I came across this article recently published by The Economist: http://www.economist.com/node/21527025 which is on game theory. It explains how game theory may be used (and how it is used) to predict human beings' behaviour in a variety of fields. Politics and negotiation are just a couple of good examples. The journalist even reports that game theory would have played a role in finding Bin Laden's hiding place last May.
Well, so far so good. The fact is, I knew about game theory because when I was working on my PhD on dispute mediation I learnt that this branch of mathematics is used in negotiation. I did not really use it for my own research; my feeling was that the emphasis on a mathematical model was hardly conciliable with my own focus on argumentation - and how argumentation can indeed change the power balance, change the individuals' priorities by helping them find out their real interests. So I sort of abandoned game theory halfway through my PhD journey.
However, who knows... In The Economist's paper, I found it particularly interesting that a game-theory based consultation may cost you a lot of money because you "must first conduct lengthy interviews with experts" to construct a reliable model. Perhaps investigating this aspect - what these "lengthy interviews" amount to and how they are used - would allow integrating game theory in an argumentation-focused study of argumentation and mediation.