Prejudice and truth about the effect of testosterone on human bargaining behaviour.

This is a great, great paper about what testosterone (T) does to behavior, which is something much subtler than most people think; I loved this paper. There's a substantial amount of literature showing that T and aggression go hand in hand, with the easy conclusion being that T causes the aggression. In actuality, when T levels are within the physiological range, it is almost always the case that things are the other way around, with the behavior driving T secretion.

But when T is indeed altering behavior, what is it doing? Folk wisdom holds that T makes one more aggressive, egoistic and antisocial. The much subtler 'social status challenge' hypothesis posits that T doesn't cause aggression, antisocial behavior, whatever; instead, it causes behaviors that preserve one's status during a challenge. Most of the time, the 'it makes you more aggressive' and the 'it helps you defend your status from attack' explanations are hard to tease apart, and the brilliance of this paper is that they are dissociated. The authors have subjects play the Ultimatum Game. Basically, one individual (the proposer) is given the power to decide how money is divided up between them and another player; the only power the other individual (the decider) has is to refuse the offer, in which case neither of them get anything. Thus, despite the seeming power of the proposer, the decider ultimately has the most power, and the proposer seriously loses status if the decider rejects their offer. So, the subjects were women who had either been given some T or placebo. Perfect: if the folk wisdom notion about T causing antisocial behavior is right, T should cause proposers to make more unfair offers. Alternatively, if the status challenge idea is right, T should make proposers increase the fairness of their offers (to avoid the subordination of being rejected). And the answer? T makes women fairer in their offers. Two important details are that a) T was not merely making these women more altruistic (which would really have been a surprise). As evidence, as 'deciders', they rejected offers at the same rate as did placebo subjects. Also, b) unsurprisingly, subjects tended to believe in the folk wisdom model where T makes you more antisocial. Thus, independently of whether they received T, the subjects who thought they'd been given T became LESS fair in their proposer behavior. So, T doesn't make you more aggressive or anti-social. It makes you defend your status more, and, in circumstances where one maintains status by being pro-social, T makes you act nicer.