2014/3/5 「虛擬遊戲角色 影響真實行為」

虛擬遊戲角色　影響真實行為

摘錄自：天下雜誌 經濟學人電子報                        2014/2/28

2014-02-14 Web only 作者：經濟學人

圖片來源：flickr.com/photos/micurs/

上個星期日，越南軟體開發者Dong Nguyen將熱門遊戲「Flappy Bird」下架；幾乎拒絕所有訪談的Dong Nguyen終於對《華街爾日報》表示，這個遊戲實在太令人沉迷，他只想製作大家可以開心地玩個幾分鐘的遊戲。

Flappy Bird結合了單純和困難，讓玩家沉迷數小時甚至數天。然而，更精緻的線上遊戲似乎不需要那麼長的時間就能影響人的行為；研究者發現，以虛擬化身為基礎的遊戲，甚至只要短短「幾分鐘」就能辦到這點。

遊戲虛擬化身是英雄還是反派，會影響玩家對自己的感覺以及他們在線上的行為。但伊利諾大學厄巴納香檳分校的Gunwoo Yoon和瓦加斯（Patrick Vargas）提出的問題非常不同：僅僅玩5分鐘的遊戲，會不會促進日常生活中的社會或反社會行為？

他們找來194名受測者，告知受測者他們將參與兩項無關的測試：一為線上遊戲可用性測試，另一項則是食品添加物盲測。第一項實驗中，受測者會隨機配予英雄、反派或中立角色，以虛擬角色對抗敵人5分鐘，再為線上角色的認同感評分。

而在「食品添加物」測試中，同一群受測者得品嚐巧克力和辣醬，再將它們倒入碗中，讓下一個（虛構的）受測者全數吃光。研究者發現，遊玩反派角色的受測者，倒入的辣醬量是遊玩英雄受測者的2倍，英雄倒入的巧克力則是反派的近2倍。中立角色似乎較為衝突，但平均來說會多倒一些巧克力。此外，受測者對角色的認同感則與結果無關──事實上，大部分人都對角色沒有強大的認同感。

後來，研究者還測試了單只是「觀看」遊戲，會不會與玩遊戲產生不同的效果。結果顯示，觀看對行為的影響小得多，站在反派這一方的人，會比玩反派角色的人少倒一些辣醬，站在英雄這一方的也會少倒一些巧克力。

僅僅5分鐘的虛擬角色扮演就足以改變人的行為，這已經夠讓人驚奇了。由結果來看，似乎就連觀看遊戲都會影響玩家。但最讓Gunwoo Yoon驚訝的是，就算玩家並不認同角色，他們在真實世界的行為仍舊與角色相符。這表示玩家可能不知道，虛擬經驗究竟對他們真實世界的行為產生了多大的影響。（黃維德譯）

©The Economist Newspaper Limited 2014

經濟學人英文原文

The Economist

Avatars and real-world behaviour

Channelling Superman

By The Economist

From The Economist

Published: February 14, 2014

Feb 12th 2014, 10:21 by P.H.| WASHINGTON D.C.

WHEN Dong Nguyen, a Vietnamese software whizz, pulled his frustratingly enjoyable game "Flappy Bird" from mobile app stores last Sunday, it left both players and industry insiders scratching their heads. Flappy Bird had swiftly become the most downloaded game on both iPhones and Android phones, and was making some $50,000 a day in advertising revenues—the kind of success most game developers can barely envisage. But Mr Dong, who has rejected almost all requests for interviews, believed Flappy Bird needed its wings clipped. "It was just too addictive," he finally told the Wall Street Journal. "I just wanted to create a game that people could enjoy for a few minutes."

A quick look at reviews suggests that Mr Dong was right about his game's addictiveness. Its engaging mix of simplicity and difficulty kept players hooked for hours and even days on end. Many grew increasingly angry with their own feeble scores, and with Mr Dong (some even threatened to kill him for making the game so hard). More sophisticated online games, however, don't seem to need hours or days to influence people's real-life behaviours. Research published in the journalPsychological Science finds that avatar-based games can do so even in the brief "few minutes" that Mr Dong had hoped his own game would occupy people.

The heroic or villainous characters that gamers adopt as online avatars have long been known to affect how they feel about themselves and how they behave offline. Online superheroes, then, are more likely to perform good deeds in real life; villains, less likely. All this, however, largely applies to long-term (that is, addicted) gamers. Gunwoo Yoon and Patrick Vargas, of the University of Illinois at Urbana-Champaign, asked a very different question: would a paltry five minutes of immersive gaming promote prosocial or antisocial actions in everyday behaviour?

To find out, 194 participants were told they would be taking part in two unrelated experiments: one a test of online-game usability, the other a blind taste test of food additives. In the first, participants were randomly assigned an avatar: either a good guy (Superman), bad guy (Voldemort) or neutral character. They then battled enemies as their avatar for five minutes, after which they were asked to rate how closely they identified with their online character.

In the "food additive" test, the same participants were instructed to taste both chocolate and chilli sauce, then pour an unspecified amount of one or the other into a bowl for a (fictional) future participant—who, they were told, would have to eat all of what was provided.

The researchers found that participants who had had spent their five minutes as villains poured about twice as much chilli sauce as those who played a hero. Heroes, meanwhile, poured almost twice as much chocolate. Neutral avatars seemed conflicted, but on balance handed out a little more chocolate. The degree to which each participant identified with their avatar appeared irrelevant to the outcome—in fact most didn't strongly identify with their online character.

In a further study, the researchers tested whether actually playing the game for five minutes produced different results from simply "observing" it—viewers were told to watch a game demonstration for five minutes and "put themselves in the shoes" of either hero or villain. Observers' behaviour was much less affected by their experience: those merely trying to take the perspective of villains poured less chilli than those playing them, while those doing likewise for heroes poured less chocolate.

That a mere five minutes of virtual role-play as either hero or villain was enough to cause people to change their behaviour—to reward or punish anonymous strangers—is surprising enough. Even brief exposure to immersive games, it seems, imbues players with a sense of agency. But what most startled Mr Yoon is that, in the real world, participants continued to behave in ways consistent with their avatars' characters even if they said they didn't closely identify with them. This means they were probably unaware of how much their virtual experience was influencing their real-world actions.

Moral: creating more games with compelling, heroic avatars would probably promote more prosocial behaviour in real life. Reincarnating Flappy Bird as Super Flappy Bird might help too.

©The Economist Newspaper Limited 2014