We talk about molecular population and evolutionary GENETICS and GENOMICS. You know, the caliper measurement of a gene's evolvability in moles.T-Rex thinks he's eliminated zombies with logic. The basic idea: Zombies depend on human brains to survive, but they also must bite humans (turning them into zombies) to create more zombies. If zombies were really good at catching humans and eating their brains, there would be no more humans and the zombie population would die off. Conversely, if the zombies had trouble getting to the human brains, they could convert humans into zombies (by biting them without eating their brains), but they would starve due to brain deprivation.
Anders Sandberg disagrees, and he's done the simulations to back up his argument. Assuming heritability of brain eating ability in zombies (you can take certain liberties with fictional creatures), Anders shows that zombie populations will evolve towards intermediate brain eating and zombie creating. Here's what Anders expects to happen over time (humans are the green line, zombies are shown in red):
The top graph shows that a zombie population that starts out with maximum brain eating efficiency will decrease in efficiency until humans become very rare. At that point, the fitness of efficient brain eating will increase and those individuals will increase in frequency. But humans become pretty rare. In fact, it looks like they're teetering near extinction. These are relative numbers, however, so it's hard to tell how the absolute human population size changes over time.
There is also a post on vampires and the coalescent promised. I'm psyched.
(Via Well Rounded Nerds.)
Comments
I suppose the obvious direction for future research would be to factor vampire and werewolves into those calculations. You also have to assume that the normal human population's ability to fend off zombie, vampire, and werewolf attacks will change over time. Is there a possible stable configuration for all four populations, maybe a synced sequence of boom and bust like the caribou-wolf equation?
Posted by: John McKay | January 15, 2007 12:46 PM
This is an issue of frequency dependant selection- from which a mixed ESS develops.
I'm not so sure about the intermediate feeding hypothesis, as it seems it is not likely a global ESS- a individual mutant fast-eater could sweep through the population and replace the intermediate.
I'd make a simple model to test this, but that would be WAY too much like work..Anders???
Posted by: Matt | January 15, 2007 1:27 PM
Reminds me of something I read about vampire population dynamics (that's a pdf).
Posted by: scienceiscool | January 15, 2007 7:16 PM
Rather than seeing this in terms of that discredited darwinism nonsense, I think we finally have proof positive for intelligent design theory.
Human brains were designed to be eaten by Zombies.
Why else would they be so tasty !
Posted by: MartinC | January 16, 2007 2:38 AM
Ah, the things you can do with Populus software and a crate of beer!
Posted by: Dr Valerie Olson | January 16, 2007 6:29 AM
I'm an undergrad with little better to do... is there anywhere can I get this Populus software for free?
Posted by: KMeck | January 23, 2007 6:32 PM
Populus is available here. But it looks like the data were analyzed in R. If you want to be an undergrad with good skills that grad schools will like, you should learn R.
Posted by: RPM | January 23, 2007 8:06 PM
..... you guys are gay.....
Posted by: a real life girl | January 20, 2009 8:15 PM