The possible role of testosterone in facilitating human aggression has been studied for about 25 years now, and is of interest to a range of researchers as it is at the interface of neuroscience and social and evolutionary psychology.

“The article is a systematic synthesis of existing knowledge in relation to a theory that has not been applied to human studies before.”

For all this time, it has remained a set of data in search of an adequate theoretical framework. The simple view that testosterone secreted at puberty increases aggressiveness, as it does in many other mammals, was not supported by the evidence, and yet correlations between testosterone and measures of aggression were reported time and again.

I happened to find an evolutionarily based theory of testosterone and behavior that had been applied to studies of birds (by Wingfield), and assessed whether the evidence on testosterone and human behavior (wider than that on aggression) fitted this theory: it did, so now we have a theory that integrates all these many studies.

As a bonus, the evidence showing associations between psychological individual differences and testosterone, also seems to fit an extension of the theory to apply to alternative mating strategies. The article has come to the attention of some very active research groups interested in testosterone and human behavior, and it has started to be cited by them.

The article is a systematic synthesis of existing knowledge in relation to a theory that has not been applied to human studies before. In a way, it does set out a new discovery, in that it demonstrates a novel way of conceptualizing the links between testosterone and human behavior, and provides meta-analytic evidence for specific parts of the theory.

The challenge hypothesis was proposed as a theory of the testosterone-behavior relationship in birds. Essentially, it starts with the high fitness costs of maintaining high circulating levels of testosterone through a male’s adult lifetime. As these are high in monogamous species, the levels secreted at puberty are only sufficient to maintain reproductive physiology and behavior, but not to facilitate aggressive behavior.

In situations of reproductive competition ("challenges") testosterone levels increase and facilitate competitive behavior in the short-term. When such aggressiveness is unnecessary, for example, in a paternal phase of the life history, testosterone levels are at a low level.

A systematic review of human studies in relation to a series of specific predictions from the challenge hypothesis showed that it fits the evidence from humans. The challenge hypothesis is based on a trade-off between mating effort and parental effort, and it can be extended to show that there is a coherent pattern of individual differences associated with higher or lower levels of testosterone, representing differential emphasis on mating or parental effort.

I have a background in research on testosterone and animal behavior, and I wrote a book on animal aggression (published in 1988), following which I wrote a review of the available evidence on testosterone and human aggression (1991).

I subsequently carried out some empirical research on testosterone, mood, and behavior, funded by the World Health Organization to Fred Wu (University of Manchester) and myself. This work has been published in 3 papers (2001, 2002, 2006), and the NBR article grew from this interest. Unusually, I didn’t have any particular problems writing the paper or getting it published.

One implication is that those with a little knowledge of the research in this area should stop attributing young male aggression to the increase in testosterone secreted at puberty. This view is often repeated in media accounts.

The reality is that the testosterone-aggression link is more subtle. Testosterone will enhance competitive and aggressive behavior when increased by competitive or challenging situations.

A range of competitions, sports in particular, will cause a testosterone surge, and this will be greater in winners than losers. This in turn will make them competitive with other males. Although their testosterone levels are much lower, the same effects of competition have been found in women.