Virtual reality has already proven itself to be a versatile technology. Alongside its best known and most obvious application, gaming, it's also been used in real estate, to help people with bereavement, to let you go on virtual vacations and, because duh, for porn.
The whole point of VR, of course, is to immerse yourself in a detailed alternative universe, whether that's something as banal as a virtual office or outlandish as a game set in space. But what about the people who populate these worlds? What about the way you yourself are depicted? A number of studies have sought to understand exactly what impact these avatars have, both on our psychological experience of VR and on our brains.
Our facial recognition abilities are pretty impressive. Our processing of faces differs from our processing of other, more arbitrary objects, something that's been backed up by EEG, MEG and fMRI brain imaging studies. It's a quick process: we recognise faces in the visual cortex at about 200 milliseconds from their presentation.
"We need to think about the fact that the external appearance of avatars can actually change the brain mechanisms involved and, ultimately, affect human behaviour."
And in VR, we experience a similar process. What differs here is what happens when an avatar looks exactly like us: during an EEG, we experience a different voltage modulation than we do when looking at an arbitrary avatar.
In other words, avatars that look like us are internalised and identified in a different way—which could have a huge impact on how VR is designed, and how we respond to it.
The author of one 2016 study told me that research in this area could have "many implications" for the future of virtual reality.
"Big companies are investing in this technology—Facebook, Google, HTC, Amazon," said Mar Gonzalez Franco, who researches avatars at Microsoft. "But now that this is reaching the market, we need to think about the fact that the external appearance of avatars can actually change the brain mechanisms involved and, ultimately, affect human behaviour."
The implications of this are manifold. In one study, participants were found to self-administer therapy far better when one of the avatars resembled Sigmund Freud, for example.
"People were embodied in a virtual body that resembled themselves, based on 3D scanning," Mel Slater, professor of Virtual Environments at the University of Barcelona told me. "They then explained a personal problem to a virtual character representing Sigmund Freud."
"They were then subsequently embodied as Freud, and saw and heard their own virtual body explaining the problem, then as Freud gave an answer. Then they were back in their own body, and heard their own answer delivered from the Freud body. And so it went on where they had a conversation with themselves."
In control condition of the study, participants were embodied in another copy of themselves, rather than being embodied as Freud.
"We found that people reached better emotional states when they acted as Freud as their own self-counsellor, rather than when it was another copy of themselves."
How we feel about a person has an impact here, too – Gonzalez Franco says that "therapy might be more effective in the case of Freud than with another avatar" because of our preconceived ideas about who Freud was and what he represents as a cultural figure.
"As technology advances, we develop our avatars to have deeper non-verbal communication."
Similarly, a 2014 study found that participants with a fear of public speaking found that fear lessened when their virtual body did not have their own face.
It's not just therapy that this technology could help improve—it could also be a boost for more traditional VR entertainment experiences.
"Until now, the entertainment industry has provided mostly canned experiences," said Gonzalez Franco. "A one-movie-suits-all-the-spectators sort of paradigm. But in the future, experiences will be personalized to a much higher degree, not only by introducing ramifications to plotlines, but also by changing how the characters look."
The use of personalized avatars could, he said, change the way that people emotionally respond to films. And it could even "democratize" entertainment—Gonzalez said that the higher degree of personalization this could inspired could "help reach currently underrepresented populations". People of colour or disabled people, for example, would see movies from the perspective of a unique – and representative – avatar.
VR companies are also increasingly aware of how avatars impact experiences. Eric Romo, CEO of AltspaceVR, tells me that the company "designs our avatars to provide the most natural communication possible".
And how avatars look is "only part of the puzzle" Romo says.
"Much of our communication when we are together is non-verbal. Eye contact, gestures, and body language, for example, help us connect, express ourselves and add meaning. As technology advances, we develop our avatars to have deeper non-verbal communication. We strive to deliver the essence of interaction while minimizing incorrect or misinterpreted behavior."
Altspace's design philosophy, Romo says, is "don't show what you don't know", a theory they've based on the work of Stanford's Virtual Human Interaction Lab. What this means in practice is that their VR avatars only show behaviour "we can measure directly".
"For example, we don't have a direct way to measure expressions, so we don't try to infer a smile or a frown based on another behavior," Romo says. "We show hands if the user has Leap Motion or Microsoft Kinect, but don't show arms unless we have a way to measure arm location and movement, such as with a Perception Neuron motion capture system."
Slater and Gonzalez also warn that designers should consider the impact of avatars before they design VR experiences.
"In the same way that a movie can have very strong and long lasting effects on the way you think and act, so does virtual reality," said Gonzalez. "Plus, the intensity of the experience will be higher."
As for the future, researchers and tech companies alike are looking to develop the capabilities of VR to improve the experience. AltspaceVR are developing technology that can "sense body movement, eye tracking and facial expression", and Slater notes that haptics will becoming increasingly important for a more cohesive and immersive experience.
"More sensing and data will inform better avatar design for more natural communication," Romo tells me.
"The more ways we can track behaviors related to communication, the better our avatars can be as a medium for natural expression and increased human connection in VR."