ReviewTestosterone and human aggression: an evaluation of the challenge hypothesis
Introduction
There are a large number of studies investigating the possible association between testosterone and aggression in humans. Two sets of well-known findings provide the background for such studies. The first is the clear association between physical aggression and the age and sex of the perpetrator. As in many other vertebrate species, young adult males show higher levels of physical aggression, violence, and homicides to members of the same sex (Archer, 2004, Daly and Wilson, 1988), than do other age and sex categories. These findings fit the view that human male aggression is, like that of many other mammals, heightened when reproductive competition is most intense.
Among birds and mammals, one widespread mechanism that increases the readiness of males to fight during phases of the life-history when reproductive opportunities are greatest is the action of testosterone on areas of the brain controlling aggressive behavior (Archer, 1988). The co-option of testosterone in the control of inter-male aggression ensures that this is enhanced at times in the life-cycle when males need to compete for receptive females or the resources necessary to attract such females, i.e. in the breeding season or after sexual maturity. The involvement of testosterone in the control of aggression has been shown in many species, from fish to mammals, but with substantial interspecies variation (Archer, 1988). Experimental studies from the first half of the last century, on animals such as the domestic fowl and the house mouse (Allee et al., 1939, Beeman, 1947a, Beeman, 1947b) have shown a causal link between the hormone and aggressive behavior.
This background provided the basis for a number of studies, beginning in 1971 (Persky et al., 1971), investigating a possible link between circulating testosterone levels and aggression in humans. The view that testosterone secreted by males at puberty and throughout adult life facilitates human male aggression, despite its superficial appeal and its incorporation into media accounts of young men's behavior, is at best an oversimplification. The consensus is that there is a weak and inconsistent association between testosterone levels and aggression in adults (Albert et al., 1993, Archer, 1991, Archer, 1994, Archer et al., 1998, Archer et al., in press, Book et al., 2001), and that the administration of testosterone to volunteers typically does not increase their aggression (O'Connor et al., 2001a, O'Connor et al., 2004).
In this article, I first outline an alternative theory of the role of testosterone in the control of male aggression to the simple view based on studies of laboratory mammals such as the house mouse. This alternative is ‘the challenge hypothesis’. Predictions derived from it are contrasted with those from the view that there is a simple causal relationship between circulating testosterone and aggression (‘the mouse model’). The evidence concerning testosterone and human behavior is then reviewed in relation to predictions from the challenge hypothesis. Because the challenge hypothesis concerns attributes other than aggression, it will be necessary to review a more extensive body of research than that concerned with testosterone and aggression. Since it also concerns bi-directional causal links, studies of testosterone levels as a consequence of changes in behavioral or environmental contingencies are evaluated. Since it also concerns the trade-off with other forms of adaptive behavior, such as paternal care, studies linking this with testosterone are also covered. Finally, the link between testosterone and overall life-history strategies is considered as an extension of the challenge hypotheses to apply to adaptive individual differences.
Section snippets
The challenge hypothesis
The challenge hypothesis was originally proposed to account for testosterone-aggression associations in birds with a monogamous mating system (Wingfield et al., 1990). It holds that there are specific context-dependent increases in testosterone levels that are associated with aggression. Testosterone levels rise to moderate levels at the start of the breeding season and these levels support reproductive physiology and behavior. During challenges to males in contexts that are relevant to
The challenge hypothesis applied to humans
The extension of the challenge hypothesis to humans involves some modification of both the original, and of Muller and Wrangham's version, since this is based on a polygynous species in which parental care is absent. Human mating systems are variable, and have been described in evolutionary terms as involving mild polygyny, together with paternal care (Geary, 2000). The range of possible mating systems, which can be linked to ecological conditions (Alexander et al., 1979, Smith, 1998) in a way
Hypothesis 1: There is no increase in aggression at puberty
The challenge hypothesis makes a clear prediction that there are two types of testosterone-behavior relationships, the first is to support courtship and mating, and the second is to facilitate competitive behavior in relevant circumstances. The first occurs in birds throughout the breeding season in seasonal breeders, and from sexual maturation onwards in non-seasonal breeders. The second involves increases in testosterone that are facilitated by reproductive competition. We would therefore
Hypothesis 2A: Men respond to sexual arousal with increased testosterone
In this section, I consider one aspect of the prediction that testosterone is increased in situations involving mating effort, that the testosterone levels of young men increase in response to sexual stimuli. Several studies have examined changes in testosterone in relation to sexual activity, or the influence of exposure to erotic stimuli on men's testosterone levels. The first of these studies (Fox et al., 1971) measured testosterone levels of a single male volunteer during and immediately
Hypothesis 2B: Men respond to competition with increased testosterone
Most of the research relevant to this question has not been designed from an evolutionary viewpoint, and therefore it does not acknowledge the centrality of competition associated directly or indirectly with mating opportunities. Studies of competition have tended to center around competitive sports and laboratory tasks that are convenient to use with undergraduates. In this section, I consider first some research involving reputation and insults, which are particularly important in disputes
Hypothesis 3: The testosterone response to challenge increases aggression
The other side to the challenge hypothesis—and indeed its whole point in adaptive terms—is that the testosterone surge should increase aggressiveness in competitive situations. The evidence for this is hard to find in studies of humans, which have tended to examine the association between levels of testosterone and aggressiveness among samples of adults (see Section 9), or have involved the impact of competition on testosterone levels.
In an extension of an earlier study on judo competitors (
Hypothesis 4: Testosterone levels are lower among paternal men
There is evidence for a consistent pattern of hormonal changes in paternal birds and mammals (at around the time of birth, which does not occur in males of non-paternal species (Wingfield et al., 1990, Wingfield et al., 2000, Wynne Edwards, 2001). To assess whether a similar pattern occurred in humans, Storey et al. (2000) obtained salivary samples from 34 couples at one of four times before and after birth, and measured cortisol, prolactin and testosterone levels. The men showed differences
Hypothesis 5: Aggressive dominance is correlated with testosterone levels
According to challenge hypothesis applied to primates (Muller and Wrangham, 2004) the result of the cumulative effects of successful challenges should be that aggressive dominance will be positively correlated with testosterone levels. Among non-human animals, dominance is based, at least initially, on physically-based contests. This is probably also the case in most pre-state human societies, and in the modern world where there is no effective rule of law. We should therefore predict from the
Hypothesis 6: Testosterone is associated with alternative life history strategies
The original challenge hypothesis concentrated on species differences in the allocation of mating versus parental effort in males, and, among monogamous species, temporal changes in the allocation of mating versus parental effort. The general principle of differential allocation of mating versus parental effort may underlie relatively enduring individual differences among males in their relative preference for life history strategies that emphasize either mating or parental effort, differences
Does the Challenge Hypothesis apply to women?
Although the challenge hypothesis is concerned with males, Wingfield et al. (2000) did note that there was considerable variation in testosterone levels between females of different bird species. When sexual dimorphism was less pronounced, as in monogamous species, testosterone levels of females relative to those of males are higher, and it is possible that testosterone may be involved in female competition. Some of the studies covered in this review relate to this possibility. There was only
The role of neonatal hormones
The challenge hypothesis addresses the role of circulating testosterone levels during adult life. However, it has long been known from studies of mammals (and later birds) that androgens influence neural structures at two times in the life history, perinatally and at puberty. The earlier influence, which occurs either prenatally or in early postnatal life, is known as an organizing effect, and typically primes the brain, rather than influencing behavior directly. The pubertal increase typically
Conclusions
Evidence from a range of different studies involving measures of testosterone and behavior indicates a reasonable fit with predictions derived from a modified version of the challenge hypothesis, originally applied to testosterone-behavior relationships in birds. The challenge hypothesis has a sounder basis in evolutionary theory than the alternative ‘mouse model’, which supposes a straightforward influence of testosterone on aggression in males from puberty onwards. The basis of the challenge
References (191)
- et al.
Testosterone supports hormone-dependent aggression in female rats
Physiol. Behav.
(1989) - et al.
Hormone-dependent aggression in male rats is proportional to serum testosterone concentration but sexual behavior is not
Physiol. Behav.
(1990) - et al.
Aggression in humans: what is its biological foundation
Neurosci. Biobehav. Rev.
(1993) - et al.
Rewarding properties of testosterone in intact male mice: a pilot study
Pharmacol. Biochem. Behav.
(2000) - et al.
Impulsive and premediated aggression: a factor analysis of self-reported acts
Psychiatry Res.
(1999) - et al.
Testosterone, cortisol, and women's competition
Evol. Hum. Behav.
(2002) - et al.
Early androgen effects on aggression in children and adults with congenital adrenal hyperplasia
Psychoneuroendocrinology
(1997) - et al.
Changes in testosterone, cortisol, and estradiol levels in men becoming fathers
Mayo Clin. Proc.
(2001) - et al.
Testosterone changes during vicarious experiences of winning and losing among fans at sporting events
Physiol. Behav.
(1998) - et al.
Endogenous plasma testosterone levels and commission errors in women: a preliminary report
Physiol. Behav.
(2001)
Plasma testosterone and pathological gambling
Psychiatry Res.
The relationship between testosterone and aggression: a meta-analysis
Aggress. Viol. Behav.
Testosterone, and winning and losing in human competition
Horm. Behav.
Men in committed, romantic relationships have lower testosterone levels
Horm. Behav.
A few good men: evolutionary psychology and female adolescent aggression
Ethol. Sociobiol.
Hormones, sex, and status in women
Horm. Behav.
Mating season aggression and fecal testosterone levels in male ring-tailed lemurs (Lemur catta)
Horm. Behav.
Testosterone, intelligence and behavior disorders in young boys
Pers. Individ. Diff.
Androgen levels and components of aggressive behavior in men
Horm. Behav.
Testosterone and aggression in children
J. Am. Acad. Child Adolesc. Psychiatry
Male and female salivary testosterone concentrations before and after sexual activity
Physiol. Behav.
Testosterone and personality among college students and military veterans
Pers. Individ. Diff.
Testosterone differences among college fraternities: well-behaved vs rambunctious
Pers. Individ. Diff.
Exploring the mind of testosterone: a bleeper study
J. Res. Pers.
Going on stage: testosterone in greetings and meetings
J. Res. Pers.
Disinhibitory sensation seeking, personality and gonadal hormones
Pers. Individ. Diff.
Early androgen stimulation and aggressive behavior in male and female mice
Physiol. Behav.
The relationship between aggressive behavior and puberty in normal adolescents: a longitudinal study
J. Adolesc. Health
Testosterone and prolactin are associated with emotional responses to infant cries in new fathers
Horm. Behav.
Neurotransmitter and endocrine modulation of aggressive behavior and its components in normal humans
Behav. Brain Res.
Neurotransmitter-neuroendocrine responses to experimentally induced aggression in humans: influence of personality variable
Psychiatry Res.
Testosterone, cortisol and mood in a sports team competition
Horm. Behav.
The measurement of dominance in pregnant women by use of the Simple Adjective Test
Pers. Individ. Diff.
Dominance and testosterone in women
Biol. Psychol.
Hormonal regulation of aggression towards juveniles in female house mice
Horm. Behav.
Marriage and fatherhood are associated with lower testosterone in males
Evol. Hum. Behav.
Endocrine and behavioral responses to aggression and social dominance in the green anole lizard, Anolis carolinensis
Gen. Comp. Endocrinol.
Alternative reproductive strategies and tactics: diversity within sexes
TREE
Changes in saliva testosterone after psychological stimulation in men
Psychoneuroendocrinology
Sexual dimorphisms and breeding systems in pinnipeds, ungulates, primates, and humans
Modification of the social order in flocks of hens by injection of testosterone propionate
Physiol. Zool.
The effects of exogenous testosterone on sexuality and mood of normal men
J Clin. Endocr. Metab.
The behavioural biology of aggression
The influence of testosterone on human aggression
Br. J. Psychol.
Testosterone and aggression
J. Offender Rehabil.
Problems with the concept of dominance and lack of empirical support for a testosterone-dominance link
Behav. Brain Sci.
Sex and Gender
Variability among males in sexually-selected attributes
Rev. Gen. Psychol.
Cited by (944)
Testosterone's role in modulating human behaviors relevant to mating and parenting
2024, Frontiers in NeuroendocrinologyMating-related stimuli induce rapid shifts in fathers' assessments of infants
2024, Evolution and Human BehaviorPrenatal testosterone synchrony in first-time parents predicts fathers' postpartum relationship quality
2023, Hormones and BehaviorA meta-analysis of physiological stress responses to natural environments: Biophilia and Stress Recovery Theory perspectives
2023, Journal of Environmental Psychology