Research article
Testosterone reactivity to competition and competitive endurance in men and women

https://doi.org/10.1016/j.yhbeh.2019.104665Get rights and content

Highlights

  • Increases in testosterone were related to relatively higher competitive endurance.

  • Decreases in testosterone were related to relatively lower competitive endurance.

  • This effect was specific to men who gained status relative to others present.

  • This effect was found in the context of dyadic as well as group competition.

  • Trait dominance predicted competitive endurance, mediated by testosterone change.

Abstract

Transient shifts in testosterone occur during competition and are thought to positively influence dominance behavior aimed at enhancing social status. However, individual differences in testosterone reactivity to status contests have not been well-studied in relation to real-time expressions of competitive behavior among men and women. This research tests the association between changes in endogenous testosterone levels during competition and performance in terms of competitive endurance. Participant sex, social presence, and relative status outcomes (e.g., winning vs. losing) are tested as moderators of this relationship. In two studies, men and women (total N = 398) competed in the competitive will task (timed weight-holding) either individually or in the presence of an opponent (Study 1) or as a team with and without the presence of a competitor team (Study 2). Results showed a positive relationship between testosterone reactivity and performance for men, particularly those who won or ranked highest among their group – with increasing testosterone predicting better performance and decreasing testosterone predicting worse performance. For women, the effect only emerged among individuals who competed in dyads and lost. In Study 2, an exploratory mediation analysis revealed that individual differences in trait dominance predicted both testosterone reactivity to competition and task performance, with testosterone reactivity (moderated by sex and status outcome) partially explaining the direct relationship between dominance-related traits and behavior. Our goal was to examine testosterone reactivity in relation to real-time competitive effort and highlight the potential role of this relationship in explaining how individual differences in trait dominance produce competitive behavior.

Introduction

Originally proposed in the context of territorial behavior in male birds, the challenge hypothesis (Wingfield et al., 1990) guided predictions about the kinds of social encounters that produce increases in circulating testosterone. These elevations in testosterone appear to promote aggressive behaviors that benefit reproduction by increasing competitive behavior required to gain access to receptive females. Conversely, high testosterone maintained over an extended period is hypothesized to be costly in terms of health and appears to inhibit or reduce the affiliative behavior required for parental care following successful reproduction (Muller, 2017; Wingfield et al., 2001). Thus, a key premise of the challenge hypothesis is that an increase in testosterone level in response to a social encounter should be transient and occur only within contexts in which relative social position and, correspondingly, access to a limited resource, is at stake. That is, elements of the social context should play an important role in moderating the relationship between testosterone reactivity and competitive behavior.

The challenge hypothesis has been extended to humans to generate predictions about testosterone reactivity to contests for social status observed in men and women (Archer, 2006; Casto and Edwards, 2016a; Geniole and Carré, 2018). This literature supports the notion that short-term elevations in testosterone can adaptively (for the purposes of competitive success) alter competitive and aggressive behavior and underlying neural processes in humans (e.g., Carré et al., 2009, Carré et al., 2013; Goetz et al., 2014; Hermans et al., 2008; Mehta et al., 2015a; Radke et al., 2015; van Wingen et al., 2009; Wagels et al., 2018; for review, Zilioli and Bird, 2017). Consonant with predictions from the challenge hypothesis, the relationship between the testosterone response to competition and dominance-related behavior that would facilitate competitive success appears to be dependent on aspects of the social environment (for review, Carré and Archer, 2018; Geniole and Carré, 2018).

For instance, a critical aspect of the competitive social environment that affects testosterone reactivity is the competition outcome, social status gained or lost as a result of implicit or explicit rules of the contest. According to the biosocial model of status (Mazur, 1985, for review Casto and Edwards, 2016a), testosterone should increase following a contest for status for those who experience relative social victory in order to promote subsequent dominance and decrease for those who experience relative social defeat in order to promote subsequent deference. Indeed, research has shown that this designation of a winner or loser (i.e., competition outcome) has been shown to moderate the relationship between testosterone reactivity and subsequent competitive behavior. For example, in men, change in testosterone during competition has been found to be positively related with an individual's choice to compete again afterwards (against the same opponent), but only for those who lost (Mehta and Josephs, 2006). Specifically, losers experiencing increases in testosterone were more likely to choose to compete again while losers experiencing decreases in testosterone were more likely to choose the non-competitive alternative. For winners, the direction of change in testosterone was not related to whether men chose to compete again or not. In a separate study, when men were asked to decide whether or not to compete again (against a different opponent), their testosterone reactivity to competition (increasing or decreasing) positively predicted the likelihood of making the competitive choice only for those who decisively won (Mehta et al., 2015b).

In another study of men, a positive relationship between testosterone reactivity to a competition and subsequent risk-taking behavior was found only among those who won (e.g., Welker et al., 2019), while a different study found that testosterone reactivity to competition was inversely related to subsequent risk aversion regardless of prior win or loss (Apicella et al., 2014). Carré et al. (2013) found that aggressive behavior following a competition for status was predicted by testosterone reactivity during the competition in both male winners and losers, but winners were more likely to increase in testosterone and displayed greater post-competition aggression than losers. No effects were found for women who were included in this study. In fact, in much of the research discussed above, women were either not included as participants (majority of prior studies) or, women were included, but no significant relationship between testosterone reactivity and the specific status-related behavior was found (Casto and Prasad, 2017). Nonetheless, these studies suggest that a testosterone increase during competition influences subsequent dominant, aggressive, or competitive behaviors dependent on shifts in relative status. However, these kinds of behaviors are also beneficial to status achievement during the competition period itself. Despite the importance of dominance and competitiveness in determining who will win or lose, little is known about how testosterone reactivity during competition relates to ongoing, time-matched status-related behavior such as competitive effort. Indeed, much of the prior research on the winner-loser effect in the laboratory has attempted to control for individual differences in competitive motivation or effort by rigging the competition to be demonstrably easier for those who “won” than for those who “lost” (e.g., Carré et al., 2013; van der Meij et al., 2010; Zilioli and Watson, 2012; for review, Casto and Edwards, 2016a). The present research is premised on the notion that the testosterone response to competition is meaningfully tied to, and perhaps even functions to regulate, naturally occuring individual differences in competitive performance.

Given that competition-related changes in testosterone are relatively short-lived (e.g., Casto and Edwards, 2016b; Henry et al., 2017; Oxford et al., 2010) and androgen effects on tissues including the brain can be quite rapid (Michels and Hoppe, 2008; Simoncini and Genazzani, 2003), it is likely that a testosterone increase during competition would be mechanistically capable of facilitating ongoing competitive effort. Although testosterone reactivity to competition has been recently linked with time-matched changes in self-reported competitiveness and observer-rated self-assuredness in men (Kordsmeyer and Penke, 2019), the extent to which testosterone reactivity to competition predicts measures of competitive behavior expressed during that competition has, to our knowledge, not been tested in humans. That is, despite an abundance of research on testosterone's relationship to dominant and competitive behavior, we still do not know whether testosterone reactivity promotes dominant and competitive behaviors within the contest period – a time when status is actively being contested and testosterone's ability to promote dominance would be most relevant to achieving the contest-related goal.

Hormone-behavior interactions are bidirectional in nature and thus, observing competition-associated testosterone change and behavior simultaneously leaves uncertainty about the direction of the relationship. Particularly in regards to rapid mechanisms of hormonal regulation interacting in coordination with sensory-cognitive and neurological processes, causal implications are particularly difficult to disentangle. Thus, the relationship is perhaps best characterized as a rapid feedback loop in which the cascade of physiological changes producing a testosterone increase flow from the awareness that one has entered a competitive contest and ongoing awareness and competitive motivation are, in-turn, affected by these hormonal changes (Casto and Edwards, 2016a, see Fig. 5 and related discussion). Causal patterns are further complicated by additional factors that modulate direct relationships within the feedback loop such as baseline individual differences in social dominance, sensitivity to status threat, the quantity of androgen receptors, and baseline levels of testosterone and cortisol (e.g., Carré and Olmstead, 2015; Eisenegger et al., 2017; Gettler et al., 2017; Vermeer et al., 2016; Zilioli and Watson, 2012).

Nonetheless, a growing body of research experimentally manipulating testosterone levels by intranasal, oral, or transdermal administration suggests that, in certain contexts, elevated testosterone causally promotes the emergence of competitive and aggressive behavior (for recent reviews; Casto and Mehta, 2019; Geniole and Carré, 2018). For example, in one study, women given supplemental testosterone were more likely to choose to compete again than women given a placebo, but only following a social victory and only if they were high in trait dominance (Mehta et al., 2015a). Exogenously administered testosterone has been shown to elevate men's perceptions of their own physical dominance (Welling et al., 2016), an effect that would likely increase competitive efforts. Additional research shows that testosterone can rapidly activate areas in the brain involved in experiences of pleasure and reward and resulting motivated behavior (Vermeer et al., 2016; Welker et al., 2015) which is consonant with reports that testosterone reactivity is associated with task enjoyment (e.g., Mehta et al., 2015b). Although the relationship between testosterone and competitive behavior is bidirectional, prior work justifies the prediction that testosterone reactivity is a driving force underlying competitive striving.

In line with previous research about the role of social context in moderating testosterone's effect on behavior (Carré and Archer, 2018), a particularly salient aspect of the competition environment is the physical presence of an opponent or opponents. While models about the dynamic relationship between testosterone and competition (e.g., the challenge hypothesis) typically assume that emergence of this relationship requires an actual social encounter, it is unclear whether this is indeed the case. That is, the physical presence of an opponent could motivate competitive efforts and may even be an essential eliciting condition for hormone-behavior relationships in competitive contexts (Roney, 2016). Despite this notion, many laboratory studies of testosterone reactivity to competition merely give the illusion of the physical presence of an opponent such as telling the participant that a competitor is in a nearby room or showing an avatar image of a person while the ‘competitor’ actions are computer generated (for review, Casto and Edwards, 2016a). It is possible that this design feature inadvertently eliminates elements of the social context provided by the physical presence of an opponent that contribute to testosterone reactivity and competitive motivation, a prediction that has not been previously tested and that we test in this research.

Another notable omission from prior research regarding social context is the intervening role of group membership and the social presence of opponent groups, a context that is likely to reflect the inter-group condition under which testosterone mechanisms for regulating competitive behavior evolved. For social species including humans and some non-human primates, contests for status often take place within the context of social groups where placement among the social hierarchy is at stake (Cummins, 2006; Hill et al., 2008; Sapolsky, 2005; Zink et al., 2008). Further, between-group dominance contests – coalitional competitions – are common among human and non-human primate species (Crofoot and Wrangham, 2010). In these instances, group membership is likely to influence testosterone reactivity to competition and competitive behavior, as members of groups must balance dominance striving with the more cooperative behavior required for within-group coalition formation and maintenance (Diekhof et al., 2014; Flinn et al., 2012; Reimers and Diekhof, 2015; Tauer and Harackiewicz, 2004). This prediction is supported by a few studies that have found that, among men, testosterone reactivity to competition is affected by whether the contest is within or between group, with attenuated responses found for the former (Wagner et al., 2002, Oxford et al., 2010, Flinn et al., 2012; see Mehta et al., 2009 for similar effects with baseline testosterone). Moreover, in between-group competitions, men who ranked the highest among their group showed a greater testosterone increase over the course of competition, but only if their group won (Oxford et al., 2010). Although never before tested in the laboratory, the physical presence of a competitor group could effectively shift focus from within-group status conflict (beating one's own groupmates) to between-group status conflict (beating members of the other group), potentially altering the ratio of dominance/cooperativeness required for success and related testosterone reactivity.

The present research was designed to explore the relationship between individual differences in testosterone reactivity to competition and competitive persistence expressed during that competition. In two studies, participants competed in the competitive will task, in which performance was determined by the level of endurance a participant expressed during a weight-holding contest for time. Additionally, we experimentally manipulated the social presence of one or more opponents. Although the task was ‘physical’ in that it involved holding a weight, there was undoubtedly a strong psychological component in that a participant had to endure substantial discomfort to outcompete others. To validate task performance as a representation of psychological motivation, we provide evidence in the present research for a positive association between performance time and self-reported individual differences in competitiveness, dominance motivation, and task-specific motivation. Additionally, we show that physical qualities associated with strength (i.e., height, weight, body mass index) do not correlate with performance and further, that the relationship between the psychological factors discussed above and performance remains after controlling for height, weight, and body mass index. Combined, these analyses suggest that trait-level characteristics related to status motivation represent underlying psychological motivations that are expressed behaviorally, at least in part, as competitive endurance above and beyond physical strength. Critically, in this task, all participants competed against an unknown set of other participants for a small monetary grand prize given to those who performed the best overall (distributed after all subjects participated). Thus, we expected that when an opponent or opponents were present, shifts in relative status as a result of having performed better (a relative win) or worse (a relative loss) would occur during the competition as relative winners would continue competing against the unknown standard of ‘best overall performance among all the participants.’ Thus, generally, we predicted that testosterone reactivity across the task would interact with the relative win/loss to predict competitive endurance since the status gained for winners would occur before the end of the competition.

Specifically, Study 1 was designed to test the relationship between testosterone reactivity and competitive endurance in men and women competing in dyads as well competing individually against an unseen group of individuals. For participants who competed in dyads, we also tested the moderating effects of relative social victory or defeat compared to one's opponent. We predicted that testosterone reactivity would be positively associated with competitive endurance such that greater increases in testosterone would relate to relatively longer performance times and likewise, greater decreases in testosterone would relate to relatively shorter performance times. Further, for those who competed in the presence of an opponent, we predicted that the positive relationship between testosterone reactivity and competitive endurance would emerge only for those who experienced relative social victory within that context.

Study 2 was designed to examine the relationship between testosterone reactivity and competitive endurance among men and women competing in teams, one team against another, or competing against unseen opponent teams. As in Study 1, we also tested the moderating effects of relative performance-based rank among the other participants present. We predicted that larger increases in testosterone over the course of competition would be associated with longer competitive endurance while larger decreases in testosterone would be associated with shorter competitive endurance but only for those who experienced relative social victory (ranked highest among the group). Both studies included men and women so that participant sex could be tested as moderator of the relationship between testosterone reactivity and competitive behavior. Although prior research, discussed above, would suggest that this relationship should only emerge among men (within specific contexts), with substantially less prior research on women, this prediction was made with the caveat that little is known about how these variables should relate in women (Casto and Prasad, 2017).

Section snippets

Participants

A sample of 190 undergraduates (131 women and 59 men, with a mean age of 19 years, SD = 1.1, range = 18 to 25) from Emory University participated in the study. Participants were recruited from the psychology department subject pool composed of undergraduates enrolled in either of two introductory psychology courses, each of which has a research participation option as a condition for satisfactory completion of the course. To minimize potential cultural influences, only U.S. citizens whose

Study 1 results

Descriptive statistics and correlations for the principal variables of the study are shown separately for men and women in Table 1 as are between-sex statistical comparisons and comparisons between women hormonal contraceptive users and non-users. As expected, men had significantly higher testosterone levels than women. Men and women did not significantly differ in testosterone reactivity to competition whether measured as ΔT%, unstandardized residual, or AUCi. Women using hormonal

Study 1 discussion

Overall, results of Study 1 demonstrate a positive but weak relationship between testosterone reactivity and task performance in men and women, independent of whether one was competing alone or in the physical presence of an opponent. Specifically, because testosterone reactivity essentially centered around zero, those who increased in testosterone endured longer relative to those whose testosterone decreased. For those who competed in the physical presence of an opponent, and thus, were

Participants

A sample of 208 undergraduates (with a mean age of 20 years, SD = 3.0, range = 18–42) from the University of Oregon participated in the study. Seventy-nine participants identified themselves as assigned male at birth. Three of the 129 participants identifying themselves as assigned female at birth reported their current gender identity as transgender and reported using hormonal therapy. These three individuals were excluded from all analyses. Two individuals assigned female at birth reported

Study 2 results

Descriptive statistics and correlations for the main study variables for men and women are shown in Table 2. As with Study 1, salivary testosterone levels before and after competition were significantly higher for men compared to women. Men and women did not significantly differ in testosterone reactivity to competition (ΔT%, unstandardized residual, or AUCi). Women using hormonal contraceptives had significantly lower testosterone levels relative to women not using any form of hormonal

Study 2 discussion

Study 2 was designed to extend evidence on the relationship between testosterone reactivity and task performance within the group context. With groups of 3–8 participants assigned to teams, the social presence of a competitor team was manipulated and the incentive structure of the competitive will task was altered to motivate both individual and team-based competitive effort. In addition to physical presence of a competitor team, the moderating effects of participant sex and social status in

General discussion

Performance in the competitive will task was defined as the amount of time a participant endures in a weight-holding contest for time. Endurance requires a combination of physical and mental components that interact to aid in the expression of competitive “effort” in nearly all contests for time. Indeed, holding a weight for time and has been previously and independently validated as a measure of mental toughness, a psychological construct in which competitive persistence is a core feature (

Conclusion

In the present research, we introduced a novel competitive task designed to determine individual differences in competitive endurance. Performance in the competitive will task reflects personal characteristics that would be advantageous for success in dominance contests in both early (primitive) and modern human social contexts. Across two studies and a relatively large sample, the present research provides original evidence of a positive relationship, in men, between testosterone reactivity to

Transparency and open science practices

All supplementary results, analysis code, and data are available on the open science framework (https://osf.io/6r7da/).

Acknowledgments

This research was funded in part by the National Science Foundation, grant No. 1451848 (to P.H.M.).

References (105)

  • J.M. Carré et al.

    Exogenous testosterone rapidly increases aggressive behavior in dominant and impulsive men

    Biol. Psychiatry

    (2017)
  • K.V. Casto et al.

    Testosterone, cortisol, and human competition

    Horm. Behav.

    (2016)
  • K.V. Casto et al.

    Recommendations for the study of women in hormones and competition research

    Horm. Behav.

    (2017)
  • J.T. Cheng et al.

    Pride, personality, and the evolutionary foundations of human social status

    Evol. Hum. Behav.

    (2010)
  • K.D. Cobey et al.

    Hormonal contraceptive use lowers female intrasexual competition in pair-bonded women

    Evol. Hum. Behav.

    (2013)
  • D.A. Edwards et al.

    Oral contraceptives decrease saliva testosterone but do not affect the rise in testosterone associated with athletic competition

    Horm. Behav.

    (2009)
  • A.B. Eisenbruch et al.

    Conception risk and the ultimatum game: When fertility is high, women demand more

    Pers Individ. Differ.

    (2016)
  • C. Eisenegger et al.

    Testosterone and androgen receptor gene polymorphism are associated with confidence and competitiveness in men

    Horm. Behav.

    (2017)
  • S.N. Geniole et al.

    Human social neuroendocrinology: review of the rapid effects of testosterone

    Horm. Behav.

    (2018)
  • S.N. Geniole et al.

    Testosterone dynamics and psychopathic personality traits independently predict antagonistic behavior towards the perceived loser of a competitive interaction

    Horm. Behav.

    (2013)
  • S.N. Geniole et al.

    Effects of competition outcome on testosterone concentrations in humans: an updated meta-analysis

    Horm. Behav.

    (2017)
  • L.T. Gettler et al.

    The role of testosterone in coordinating male life history strategies: The moderating effects of the androgen receptor CAG repeat polymorphism

    Horm. Behav.

    (2017)
  • S.M. Goetz et al.

    Testosterone rapidly increases neural reactivity to threat in healthy men: a novel two-step pharmacological challenge paradigm

    Biol. Psychiatry

    (2014)
  • E. Gonzalez-Bono et al.

    Testosterone, cortisol, and mood in a sports team competition

    Horm. Behav.

    (1999)
  • A.C. Hahn et al.

    A longitudinal analysis of women’s salivary testosterone and intrasexual competitiveness

    Psychoneuroendocrinology

    (2016)
  • A. Henry et al.

    Performance during competition and competition outcome in relation to testosterone and cortisol among women

    Horm. Behav.

    (2017)
  • E.J. Hermans et al.

    Exogenous testosterone enhances responsiveness to social threat in the neural circuitry of social aggression in humans

    Biol. Psychiatry

    (2008)
  • T.L. Kordsmeyer et al.

    Effects of male testosterone and its interaction with cortisol on self-and observer-rated personality states in a competitive mating context

    J. Res. Pers.

    (2019)
  • P.H. Mehta et al.

    Testosterone change after losing predicts the decision to compete again

    Horm. Behav.

    (2006)
  • P.H. Mehta et al.

    When are low testosterone levels advantageous? The moderating role of individual versus intergroup competition

    Horm. Behav.

    (2009)
  • P.H. Mehta et al.

    Exogenous testosterone in women enhances and inhibits competitive decision-making depending on victory–defeat experience and trait dominance

    Psychoneuroendocrinol.

    (2015)
  • G. Michels et al.

    Rapid actions of androgens

    Front. Neuroendocrinol.

    (2008)
  • E.R. Montoya et al.

    How oral contraceptives impact social-emotional behavior and brain function

    Trends Cogn. Sci.

    (2017)
  • M.N. Muller

    Testosterone and reproductive effort in male primates

    Horm. Behav.

    (2017)
  • S. Otten

    The Minimal Group Paradigm and its maximal impact in research on social categorization

    Curr. Opin. Psychol.

    (2016)
  • J. Oxford et al.

    Hormonal responses differ when playing violent video games against an ingroup and outgroup

    Evol. Hum. Behav.

    (2010)
  • M. Pearson et al.

    Menstrual cycle and competitive bidding

    Games Econom. Behav.

    (2013)
  • S. Prasad et al.

    Unstable correspondence between salivary testosterone measured with enzyme immunoassays and tandem mass spectrometry

    Psychoneuroendocrinology

    (2019)
  • J.C. Pruessner et al.

    Two formulas for computation of the area under the curve represent measures of total hormone concentration versus time-dependent change

    Psychoneuroendocrinology

    (2003)
  • D.A. Puts

    Mating context and menstrual phase affect women’s preferences for male voice pitch

    Evol. Hum. Behav.

    (2005)
  • J.R. Roney

    Theoretical frameworks for human behavioral endocrinology

    Horm. Behav.

    (2016)
  • L.A. Rudman et al.

    Backlash effects for disconfirming gender stereotypes in organizations

    Res. Organ. Behav.

    (2008)
  • O.C. Schultheiss et al.

    Implicit power motivation predicts men’s testosterone changes in implicit learning in a contest situation

    Horm. Behav.

    (2002)
  • B.C. Trainor et al.

    Opposing hormonal mechanisms of aggression revealed through short-lived testosterone manipulations and multiple winning experiences

    Horm. Behav.

    (2004)
  • L. van der Meij et al.

    Testosterone responses to competition: the opponent’s psychological state makes it challenging

    Biol. Psychol.

    (2010)
  • J.D. Wagner et al.

    Hormonal response to competition among male coalitions

    Evol. Hum. Behav.

    (2002)
  • K.M. Welker et al.

    A comparison of salivary testosterone measurement using immunoassays and tandem mass spectrometry

    Psychoneuroendocrinology

    (2016)
  • L.L. Welling et al.

    Exogenous testosterone increases men’s perceptions of their own physical dominance

    Psychoneuroendocrinology

    (2016)
  • I. Wiegratz et al.

    Effect of two oral contraceptives containing ethinylestradiol and gestodene or norgestimate upon androgen parameters and serum binding proteins

    Contraception

    (1995)
  • S. Andersen et al.

    Gender, competitiveness, and socialization at a young age: evidence from a matrilineal and a patriarchal society

    Rev. Econ. Stat.

    (2013)
  • Cited by (25)

    • Hormonal contraceptives as disruptors of competitive behavior: Theoretical framing and review

      2022, Frontiers in Neuroendocrinology
      Citation Excerpt :

      One of the four behavioral studies reported no significant difference between HC users and NC women, though the direction and size of the effect was consistent across all laboratory contexts (Casto et al., 2020, study 2). Taken together, Bradshaw et al. (2020) and Casto et al. (2020) demonstrate that HC users have lower persistence across cognitive and competitive physical tasks than NC women, thus strengthening the generalizability of these effects. One factor which may influence dominance-based status seeking, particularly for studies that assess intrasexual competitiveness, dominance and dehumanization, is relationship status.

    • Testosterone fluctuations in response to a democratic election predict partisan attitudes toward the elected leader

      2021, Psychoneuroendocrinology
      Citation Excerpt :

      The present research also examined the association between testosterone responses and attitudes toward the winner among the winner’s supporters. According to the biosocial model of status, testosterone responses to competition should promote dominance directed toward opponents in pursuit of social status (Carré et al., 2013; Casto et al., 2020; Oyegbile and Marler, 2005; cf. Apicella et al., 2014). For supporters of a winning political candidate, attitudes toward the winner are expected to reflect attitudes toward an ingroup leader, not an opponent.

    View all citing articles on Scopus

    This paper belongs to special issue “Challenge Hypothesis 30th”.

    View full text