Volume 5, Issue 4, https://doi.org/10.1037/tmb0000146
This special collection on “Advances in Research on Learning in Immersive Virtual Reality” edited by Jeremy N. Bailenson and Richard E. Mayer provides the latest installment of empirical research on learning academic material in immersive virtual reality (IVR). The articles were written by a group of scholars who have well over a century of experience of combined empirical work specializing in understanding the processes and outcomes of using the medium. The collection includes three genres of IVR research: media comparison studies, which compare learning in IVR versus learning the same content with conventional media; value-added studies, which compare learning in IVR versus learning in the same IVR environment with one feature added; and consequences studies, which observe learning in IVR over an extended time period. The research in this special collection yields three research themes: (1) media comparison studies show that learning in IVR is not always more effective than learning with conventional media, (2) value-added studies show that there are some ways that make learning in IVR more effective and some ways that do not, and (3) consequences studies show that it may be useful to examine changes in learners over longer terms. Overall, progress is being made in understanding how learning works in virtual environments.
Funding: Preparation of this article was supported by Grant N00014-21- 1-2047 from the Office of Naval Research.
Disclosures: The authors have no conflicts of interest to disclose.
Open Access License: This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND). This license permits copying and redistributing the work in any medium or format for noncommercial use provided the original authors and source are credited and a link to the license is included in attribution. No derivative works are permitted under this license.
Correspondence concerning this article should be addressed to Richard E. Mayer, Department of Psychological and Brain Sciences, University of California, Santa Barbara, Santa Barbara, CA 93106, United States. Email: [email protected].
When Ivan Sutherland first introduced the concept of virtual reality to the academic community, one of the main motivations for building his “mathematical wonderland” was to more easily allow learners to engage with complex materials (Sutherland, 1965). Three decades later, academic scholars published the first research examining conceptual learning in immersive virtual reality (IVR; which we shorten to VR in this article) in a mainstream education journal (Moreno & Mayer, 2002). In retrospect, the findings from that pioneering study presciently summarize the field of learning in VR today—the medium succeeds at motivating and engaging learners, but high presence does not necessarily translate into learning gains. And today, while scholars continue to explicate the nature of how VR can add value to curricula, technology companies are leaning into the large-scale implementation of VR headsets in classrooms, with billions of dollars being focused on immersive educational initiatives across Silicon Valley (Bailenson et al., 2024).
The goal of this special collection is to help scholars, practitioners, and policymakers better understand when and why VR adds value in classrooms, and when it does not. We recruited a group of scholars who have well over a century of experience of combined empirical work specializing in understanding the processes and outcomes of using the medium. Specifically, as summarized in Table 1, these studies can be sorted among three genres of research on learning in VR: media comparison research, value-added research, and consequences research (Mayer et al., 2023). Media comparison research involves comparing the learning outcomes and processes of students who learn the material in VR with those who learn the same material with conventional media such as on a desktop computer. This work can provide useful information about when it makes sense to convert conventional lessons into VR. Value-added research involves comparing learning outcomes and processes of students who learn from a lesson experienced in VR with those who learn the same lesson in VR with novel features added such as with some form of pretraining. This work can provide useful information about how to increase the effectiveness of learning in VR. Consequences studies involve examining the changes in learning outcomes and processes over the course of learning in VR. This work can provide useful detailed information about the mechanisms underlying learning in VR. In this overview, we examine examples of research from this special collection that fall within each of these genres.
Three Genres of Research on Learning in Virtual Reality | ||
---|---|---|
Name | Description | Example |
Media comparison | Compares learning in IVR with learning the same content with conventional media | Do people learn a technical procedure better in IVR or from watching a video? |
Value-added | Compares learning from a lesson in IVR with learning from the same lesson with one feature added | Do people learn better in IVR if they receive pretraining on the usefulness of IVR? |
Consequences | Examines changes in learners as they learn in IVR over an extended time period | Do people develop a better sense of community if they learn together in IVR over several weeks? |
Note. IVR = immersive virtual reality. |
A theme of the media comparison studies in this special collection is that there is no consistent evidence that learning in IVR is more effective than learning with conventional media. Table 2 summarizes four examples of media comparison research from this special collection. When looking at the comparison studies we find that VR sometimes outperforms control conditions, but when doing so, the effect size tends to be small and not consistent across dependent variables. For example, Rawski et al. (2022) compared sexual harassment bystander intervention training videos either immersively using Google Cardboard or displayed on traditional 2D screens. VR outperformed the 2D display on some variables, for example, intention to act in the future. However, on others, VR fared worse than video, for example, the amount of time spent practicing, and on still others such as knowledge acquisition, there were no differences between conditions. Similarly, Queiroz et al. (2022) presented two studies that examine both knowledge acquisition and self-efficacy for middle school girls who learned about science either in IVR or on a 2D screen. In one study, VR outperformed video on self-efficacy but not learning, and in the other, VR outperformed video on learning but not self-efficacy. Johnson et al. (2022) found that students learned a mechanical maintenance procedure just as well on a desktop computer as in IVR. They also provided some support for individual difference effects, showing the interaction between spatial abilities and VR training in which VR training can create modest positive effects on learning mechanical maintenance skills for students with low spatial skills. Petersen et al. (2022) found that students learned pipetting skills better with an in-person instructor than with IVR. Indeed, when examining the studies in this category, one can easily agree with the conclusions drawn from Petersen et al. (2022): “VR should not replace traditional ways of teaching scientific procedures. Rather, it can be a complement to traditional teaching that can increase accessibility.”
Media Comparison Studies | |||
---|---|---|---|
Source | Content | Learner | Control |
Rawski et al. (2022) | Antisexual harassment knowledge and skills | Adults | Video |
Queiroz et al. (2022) | Ocean ecology knowledge | Middle schoolers | Video |
Johnson et al. (2022) | Mechanical maintenance procedural skills | Adults | Desktop simulation |
Petersen et al. (2022, Experiment 2) | Pipetting procedural skills | High schoolers | Face-to-face |
A theme of the value-added studies in this special collection is that there are some techniques that boost the effectiveness of learning in IVR and some that do not. Table 3 summarizes four examples of value-added research from this special collection. Plechatá et al. (2022) investigated the effect of a VR simulation that focused on the impact of food choices on the natural environment in a sample of middle school students. The intervention resulted in a significant pre- to posttreatment increase on all measured variables: intentions, knowledge gain, and knowledge transfer with a moderate-to-large effect size. Moreover, the authors varied whether or not participants were allowed to change the future by making more pro-environmental decisions and demonstrated that adding the manipulation increased pro-environmental intentions and knowledge transfer. In addition, implementation matters, and how one prepares users for a VR experience can have an impact. For example, Ferdinand et al. (2023) showed that how one frames the medium can change the attitudes of learners subsequently. Petersen et al. (2022) found that adding pedagogical agents into an immersive virtual environment did not improve the learning of pipetting skills. Johnson et al. (2022) found that students learned mechanical maintenance procedure skills in IVR just as well when they interacted with voice-based commands as with gesture-based commands.
Value-Added Studies | |||
---|---|---|---|
Source | Content | Learner | Variable |
Plechatá et al. (2022) | Climate change knowledge and beliefs | Middle schoolers | Added simulation |
Ferdinand et al. (2023) | Biology knowledge | High schoolers | Pretraining on the usefulness of VR |
Petersen et al. (2022, Experiment 1) | Pipetting procedural skills | Adults | Active instructor versus hand only |
Johnson et al. (2022) | Mechanical maintenance procedural skills | Adults | Interact via voice or gesture |
Note. VR = virtual reality. |
A theme of consequences studies in this special collection is that there is something to be learned by observing changes in learners over the course of extensive experience in VR. Table 4 summarizes two examples of consequences research from this special collection. Han et al. (2022) presented field study data that highlight how to overcome the complexity of this media, demonstrating that students need to master “learning how to use VR before learning with VR.” Bondie et al. (2023) further demonstrated this point, showing how previous experience within a domain moderates how VR training is received.
Consequences Studies | ||
---|---|---|
Source | Content | Learner |
Han et al. (2022) | Learner ratings of social connection | Adults |
Bondie et al. (2023) | Teacher feedback quality | Teachers |
In line with Moreno and Mayer’s (2002) foreshadowing research on the limited value of learning academic content in IVR, the research in this special collection yields three research themes: (1) media comparison studies show that learning in IVR is not always more effective than learning with conventional media; (2) value-added studies show that there are some ways that make learning in IVR more effective and some that do not; and (3) consequences studies show that it may be useful to examine changes in learners over longer terms.
On the theoretical side, there is a need for a comprehensive theory of the cognitive, social, affective, and motivational processes involved in learning in VR. On the practical side, there is a need to determine the conditions under which learning in VR is most effective, including determining which features boost learning in VR, which kinds of material and which kinds of learners benefit most, and how to embed VR learning experiences within the context of existing lessons and courses.
This special collection offers an impressive array of research, with variance in participants (children, college students, professionals), type of VR hardware (phone-based, room-scale, 2D desktop), type of content (computer graphics, 360 video, real-time avatars), type of learning (science, technology, engineering, and mathematics, mechanical, soft skills), and outcome measures (recognition, recall, learning transfer, self-efficacy). In this sense, the studies are incredibly different from one another. But the findings across the studies offer similar themes—who learns, in what context, and how learning is measured, matters. As VR continues to become integrated into classrooms around the world, scholars should continue on this nuanced path of research. The research presented in this special collection offers a worthwhile step in understanding how to make the most of learning academic content in VR.