Indeed, Austin and his collaborators found a single gene that controls bacteria "selfishness." If it's off, bacteria slow down their metabolism and reproduction rate when they sense their environment has been depleted of nutrients. This prevents them from completely destroying their living space. However, if this gene is turned on ("expressed" is the technical term) the bacteria go right on eating until nothing is left. They even develop the ability to feed off of other dead bacteria.
Interesetingly, the gene is off by default when bacteria are found in the wild. But if you put them in a petri dish, mix them together, and cut off their food supply, you rather quickly (after only about 4 days!) see selfish mutants emerge. These mutants rapidly consume all the remaining food, including each other, and then starve.
This is an interesting conundrum. The petri dish situation seems pretty dire: first the cheaters win, and then everyone loses. This is another prisoner's dilemma situation: cheaters seem to have the advantage over the self-restraining altruists, but if everyone cheats then everyone is worse off.
On the other hand, bacteria in the wild exercise restraint, so there must be something different going on in the wild than in the petri dish.
Intrigued, Austin and his colleagues set up a different experiment. They designed an artificial landscape conatining many differnt chambers in which the bacteria could isolate themselves. Food sources were spread unevenly through the landscape. They also found a way to "manufacture" the selfish bacteria by fiddling with their DNA, and they dyed them a different color from the altruists to discern the interactions between the two.
In this situation, the altruists and the cheaters managed to coexist by segregating themselvs. The altruists gathered in dense clumps (and lived in harmony?) while the cheaters spread out sparsely (they don't even like each other!) around the altruists, occasionally gobbling up a dead one. Somehow, the altruists are able to segregate themselves in such a way that the cheaters can't steal their food; a marked contrast to the first experiments in which the bacteria were continually mixed together. Here's what this segretation looks like within two of the "chambers":
The chamber on the left, which is nutrient-poor, contains mainly cheaters waiting for others to die. The nutrient-rich chamber on the right contains "patches" of altruists and cheaters, never fully mixed. You can't see it from the picture, but the green altuists are very densely clumped and the red cheaters are spread apart from each other.
The possible life lesson here is that altruists can exist in a society with cheaters if the altruists can segregate themselves to form (utpoian?) communities. If there is forced mixing between the two groups then, unfortunately, it all ends in tragedy.
A very similar lesson can be found in the work of Werfel and Bar-Yam, but that's a story for another time.
"This is another prisoner's dilemma situation: cheaters seem to have the advantage over the self-restraining altruists, but if everyone cheats then everyone is worse off."
ReplyDeleteI'm not sure I read it that way. In this case cheaters have the advantage to a point but then they cause their own annihilation. The critical factor doesn't seem to be whether everyone is cheating - it seems to be whether cheating is a suitable equilibrium considering the source and level of food. Or did I misread it?
You're paper's in my pile of things to read. I won't be able to provide any useful feedback but it looks like I'll understand it clearly enough to find it interesting.
I was talking with Justin Werfel the other day, and it seemed to both of us that the people who study altruism via game theory (Axelrod, Nowak et al.) aren't communicating very well with the people doing kin selection or other approaches. For example, Hauert and Szabo, in a paper confusingly entitled "Game Theory and Physics" (2005), look at how distributing agents across space affects the outcome of repeated Prisoner's Dilemma games. Just as in the Rausch/Werfel/Bar-Yam work, distributing your population across space (so that it's not uniformly mixed) allows new patterns to arise: what's going on over here can be separate from what's happening over there. However, the people who do this with game theory don't know about the people doing simulations of predator-prey interactions, or vice versa.
ReplyDeleteThis is particularly unfortunate because Page and Nowak (2002) suggest that evolutionary dynamics can be unified, i.e., that the equations used in these different approaches can be mapped onto one another.
In my mind, three important questions stand out:
1. Is there a general way of understanding the transition between "well-mixed" and "spatially separated" systems, in particular with regard to adding long-range connections to a spatial lattice? This would decrease the clustering coefficient, two neighbors of a given node would be less likely to be directly connected themselves; one could possibly write the analog of the Ginzburg criterion for when the mean-field description, applicable to the well-mixed case, ceases to be valid.
2. Is there a connection between distribution across space and behavior in time? An interesting result of Werfel and Bar-Yam, for example, is that behaviors which benefit a lineage in the short term become disastrous in the medium or long term — an overzealous predator or a selfish mutant pathogen destroys all the resources in its vicinity, eventually ruining its local environment so that its descendants cannot survive. (This is how the system guards against "cheating", allowing something vaguely reminiscent of "group selection" to emerge, although personally, I do not find the terminology of group selection very helpful.) I think biologists would say that the predators have an "evolutionary stable strategy" (ESS) in the short term which fails in the long term. In other words, the ESS concept has to be augmented. The game-theoretic analog of the ESS is the Nash equilibrium, so what we're talking about is a situation that is no longer "quasi-static", and the behavior of the actors itself changes the environment such that what looks like a stable equilibrium stops being one as time goes on.
For elaborations on the "quasi-static" assumption, described without reference to game theory, see Goodnight et al., "Evolution in spatial predator-prey models and the prudent predator: The inadequacy of steady-state organism fitness and the concept of individual and group selection" Complexity (27 Feb 2008), of which a PDF copy can be found here.
3. When I discussed the Werfel and Bar-Yam paper with an actual biologist, she brought up the issue of "kin recognition". If Alice finds that she's close kin to Bob, her behavior can change: this could mean not mating with Bob, or acting more altruistically towards Bob than to other members of the population. Spatial distribution does this job for you, to an extent, because in systems like the Werfel/Bar-Yam model, the organisms near you are often related to you; however, the timescale issue in point 2 means that what happens in that model is not, strictly, old-school kin selection. I'm curious, therefore, about how kin recognition might break the mean-field assumption, "unmixing" the population.
@AP-I think we agree. The advantage gained by the cheaters is only in the very short term. But it's enough of an advantage that they take over the altruists before they annihilate themselves. Tragedy of the commons might be a better label for this type of situation than prisoner's dilemma.
ReplyDelete@blake-I'm a bit surprised by this disconnect you see between game theory approaches and predator-prey approaches, given that one of the first reasearchers to look at spatial prisoner's dilemma was ecologist Robert May (example), who also studied predator-prey systems extensively.
I think the questions you ask are very good ones. Clearly, the issue is not so much physical (spatial) separation as segregation--the altruists need to be able to congregate unmolested by cheaters. By varying the possibilities for segregation, I'm pretty sure you'd see a phase transition from cooperative stability to inevitable destruction.
I had a thought while walking home yesterday evening, about a way in which kin recognition might not be enough to unmix a population. Consider a well-mixed test tube full of predator and prey microorganisms. Different genetic varieties can exist, and in principle, these varieties can alter their behavior based on the population fraction which shares their own genotype, but they're all competing for the same food resources. Consequently, disasters will be global, rather than local.
ReplyDeleteYeah, I think that's right. Kin recognition allows for a kind of altruism (if you can call it that), but not for restraint in consumption/reproduction rates. For that you need some kind of separation of resources (spatial separation being the main example.)
ReplyDeleteApropos our journal-club discussion this afternoon, somebody with an institutional subscription should download A. Gardner and S. West, "Social Evolution: The Decline and Fall of Genetic Kin Recognition" Current Biology 17, 18 (2007).
ReplyDelete