Anyone who has ever had a bad hair day, when looking like a latter-day Medusa makes you feel cranky and antisocial and plodding, can sympathize with the Oakland Raiders—and not because the players get helmet hair. The Raiders alternated between mostly black and mostly white uniforms, depending on whether they were playing at home or away. Knowing that appearance affects people's mood and outlook, psychologists wondered whether uniform color influenced the Raiders' aggressiveness. Using data from the 1970s and 1980s, they found that the team racked up way more penalty yards—a measure of aggression—when they wore black than when they wore white, for infractions both minor (encroachment) and major (roughing the kicker). The pattern held even when the scientists took into account different conditions and styles of play at home and away. But while the 1988 finding has become a classic in psychology, the explanation remains controversial. Do referees, because of black's cultural baggage, see black-clad players as meaner and badder than those in, say, baby blue? Or does wearing black make players see themselves as tougher and meaner—and therefore cause them to play that way?
Jeremy Bailenson and Nick Yee of Stanford University had this and other classic studies in mind when they started wondering about the effect of being able to alter one's appearance. They weren't going to study wardrobe choices, however. Their quarry is avatars, digital representations of players in such games as Second Life. "Your physical appearance changes how people treat you," says Bailenson. "But independent of that, when you perceive yourself in a certain way, you act differently." He and Yee call it "the Proteus effect," after the shape-changing Greek god. The effect of appearance on behavior, they find, carries over from the virtual world to the real one, with intriguing consequences.
In one Stanford study, volunteers were assigned avatars who ranged from attractive to plain. It is one of life's inequities that the world sees attractive people as possessing a long list of desirable traits, including honesty, generosity and kindness. Perhaps as a result, people judged attractive are more self-confident than ugly ducklings, and so tend to be extroverted. Using a virtual-reality headset, the volunteers—actually, their avatars—walked across a room to interact with another avatar. Those with attractive avatars got within three feet of the stranger; those with homely ones kept almost six feet away. How much "personal space" one needs is inversely proportional to self-confidence, which having an attractive avatar increases. When the stranger asked the players to "tell me a little about yourself," good-looking avatars revealed more: feeling attractive increases self-esteem and therefore friendliness.
The Proteus effect spilled into the real world. After their virtual-reality session, players were shown photos from an online dating site and asked to pick those who "would be interested in you." Players who had been assigned attractive avatars picked more-attractive candidates than did players (of equal pulchritude in real life) who had been represented as homely avatars. Male players were also asked to enter personal information for an online dating site. In this situation men routinely inflate their height by an average of one inch. But those who had had an attractive avatar told the truth.
Western society sees taller people as more competent and having greater leadership potential than shorter people. The effect of virtual height, too, bleeds into real life, the Stanford scientists find. After their avatar roamed through a virtual world, players took seats in the real world to play a split-the-loot game. Player One proposed ways to divide $100 which Player Two could accept or reject, in which case neither player got anything. People tend to reject unfair offers, even though accepting a lopsided $99/$1 split leaves them objectively better off than walking away in a righteous huff. Players fresh from being a tall avatar showed their cockiness, proposing, on average, $61/$39 splits. People with short avatars averaged offers of $52/$48, Yee and Bailenson reported last year. When it was their turn to weigh an insulting $75/$25 offer, players with tall avatars rejected it 62 percent of the time; those with short avatars stood up for themselves a wimpish 28 percent of the time.
The goal of the research isn't to tell Second Lifers they'll have more virtual friends, money, power and other goodies if they create a hot, young, powerful avatar. Most do so anyway. Instead, avatars might serve therapeutic purposes, helping those with social phobia, say, become more confident and friendly in real life. The work also underlines the power of new media to affect our behavior: players who roamed a virtual world as a KKK-clad avatar felt more aggressive than they did before playing the game, while those whose avatar wore a doctor's coat scored higher on a test of friendliness. It's not clear how long the spillover to the real world lasts. But even if it's only a few hours the potential is impressive: online players spend, on average, 20 hours a week as their avatar.