Computer games can predict outcomes of real world crisis

During my ongoing literature review I often discover interesting facts about things I’ve never thought about. Sometimes I can connect these facts with my own observations: The result is mostly a completely new idea why things are as they are. Maybe these ideas are new to you, too. Therefore I’ll share my new science based knowledge with you!

This week: This time, I highlight another potential of huge virtual worlds: the potential to predict outcomes of a real world epidemic.

In „Finding of the week 32″[1] I highlighted how the huge dimensions of MMORPG worlds create a virtual environment that almost mirrors our real world. Scientists and in some cases learners can take advantage of these virtual environments and observe real world mechanisms. Additionally, these environments allow their observers to easily trace back the causes of the outcomes, because they are self-contained environments.

Another interesting fact these environments can be used for is to observe the dynamic of huge masses of people in a highly populated area. According to Lofgren and Fefferman[2] this can be useful to predict the spread of an epidemic. Back in the year 2005, a new World of Warcraft raid instance was released that caused due to a bug the death of many players.

During the fight against the final boss of this particular instance, players could get infected by a disease that spreads over to nearby players or player’s companions. Although the disease should remain in the instance, players infected with it were able to zone out of the instance and enter hugely populated areas like the capitals of Azeroth. The boss fight mechanic suddenly became an epidemic causing the death of many players being in these areas and won’t stop to spread until the developers of the game did a server reset.

This unexpected event had shown the potential of virtual environments to predict the dynamic of masses under such conditions. Although the World of Warcraft epidemic wasn’t planned and the developers never gave out data to analyze the spread of the disease, the potential for simulations of epidemics in such virtual worlds was obvious. The disease had infected most of the inhabitants of highly populated areas within a short amount of time.

The spread of a disease could also be observed during the pre-launch event of Wrath of the Lich King, the third World of Warcraft addon. During the event suspicious crates and infected roaches appeared in the capitals of World of Warcraft. Players who got too close to these crates were infected with a disease causing them to become a zombie after a certain amount of time. Apart from attacking other players, zombies could explode and cause nearby players to get infected, too. Players in highly populated areas were more in danger being infected than players in area with a low density of population.

Suspicious crates

But how can a game event share similarities with possible real world epidemics? On the one hand, players live a parallel life in these worlds. MMORPGs simulate an ongoing world that even goes on when the player isn’t online. In this endless world, players have a daily life, they buy and trade goods and meet with their ingame friends. Most of these activities require players to visit highly populated areas where, in case of an epidemic, the danger is high to be infected by the disease and to increase the spread of the disease.
This virtual daily life has some similiarities to our real world daily life: we need to go to highly populated areas. Shopping at a grocery story, going to the movies, going to work and traveling by plane or train require us to visit highly populated areas like the auction house in World of Warcraft.

Empty market place in Stormwind during the Zombie-Invasion

Apart from the daily business, players also do have a high attachment to their avatars. Over time, the avatars aren’t a tool to explore the world anymore. Instead, the player becomes the avatar and starts to feel like the avatar. This attachment increases the immersion of the game world and the zombie plague becomes a major threat to the player himself. The epidemic of the zombie plague becomes a global crisis in the game, because players start to avoid highly populated areas to protect themselves the same way, as we wouldn’t leave a safe spot in case of a real crisis. This however starts to highly influence the trading and results in a shortage of goods.

The combination of a high attachment to the own avatar and the resulting immersive role-playing causes even a game epidemic to reach possible real world dimensions and results of real world epidemic. The immersive gameplay becomes a simulation of our daily life and allows scientists to predict a possible outcome of such a crisis.
However, this isn’t as easy to achieve as it sounds. The deep immersion of the role-playing needs such huge virtual worlds as they exist in World of Warcraft and has to grow over time. Creating a meaningless virtual world wouldn’t create the same results during a virtual epidemic, because players wouldn’t be as attached to their avatars as in World of Warcraft. Additionally, games like World of Warcraft should be played because they make fun and because they provide an authentic environment. Tossing in a virtual epidemic without any connection to the game world could possible destroy the immersion.
Furthermore, the game industry makes their money by providing this immersive environment. Creating a virtual epidemic for real world research and giving out sensible data about the own players would be a high risk for these companies to lose the trust of their consumers.

To wrap things up, immersive virtual worlds do have a high potential to provide critical data to predict the outcome of real world dynamics. Unfortunately accessing the data is very difficult, because game companies can’t give access to sensitive data about their own players and won’t risk to destroy their game world by adding features just for science. However, if an in-game event as the above-mentioned pre-launch event takes place, scientists can do some sample observations just by playing the game.

References

[2] Lofgren, Eric T.; Fefferman, Nina H. (2007): The untapped potential of virtual game worlds to shed light on real world epidemics, in: The Lancet infectious diseases, 7 (9), pp. 625 – 629.