Students who used immersive virtual reality (VR) did not learn significantly better than those who used two more traditional forms of learning, but they vastly preferred the VR to computer-simulated and hands-on methods, a new Cornell study has found.
“We didn’t know exactly what we were going to see,” said Jack Madden, M.S. ’17, doctoral student in astronomy in the College of Arts and Sciences (A&S) and first author of “Ready Student One: Exploring the Predictors of Student Learning in Virtual Reality,” which published March 25 in PLOS One. “But it’s amazing that this brand-new technology performed just as well as these tried-and-true methods that are used today in classrooms. So at least we’re not harming students by using VR.”
Though the virtual reality experiment didn’t change learning outcomes overall, the researchers found that students with more video game experience learned better using VR than those with little video game experience – a finding that correlated closely with gender.
The study – which has new implications as learning around the world shifts online to combat the spread of coronavirus – aimed to take a step toward determining whether new educational technology tactics, while popular, are actually effective.
“There’s been a big push for enhanced technology in classrooms,” Madden said. “I think we can be in awe of these fancy, shiny devices and it might feel like they’re helping, but we need to know if they actually are.”
The experiment recreated a demonstration commonly used in astronomy classes to help students understand the phases of the moon. The concept of the phases of the moon is a challenging one in astronomy, since comprehension requires students to place themselves in perspectives of the sun, moon, and Earth system that are generally inaccessible, according to the paper.
In the traditional method, a participant, representing Earth, holds a short stick topped with a ball representing the moon. A nearby light mimics the sun. Participants keep the ball at arm’s length and spin in a circle to create an illumination pattern that illustrates the moon’s phases.
In the study, the researchers directed one-third of the participants to use that traditional hands-on method, and one-third to use a desktop computer simulation allowing them to manipulate their viewing positions and perspectives as well as the progression of time, which was synchronized with the three bodies’ orbits and rotation.
Another one-third of the students used the VR option, which hewed to the hands-on experiment as closely as possible while allowing participants to use a headset and controllers to move forward and backward in time, change the moon’s orbit and choose different viewing positions.
The 172 participating students were given a pre-test before being randomly assigned to one of the three experiments, and then re-tested after the simulation to see what they’d learned. After the second test, all participants tried the other two simulations. The participants were also asked for their major, gender, and experience with video games and virtual reality.
Students’ scores on the tests increased 25% after doing one of the simulations, regardless of which one they used, the study found. There was no significant difference among students with VR experience or among majors, though students in science fields slightly outperformed those in nonscience fields.
Males were far more likely to have video game experience, the survey found, and also learned more in the VR simulation, suggesting that either gender or prior video game experience could impact the success of VR-based learning. Reviewing prior work, the researchers found that video games requiring players to navigate 3D spaces are more popular among males than females.
“This is an interesting finding, because it could potentially imply that if you can provide learners with that experience, then you could show broad benefits from immersive learning,” said co-author Andrea Stevenson Won, assistant professor of communication and director of the Virtual Embodiment Lab in the College of Agriculture and Life Sciences. “However, more study is definitely needed.”
“If you’re unfamiliar with navigating this kind of 3D space, you’re not going to learn as well in it, so that could be a barrier,” Madden said. “One of the conclusions of our work is that we need to do a better job of asking questions around things that might be gendered, like video game experience. There’s a lot of finer detail you need to know to make VR learning successful.”
The researchers also found that seeing the earth from space did not change students’ environmental attitudes, asking participants whether they agreed with statements such as “When humans interfere with nature it often produces disastrous consequences” and “In order to protect the environment, we need economic growth” before and after the VR experience.
Around 78% of the students preferred the VR model to the other two, using phrases such as “easier to visualize,” “more fun” and “more immersive.”
“Having an overall space to see where everything is helps a lot,” one student wrote. “Even in class, I still had a hard time understanding what they are talking about in concept. But I think I learned a lot in VR and being able to manipulate the environment on my own accord. It seems more engaging than the other two methods.”
The study’s co-authors are Natasha Holmes, the Ann S. Bower Assistant Professor in A&S; Jonathon Schuldt, associate professor of communication; and communication doctoral students Swati Pandita and Byungdoo Kim. The research was supported by Oculus Education.