If TV producers ever made a version of America’s Funniest Home Videos about virtual reality fails, it would be crammed full of folks taking spills while wearing VR headsets, walking into walls while thinking they’re walking in open space, and knocking out bystanders with controllers.
YouTube has no shortage of videos in this department, with suitably hysterical titles like “Funny VR Fail Compilation!” and “VR Freakouts and Fails: Best of the best for VR REACTIONS AND LAUGHS.”
Andreea-Anamaria Muresan, a Ph.D. fellow in Human-Centered Computing at the University of Copenhagen, is documenting these pratfalls. And in the process, she is hoping to make the future of virtual reality better and, by extension, presumably less LOL-worthy.
“To understand what types of breakdowns happen when people use VR in their homes, we analyzed 233 VR fails videos on YouTube,” Muresan told Digital Trends. “We simply looked at what the community defined as a VR fail and we did not apply that label ourselves.”
Muresan and her colleagues — including Emily Dao from Monash University in Melbourne, Australia, Jarrod Knibbe from the University of Melbourne, and Kasper Hornbæk from the University of Copenhagen — created a framework to categorize VR fails and explain why they take place. By applying their insights, they believe that it will be possible for VR developers to build better, more immersive virtual environments with added awareness of all the ways they could possibly go wrong.
The challenges raised by VR disasters
“During our analysis, we assumed two perspectives over the videos,” Muresan said. “First, we saw these breakdowns as a disruption of the VR experience and something to be avoided. VR fails videos in which people damaged their equipment, collided with objects or accidentally removed their headsets are just a few examples of these breakdowns. Secondly, we looked at the positive aspects of VR fails, which often brought the social aspect of VR at the forefront of the experience. People gathered to see their friends play, shared their enjoyment, and even caused some of these ‘fails’ by playfully engaging with the VR users.”
One of the biggest sources of VR fails the team discovered were instances in which VR users collide with walls, furniture, or spectators. These were typically motivated by fear reactions: Scared users, such as those reacting to an object rushing toward them, responding with an exaggerated, uncontrolled movement.
One of the inherent problems with virtual reality is that, ultimately, you might perceive yourself as being in a virtual world, but you are, in fact, in a real one.
Of course, cataloging funny VR fails is one thing — Amusing work if you can get it — but the goal of this work isn’t to make it easier to search for amusing virtual reality errors; it’s to help developers improve VR experiences.
“We see these fails as opportunities to design novel interactions which either serve to prevent breakdowns or leverage the ways people engage with this technology,” Muresan said.
This is easier said than done. One of the inherent problems with virtual reality is that, ultimately, you might perceive yourself as being in a virtual world, but you are, in fact, in a real one. You may don a headset and transport yourself to the magnificent desolation of the moon for a bit of escape. However, you’re still in the same cramped studio apartment that can be crossed in five-and-a-half steps. Want to be able to walk further in virtual reality without colliding with an obstacle? Rent a bigger apartment.
That’s not the only possible solution, of course. Researchers have investigated other ways to simulate infinite walking in VR. One intriguing solution developed by researchers in Japan tricks the brain into thinking it’s walking in a straight line when it’s actually walking in giant circles. It’s a smart approach that could create a limitless space in which a person could walk forever without hitting a wall, but nonetheless requires a “play space” of 16 by 22 feet in order to pull off the illusion. An alternative, created by researchers from New York’s Stony Brook University, Nvidia, and Adobe, works by exploiting saccades, the quick eye movements that occur when people are looking at different points in their field of vision, such as when scanning a room.
Muresan’s proposal is a bit different: Rather than “hack” the user, why not change the VR experience to accommodate potential obstacles?
What can developers do about it
“In our design implications section, we give a few very specific and pragmatic examples of how to prevent breakdowns and how to design for spectator engagement,” Muresan said. “For example, we suggest allowing users to make more complicated boundary spaces to prevent collisions, such as considering overhead objects. Another approach here would be to dynamically change game elements. In this case, the VR application detects when players are close to going outside the boundary and steers them away.”
Depending on the space available, the VR title could, for instance, swap out swords (which require big swinging arm movements) for shields. This would change a player’s behavior without having to compromise the fidelity of the virtual world.
Whether this can be feasibly implemented remains to be seen. Muresan pointed out that certain headsets, like the Oculus Quest, are already kitted out with cameras and use inside-out tracking, which could potentially be repurposed for depth perception or object detection. Alternatively, it might be possible to employ dedicated cameras or lidar sensors for this functionality.
However, as she noted, this alone may not be enough. “I think more research is needed to understand [the] types of behaviors that lead to collisions and how to prevent them effectively. Because of this, some of our suggestions for design implications focused on prevention.”
To their credit, Muresan and team aren’t shying away from the challenge. They next aim to implement some of their design proposals in prototype games. “I believe our work gives developers insight into how people use VR at home,” said Muresan. “Most research looks at VR in highly controlled lab environments, whereas people’s homes are more dynamic, full of furniture, and, sometimes, children and pets running around. We are giving designers an idea of how to fit this technology into people’s busy home lives and how to share it with their loved ones.”
A paper describing the work was recently presented at the 2021 CHI Conference on Human Factors in Computing Systems.
