Part of the problem here is that even if testosterone had absolutely no effect on aggression at all, we would still see people taking testosterone injections acting more aggressive. Why? Because the common belief is that testosterone will make you more aggressive. Give them saline and tell them it’s testosterone and they’ll start bumping people in the street as well.
To test whether there is an actual effect going on here, they’d need to look at what how two different groups of FtM transsexuals respond when one is placed on a placebo, and one given testosterone. The article linked to by Gwern discusses this effect of perception on behavior:
Folk wisdom holds
that testosterone causes antisocial, egoistic, or even aggressive behaviors in humans. However, the correlational
studies discussed above already suggest that this simple
folk view probably requires revision [34,56]. A recent placebo-controlled testosterone administration study found
support for the idea that the testosterone–aggression link
might be based upon ‘folk’ views: individuals given placebo
who believed they had been given testosterone showed less
fair bargaining offers compared with those who believed
that they had received placebo, thus confirming people’s
stereotypes about the behavioral effects of testosterone.
More importantly, however, when statistically controlling
for this belief of treatment assignment, one acute dose of
testosterone in women increased the fairness of proposers’
bargaining offers in an ultimatum game [13] (Figure 3).
Part of the problem here is that even if testosterone had absolutely no effect on aggression at all, we would still see people taking testosterone injections acting more aggressive. Why? Because the common belief is that testosterone will make you more aggressive. Give them saline and tell them it’s testosterone and they’ll start bumping people in the street as well.
To test whether there is an actual effect going on here, they’d need to look at what how two different groups of FtM transsexuals respond when one is placed on a placebo, and one given testosterone. The article linked to by Gwern discusses this effect of perception on behavior: