Testosterone as a personality variable

The growth of personality research has led to the integration of consistent variation of individual behaviour in multidimensional approaches including physiological variables, which are required to continue building a more comprehensive theory about coping strategies. In this study, we used wild-caught males of Ctenomys talarum (tuco-tucos), a solitary subterranean rodent, to assess the relationships among personality traits and several physiological variables, namely stress response, testosterone, immunity, and energy metabolism. Subjects (n = 21) were used in experimental tests assessing behaviour, energy metabolism, testosterone levels, inflammatory cell-mediated and humoral immunity, and stress response to a simulated predator attack. The structural equation model explained a moderate portion of the variance of personality behaviours related to activity (52%), boldness (35%), and socioaversion (30%). More active and bold individuals showed higher oxygen consumption. While those subjects had lower baseline cortisol levels, there was no relationship between cortisol levels of the stress-induced response. Cell-mediated immune response was related to activity levels. Finally, testosterone only affected boldness. Despite some of these relationships diverge in direction to predicted ones, overall they support the existence of coping styles in male C. talarum; and are discussed in the light of current hypotheses and particular behavioural and ecological traits of tuco-tucos.

The results of a recent study, Félix et al., 2020, give new information about the behavioral and endocrine correlates of individual differences in the potential for androgen response in male cichlid fish. We think the study raises issues that are pertinent to the study of hormones and competition in other species, particularly humans. Focusing mostly on androgen reactivity to social challenge, we emphasize the importance of inter-individual variability in physiology, personality, and motivation in studies of hormone responses to social encounters. Additionally, we give special attention to matters of “repeatability” and the timing of hormone sampling. We conclude with an appreciation of the value of comparative analysis in behavioral endocrinology.

Thirty-five women participating in one or more intramural flag-football games provided saliva samples before, immediately after, and 10 min after competition and completed an after-competition questionnaire appraising their own performance during the game. As seen in other studies of elite athletes, these recreational athletes, on average, showed significant elevations in testosterone (T) and cortisol (C) across the competition period – the “competition effect”. In winners and losers, T levels at all time points measured were positively related to athletes' appraisals of their own individual performance. Results from this study show that the competition effect for T and C is evident in recreational women athletes and provide preliminary evidence about the relationship between cognitive appraisal and competition-related T levels.

This study assessed sexual orientation and psychobiological stress indices in relation to salivary sex hormones as part of a well-validated laboratory-based stress paradigm. Participants included 87 healthy adults that were on average 25 years old who self-identified as lesbian/bisexual women (n = 20), heterosexual women (n = 21), gay/bisexual men (n = 26), and heterosexual men (n = 20). Two saliva samples were collected fifteen minutes before and fifteen minutes after exposure to a modified Trier Social Stress Test to determine testosterone, estradiol, and progesterone concentrations via enzyme-immune assaying. Mean sex hormones were further tested in association to stress indices related to cortisol systemic output (area under the curve with respect to ground) based on ten measures throughout the two-hour visit, allostatic load indexed using 21 biomarkers, and perceived stress assessed using a well-validated questionnaire. Results revealed that lesbian/bisexual women had higher overall testosterone and progesterone concentrations than heterosexual women, while no differences were found among gay/bisexual men in comparison to heterosexual men. Lesbian/bisexual women and heterosexual men showed positive associations between mean estradiol concentrations and allostatic load, while gay/bisexual men and heterosexual women showed positive associations between mean testosterone and cortisol systemic output. In summary, sex hormone variations appear to vary according to sexual orientation among women, but also as a function of cortisol systemic output, allostatic load, and perceived stress for both sexes.

The dual-hormone hypothesis posits that the effect of testosterone on social behavior is moderated by cortisol. The present study tested this hypothesis with a competitive reactive aggression paradigm in 53 healthy undergraduate women. Salivary cortisol and testosterone were assessed at baseline. Participants were personally insulted and subsequently given the opportunity to retaliate by administering blasts of white noise to the provocateur. Participants were randomly assigned to win or lose the aggressive competition. Basal testosterone positively predicted reactive aggression and state dominance, but only among participants with high concentrations of basal cortisol. The corresponding, reverse pattern was found for state submissiveness. Winners also had higher concentrations of testosterone than losers following the aggressive competition. We discuss the role of heightened reactivity to social provocation as a possible explanation for these effects.

Individuals within species differ in their behavioural reactions to the environment. Consistent individual differences in these behaviours (personality traits) are often correlated and known to be under natural selection. These differences are frequently associated with variation in physiological traits, such as endocrine profiles. For example, variation in circulating testosterone levels is associated with variation in several personality traits and has been hypothesized to be a marker for personality in humans and rodents. The importance of testosterone in controlling both behavioural strategies and individual physiological differences suggests that direct selection on personality traits might cause pleiotropic selection on the physiological mechanisms underlying these traits. To test this hypothesis, we quantified levels of plasma testosterone levels and measured phytohaemagglutinin (PHA)-induced immune responses of male great tits, Parus major, in lines artificially selected for diverging levels of avian personality (‘fast’ and ‘slow’ exploratory behaviour). We found that testosterone levels were highly repeatable within individuals and fluctuated predictably over the season. Contrary to our expectations, ‘slow’ explorers had consistently higher levels of baseline testosterone and higher immune responses than ‘fast’ explorers. These results show that phenotypic selection for variation in personality traits corresponds to consistent differences in hormone profile and immune function, but that higher aggression levels do not need to be associated with higher baseline testosterone levels. Our results confirm that personality traits have evolved as a result of selection on both the underlying controlling physiological mechanisms and the phenotypic traits.