Photo of woman testing virtual reality headset.

Photo by betto rodrigues, Shutterstock: A woman tries a virtual reality headset during a recent virtual reality exposition at the Los Angeles Convention Center.

Stanford News - June 22nd, 2016 - by Bjorn Carey

Last week, the Federal Communications Commission won a historic case and established that high-speed internet service should be defined as a utility, and as such access should be available to all Americans on the same basic terms. The next day, the chairman of the FCC visited Stanford’s Virtual Human Interaction Lab (VHIL) to learn more about how online virtual reality experiences could affect future infrastructure and policy.

“Virtual reality shouldn’t have gatekeepers,” said Tom Wheeler, chairman of the FCC. “It starts with an internet that is fast, fair and open.”

At the VHIL, Jeremy Bailenson, the Thomas More Storke Professor of Communication, explained his research showing how virtual experiences can be so immersive that, if carefully guided, users can come out from under the headset and have a more empathetic view of the world. Through building diverse virtual environments that create strong mental connections with users, Bailenson and his research colleagues have gained considerable insight on some virtual reality “best practices” and the technological demands of delivering immersive experiences.

After being wowed in simulations where he could fly like Superman, block hockey penalty shots and role-play as people of another race or gender, Wheeler got down to the brass tacks. One significant concern about building out virtual reality experiences to the masses is whether infrastructure could handle the bandwidth surge, Wheeler said. There are a few delivery options on the table, Bailenson explained. One option involves streaming 360-degree video – museum tours, video games, interactive films, huge dance parties, etc. – over the internet that users could virtually explore.

Such an effort, however, would require significant bandwidth. For that reason, Bailenson favors a scenario where users download 3-D content of an environment to a local computer, and the only data sent over the internet is the tracking information of the user’s body actions – where you look, what your hands touch, where your feet walk. Sending small packets of tracking data would lessen the burden on the network and also allow for faster, smoother response times, improving the immersive feeling of the experience.

“You would have big bursts of data transfer as people download the models, which can be photo-realistic, but then you would be transmitting only the tracking data, which is super-efficient,” said Bailenson, who is also a senior fellow at the Stanford Woods Institute for the Environment. “I think the 3-D model platform will win out.”

Another area of concern for Wheeler was privacy, from both user-to-user and user-to-business perspectives. Virtual reality can provide the illusion of privacy, Bailenson said, but sophisticated users – or computers – can actually infer a lot of personal details from data collected in virtual spaces.

Virtual reality systems track a person’s movements down to fractions of a millimeter, and these actions – both the way you move as well as the way you react to stimuli – create a sort of virtual footprint. Bailenson’s work has shown that this footprint can be traced back to individual user profiles with decent reliability. Furthermore, researchers can predict attitudes and future behavior by watching these data sets. The implementation of machine learning algorithms could reveal even more information from the body tracking data.

“Virtual reality technology is becoming incredibly immersive, to the point where we’ve shown that your brain processes it in much the same way it does real-life experiences,” Bailenson said. “We absolutely need to consider how this medium will affect people.”

Originally published at Stanford News