In 2005, a pandemic wiped out cities, painting the streets white with skeletons. Those who were asymptomatic spread the disease to thousands, and within a day, the world ground to a halt. This happened in the video game “World of Warcraft.” Despite the pandemic lasting only three days and being virtual, it gave epidemiologists and sociologists real-world insights into human behavior and disease modeling.
“World of Warcraft” is a massive multiplayer online role-playing game where players create a character and acquire items, skills, and animal companions that enhance their character’s combat abilities. On September 13, 2005, the developers added a new region with a new Boss known as Hakkar. Hakkar had a unique way he could attack players: Corrupted Blood, a status effect that caused players to lose health over time and would spread from player to player. High-level players needed to have either passive healing, which allowed them to outlast the damage, or abilities that allowed them to remove the effect. Defeating Hakkar also removed the effect.
In an ideal world, newer players who do not have access to the boss would never get the effect. Unfortunately, the animal companions of high-level players could also contract Corrupted Blood and be “dismissed.” Dismissing a pet prevents them from taking damage and losing the Corrupted Blood effect. Oblivious players resummoned their companions in crowded areas, bringing the plague outside the boss fight. Many players died instantly, but those who did not teleported away after contracting the disease. It only took one player teleporting from one city to another for the virus to spread.
To understand the danger a given disease poses, epidemiologists measure its infectiousness by calculating R0 (r-naught), which is the average number of people one person will infect. While the value for Corrupted Blood is unknown, lowball estimates place it solidly above ten, three times as infectious as COVID-19. The disease would also linger in specific locations on non-player characters (NPCs). Developers program these characters not to take damage, which prevents bad actors from killing them and blocking story progression for other players. However, this allowed Corrupted Blood to spread from NPC to player endlessly, even after killing nearly all of the susceptible population.
As stated by Lofgren in 2007, “It is physically impossible, financially prohibitive, or morally reprehensible to create a controlled, empirical study where the parameters of the disease are already known before the course of epidemic spread is followed.” While computer simulations that track infection exist, such as Transportation Analysis Simulation Systems or Epidemiological Simulation Systems, they are programmed based on human behavior outside of an outbreak.
During the “Corrupted Blood Incident,” many people avoided logging on for the duration of the outbreak, while others took a more active stance. One group exploited the situation and spread the disease for their amusement. The behavior is reminiscent of looters taking advantage of chaos for their own gain during COVID-19. The other group took it upon themselves to heal low-level players and make announcements about the spread of Corrupted Blood across the server. They would create first aid stations outside cities and help new players evacuate cities. These responses shed light on an aspect of human behavior that is underrepresented in current models.
There are limitations to using player behavior to model real-world action, as many may not act the same way in the real world. Teleporting also made it impossible to quarantine the disease, which is not a problem in real-world modeling. Since this pandemic, there have been other pandemic simulation games, such as “Plague Inc.,” but without the human behavior of individual actors influencing the spread of infection, it will be hard to find a virtual example as enlightening.
