XR for Teaching and Learning: Year 2 of the EDUCAUSE/HP Campus of the Future Project

Factors That Influence Learning

The research question motivating this study is: What factors influence the effectiveness of XR technologies for achieving various learning goals? Before an XR technology can be used to achieve anything, however, it has to be adopted for use.

Factors That Influence the Adoption of XR

"What can we do in XR that we can't do otherwise?"
Meredith Thompson, MIT

Two factors emerged as critical for the adoption of XR in pedagogy. The first is that the XR application must fit into instructors' existing practices. This is of course no different from the adoption of any other new technology: we all use new tools for old uses at first, until we figure out what the new tool is capable of. This is a central finding of research on diffusion of innovations: To be adopted, an innovation must be compatible with whatever existing systems are in place.1 And to be adopted not just by one instructor but more widely, across a curriculum or a field, an XR application must fit into existing instructional methods. Many disciplines (such as nursing) have existing standards, or at a minimum existing curricula and instructional methods (as in biology), and XR must fit into such existing standards and practices. Over time, those standards and practices will change, influenced at least in part by XR. But at least at the beginning, an innovation must fit into existing ways of doing things.

The second factor influencing the adoption of XR is cost. Again, this is basic diffusion of innovations stuff: the cost of adopting an innovation cannot be too high. Cost may take the form not only of money but also of the time required to scale the learning curve, the cognitive load of using it, etc. For XR, all of these costs figure into instructors' calculations. Commercial XR applications, such as anatomy simulations published by traditional textbook publishers, may be quite expensive, and if a commercial simulation does not match the instructional goal, then an instructor or a program is unlikely to spend the money. (This was part of the motivation behind the Yale Department of Neuroscience developing its own AR application.) Even if an XR application does match the instructional goal, it must still be cost-effective—as easy to deploy, learn, and use as non-XR alternatives, for a similar or preferably lower cost. (This was part of the motivation behind the Morgan State project developing XR simulations for nursing education rather than using established simulations with standardized patient–actors.)

Factors That Influence the Effectiveness of XR

Only after an XR technology has been adopted for use in a course is it meaningful to ask what factors influence its effectiveness for achieving the learning goals of that course. Several factors seem to influence the effectiveness of XR, though further empirical research is needed on all of these.

The Fidelity and Realism of the Simulation

The more authentic an experience a simulation provides, the more valuable it is as a teaching tool. This is especially true in skills-based learning, such as nursing, where XR simulations are in competition, so to speak, with existing high-fidelity simulations such as those with standardized patient–actors. But this is also true for simulations in fields that are traditionally taught as more conceptual knowledge, the difference being that the contents of these simulations are outside the realm of human experience, so fidelity and realism must be defined differently. Humans have experience of electromagnetism, for example, but that experience is somewhat indirect: we have all played with magnets, but we can't see or feel magnetic fields directly. A simulation of electromagnetism (figure 5) must match current scientific understanding and must be updated regularly to keep it accurate as disciplinary knowledge changes. Equally important, an XR application designed as a teaching tool must be authentic; however, that authenticity is defined not by how well the simulation mirrors the physical world but rather by how well it supports "embodiment." To a certain extent, the XR user must engage in willing suspension of disbelief when entering a simulation, just as with any media content. The difference is that, for an educational simulation, the "fictional" world of the simulation must not only be internally consistent, as it should be with any story, but also be consistent with the physical world with which the user is familiar.2 To be useful as a teaching tool, a simulation must be both accurate and convincing.

A student using Electrostatic Playground
Figure 5. A student learning about electromagnetism using Electrostatic Playground
Image courtesy of John Belcher, MIT Department of Physics

The Ease of Use of the Hardware and Software

New technologies are often complicated or difficult to use. Over time and through multiple versions, interfaces change as developers learn more about use cases and usability. This has certainly been the case for XR technologies. We are already seeing some standardization on common and easy-to-use interfaces, such as arranging blocks of text in VR simulations in scrollable columns rather than in long lines (to avoid blurring at the edges of the user's field of view) and designing mobile AR applications to be operated one-handed (because the other hand is holding the smartphone). Interface development is even more critical for accessibility, such as captioning for users who are deaf or hard of hearing, and for controllers that are usable for those with mobility impairments. Generally standardized interfaces make a technology easier to use, as the interface is likely to be familiar even to a newcomer. (The iOS and Android smartphone operating systems have obvious differences, for example, but both rely on the desktop and icon metaphors that became familiar with the Apple Macintosh in the mid-1980s.) The ease of use of XR technology strongly influences its effectiveness for learning. A critical issue for the XR simulations being developed at Morgan State, for example, is that it is difficult, maybe impossible, to simulate fine motor actions, such as giving an injection, using a handheld controller. Morgan State—and probably any XR-based medical training—requires hand tracking, for example a VR glove or a Leap Motion controller.

Providing Something Not Available Any Other Way

An XR technology must enable the student to learn in a way that is not possible using any other media; otherwise, there is little incentive for either the student or the instructor to adopt the XR alternative and little reason to believe that it will be more effective than traditional instructional tools. XR does, however, enable learning in new ways: learning about cellular biology by exploring a cell as if it were a physical space, for example, or learning medical techniques by practicing them more times than would be possible in the physical world, or learning about architecture by developing data visualizations of real-time interactions between people and spaces. The physical world has inherent limitations: for example, a human being cannot shrink to fit inside a cell. XR technologies make it possible to overcome these limitations, thereby creating new learning opportunities.

An Increase in Students' Time-on-Task

Several interviewees for this project noted that XR technology motivated students to engage with course material for longer. It is not possible for this study to quantify the amount of this increase, as we didn't have access to the students at participating institutions during the course of this project to collect data on their time spent on coursework. However, this finding is consistent with other research on blended learning. Other researchers have found that increased time-on-task leads to increased performance on learning outcomes and that the use of blended learning techniques leads to increased time-on-task, due to the use of additional learning materials and additional opportunities for collaboration around those materials.3 These findings seem to hold for XR technologies as well, suggesting that XR is a valuable contribution to blended learning.

A Spirit of Experimentation

The student, the instructor, and those providing support to the student and the instructor all must possess a spirit of experimentation. Even though XR technologies are starting to see some standardization, both hardware and software are still in a period of rapid development. Several interviewees for this project stated that some of the best student work with XR was the most unexpected. This applied to students, both individually and in groups, who went above and beyond the course assignment, or who were working on a specific project but were given free rein in how to accomplish it, or who simply developed their own idea for a project and realized it outside of any course. These students learned a great deal about both the technology and the subject matter; but just as important, they learned how to learn. They engaged in complex problem-solving and interacted deeply with their subject matter. For students to be able to engage in this kind of self-directed learning, they need freedom and flexibility, which requires buy-in from both the instructor and the institution. Instructors must be willing to let their students experiment in the context of a course or an assignment. Even more basic, though, is that XR technology must be freely available for students to use—for example, in a lab or a makerspace. The security of the hardware is a concern, but the lighter the touch, the better. One of the most important factors influencing the effectiveness of XR for learning is simply access to the technology. From there, instructors and institutions should follow where students lead.

Notes

  1. Everett M. Rogers, Diffusion of Innovations, 5th ed. (New York: Free Press, 2003).

    ↩︎
  2. J.R.R. Tolkien, "On Fairy-Stories," in J.R.R. Tokien, The Tolkien Reader; and John McCoy and Tomer Ullman, "Judgments of Effort for Magical Violations of Intuitive Physics," PLoS ONE 14, no. 5 (May 23, 2019).

    ↩︎
  3. Means, Toyama, Murphy, Bakie, and Jones, Evaluation of Evidence-Based Practices in Online Learning.

    ↩︎