[As a commenter notes, it’s surprising this hasn’t been investigated sooner; the story is from Ars Technica and more information, including the article abstract, is available from Purdue University’s coverage. –Matthew]
Virtual noses keep real-world VR sickness at bay
Simulation sickness solution may have been sitting right in front of our faces.
by Kyle Orland – Mar 25, 2015
As the new wave of virtual reality headsets barrel ever closer to consumer reality, the effects of “simulator sickness” on a significant portion of the population remain a concern. A group of researchers at Purdue University say they’ve found an easy way to mitigate this effect by adding one bit of reality that most VR simulations leave out: a virtual nose sitting persistently at the corners of your vision.
Offering a fixed object that doesn’t shift as you move around a virtual world has been shown to help anchor many VR users, reducing the apparent difference between visual and sensorimotor stimuli that can lead to simulation sickness. That’s useful for VR experiences that can insert a virtual cockpit or vehicular frame around the user. A virtual nose, though, has the potential to be much more generalizable to any VR experience that takes place from a first-person perspective.
The Purdue study divided 43 undergraduate volunteers into two groups. The first group went through two unmodified virtual reality demos on Oculus Rift development kits, while the other went through the same demos with the virtual nose placed where a real nose would appear in front of the environmental view. The “nasum virtualis” group lasted an average of 94.2 seconds longer in a simulated walk-around on a Tuscan villa before feeling sick and lasted 2.2 seconds longer on average in a roller coaster simulation.
In addition to the objective “time-to-sickness” measure, researchers also asked subjects to subjectively measure their levels of physical distress after the demos. The group with the virtual nose reported 13 to 14 percent less discomfort on this survey measure, depending on which simulation was run.
“It raises a lot of questions”
Purdue computer graphics Professor David Whittinghill, who led the research, tells Ars that the size of the effect, and the initial sickness reported, varied widely across the subject pool. There were some users that showed no signs of nausea on both demos, and two subjects had to “tap out” almost immediately due to sickness. Overall, though, despite the small sample of overwhelmingly male undergrads (only three volunteers were women), Whittinghill said the results were “not bad, statistically speaking. More is always better, but… it’s a really large effect.”
Surprisingly, Whittinghill said, none of the 20 subjects that went through the demos with the nose actually noticed that it was there until they were told by researchers after the fact. Whittinghill theorized this might be due to the “change blindness phenomenon” that leads our perceptual systems to filter out visual information that doesn’t change over time. Since the virtual nose is projected so close to the face, and in the corner of the peripheral vision, the visual signal of the nose barely changes even as your eye naturally wobbles and darts around the screen. Yet the reduced nausea results suggest there’s some value to having the nose there, even if it’s not being consciously perceived.
Whittinghill said his presentation of preliminary results at the Game Developers Conference earlier this month attracted the attention of Oculus CTO John Carmack, who shared a brief chat with Whittinghill after his own VR talk. “He was intrigued… it hadn’t occurred to him and he thought he might give it a shot,” Whittinghill recalled.
After pushing forward for rapid publication of the initial study, in the longer term Whittinghill said he’d like to expand the research to see if the type of nose being simulated affects the results. That means adding support for noses of different shapes and hues, or even photographing subjects to recreate their own noses in a virtual environment.
“I wish we could say we painstakingly thought this out, but [we just] modeled a nose that kind of looked like our own,” Whittinghill said of the research so far. “We’re just four white dudes, we brought that ethnic bias to the table.”
Whittinghill said he’s also interested in seeing if the presence of other fixed objects would have similar effects, or if a virtual nose is particularly effective at reducing simulator sickness. “Is it the presence of anything in your visual system or does it have to be a nose? Does it have to be like your own nose?… There are a lot of fine details, and it raises a lot of questions. I think we’re going to have our hands full.”