Pages

Thursday, December 4, 2014

A Trust-building Camp for Sticklebacks



            Sometimes we let the activities of a single individual color our perception of the group as a whole. In watching their actions, we realize that anyone could act like that, assume they probably will, and react accordingly. This is often not the case, though. Despite the ticket you got, cops aren't all bad people. Despite your ex, members of the opposite sex are probably decent as well. Despite the fact that that guy just emptied the 'give a penny, take a penny' jar, someone filled it up in the first place. The point is that we allow single instances to influence all of our future interactions with people that we classify as similar.
                         When we look at game theory, activities like the stag hunt or prisoner's dilemma involve independent decisions of two people, allowing them to help or hinder each other. For those that aren't aware of these games, they work much in the fashion of rock-paper-scissors, where you both reveal your actions at the same time, trying to predict and account for the actions of the other. There are two choices in each of the games: to work together (cooperate), or to look out for yourself (defect). In the stag hunt, working together has double the payout of going it alone, but the reward is only distributed if your partner chooses to work together as well. Working alone, on the other hand, gives you a smaller payout no matter what your partner does, so the game revolves around trying to predict how far you can trust each other. This is amplified in the prisoner's dilemma, with harsh penalties rather than smaller rewards. When your partner has betrayed you every round this game, you can get a pretty good idea of what they're going to do next. While there are a number of "optimal" ways to play these games, if you sit someone down and have them start playing, they'll adapt to the situations they've seen, fitting into the atmosphere of either cooperation or distrust.
[1] The choices in the Prisoner's Dilemma

            Our ability to predict others' actions and account for them is all well and good, but is it really that unique? A group of researchers have recently done a study on three-spined sticklebacks to see how their previous interactions with conspecifics alter their decision-making behavior. Fish have a certain exploratory behavior: when they're in a new location, they try to get a feel for it by poking out of cover and swimming around a little in the exposed area, getting the lay of the land, as it were. With their massive schools and shoals, we all know that fish are generally fantastic proponents of the 'safety in numbers' technique, and this exploratory behavior is a similar situation. One fish will begin to explore and 'lead' the way out into exposed areas, and a second will follow, providing an extra pair of eyes on the lookout for predators.
             The collected sticklebacks were placed into the testing facility in pairs, which was part covered and part exposed. Their behavior in exploring the new territory was, optimally, done together, providing an overall safer experience for both of them, darting out and having twice as many eyes scanning for danger. This wasn't always the case, though; sometimes the second stickleback didn't follow, leaving the first to fend for itself. The testing itself was done in two phases: the first was done in isolation, determining the 'boldness' or 'shyness' of the fish by how much it ventured out on its own, but this was followed by pairings of fish, where they were able to support each other and exhibit their shoaling activity.
            All of this action was automatically tracked by camera, allowing the researchers to track the amount of time the fish spent exploring, and whether they led the expeditions or followed on them. Combining these bold/shy classifications with the fish's behavior, they were able to watch how these sticklebacks altered their behavior based on their previous interactions. It turned out that the boldness or shyness of an individual determined how it treated and responded to its partners. The bolder fish tended to rely mostly on themselves and their previous experiences, spending more time exploring the bolder they were, and the shyer their previous partners were. The shyer fish tended to follow, ignoring any previous interactions, responding to the boldness of their previous partner by spending more time out of cover the bolder their current partner was. The bold and the brave tend to lead, and the shy tend to follow, a not wholly unexpected result. However, bold fish behaved as if they were judging their current partners based on experience with previous partners. The shyer fish, in contrast, were more susceptible to social cues from their current partner. Shyer partners meant that they had to take the lead, while a bold partner would allow them to be much more lax.
[2] Does superman care if he has backup?

            It turns out that even fish learn what to expect from each other based on the behavior of previous partners. As ideal a world it would be without these assumptions, there is a sound ecological basis for it: classification is an important skill. This means that it's rather unlikely that undue and premature distrust of others will ever go away. Sadly, it looks like the tragedy of the commons will remain applicable until we can all trust each other not to take advantage of it, like that one guy did that one time and we all know everyone else would given the chance.

Sources:
Jolles, J. W., A. Fleetwood-Wilson, S. Nakayama, M. C. Stumpe, R. A. Johnstone, and A. Manica. 2014. “The Role of Previous Social Experience on Risk-Taking and Leadership in Three-Spined Sticklebacks.” Behavioral Ecology 25 (6): 1395–1401. doi:10.1093/beheco/aru146.
Media:
 [1] prisoner’s dilemma, Art, from Encyclopædia Britannica Online, accessed December 03, 2014, http://www.britannica.com/EBchecked/media/85154/Table-4-The-prisoners-dilemma-is-a-well-known-problem
[2]Alex Ross, Superman, Wikimedia commons upload, December 2008, Used under Fair Use: http://en.wikipedia.org/wiki/File:Superman_by_Ross.jpg

No comments:

Post a Comment