Case Study: VR in the Classroom

​1. The Student Experience


In 2018, 3.78 million students were enrolled in the 8th grade of U.S. public schools. Students in this age group experience introductions to both the life sciences and physical sciences. As an example, let’s consider a hypothetical typical physical science classroom. In such classroom, a 13-year-old’s (let’s call her Katie) first experience of a lecture introducing chemistry is likely to include a fabulous demonstration of elephant toothpaste, followed by a dry explanation of how solids are not the same as liquids, which are also not the same as gases.



While explosions and foam are flashy and exciting, they aren’t personal; that is, they don’t draw a clear link between the student’s personal motivations and the goals of the classroom. Without that personal connection, the subject matter as a whole may seem irrelevant to the student's everyday life, often leading to the same questions many of us have asked ourselves at some point while in school:

Why am I learning this and why should I care? How is this knowledge going to be useful in my everyday life?

Though they can be fun for a moment, these types of activities are unlikely to leave a long-lasting impression on a student that's distracted by a difficult life at home, or a student who has not yet made the mental leap in understanding how topics covered in school can be relevant to their future. Unsurprisingly, STEM topics which are complex or abstract suffer significantly from this perceived lack of personal relevance, which could explain why over time students seem to lose interest in science.


To make matters worse, students now find themselves in a worldwide accelerating attention economy. The rise of smartphones and broadband internet has made it easier than ever to access quality entertainment. As a result, expectations for what qualifies as interesting content have gone up. Schools are now more than ever in a race to hold the interest of students, with teachers often reporting in our research interviews that it's becoming increasingly difficult to retain students’ attention.


We therefore think it's imperative that further research be carried out to assess alternative mechanisms for engaging students in and outside of classrooms, and how to deploy those at scale, especially if these can help ignite student interest for topics which can be seen as abstract or complex, as is the case with the majority of STEM subject matter.


2. Let's talk about games


In recent decades, videogames have evolved to become one of the most engaging mediums, with hits like Fortnite and Minecraft dominating pop culture. Still, despite their modern omnipresence, it's not uncommon for videogames to still occasionally be thought of as trivialities whose only value is entertainment.


In practice, however, trivialities are only a subset of the increasingly large domain of videogames. For modern game designers, whether the software they work on is for entertainment, education, advertising, neural network training, or any of the other disciplines that have taken on some “gamified” aspects over the years, the common definition of a “videogame” is as a digital experience featuring challenges and rewards. The common definition of a “serious game” to that group is simply “a game which does not have entertainment as its primary purpose”. This illustrates the industry view that entertainment is not the essential nature of a videogame: interactivity is. So while it is true that the videogames industry is predominantly focused on entertainment, the techniques and standards it has developed for communicating skills to users, testing them, and keeping them interested have turned out to be extremely powerful, and applicable to every kind of learning.


In 1974, the Minnesota Educational Computing Consortium bundled a game called The Oregon Trail with the Apple II microcomputer, which schools were beginning to buy at the time to support their students’ tech literacy. As a result, a whole generation of students learned the hazards of westward migration for life. Having sold 65 million copies since then, The Oregon Trail remains a cultural icon to this day because of how uniquely engaging it was for its time.



From an evidentiary standpoint, the superior effectiveness of “serious games” as learning systems is a well-known result in learning science research, with an abstract from a decade ago in the Journal of Educational Psychology suggesting that it is the assumption of the literature (confirmed in their review) that serious games result in better learning and retention than conventional instruction.


For example, in a study of Outbreak @ The Institute, Eric Rosenbaum and his team found that having students take on unique identities (such as a technician, public health expert, or doctor) heightened players’ engagement in the learning process. Edward Dieterle presented similar findings in a study of River City, noting that the high level of autonomy in videogames allows students to customize gameplay so that it addresses, at least in part, their cultural norms. Sasha Barab and Chris Dede came to similar conclusions, noting that inimitable identities allow students to disassociate from derogatory perceptions of their physical appearance and ability levels.


Several studies have also examined how video games affect students’ participation in STEM discourse. For example, in studies of the videogame Whyville, Yasmin Kafai and her team found that gameplay encouraged students’ participation in scientific arguments and led to the use of higher-level vocabulary words (e.g. contamination) than they would use in everyday conversations.


Connolly, Boyle, MacArthur, Hainey, and Boyle examined empirical literature on serious games related to learning, skill enhancement and engagement. The findings revealed that gameplay is linked to a range of perceptual, cognitive, behavioral, affective, and motivational impacts and outcomes. The most frequently occurring outcomes and impacts were knowledge acquisition, content understanding, affective, and motivational outcomes. Wouters, Van Nimwegen, Van Oostendorp, and Van der Spek had similar findings during a meta-analysis on the cognitive and motivational effects of serious games. The researchers noted that serious games had a statistically significant effect on learning. Other research by Marino, Israel, Beecher, and Basham found that students and teachers identified a clear preference for learning about STEM using video games.


When viewed collectively, this evidence supports the notion that educational video games may be an underdeveloped medium for engaging these groups in the scientific community.


3. Let's talk about Virtual Reality


In a famous 1993 episode of The Simpsons, Lisa Simpson, the show’s cartoonishly perfect student, has a literal dream, from which she rudely awakens, where her class uses Virtual Reality to visit Genghis Khan. In 2011, VR’s resurgence in the public imagination came with speculative science fiction hit Ready Player One, which envisioned a virtual planet of cathedral-like classrooms and infinite library halls. Since VR was first conjectured, its use in education has been the subject of anticipation. Today, VR is increasingly commonplace as an enterprise tool and a consumer platform, but it has yet to leap into the classroom, which has historically struggled to get modern technologies into students’ hands.


When describing modern, effective methods of teaching, the Universal Design for Learning (UDL) framework emphasizes the creation of high degrees of student agency and customizing learning to each students’ needs. However, in the current reality of budgeted public schooling, it’s nearly impossible to redesign classroom environments in response to individual students. One potentially scalable, plausible way to achieve such a thing would be to use virtual reality (VR), a new development in education technology which shows significant promise, and by definition allows each student to have a personal, customized space, designed for specific activities.


In a 2014 review of 13 studies, Merchant et. al. found a positive effect on both learning and engagement from trials of VR education software, with a particularly strong effect for VR learning games. The compelling feeling of being transported to another world and the capacity for UDL-conformant interactivity across multiple sensory domains both may explain the engagement advantages of virtual reality.





3. When's the best time to intervene?


As a guide to addressing these issues, in a 2015 survey of 24,000 students, Mangu et al. found grades 7th through 11th to be key time periods for building self-efficacy in math and science, as well as for influencing STEM as a career interest.


​1. Problem Overview


Based on a body of theoretical and empirical support for the use of VR – and more broadly, gaming – in learning, [2, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16] we hypothesized that, despite the broad enthusiasm of educators around the topic of VR, [3] the lack of widespread adoption for engaging VR technology in the classroom must be the direct result of technical, financial and social barriers that have yet to be addressed and which we could potentially address. As a result, we focused on the following two areas of concern:​


  1. What experiences are students missing, if any, that a VR platform could address to improve sustained personal interest in STEM topics, and how can we deliver on that?

  2. What limitations are preventing schools from leveraging the apparent benefits of VR to improve student engagement at a significant scale, and how do we overcome those limitations?

These two questions guided our research and development efforts.


​2. The Minimum Viable User Experience, According to Interviewees


The customer interviews helped us form a set of standards that define a usable teaching tool. Three primitive criteria were identified:

  • Is the teaching tool reasonably affordable and easy to use in the classroom?

  • Can the teaching tool accommodate the whole classroom, i.e. can every student participate?

  • How versatile is the teaching tool, i.e. can the teacher curate, adapt and pace its learning content such that it can better suit the needs of the classroom curriculum?

Up until now, VR solutions have failed to meet these criteria, despite showing significant promise in terms of engagement and learning benefits. [2, 16] As a result, schools have been unable to adopt VR technology for frequent, non-novelty usage. As a converse example, a tool like YouTube meets all of the criteria: it’s free and easy to use; the whole classroom can be involved simultaneously; and the teacher can easily curate the kinds of videos to play in the classroom, as well as upload their own. YouTube, however, does not offer the same level of student agency and hands-on interaction a VR experience would, thereby impacting student engagement.


3. VR hardware in the Classroom


​ There have been three grades of modern VR hardware: 1) Tethered VR headsets, which rely on the power of an expensive computer to allow for complex simulations (problems: complicated setup; not affordable). 2) Smartphone VR cases, which can offer a wireless experience, but require a high-end phone (problems: limited tracking; limited interaction). 3) Standalone VR, which is completely self-contained hardware, including its own screen, tracking, and processor, requiring no external computer, smartphone or complicated setup. Standalone VR allows for the most affordable, high-interaction experiences.



Historically, 6-degrees-of-freedom (6DOF) VR software has always been tethered, thereby requiring an expensive computer, access to a large area of physical space and a lengthy and cumbersome setup:



​ This experience has always been in direct contradiction with two of the primitive usability criteria previously identified as necessary for the minimum viable user experience (MVUX):

  • Is the teaching tool reasonably affordable and easy to use in the classroom?

  • Can the teaching tool accommodate the whole classroom, i.e. can every student participate?



Unlike previous 6DOF VR headsets, the Oculus Quest is a completely untethered standalone unit, requiring no external PC. This makes the device significantly more affordable and easy to deploy in a classroom. Its consumer price-point of $399 makes it comparable to the cost of a tablet, which is an investment most schools are comfortable with. 18 However, hardware accessibility alone is not sufficient to make a 6DOF VR experience functional in the classroom. Classrooms have limited space, and moving desks is disruptive. In order to be truly accommodating of traditional learning environments, software solutions were required to enable a functional experience in situations involving arbitrary numbers of VR headsets, limited physical space and diverse users. To achieve this flexibility, Tablecraft has been designed to support seated, standing, low-space and room-scale play styles, allowing players to interact with the virtual world using minimal physical movement or real-world space requirements, as seen fit. Other accessibility features include support for varying player heights, as well as some visual and hearing aid options, with more to be added over time.


In our final Phase I study, the hundreds of students in the sitting and standing groups seemed to report no significant differences in level of engagement, despite many interactive features being placed far away. Our research team, by contrast, reports that sitting students were significantly easier to manage.



​ The left half of the picture above highlights a group of students, some of them sitting, others just standing, but all of them simultaneously wearing Oculus Quest ($399) devices to use Tablecraft in limited physical space conditions. This addressed the first minimum viable user experience (MVUX) criterium:


  • Is the teaching tool reasonably affordable and easy to use in the classroom?


Coincidentally, the right half of the picture above highlights the solution to the second MVUX criterium, which is further examined in the following section of this report:


  • Can the teaching tool accommodate the whole classroom, i.e. can every student participate?


5.2. Whole-Classroom Participation


​The biggest problem with the typical usage of VR in school is that the majority of students end up being left out. In early playtests we’ve conducted before we had the ability to accomodate the whole classroom, students not in VR would always end up shouting out suggestions for things that the students in VR could be doing. This created a lot of enthusiasm in the classroom, but it also reflected a need worth addressing: every student wants to participate in the action, even if they’re not in VR themselves.


At first, we tried increasing classroom participation by developing a fully-featured PC version of Tablecraft . Our theory was that by giving schools with limited numbers of VR headsets a fully-featured Tablecraft on PC, their students would still be able to engage with the best of what the platform has to offer. However, this turned out to not be the case. The tangible environment and intuitive interactions carefully handcrafted to keep users engaged in VR are less effective when users have to operate a PC keyboard and mouse. Keyboards have a lot of buttons, and any sort of 3D software like Tablecraft can quickly become overwhelming for certain users if they have to memorize which buttons control camera movement and which ones control other features, especially if users have no prior gaming experience. Furthermore, this solution failed to address the way students and teachers have always most enjoyed Tablecraft: collaboratively. The real solution would have to operate on the expectation of a small number of headset users interacting asymmetrically in real-time with a larger number of non-VR users. As a result, we pivoted and simplified our non-VR interaction scheme to make it compatible with the intuitive interface of phones and tablets, which virtually all students have access to and know how to use, and then developed a planned peer-to-peer online framework that allowed students on those mobile devices to join the labs of their peers who are in VR.


In the new paradigm, Tablecraft can support arbitrary numbers of VR headsets and users. With Tablecraft ’s free iOS and Android app, non-VR users can search for and join the VR labs of those connected to the same wireless network (even when there’s no internet access). In these shared labs, players are able to perform tasks for each other, share creations and discoveries, exchange access to virtual tools, and take advantage of the social space to explore gameplay ideas with one another as they experiment together or individually at their own pace. Using the same app, teachers, who cannot be in VR for reasons of supervision, are also able to easily jump from lab to lab, see each lab’s statistics, communicate with the students in VR and even help students with their virtual activities.



Both mobile and VR users have their own avatars in the virtual space, which creates a true feeling of mutual presence. This allows players to wave and speak directly to their peers as they see each other in the virtual world, which is a memorable social interaction and greatly mitigates VR isolation. In classroom environments, Tablecraft’s mobile app lets everyone participate in the VR activities even if there is only one user in VR. Outside the classroom, it allows friends and family to also participate.


5.4. Content Customization


The third aspect of the minimum viable user experience that was identified in customer interviews involved teachers’ desire to apply the tool in flexible ways and at variable points in their curriculum. This need was specifically formalized as:

  • How versatile is the teaching tool, i.e. can the teacher curate, adapt and pace its learning content such that it can better suit the needs of the classroom curriculum?

For a concrete example, consider a teacher who wants a short class activity to focus on the many uses of Carbon, because they happened to mention, at the end of their lecture, that there was surprisingly little difference between a diamond and graphite. In this scenario, the teacher might want every student to have unlimited access to Carbon, allowing them to experiment with it.


This level of flexibility is actually one of the big advantages of virtual reality. Whereas customizing real-life interactive activities can have logistical barriers, a requested setting like the one above is intrinsically possible in virtual reality. During Phase I, we have accumulated a few such requests, and we

are expecting to implement them all during Phase II.


6. Research Outcomes so far


In Phase I, we partnered with a team of experts in the fields of learning sciences, STEM education, and school psychology at the University of Central Florida (UCF) to conduct a preliminary evaluation of the product’s usability, engagement potential, and STEM content. The following research questions guided the Phase I product evaluation:

  1. To what extent can students use the product with minimal or no external help?

  2. To what extent can educators use the product with minimal or no external help?

  3. To what extent can the whole classroom participate in the VR experience?

  4. To what extent are students engaged in the subject matter while/after using Tablecraft?

  5. To what extent do students and educators see a fully realized Tablecraft as a learning tool?

  6. To what extent are school wireless networks ready to support a multiplayer VR environment?

The research questions, evaluation protocol, consent and assent process, inclusion and exclusion criteria, and analysis procedure were approved through the University of Central Florida IRB.



Hello


Hello again

//  more shenanigans_