Gender differences in testosterone and cortisol response to competition
Introduction
The role of testosterone and cortisol in competition has attracted substantial attention for more than two decades. Studies have focused on a variety of competitive events including wrestling (e.g. Elias, 1981), tennis (e.g. Booth et al., 1989), chess (e.g. Mazur et al., 1992), judo (e.g. Salvador et al., 1999), and rugby (e.g. Bateup et al., 2002), as well as on competitive tasks in laboratory settings (e.g. Gladue et al., 1989, Mazur et al., 1997). Guiding the research is a seminal theoretical model of gaining, maintaining, and losing status than entails both dominance and stress (Mazur, 1985). The essence of the model is that to gain and maintain status, the competitor must out-stress the opponent. This is achieved through dominant behavior (e.g. staring, threatening gestures or speech, physical aggression, or making a superior play). The frequently observed association between androgens and dominance behavior has led researchers to focus on testosterone. Stress, as indexed by cortisol (a reliable measure of hypothalamic–pituitary–adrenal axis activation) is important both in preparing for competition as well as assessing resilience to stress created by the opponent. Mild increases in cortisol prepare individuals for action and lower cortisol levels or reactivity may indicate more resilience to stressful situations (Stansbury and Gunnar, 1994). These factors led researchers to include cortisol in competition studies. Sporting events are the venue of choice to study hormones and competition because the setting is highly organized and monitored, the rules are clear, and measures of performance, such as winning and losing, are well defined.
The purpose of this study is to address three questions that will significantly extend knowledge of the biosocial model as it pertains to gaining, maintaining, and losing status. (1) Are the associations between testosterone, cortisol, and competition the same for women as they are for men? Preliminary studies provide a mixed picture. (2) Is the link between hormones and competition the same for inexperienced competitors as it is for elite players? Elite competitors have been the subjects of most of the research to date. (3) Are the associations between hormones and competition related to cognitive activity? To date few studies examine the links between hormones and constructs such as attitude, mood, and mental preparation in the context of competition. In cases where they have been studied, few associations were detected. Hormone research has focused on three time bound phases of competition: pre-event, event, and post-event phases. In keeping with prior studies we examine hormone production and hormone-behavior associations by event phase.
Section snippets
Anticipation of competition during the pre-event phase
Competitors anticipate an event by engaging in a wide variety of efforts to mentally prepare for competition. There may also be physical preparation for some events but this is limited in order to conserve energy. Many hormone researchers define the pre-event period as the 24-h immediately preceding competition. This time frame is defined so that biological samples can be obtained at the same time as the event on a non-competition day, allowing observation of changes in hormones uncompromised
Reaction to competition during the event phase
To our knowledge, few studies have examined hormone change and performance while an event was in progress. Rather, the focus has been on post-event indices of performance, such as winning or losing, and the moods associated with these outcomes. Higher testosterone has been associated with winning and elation immediately following the event and lower testosterone associated with losing (Booth et al., 1989, Elias, 1981, Mazur et al., 1992). Winning is not consistently associated with increased
After the competition during the post-event phase
Research on the post-event phase focuses on the amount of time that elapses between the end of the event and when hormones return to levels appropriate for the time of day. For example, testosterone has been observed to decline more slowly among winners than losers (Booth et al., 1989). This finding has been interpreted as having functional significance. Winners often face immediate challenges from others. The higher testosterone keeps them prepared to deal with new competitors. For the losers,
Hypotheses and predictions
The study described here explored three basic questions as they apply to each phase of competitive events. First, it examined sex differences in hormone production throughout the course of a competitive event. We predict that a similar pattern of cortisol production will be observed in both sexes, but that there will be marked differences in the pattern of testosterone production during the competition. Second, we studied the extent to which previous competitive experience was linked to
Methods
Competitive ergometer rowing is an ideal context in which to study associations between gender, experience, cognitive activity, performance, and hormone production. Although the majority of rowing competitions occur outdoors, indoor competition using a machine that simulates rowing is a regular occurrence during the winter months. Such competitions challenge individuals to row 2000 m in as brief a time as possible. Rowers are ranked according to race time with the fastest achieving first place.
Analytical strategy
For the pre-event phase, time-matched baseline and pre-competition samples were compared to determine whether hormone levels were elevated in anticipation of the competition. During the event phase, reactivity was operationalized as the extent to which salivary hormone levels increased from the pre- to the 20-min post-competition sample and maintenance was defined as the change from 20 to 40-min post-competition sample. Within each event phase, analyses were conducted in two steps. First, group
Discussion
The overwhelming pattern was that the behavioral correlates of, and endocrine responses to, this competitive setting varied significantly by gender. The nature of the gender difference was complex. The pattern was influenced by whether athletes had prior competitive experience; whether the measurement was recorded in anticipation, in response, or after the competitive event; and which particular hormone was involved. This complexity raises many new research questions and suggests that revisions
Implications for the biosocial model of status
Taylor et al. (2000) did not specifically discuss biobehavioral responses to competition, nor did they implicate cortisol or testosterone in their ‘Tend and Befriend’ treatise. They focused their argument primarily around oxytocin. Nevertheless, in this and other studies (Bateup et al., 2002), we report relationships that seem consistent with some of the ideas in Taylor et al. (2000). Particularly relevant is the pattern of gender-related differences in the association between social bonding
Future directions
Although the study made some important contributions, it could be improved with a larger number of competitors and events. Such an increase in numbers would provide the statistical power to more completely explore individual differences within subgroups defined by the factorial combinations of gender and experience level studied here. In addition, because intense physical activity even in the absence of competition can influence hormone levels (Suay et al., 1999), it is possible that some
Conclusion
In conclusion, the present study has contributed uniquely to our understanding of hormone-behavior relationships. More than 50 years of basic research reveals that the mechanism linking hormones and behavior is very complex. Contemporary behavioral endocrinologists assume that in humans the link is not simply a biological ‘cause-and-effect’ mechanism, rather it is best described as a bi-directional association that is highly dependent on intrinsic individual differences in social perception,
Acknowledgements
This research was supported in part by the Behavioral Endocrinology Laboratory and the Population Research Institute (PRI) at The Pennsylvania State University. We wish to acknowledge Stefanie Bachman, Erica Schreffler, Michelle Gandhi, Linda Boseska, Kelly Weaver, Christine Cohen, and Ann Morrisey for their assistance with data collection. Thanks are due to Mary Curran for biotechnical support with immunoassays. We are grateful to the Penn State Crew Team members who participated in the
References (36)
- et al.
Testosterone, cortisol, and women's competition
Evolution and Human Behavior
(2002) - et al.
Endogenous testosterone and competition: the effect of ‘fasting’
Steroids
(1993) - et al.
Testosterone, and winning and losing in human competition
Hormones and Behavior
(1989) - et al.
Glucocorticoid regulation of diverse cognitive functions in normal and pathological emotional states
Neuroscience and Biobehavioral Reviews
(2003) - et al.
Testosterone, cortisol, and mood in a sports team competition
Hormones and Behavior
(1999) - et al.
Testosterone, status, and mood in human males
Hormones and Behavior
(1980) - et al.
Sex difference in testosterone response to a video game contest
Evolution and Human Behavior
(1997) - et al.
Winning, losing, mood, and testosterone
Hormones and Behavior
(1992) - et al.
Correlating testosterone and fighting in male participants in judo contests
Physiology and Behavior
(1999) - et al.
Anticipatory cortisol, testosterone and psychological responses to judo competition in young men
Psychneuroendocrinology
(2003)