You’re fast enough, you’re smart enough, and doggone it, you can kill zombies
In the event of a zombie apocalypse, you might want to befriend a Frisbee golf pro — the skills may transfer to record-slinging, a decapitation technique favored in Shaun of the Dead. Or if you prefer a traditional zombie-slaying approach, recruit a baseball player.
Choose wisely, as skill is essential to human survival in the face of a zombie epidemic. Of all the mythical characters threatening humankind, the walking dead are most like an infectious disease, says Robert J. Smith? (the question mark is part of his name) of the University of Ottawa. Modeling how a zombie infestation would play out — like assessing a swine flu outbreak — can illuminate which counterstrategies work best, even when there’s little data and little time to gather more.
“Zombies are as random and capricious as any disease,” says Smith?, a mathematician specializing in epidemiology.
Actually, zombies are worse than many diseases. A few years ago, Smith? and colleagues used an epidemiological lens to examine zombie infiltration. Their model allowed for factors that might give humans an edge, such as the possibility of quarantine and a time delay, a latency period before the infected became zombified. Nonetheless, simulations based on a city of half a million people found that zombies would completely take over in less than four days. This doomsday scenario emerges from basic zombie biology: Once the dead can be brought back to life (or at least undeadness), everything tips in favor of more dead.
“In the end, all roads lead to zombies,” Smith? says.
Other researchers found the outcome distressing. “It painted a bleak picture for the survival of humanity under most scenarios,” mathematicians Jennifer Badham and Judy-anne Osborn note in a paper in the upcoming book Mathematical Modelling of Zombies. They decided to examine a zombie takeover from another perspective, incorporating individual human features such as speed and the ability to become highly skilled at killing zombies.
Badham and Osborn, of the Australian National University in Canberra, constructed an agent-based model, an approach starting at the bottom, rather than from the bird’s-eye system-wide view of epidemiological studies. Agent-based models, as the name implies, are based on individuals with agency — the capacity to act. Incorporating information about individual agents’ decision making can reveal how nuances may affect the system as a whole.
Thanks to advances in computing power, agent-based modeling is now used widely, for exploring the behavior of everything from cells to stock market investors. And, fortunately, zombie-human interactions. Unlike the epidemiological model, the agent-based model includes variables such as initial human skill, improvement through experience and human speed, variables that can tip the field in humankind’s favor.
Agents interact within a given space. When a human and zombie end up at the same location, they fight — a critical event that has different outcomes depending on whether the human is, say, a record-thrower or chainsaw-er extraordinaire.
When humans have no skill or speed (that is, are like their homogeneous zombie foes), some humans still survive at very low zombie densities. Such “success” is achieved mainly because humans have spatial awareness and are smart enough to run away. And until there is a critical mass of zombies upon which humans can refine their zombie-slaying skills, the outlook is still bleak. With enough zombies, humans gain experience with each kill, and the odds of success surge.
If humans are both faster than zombies and can become really skilled, people can defeat the hordes, even at really high zombie densities — contrary to the conclusion of the epidemiological analysis. “We have been able to show that the grim conclusions for the future of the human race in the event of a zombie attack …were unnecessarily pessimistic,” write Badham and Osborn.
The comforting result from the agent-based approach is fitting, as it highlights what differentiates humans from our zombie brethren. At a fundamental level, agent-based models recognize the individual human, each with his or her own quirks and beliefs and the autonomy to act on those beliefs. We are not listless, homogeneous beings who move at a constant pace seeking only to eat brains. Humans have different experiences and skills, and the right mix of the two could help in a war to end all zombie wars.
As zombie-slayer Tallahassee notes in Zombieland, “Mama always told me, ‘Someday, you’ll be good at something.’ ”
Some have suggested that the appeal of the zombie genre is that it encapsulates the individual’s fight against the machine, the rat race, the creeping worry that we all live a sad, zombie life from which we can’t escape. Well, thankfully, with enough practice or previously acquired expertise, we can kick zombie ass and assert our own rage against the machine while we’re at it.
SN Prime | August 8, 2011 | Vol. 1, No. 8
