This is an archived story; this page is not actively maintained. Some or all of the links within or related to this story may no longer work.
For the latest University of Minnesota news, visit Discover.
University of Minnesota
March 10, 2009
Online games can reveal a lot about human behavior, says computer science and engineering professor Jaideep Srivastava.
Online gaming opens new doors to understanding human behavior
By Deane Morrison
In the fantasy game EverQuest II, players can pit their swords against a dragon's fangs, form alliances against evil warlocks, or go on a quest to win power and treasure. Sound like the real world?
It sure does to Jaideep Srivastava. After a study of "EQ2," the University professor of computer science and engineering and his social scientist colleagues concluded that such massively multiplayer online (MMO) games represent a microcosm of human society. But unlike real society, MMOs can reveal much greater detail about human social behavior because a computer tracks every move the players make.
The more than 300,000 EQ2 players may undertake their quests alone or as part of a guild of fighters, healers, weapon makers, tradespeople, or other sorts. As they achieve goals and gain points, they move to a higher level of play and get "a lot of social approval from other people," Srivastava says. They can also talk to other players in real time, shift alliances, and generally interact socially in most of the conventional ways.
"A quest is sometimes like a pickup game of basketball," says Srivastava. "Sometimes two people will go to a court and join in play. Or teams or clubs will decide to schedule a quest at a specified date and time." And people play games like EQ2 for familiar motives: to achieve something, to socialize, and to relax.
"Now we have a world where behavior is the same as real life, but also observable and measurable to a fine granularity," says Srivastava. "It's a highly instrumented environment where we can study things we never could in the real world and potentially learn things with applications in the real world."
In researching human behavior, social scientists often rely on surveys, Srivastava elaborates. But besides cost, surveys are subject to a self-reporting bias that makes people try to appear better than they are, even in anonymous settings.
For example, suppose one or more guilds or teams undertake quests of the same difficulty, but the groups have different organizational structures. Which team will succeed more? That kind of data gives social scientists clues as to how real structures behave.
"Now we have a world where behavior is the same as real life, but also observable and measurable to a fine granularity."
"Or," Srivastava says, "There's an organizational structure doing something and there's a crisis such as a member dying. What type of structure recovers faster, and how do they recover and regroup? This tells us something about how people form alliances."
This kind of window on behavior compares with Google's recently demonstrated ability to determine where various health conditions were prevalent, Srivastava says. Google had only to analyze the key words in searches originating from different parts of the globe.
"Google is a huge monitor for global minds," he says. And as for online gaming, "My social science colleagues are excited that the Web is a place where people interact. They say it's like the Hubble Space Telescope or an electron microscope for the social sciences because you can see social structures and behaviors so well."
One reason Srivastava and his colleagues trust EQ2 to mirror actual life is that in their study of 7,000 players (in which they used—you guessed it—surveys), the responses revealed predictable motivations. Men tended to play for the chance to achieve something. Among women, a minority played for achievement the same as men, but the chance to socialize was the strongest motive.
The survey also showed that when playing the game with a romantic partner rather than a friend, women reported greater satisfaction but men reported less.
"Given that women like the game's social aspects, women saw this as time to spend with their romantic partner," Srivastava says. "But men tended to want achievement more than socialization with a romantic partner."
Whatever the motive, EQ2 players trade in a goodly chunk of their real-world lives for the game: on average, players spend 26 hours a week at it, Srivastava says.
He would like to extend studies of online behavior to examine how people read various compositions, noting how much text they read at once, what parts they reread, where they put bookmarks, and so on.
“With e-books like Amazon’s Kindle, which can record detailed data on reading behavior, such studies are now possible. It could help authors understand how their composition was viewed by readers,” says Srivastava.
The research has attracted funding from the U.S. Army, because understanding such social structures and behaviors can be helpful in training. Srivastava and his colleagues are also funded by the National Science Foundation's Human and Social Dynamics program. Working with him are organizational theorist Noshir Contractor of Northwestern University, sociologist Scott Poole of the University of Illinois, Urbana-Champaign, and social psychologist Dmitri Williams of the University of Southern California. EQ2 is a product of Sony Online Entertainment.