seX & whY Episode 8: Influence of Testosterone and Cortisol on Decision Making

Jeannette WolfePodcast Episodes

Show Notes for Podcast Eight of seX & whY

Host: Jeannette Wolfe

Guest: Dr. John Coates

Topic: The Influence of Testosterone and Cortisol on Decision Making, With Neuroscientist Dr. John Coates

Dr. John Coates is a neuroscientist and author of The hour between dog and wolf- how risk taking transforms the body and mind.  He is an ex-trader and now runs Dewline Research. He studies how subtle unconscious changes in an individual’s physiology can shift their decision making and is particularly interested in the roles of testosterone and cortisol.  He is specifically focused on how the fluctuation of these hormones might influence volatility in the stock market. As it appears that both successful traders and emergency medicine are required to make high impact decisions in novel and often unpredictable situations, I think there is much we can learn from his work and I am thrilled he could join us for this discussion.

Before we delve in, I’d like to remind folks that my interest in this material is to better understand how individuals and teams can optimize their performance under stress. The material we are covering in this podcast- the possible influence of sex hormones on decision making- is undoubtedly going to make some listeners uncomfortable. I truly believe, however, that this topic is important and deserves an honest and curious appraisal. To be absolutely clear, I do not believe that there is a better sex equipped with a better brain, rather that there are simply different neurobiological ways that different brains use to approach and complete similar tasks. My goal here, is for us to develop better insight into how we individually react under different high stress scenarios. Hopefully, we can then use this information to explore new ways to play up our individual strengths and mitigate potential vulnerabilities. Let’s get started.

Over the years, Dr. Coates and his team have  conducted some pretty interesting “field work” studies especially his 2008 study on London short traders. In that study his team took twice daily saliva samples in 17 male traders over an 8 day period and found:

  • Both cortisol and testosterone levels varied greatly throughout the study
    • Mean daily cortisol levels increased as much as 400%
    • Afternoon cortisol levels increased as much as 500% (in an unstressed individual cortisol typically peaks in the early morning.)
  • Elevated AM testosterone levels correlated with afternoon profitability
  • Elevated cortisol levels correlated with market volatility (but interestingly not with simple losses)

Since then he has done several additional studies and concludes that the only way to really understand the bubbles and crashes of the stock market is by better understanding the human physiology of the traders.  Here are some of his take home points.

  • An individual’s risk preference is probably far more dynamic than previously believed and is impacted by subtle, unconscious, shifts in physiology
  • Individuals can have different risk preferences in different domains (participate in dangerous hobbies but are conservative with their finances)
  • Individuals with increased interoceptive awareness may be quicker to recognize anomalous blips of data buried within piles of “expected” information. This may contribute to the phenomenon of a “gut instinct”
  • Hormonal fluctuations likely contribute to risk preferences
    • Increasing testosterone levels likely shifts risk preferences to make individuals more open to riskier endeavors
      • Young males in competitive situations may be particularly vulnerable as they have significantly higher levels of baseline testosterone than women and older men
      • This risk shift is likely even more dramatic in individuals taking unnecessary testosterone supplementation (which is now a 2 billion dollar industry with 2/3 of the individuals who use testosterone not having a medically indicated reason for taking it.)
    • Increasing cortisol levels (in particular chronically increased levels) likely shifts risk preferences in the opposite direction and makes individuals act more risk adverse.
    • As these hormonal shifts are occurring unconsciously, it is difficult for individuals themselves to recognize their behavioral shift and depending upon the situation external safeguards (perceptive team members, monitoring systems) could be helpful.

“Winner’s Streaks”

– In the research community there is still some controversy as to whether this phenomenon even exists or if such streaks simply represent statistical outliers that are selectively remembered due to their unusualness.

– Coates strongly believes that winner’s streaks are real and are crucial to understanding behavior under certain circumstances.

– There is good data in the animal kingdom to suggest that if two male animals are in a competition and if their size, motivation (i.e. being hungry versus well fed) and baseline aggression are all controlled, that the animal who wins that encounter will be statistically more likely to go on and win their next competitive encounter.

Some theories as to why this might occur:

  • Actual competition gives each opponents and idea of how they might stand in future altercations
  • Winners self-perception of their strengths increases, and they become more comfortable with additional confrontation
  • The initial victory may physically increase the winner’s resources allowing it to go into its next encounter with an advantage (i.e. access to more food increases its size)
  • A potential physiological contributor to a winner’s streak may be real time fluctuations in an individuals’ testosterone levels (and possibly a change in the sensitivity of their testosterone receptors). Although many things can cause fluctuations in testosterone levels, two things that appear to consistently elevate it are competition and winning.

Over a period of time, consistently elevated testosterone levels might offer an advantage by increasing:

  • muscle mass
  • hemoglobin/oxygen capacity
  • confidence, persistence and increased risk taking
  • desire to seek out novelty

Like most hormones, however, testosterone’s effects likely plot out on an inverted U shape curve in that depending on the circumstances:

  • small increases of testosterone levels might be advantageous as a slight increase in risk tolerance may lead to increased reward
  • at some point, however, risk becomes excessive and becomes a disadvantage
    • in animal research this may lead to:
      • patrolling of unrealistically large areas
      • increasing exposure to dangerous situations
      • increasing fighting
      • neglecting parental duties
      • loss of energy stores
    • Research in humans shows that increasing testosterone levels
      • Increases risk preference
      • Quickens reaction time
      • Defaults to automatic thinking
    • In high levels, especially if given exogenously can lead to
      • Euphoria
      • Mania
      • Impulsivity
      • Sensation seeking


Specific research done by Coates and his team

Tennis experiment

Question addressed: Are “winning streaks” a real phenomenon or simply statistical outliers?

What they did- Looked at large data base of historical tennis matches in which players who were similarly ranked went into an extended tiebreaker involving more than 20 points in the first set and in which the winner was determined by only two points. (They did this to essentially try and show that on the day of their competition that not only were both players similarly ranked but that they were also playing at a similar level- i.e. both were having a “good day”)

Results- Men (N=235 matches) who won their first set were 60% more likely to win second set but no significant difference in second set victory was found amongst women (N= 140), suggesting that this might be driven by testosterone as women have about 5-10% level of men.

Cortisol study

In this study Coates and his team were interested in how an acute and a chronic elevation in stress hormones might affect risk preference. Using data from one of their previous studies which showed that during a period of increased market volatility that traders had a 68% increase in their daily cortisol levels, they went back to the lab to try and replicate this finding and then test decision making in a more controlled environment.

What they did: randomized double-blind placebo controlled cross over-study involving 20 men and 16 women. In treatment arm, volunteers were given weight- based hydrocortisone 3x a day for 8 days to mimic cortisol increases seen in traders. All participants played a lottery style game in which they could choose an option in which they had a lesser chance of winning but a higher pay out if they did, or a less risky option in which they had an overall increased chance of winning but at a lower expected payout. The game was played after acute and chronic dosing.

Findings- they did not find a difference in risk preference amongst volunteers after they received their initial hydrocortisone (as an aside, the literature on risk preference after acute cortisol increase is somewhat inconsistent) but in this study they did find that after 8 days of taking exogenous steroids that individuals became much more risk adverse and that men were affected more so than women.

Thoughts as to why chronically elevated steroids change our decision making

  • Physical changes occur in the hippocampus that impair normal functioning (neurogenesis is suppressed and dendritic spines are reduced )
  • Similarly, changes also occur in the prefrontal cortex
    • Negatively affect working memory
    • Decrease attentional control
    • Impair behavioral flexibility
  • The amygdala, on the other hand, revs up, causing increased dendritic connections and increase corticotropin releasing hormone gene expression
  • Bundled all together this may lead to:
    • Increased focus on imagined threat
    • Increased risk of anxiety, depression, and learned helplessness
    • Shift to habitual behavior and decreased motivation to try novel action

Using this data, Coates theorizes that prolonged periods of financial uncertainty in the stock market likely cause traders’ cortisol levels to increase and stay increased leading to an aversion to risk or an “irrational pessimism” that left unchecked can lead to a bear market.

Finally, attached below is a reference to a recent review article that Dr. Coates wrote summarizing his theories as to the relationship between cortisol and testosterone on bull and bear markets and emphasizing the importance of field work in scientific discovery and refinement.

To learn about some complementary research being done at Wharton check out this interview with Gideon Nave and Amos Nadler in which they discuss their recent work evaluating decision making in men using exogenous testosterone. They found that that although certain cognitive functions appeared unaffected (like doing math problems), men who were given testosterone gel were more likely to rely on their gut instinct when answering questions. Which, again, depending upon the circumstances could be potentially helpful or harmful.

Coates, J. M., & Herbert, J. (2008). Endogenous steroids and financial risk taking on a London trading floor. Proceedings of the National Academy of Sciences of the United States of America, 105(16), 6167–72.

Kandasamy, N., Hardy, B., Page, L., Schaffner, M., Graggaber, J., Powlson, A. S.,Coates, J. (2014). Cortisol shifts financial risk preferences. Proceedings of the National Academy of Sciences of the United States of America, 111(9), 3608–13.

Page, L., & Coates, J. (2017). Winner and loser effects in human competitions. Evidence from equally matched tennis players. Evolution and Human Behavior.

Coates, J., & Gurnell, M. (2017). Combining field work and laboratory work in the study of financial risk-taking. Hormones and Behavior, 92, 13–19.