en RA y RV ofrecen eficacia y total inmersión, mientras la RM fusiona el mundo real con el virtual para tener experiencias superpuestas. Sin embargo, son escasos los estudios que examinen los factores pedagógicos y los beneficios potenciales del aprendizaje colaborativo en estos entornos. Este trabajo tiene como objetivo realizar una revisión sistemática de la literatura para explorar el impacto de las experiencias colaborativas facilitadas por AR, VR y MR. Para la selección de estudios se ha elaborado la revisión sistemática rápida siguiendo los protocolos de la declaración PRISMA 2020. En esta revisión, se analizaron 62 estudios centrados en el aprendizaje colaborativo a través de AR, VR o MR, comprendiendo 7 revisiones teóricas, 21 estudios experimentales y 34 estudios observacionales. Se han seguido los criterios expuestos en la Guía para la Evaluación de Criterios de Calidad del Instituto Nacional del Corazón, los Pulmones y la Sangre (NHLBI, 2020). Las limitaciones técnicas y la falta de diseño instruccional obstaculizan la usabilidad de la realidad virtual y la realidad aumentada en la educación. Se sugieren sesiones del mundo real y una interacción adecuada para monitorear y manipular contenido virtual. La integración de tecnologías de RM en la educación superior puede mejorar la práctica educativa, pero persisten desafíos. Se necesitan más investigaciones para evaluar su eficacia en el aprendizaje colaborativo. Al comprender los desafíos y oportunidades que presentan estas tecnologías en la enseñanza de materias científicas, los formuladores de políticas y los educadores pueden desarrollar estrategias efectivas para aprovechar las tendencias actuales y fomentar un cambio educativo significativo.

Palabras clave: realidad aumentada; realidad virtual; realidad mixta; colaboración; educación; aprendizaje.

1. INTRODUCTION

Use of augmented reality (AR), virtual reality (VR), and mixed reality (MR) in education promises new teaching and learning models that cater to the needs of 21st-century students. There has been a striking growth in electronic learning, as distance education based on digital platforms has become popularised. Thanks to this, there has been a large increase in innovation in the educational sector. There is clearly growing interest in the use of AR, VR, and MR in education, backed by studies that have demonstrated their effectiveness for improving information retention, facilitating active learning, and providing immersive contextualised experiences (Ke & Hsu, 2015; Lindgren et al., 2016; Mosher & Carreon, 2021; Vasilevski & Birt, 2020; Veer et al., 2022; Wang et al., 2021; Watson & Livingstone, 2018; Webb et al., 2022). However, despite these advances, there are gaps in the understanding of how these technologies can be exploited effectively in collaborative learning environments. Consolidation of existing knowledge about the impact of collaborative experiences using AR, VR, and MR in education is needed. Furthermore, the importance of analysing and critically assessing the quality of the available studies – experimental and observational – to provide a complete and rigorous overview of the current state of the art in this area is apparent.

The technique of AR-based learning is an active-learning method as it can quickly and efficiently turn the information acquired into long-term memory (Santos et al., 2014). Learning materials in the form of videos or as part of a simulation or a game are available for use by instructors to involve the students. According to Wanis (2019), AR combines real and virtual objects in its own environment and it is known as a technology that has the capacity to superimpose virtual objects in our real world in real time. Virtual reality, in turn, is generally known as a totally immersive computer technology that enables the user to interact with the digital working space in a unique way. By comparison, MR combines elements of the real and virtual worlds to create an immersive experience. It uses technologies such as AR and VR to superimpose digital objects in the physical environment.

According to Webb et al. (2022), many previous pieces of research have suggested that computer simulations can support the learning of difficult concepts by secondary students. Elmqaddem (2019) argues that, in the sector of education and training, it permits engineers, for example, to learn new procedures in real conditions. When encountering a new device, the learner can discover the disassembly procedure step-by-step, seeing the instructions appear in real time. HoloLens, for example, enables medical students to manipulate and visualise the human body with unprecedented precision. In the cultural sphere, augmented reality applications enable tourists or visitors to museums to discover the history of places or works just by pointing their smartphone’s camera in its direction. According to Ke and Hsu (2015), Wikitude, a location-based mobile AR application, uses the GPS incorporated into mobile devices to track users’ real locations and present contextually relevant virtual information about surrounding points of interest (for example, buildings, parks and shops). AR applications can also function without location restrictions and use images and objects from the real world as "triggers" to activate the superimposition of digital information to support learning. For example, the Aurasma application enables its users to see "Aura", multimedia artefacts that can be animations or video clips, by pointing their mobile devices at a designated trigger in the real world.

With regards to collaborative learning, mobile computer supported collaborative learning (MCSCL) is the practice of groups of individuals creating meaning in the context of a joint activity mediated through mobile computing (Ke & Hsu, 2015). In a recent review of empirical studies of MCSCL, Hsu and Ching (2013) found multiple ways in which mobile information mediated the creation of meaning in a joint activity.

In particular, wirelessly connected mobile devices can: 1) facilitate the exchange of information and the provision of instant commentaries; and 2) provide people with different parts of a group learning task and coordinate the interaction orientated towards the task. However, although Webb et al. (2022) suggest that virtual reality has benefits for learning, they conclude that there are insufficient studies examining what pedagogical factors can be important, or whether collaborative learning in these settings can be beneficial (Merchant et al., 2014). Whewell et al. (2022) explored different combinations of immersive digital technologies through a connected community, noting that both connectivism and constructionism underline the importance of constructing social relations between students in a learning context. Alalwan et al. (2020) indicate that understanding the challenges posed by use of virtual reality and augmented reality in teaching science subjects would provide the means for educational policymakers to support the necessary measures to reflect effectively on current trends and thus bring about educational change.

The main aim of the present work is to carry out a systematic review of literature on the impact of collaborative experiences through the use of AR, VR, and MR. This review proposes a discussion of the objectives and contributions of the studies included in the review, as well as the limitations and challenges facing the implementation of these technologies in collaborative learning environments.

Likewise, the specific objectives of the review are to provide an umbrella review that establishes the current state of the art of the topic, to analyse and assess separately the quality of the experimental and observational studies included in the review, and to discuss the objectives and principal contributions of the studies included in the review and the limitations of the use of AR and VR in collaborative settings.

2. METHOD

The systematic review followed the protocols of the PRISMA 2020 declaration (Page et al., 2021) in the selection of studies.

2.1. Eligibility criteria

The following inclusion criteria were used: studies written in Spanish and English; peer-reviewed manuscripts; works with full-text access. As exclusion criteria, doctoral theses and manuscripts from unreliable sources were rejected. The sample for the systematic review comprises studies that have used VR, AR or MR in collaborative teaching experiences, and studies that have theorised about their possibilities in the educational setting.

2.2. Information sources and search strategy

The search for studies was done on 23 May 2023 in two of the most important online databases in the areas of social sciences and health sciences: ISI Web of Science and SCOPUS. By editorial decision, the search for studies was updated for the final time on 13 May 2024. Table 1 shows the details of the searches in JCR and in SCOPUS, using the following search term: [collaborative learning AND VR OR virtual reality OR AR OR augmented reality OR MR OR mixed reality]. The search in SCOPUS was filtered by title, abstract and keywords, and in ISI Web of Science by TOPIC, since 2009. A quantitative ad-hoc table was prepared, ordered by year and assigning a record number to the studies that permitted their subsequent selection in the classification phase.

2.3. Selection of studies and data extraction process

During the review of the full text, the reviewer assessed the eligibility of the studies in accordance with the predefined inclusion and exclusion criteria. In this case, as one single person carried out the selection process in the study, it was important to ensure that the process was impartial and exhaustive. This was achieved by having a second reviewer independently examine a randomly selected subset of the studies to guarantee consistency in the selection process. With regards to the data extraction process, the data were extracted using a standardised form to guarantee consistency and precision. In this case, the data extraction was done by one single person and recorded in an Excel spreadsheet, which included information about the design of each study, the characteristics of the participants, the interventions, the results and the risk of bias, among other relevant aspects, helping ensure that the data would be complied in a consistent and precise way and thus improving the quality and reliability of the systematic review.

Moreover, to analyse the scientific production, the Guidance for Quality Assessment Tool for Systematic Reviews and Meta-Analyses of the National Heart, Lungs and Blood Institute (NHLBI, 2020) was used. The NHLBI guidance is a well-established and widely recognised tool in the field of health and medicine. Its use provides credibility and reliability to the assessment of the quality of the systematic review. It also provides a standardised and systematic framework to assess the quality of scientific evidence, including considerations on the methodology of the studies included and the clarity and coherence in the presentation of the results among other relevant aspects for the thematic area of the present review.

The Quality Assessment of Controlled Intervention Studies scale was also used to assess the original experimental studies in which an intervention or treatment is implemented and is compared with a control group. This scale centres on various key methodological quality domains, such as study design, random allocation, allocation concealment, blinding of participants and assessors, comparability of the groups at the start of the study, handling of deviations from the protocol and the attrition rate, among others. Each domain is assessed against a series of specific criteria, and a score is assigned to each study according to whether these criteria are fulfilled. This allows a quantitative assessment of the quality of the studies included in the review.

Finally, the Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies scale was used to assess the original observational studies, such as cohort studies and cross-sectional studies, that did not involve experimental interventions. Like the scale for experimental studies, this tool centres on various domains of methodological quality that are relevant to observational studies, such as study design, the size and representativeness of the sample, the validity and reliability of the measurements, the handling of confounding factors and biases, and the follow-up of participants. Each domain is assessed by means of specific criteria, and a score is assigned to each study according to its fulfilment, which allows a quantitative assessment of its quality.

Both tools are designed to be used by systematic reviewers as part of the process of critical assessment of scientific literature. They allow a standardised and rigorous assessment of the methodological quality of the studies included in the review, something that contributes to the validity and reliability of the findings of the review. To measure the quality of the experimental studies and of the controlled intervention or observational and transversal cohort studies, the quality criteria from the NHLBI (2020) quality assessment tools were followed, using cutoff points that define a quality scale of 1 to 4 stars depending on the number of criteria that were fulfilled [1 star, ★☆☆☆ (< 3), 2 stars, ★★☆☆ (3–6), 3 stars, ★★★☆ (7–10), 4 stars, ★★★★ (11–14)]. The NHLBI quality assessment tools are not designed to assess the quality of articles in the field of education. However, responses provided above are based on the sections of the tool that are applicable to this review.

3. RESULTS

The search gave 3227 results in SCOPUS (1656 conference proceedings, 978 journal articles, and 455 books), resulting in 36 reviews, 610 articles, 142 book chapters, and the rest unspecified. The search in WoS with the same search string, seeking words in the search fields by “title”, gave 75963 results, leaving 2497 for the area of Educational Research in Education. In total, in the first phase of the identification, 5724 registers were identified. Table 1 shows the details of the searches in JCR and in SCOPUS.

In the second phase of classification, the search for articles was filtered by open access, and registers that did not comply with the inclusion and exclusion criteria were eliminated, giving a total of 696 results in SCOPUS and 815 results in WoS, leaving a total of 1511 registers. In the eligibility phase, and after reading the abstract, 1274 studies were rejected because they did not fulfil the inclusion criteria and were not related to the thematic line of the systematic review, resulting in 237 studies. Finally, in the article selection for review phase, 62 works were selected. Figure 1 is a flow chart of the article selection process.

This review analysed 62 studies centred on collaborative learning through AR, VR, and MR, including 7 theoretical reviews (Table 2, Annexe 1), 55 original studies, 21 experimental studies (5 VR, 8 AR, and 8 MR) (Table 3, Annexe 2) and 34 observational studies (10 VR, 5 AR, and 19 MR) (Table 4, Annexe 3). Taking the matrix of the 55 original studies included in the review, 15 studies (28 %) analyse the impact of collaborative work through VR, 13 studies (23 %) analyse the impact of collaborative work through AR, and 27 studies (49 %) analyse the impact of collaborative work through MR.

Figure 2 compares the percentages of fulfilment of each of the 14 assessment criteria from the Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies to assess original observational studies on the one hand and, on the other, from the Quality Assessment of Controlled Intervention Studies to assess the original experimental studies.

Figure 3 compares the quality of the observational and experimental studies included in the systematic review.

It is apparent that the observational studies displayed greater quality than the experimental studies, taking into account the 14 assessment criteria from the two NHLBI tools. In decreasing order, 52 % of the observational studies scored 3 stars, 33 % scored 4 stars, 9 % scored 2 stars, and 6 % scored 1 star. On the other hand, 63 % of the experimental studies obtained 1, 26 % scored 2 stars, 11 % scored 3 stars, and 0 % scored 4 stars.

4. DISCUSSION

4.1. Objectives and principal contributions of the studies included in the review

The studies are centred on exploring the benefits and applications of AR and VR in education, as well as in collaborative learning. Many of them also seek to comprehend how these technologies can improve social interaction, creativity and the idea creation in collaborative learning (Rodríguez et al., 2018; Sanabria & Lizárraga, 2017). Some of the studies included in the systematic review have the aim of exploring, developing and validating the effectiveness of VR and AR in collaborative learning (Ke & Hsu, 2015; Lindgren et al., 2016; Veer et al., 2022; Watson & Livingstone, 2018). These studies seek to investigate the ways these technologies can improve learning experiences, foster collaboration and improve academic performance in different areas, such as biology, technological education, geography and engineering, focussing on different age groups.

The studies also centre on the development of collaborative interfaces in VR and AR environments, as well as on identifying the effects of haptic systems and 3D models on learning (Kucukyilmaz & Issak, 2019; Webb et al., 2022). Furthermore, some studies have compared the results of collaborative learning in virtual reality and augmented reality environments with traditional text books. Some studies have sought to compare the performance of students in virtual settings with the use of conventional text books (de Back et al., 2020). Others seek to develop improved tools with AR and VR and assess their effectiveness in learning (Sanabria & Lizárraga, 2017), and others explore how technical limitations can be overcome to improve the experience of collaborative learning in AR and VR environments (Alhumaidan et al., 2018; Ahsen et al., 2019; Bekele et al., 2021; Jovanović & Milosavljević, 2022; Rusli et al., 2023)

Although some of the objectives of the studies that are centred on AR and VR can complement one another, there are some notable differences. Studies that centre on AR generally seek to examine how AR technology can improve the learning experience, for example, how AR can be used to improve comprehension of abstract concepts or to improve interaction between students and learning materials. Studies have also explored how AR can improve creativity and collaboration in class (Chung et al., 2021; Matcha & Rambli, 2013; Sanabria & Lizárraga, 2017; Strada et al., 2023).

In contrast, the studies that centre on virtual reality generally seek to examine how virtual reality technology can improve the immersive learning experience and provide students with a simulated environment for experiencing real-life situations (de Back et al., 2020; Lindgren et al., 2016; McArdle et al., 2011). Studies have also been done to explore how virtual reality can improve collaboration and problem solving among students (Wang et al., 2021; Webb et al., 2022).

In summary, augmented-reality studies centre on how the technology can improve comprehension and interaction, while virtual-reality studies centre on providing an immersive simulated learning experience. In conclusion, these studies underline the importance of mixed reality in collaborative learning and they suggest that these technologies have the potential to improve collaborative learning and students’ academic performance significantly. They also show that technologies can improve social interaction and creativity, significantly improving learning and collaboration in a close future.

With regards to the principal contributions of the articles reviewed, AR and VR are technologies with great potential in different areas of application, such as entertainment, tourism, architecture, medicine, education and industry. In particular, immersive virtual reality improves collaborative learning, especially in spatial awareness and navigation, and it can improve the participation of students in a collaborative problem-solving task and improve collaborative group communication (de Back et al., 2020). It is also worth mentioning the potential of collaborative virtual environments (CVE) for supporting socio-communicative interactions in people with autism spectrum disorder (ASD) (Wallace et al., 2017). Moreover, technological pedagogical content knowledge (TPACK) is important for teachers and students in designing and implementing educational technologies. In this sense, the design of mobile augmented reality devices tended to be better at promoting technological knowledge competences, while the visualisation on mobile phones with AR technology appeared to be better at supporting the development of content knowledge, in both cases increasing students’ scientific interest and collaboration skills (Ke & Hsu, 2015).

AR and VR technologies have great potential to improve collaborative learning and socio-communicative interaction in different areas of application, but it is necessary to continue researching the use of these technologies in real educational environments and to assess their efficacy for improving learning and group collaboration (Wallace et al., 2017). AR and VR are also effective tools for improving the education and understanding of the sciences, transforming passive learning into attractive interaction with 3D objects and improving students’ motivation for the intervention (McArdle et al., 2011; Stromberga et al., 2021; Tüzün et al., 2019; Watson & Livingstone, 2018; Webb et al., 2022)

As for learning social skills, intervention through AR/VR has proven to be effective in the majority of the studies, which suggests that combining these technologies can provide a more effective intervention (Mosher & Carreon, 2021). In the field of medicine, AR and VR also improve learning of practical skills and comprehension of anatomy. In particular, MR has shown value for facilitating meaningful learning opportunities and improving students’ knowledge of dynamic systems and control concepts (Richards, 2023). Likewise, VR and AR technologies are valuable tools that complement traditional teaching methods, providing an attractive and motivational educational experience. Nonetheless, their efficacy depends on various factors and requires more research.

4.2. Limitations of the use of AR and VR in collaborative settings

According to Fu and Liu (2018), because of the specific nature of construction projects, the use of AR/VR (for example, investing in hardware and simulation of construction) can result in a high cost and time spend. For virtual reality, the main drawbacks are an effect of limited immersion, inadequate sensory feedback (that is, the virtual sense predominates with less presence of the sense of hearing, haptic sense and others), lack of comfort for users (motion sickness, difficulties with the body), and the cost or accessibility of this technology. On this line, according to Elmqaddem (2019), better accessibility of VR and AR is the primary element that should help to popularise these technologies. Google was the pioneer in proposing low-cost virtual reality platforms and headsets, starting with Google Cardboard in 2014 (the first virtual reality platform and virtual reality headsets). In 2016, Google presented a new virtual reality platform and headset called Daydream (which costs under 50 euros). Given that the high importance of accessibility of virtual reality, Oculus has been working on a project for new displays. To benefit from higher quality virtual reality headsets such as Oculus Rift, an expensive computer is needed (that meets the needs of Oculus Rift). And Oculus Rift itself, cost 700 euros. According to Miller et al. (2020) portable headsets, like Microsoft HoloLens, are increasingly common and are an example of new technologies that are mature for further research from an educational research perspective. Microsoft HoloLens is a portable computer running Windows 11 that allows interaction with MR. The head-mounted display (HMD) superimposes three-dimensional (3D) virtual images on the real world, enabling users to interact with these holographic projections through voice commands and gestures. According to Sonntag and Bodensiek (2022), most of the research in this field is still limited to tablet-based augmented reality (AR), although MR through HMDs offers advanced possibilities to support students in laboratory-based learning activities: use of an HMD allows experimentation with the hands free, virtual representations or simulations can be shown and perceived three-dimensionally and superimposed on the physical environment more naturally, and intuitive gesture-based interaction with the virtual elements. Therefore, MR-HMD is able to combine laboratory learning with the advantages of interactive computer simulations such as PheT. The few studies that explicitly use MR-HMD indicate a positive impact on conceptual knowledge (Scaravetti & Doroszewski, 2019; Sonntag et al., 2019). The MR-HMDs recently available on the market, such as Microsoft HoloLens 2, which are used in the study by Sonntag and Bodensiek (2022), feature integrated eye tracking functions.

On the other hand, McArdle et al. (2011) suggested that while CLEV-R provides asynchronous access to reading material through the library, synchronous reading is where most learning occurs. As it offers both forms of e-learning, students can decide which they find more beneficial. CLEV-R also offers few opportunities for asynchronous online communication, such as message boards, email and forums. These can easily be incorporated into the design and form the focus of future studies to determine how they compare with the real-time communication that CLEV-R currently offers, and, ultimately, improve the students’ learning experience. However, there are technical limitations on the use of MR technologies, especially in the AR subcategory, such as response time in collaborative environments. There will also be difficulties replicating MR experiments in real educational environments and scenarios, and an absence of assessment processes in collaborative educational environments that incorporate technologies in the continuum of virtuality with the introduction of new optical head-mounted displays (OHMD), such as HoloLens, Oculus Rift, and HTC Vive.

As for the difficulties of applying these technologies in an educational context, Jovanović and Milosavljević (2022) suggest that the educational aspects and problem solving through 3D avatars aspects could be a good way of personalising the experience and attracting users so that they support the platform. Nonetheless, monitoring students’ activities is difficult because attention is centred on the virtual environment. It is suggested that real world class sessions are more reliable for the activities of the students. Alalwan et al. (2020) indicate that there is a shortage of guidelines for VR, with a lack of long-term participation in AR, a lack of instructional design, a lack of competence, and a lack of focussed attention in VR and AR being the most important factors to address to ensure its usability in the educational system.

As for interaction, according to Wanis (2019), VR technology is centred on replacing reality with a complete virtual environment. To manipulate virtual content in the AR and VR work space, adequate interaction is needed. Interaction is one of the most studied topics in research into virtual reality and augmented reality. Interaction with AR should work well when it is a case of using real objects to manipulate virtual objects.

The integration of MR technologies and mobile visualisation in higher education is an important contribution that can improve educational practice. AR, VR, and MR have the potential to improve scientific comprehension and education, and the combination of these technologies can provide a more effective intervention in the teaching of social skills. Although use of these emerging technologies in education has benefits, there are also challenges that must be faced, such as the lack of competence, limited instructional design, the lack of focussed attention, the lack of time, and limited environmental resources.

5. CONCLUSIONS

Augmented reality (AR) and virtual reality (VR) technologies have emerged as powerful tools in the field of education, especially for fostering collaborative learning. The studies reviewed reflect a varied focus, from exploring their benefits and applications to development of collaborative interfaces and comparison with traditional teaching methods. The following conclusions stand out in this review: