The Proteus Paradox (20 page)

At the lab, we wondered if giving someone a taller avatar would make them more confident. We again put participants into a virtual environment modeled on the physical lab and had them interact with a stranger. But instead of using an introductory prompt, we had participants in the study play a bargaining game with the confederate. We asked the participant and the confederate to share a hypothetical pool of $100. One would offer a split, and the other would either accept or reject the split. If the both parties accepted the split, they would share the money accordingly. If one rejected the split, neither would receive any money. There were four rounds of this game. The participant would make the split in the first round, and we
scripted the confederate to accept this split every time. In the second round, the confederate would make an even 50–50 split. In the third round, the participant would make the split again. But in the final round, the confederate would make an unfair split of 75–25 in his or her own favor. This negotiation exercise would reveal two things: How aggressively does the participant bargain? And what does he or she do when an unfair offer is made?

The participants in the study were given avatars that were either the same height, ten centimeters (four inches) taller, or ten centimeters shorter than the confederate's avatar. All participants offered near-even splits in the first round, so there was no difference between the height conditions. But during the third round, people given taller avatars made an offer (in their own favor) that was on average almost $10 more than people given shorter avatars. And when the confederate made the unfair offer in the final round, people given taller avatars were half as likely to accept it as people given shorter avatars. As with the attractiveness study, we programmed the virtual world such that the confederate was blind to the participant's virtual height (with the help of a little trigonometry). And when asked to guess the goals of the study, again no participant mentioned that avatar height had been manipulated.
¹⁴

Together, these two studies show that subtle manipulations in avatar appearance have dramatic effects on how people interact with each other in virtual worlds. And these effects occur rapidly, fewer than sixty seconds after being in a new digital body. We think of avatars as things of our own creation, digital clay that can be endlessly shaped. But avatar creation is actually a two-way process. The fact is that our avatars change us in turn. As discussed in the Introduction, Bailenson and I labeled this phenomenon the Proteus Effect after the Greek sea god who was capable of assuming many forms.

These two studies also led us to think about online gamers who spend on average twenty hours each week in virtual worlds. The lab studies showed how avatars change behavior within the virtual world, but what happens when people step back into the physical world? Does the Proteus Effect immediately disappear once the user logs out, or does it transfer to subsequent interactions outside the virtual world? We decided to rerun the attractiveness study with a twist. After coming out of the immersive virtual environment, the students participated in a separate study on online dating among college students. This study took place in front of a standard desktop computer, and the students created profiles on a mock online dating website geared toward college students. After creating their profile and answering some basic personality questions, they saw a grid of nine photographs of college students (from a different university) and selected the two people from the grid they were most interested in meeting. We found that study participants who embodied an attractive avatar in the virtual environment subsequently selected more attractive people from the grid to meet. Even when the students were no longer in a virtual world and interacting using a different technology altogether, their avatar's appearance still mattered.
¹⁵

In 2008, the McKinsey Global Institute estimated that two-thirds of baby boomers born between 1945 and 1955 have not saved enough money to maintain their lifestyle in retirement. The 2012 Retirement Confidence Survey revealed that more than half of current workers have not tried to calculate how much money they will need to save for retirement. It is not easy to defer immediate gratification for an incredibly intangible future thirty or forty years down the road. The
desires and payoffs of the present moment are always concrete and known—a bigger TV, Starbucks lattes every morning, a fancier car. The desires of your future, retired self are a vague abstraction at best, especially for young college grads starting their first job. This is what makes it so difficult to save money for retirement. It requires the discipline to defer an immediate, known, concrete reward for an undefined and unknown benefit decades in the future.
¹⁶

Hal Hershfield, a professor at the New York University Stern School of Business, used avatars to bring the future forward. Instead of putting participants in a different person's body, Hershfield put undergraduate students in avatars modeled from photographs of their faces. In half of these self avatars, Hershfield left the faces unaltered. In the other half, he digitally aged the faces to realistically add white hair, wrinkles, and sagging skin. As in the attractiveness study, the students saw themselves in a virtual mirror. The students then had to imagine that they had just received a windfall of $1,000. Their task was to allocate that sum among four options: buying a gift for someone special, investing in a retirement fund, planning an extravagant occasion, and putting the money into a checking account. The students who were given avatars of their seventy-year-old selves put twice the amount of money into their retirement accounts compared with students in avatars of their current selves. By bringing the future forward, Hershfield took an intangible, older self and made it visible and salient to college students.
¹⁷

Instead of using virtual worlds to put people in fantasy bodies, Hershfield's study highlighted the power of using virtual worlds to hijack someone's body and face. This painless virtual plastic surgery allows us to change how we think about ourselves and our future. Jesse Fox, a professor of communication at the Ohio State University, took this idea in a different direction. Virtual worlds allow us to
change not only how an avatar looks but also how an avatar behaves. In a virtual world, we can replicate your body and then digitally animate it to do anything we want, including things we've never done in the physical world. But why would we want to make digital doppelgängers?

One important way humans learn is by watching other people. Not only do we learn new behaviors by watching others, but we also learn the expected outcomes of a behavior. After all, it would significantly diminish our species' survival if we each individually had to poke our fingers into an electric socket to learn that it is a bad idea. Thus, when we see Peter get praised for playing the piano well, we learn that playing the piano is one way of being rewarded. This in turn makes it more likely we'll go learn to play the piano. But watching someone else is not sufficient. It's important that we are able to identify with this other person. The more similar the person is to us, the more certain it is that our outcomes will match. Thus, if Peter turns out to be a young piano prodigy who has been composing symphonies since he was twelve, most of us probably wouldn't be rushing to learn the piano. But if Peter is roughly our own age, with a similar social background, and is playing quite well after learning the piano for only one month, we may be intrigued enough to take piano lessons. Of course, it is difficult to create the perfect model for a specific person. At best, we may provide an inspiring figure that is of the same gender and roughly the same age.

In contrast, in a virtual world, we can create the perfect model of any person by creating an avatar of him or her. In Fox's study, she brought students into the immersive virtual environment and they watched their digital doppelgängers either run on a treadmill or stand idly in the virtual lab room. In other words, Fox created a digital version of the students that they had no control over. She
found that students who had watched their doppelgängers run on the treadmill were more likely to exercise over the next twenty-four hours. Your digital doppelgänger in a virtual world can change how you behave in the real world.
¹⁸

In the chapter on superstitions, we saw that people are polite to computers because our brains do not differentiate between digital media and human companions. We treat computers as people because technology outpaces human evolution. And superstitions are triggered among online gamers the same way that superstitions are triggered in Skinner's pigeons. This is the same logic that underlies all the studies in virtual environments presented in this chapter. Our brains are harried, overworked processors, besieged by constant information deluges. The automated rules that our brains use to cope with making decisions in everyday life don't care about the media we're using. Our brains make sense of virtual people using the same rules they use for making sense of physical people. When a person looks similar to us, whether we're face-to-face or in a virtual world, we automatically like him or her more.

The difference, of course, is that virtual worlds are infinitely malleable in a way that the physical world is not. There's very little I can do to make myself look more like someone else in the real world, but in the virtual world, this is trivial. Would you like that order of face stealing at 20 percent or 30 percent? This malleability is what allows our cognitive heuristics to be mercilessly hijacked in virtual worlds. Whether it is a manipulation in face stealing or avatar appearance, these undetected changes cause dramatic shifts in behavior both inside
and outside of the virtual world. Even the object of our own creation in virtual worlds, our own avatars, can become a tool for behavioral modification. These fantasy worlds of escape can ironically make us susceptible to powerful psychological tools that silently modify our behaviors and attitudes.

In this chapter, I've documented a variety of tools in virtual worlds that can modify our thoughts and behavior, but that doesn't mean these tools can operate only in virtual environments. Of course, these manipulations are easier to deploy in virtual worlds, but a bank or an investment company could generate a customized postcard advertisement with a digitally aged photograph of you if they had access to a reasonably clear photo of you. And these days, especially with the use of social networking sites and Google, it's quite straightforward to find an image of someone. This means that face morphing and doppelgänger techniques could be used in print ads sent to you in the mail or on a banner ad that appears when you're browsing on a website. As Bailenson and I conducted these studies and identified different techniques, we wondered whether people could learn to guard themselves against these persuasion methods—a sort of psychological inoculation. I'm pessimistic for two reasons. First, in many of our studies, participants were not able to detect the subtle manipulations. It is difficult to guard against something that isn't easily detected. And second, most people do not recognize or guard against the plethora of psychological tricks already in use in contemporary advertisements. In the chapter on gender stereotypes, I mentioned the concept of guilt-soothing indulgence in some products targeted at women. Perhaps the best example is the very successful marketing ploy that transformed a cheap and common gemstone into a costly engagement artifact: the diamond ring. The more others
know about you, the more tailored and targeted their manipulations can be. They could generate a doppelgänger that not only looked like you but also exhibited similar personality traits. In fact, virtual worlds can be used to infer someone's personality. In the next chapter, I describe how this can be done.

Whether it's guilds or raids, online gamers constantly face sudden and stressful decisions. And the decisions they make would seem to be very character revealing. Gaining access to such in-game information, however, is another story. When
World of Warcraft
launched in 2004, Blizzard, the game's developer, allowed players and third-party developers to modify interface elements via a scripting language. Of course, players could not make their characters indestructible or create a sword that killed anything it touched, but these addons modified the game interface in interesting and important ways. Some modifications added new buttons and charts on players' screens; others worked silently in the background. There were healing helpers that automatically targeted and cured players who had been poisoned. There were combat add-ons that tracked the average damage output of each team member. There were quest helpers that plotted the most efficient route to complete your current quests. And there were auction helpers that identified and snagged good deals for you in the auction house. None of these add-ons changed the rules or mechanisms of game play, but they provided visibility into the large number of variables being tracked by the game. In order to
work, these add-ons had to access Blizzard's server-side data. And this is how the balance of data accessibility began to shift: for the first time, a game company allowed public access to server-side game variables.

What data researchers could access, however, was severely limited. For example, combat helpers could track data only from players in their group. Add-ons could access data only if the character was logged into the game. And add-ons could never access data outside the server the character belonged to. So in theory, everyone's combat data were accessible, but in practice, an add-on had access to only a tiny slice of those data.

In 2005, three research scientists at the Palo Alto Research Center—Nicolas Ducheneaut, Bob Moore, and Eric Nickell—had created a research group studying social interaction and communities in online games called PlayOn. They had created a census add-on tool for
World of Warcraft
. The tool used the “/who” command in the game to return a list of characters who fit certain search parameters. This in-game command let players search for active characters on their server by using a name, name fragment, character class, or character level. Among other things, the command allowed players to search for potential teammates nearby. Instead of searching for a specific character, the PARC census tool returned a list of all active characters on the server. The parameters you could use in searches were also data that the command returned. Thus, the census tool gathered seven variables from every character currently logged on: name, level, class, race, geographical zone, name of guild (if any), and whether the character was currently grouped with other players. The PARC team gathered data from five separate servers at fifteen-minute intervals, twenty-four hours a day, seven days a week. Altogether, the
census tool recorded data related to over two hundred thousand unique characters.