Arts & Culture / Mosaic / January 25, 2012

Games predict real diseases

Epidemiologists face the difficult challenge of determining public reaction to a widespread epidemic without the ability to observe such an event.

Some Knox students see video games, such as World of Warcraft (WoW), as an interesting solution to this problem.

Millions of players interact online in WoW, a massive multiplayer online role-playing game. Players travel around the world, teaming up to complete more and more difficult challenges, according to senior Mike Kolbeck, a former WoW player.

“There’s a lot of potential for these types of games to be used for research purposes,” as a model for real world epidemics, Kolbeck said.

His interest in this theory was sparked by an accidental epidemic that broke out in the game in 2005. A newly introduced monster spread a “corrupted blood” infection, which spread to nearby players and killed weaker players, while stronger players were able to survive, according to BBC.

Blizzard, the creators of WoW, intended the disease to be limited to players near the monster, but the disease was accidentally spread beyond that, sweeping through cities and densely populated areas.

Kolbeck, who recently discovered the incident on the Internet, called it an “almost perfect model for the spread of real illness.” Panic spread through the game, quarantines were attempted with varying degrees of success, players fled from capital cities, people paid other players with healing powers to keep them alive and some players even intentionally spread the disease, acting as bioterrorists might.

According to rumor, the records from this incident were given to the Centers for Disease Control and Prevention (CDC) for research purposes.

Although models are often used to simulate epidemics, a video game has some advantages over a computer-generated simulation.

“You’ve actually got a person playing each of these characters,” Assistant Professor of Computer Science Jaime Spacco said, so you “can get unpredictable reactions that you couldn’t get if the model designer didn’t think of them.”

According to Spacco, any epidemic model would “have to be unexpected. Because WoW is not about plague … you probably got some really genuine reactions that are really valuable.”

There are also some aspects of the game inconsistent with reality that would limit the usefulness of video games.

“The stakes aren’t really that high,” Spacco said, explaining that death in WoW is more of an inconvenience, as characters are regenerated after death, so you “might not see the same level of desperation.”

Junior Anna Novikova created a simulation of a zombie outbreak in Galesburg as part of the Seminar in Mathematics last year, which focused on simulation.

Although she finds the idea of a video game model interesting, as it allows for a wider range of results, she pointed out it’s limitations as well.

“Their behavior is sort of predictable but it doesn’t mimic human behavior,” said Novikova, explaining that WoW characters travel large distances for quests, while people in real life often follow routines and only travel short distances to work, school or other nearby locations.

“The way healthcare is delivered in a MMORPG is different,” she added, saying the lack of hospitals, for example, would impact the results because it is a crucial place for an epidemic, especially the zombie outbreak, which she studied, to spread.

It is not a perfect model, but data gathered on video games has the potential to provide a new perspective on epidemiological research.

“I question in many ways the external validity of any video game study, but it’s an interesting experiment. We just have to be really honest about the shortcomings of that kind of approach,” Novikova said.

Gretchen Walljasper

Bookmark and Share

Previous Post
Art made public, by the public
Next Post
Pandas across the pond


Leave a Reply

Your email address will not be published. Required fields are marked *