Playing videogame or virtual reality avatars whose race is different from yours could reduce your racial biases, according to a new study. Now, scientists are using this insight to explore how technology could help build empathy and reduce tension between different groups of people.

In recent years, numerous studies have shown just how easy it is to trick your brain into taking ownership of a physical or virtual body not your own. For example, a couple years ago researchers found they could make people believe they had a third arm by doing little more than placing a rubber, human-looking arm right next to their real arms. And in 2008, scientists demonstrated the body-swapping illusion, where study participants felt as though they were in the body of a mannequin or another person.


These experiments and others left Mel Slater wondering if there are any possible consequences to the body ownership illusion. "Does the type of body you think you inhabit determine the way you behave and the way you think?" says Slater, a professor of virtual environments at the University of Barcelona in Spain.


In particular, Slater is interested in "body semantics" — the idea that a person's body type has intrinsic meaning, which may come from personal experiences or from society. The major result of body semantics is that you make automatic associations about people based on their bodies. So if you see people who appear old, you may automatically think they're physically weak or cognitively impaired, he says. But can these ideas and associations be altered if you virtually put yourself into someone else's shoes?

To find out, Slater and his colleagues decided to focus on race. Specifically, they wanted to look at implicit racial biases — those racial biases you may not even be aware you have. Unlike outright (explicit) prejudices, implicit biases may manifest in unconscious behaviors, such as when choosing whom to sit next to on a bus or train, Slater says.

The researchers began by recruiting 60 light-skinned Spanish women from the University of Barcelona, and giving them a racial implicit association test. The test measures your implicit racial biases by forcing you to make rapid associations between concepts and representations of white and black people. In the test, if you pair up white faces with positive words and black faces with negative words faster than you pair up white faces with negative words and black faces with positive words, that would indicate you have an implicit bias, Slater says.


Three days following the test, the team divided the participants into four groups. In a virtual world, one group embodied a white avatar, a second group embodied a black avatar and the third group embodied a purple "alien" avatar. A collection of devices tracked the participants' real-body movements, which were then translated into the virtual world. And the participants could see themselves — as their avatar — in a virtual mirror and when they looked down. The fourth group also saw dark-skinned figures in the virtual world, but they didn't "embody" the characters, as the mirror bodies moved independently of their own real-world movements.

The researchers allowed the women to look around their virtual environment and explore their virtual bodies for 5 minutes. After that, six black virtual women and six white virtual women walked by the participants closely, one by one, invading their personal space. "We didn't want to give them any task to do because it could affect the outcome of the experiment," Slater says. "But we wanted to make them feel some kind of emotional arousal."

Finally, the participants retook the implicit bias test, as well as an exit questionnaire, which assessed several variables of the study, including the extent of virtual body ownership and the nervousness resulting from the virtual passersby.


The scientists found that the virtual reality experiment reduced the participants' implicit racial biases, but only if they embodied a dark-skinned avatar. Interestingly, how nervous they felt in response to the virtual people affected how strong the effect was — people who were more unnerved showed a greater reduction in implicit biases. "Their nervousness was a sign of how realistic they were taking things to be," Slater says, suggesting that these women felt more "present" in the virtual world.

"This study looks very good and it is promising in terms of the reduction of racial bias," says psychologist Jim Blascovich, who's the director of the Research Center for Virtual Environments and Behavior at the University of California Santa Barbara. Blascovich, who wasn't involved in the research, notes that other labs need to replicate the study's findings with other, more representative population samples. But that shouldn't be too difficult. "One nice thing about the methodology is that they can essentially send their digital, virtual reality world to other investigators, who could employ it and have other subjects in other labs go through exactly the same experience."


For Slater, the next step is to see just how long the effect lasts, since the experiment only measured the women's implicit racial biases directly after the virtual reality experience. "But we are also interested in exploring the limits of how this body semantics works," he says. "We don't have an explanation for why it works and we don't know how far we can push it."

In another recent study by Slater's group, the researchers showed that body semantics might be able to influence behavior, at least while in the virtual world. In that experiment, one group of white people embodied a dark-skinned, casually dressed avatar "looking like Jimi Hendricks," while the other group's light-skinned avatar wore a formal suit, Slater says. They had the participants play drums in the real-world (which translated into the virtual world) and measured how much they moved their body. Participants in the casual avatar were much more expressive in their drumming.


Harry Farmer, a psychologist with the Lab of Action and Body at Royal Halloway, University of London, says that the setting in which virtual embodiment takes place is very important. In a 2009 study, scientists had white people embody black avatars and then carry out job interviews — their racial biases actually increased, perhaps because the situation elicited negative stereotypes about black people. "This opens up the possibility that if one was embodying a black avatar in a first person shooter or GTA [Grand Theft Auto] style game, this could increase negative associations towards black people by enhancing the stereotype of black people as violent," he told io9 in an email.

Farmer notes that Slater's current work compliments other recent research that looked at how embodiment can positively change attitudes. In a couple of recent studies, one of which Farmer conducted, scientists showed that implicit racial biases decreased under the rubber hand illusion, where participants felt that a rubber hand was part of their own body (biases only changed when white people felt ownership over black rubber hands). But a major caveat to the work — including Slater's — is that the studies only looked at short-term changes, he says. Moreover, they measured unconscious biases, so it's not clear if embodiment can change real-life behaviors towards other races.


With that said, Farmer adds:

The virtual reality technique used in this study does, however, offer a chance to test such questions in the lab by allowing a controlled interaction between participants and other avatars with different skin colours, and the chance to collect a wide variety of data about how participants interacted with people.

Slater says that immersive virtual reality may become a very powerful tool if his work and others continue to show that the type of body you inhabit can (positively) influence ingrained attitudes and behavior. For example, if people are hostile towards one another, you could virtually put them into one another's shoes. "It may be possible that by giving people the experience of being on the other side, it will turn down their hostility," he says. "Virtual reality could have practical applications in empathy building."


On the other hand, it won't be possible to completely control the uses of virtual reality in the near future. If virtual reality becomes commercial, people will begin to take control of other types of bodies under scenarios that may change their attitudes or behaviors in unexpected ways. "As virtual reality technology improves and the cost of it decreases, it is inevitable that 'VR arcades,' where people are fully immersed in a game world and embodied in their avatar, will emerge," Farmer says. "This development has the potential to alter, for better or worse, how people view both themselves and member of other groups."

Slater and his colleagues detailed their new virtual reality study in the journal Consciousness and Cognition.


Images and video via Mel Slater.