Testosterone and domain-specific risk: Digit ratios (2D:4D and rel2) as predictors of recreational, financial, and social risk-taking behaviors

https://doi.org/10.1016/j.paid.2010.07.003Get rights and content

Abstract

Prenatal testosterone has important effects on brain organization and future behavior. The second-to-fourth digit length ratio (2D:4D), a proxy of prenatal testosterone exposure, has been linked to a wide variety of sexually differentiated dispositions and behaviors. We examine the relationship between digit length ratios (2D:4D and rel2, the length of the second finger relative to the sum of the lengths of all four fingers) and risk-taking behaviors across five domains: financial, social, recreational, ethical, and health. In a sub-sample of male Caucasians (ethnically homogeneous), lower rel2 was predictive of greater financial, social, and recreational risk-taking, whereas lower 2D:4D was predictive of greater risk-taking in two domains (social and recreational). In the full male sub-sample (ethnically heterogeneous), the only significant correlation was a negative association between 2D:4D and financial risk. A composite measure of risk-taking across all five domains revealed that both rel2 and 2D:4D were negatively correlated with overall risk-taking in both male sub-samples. No significant correlations were found in the female sub-samples. Finally, men were more risk-seeking than women across all five contexts.

Introduction

While individuals vary a great deal in their tendency to take risks, men tend to engage in more risky behavior than women across a variety of contexts (Byrnes et al., 1999, Wang et al., 2009). Male financial investors, for instance, tend to weigh risk attributes less heavily and recommend riskier choices when building financial portfolios compared to their female counterparts (Olsen & Cox, 2001). Likewise, men are more risk-seeking in gambling tasks (van Leijenhorst, Westenberg, & Crone, 2008) and are more likely to succumb to pathological gambling (Saad, 2007, chap. 6; Volberg, Abbott, Ronnberg, & Munck, 2001). Furthermore, greater physical risk-taking among men leads to higher probabilities of mortality from motor vehicle accidents and homicide (Owens, 2002). Men’s greater penchant for risk-taking has sparked an interest in exploring the role that testosterone plays in risky behavior. In the current paper, we examine the association between a proxy of prenatal testosterone exposure (digit length ratio) and risk-taking behavior across several domains.

A few studies have investigated the link between circulating testosterone and risk-taking propensity. For instance, testosterone levels have been linked to risky antisocial and delinquent behaviors in adolescent boys (Rowe et al., 2004, Vermeersch et al., 2008), and health-related risk-taking in adult males (smoking, having multiple sex partners, drug use, and alcohol abuse; Booth, Johnson, & Granger, 1999). A popular vein of inquiry in the testosterone and risk-taking area has been to examine risk-taking within the financial realm. Coates and Herbert (2008) showed that testosterone levels of male financial traders measured in the morning were predictive of their profitability that day. White, Thornhill, and Hampson (2006) demonstrated that circulating testosterone was positively related to new venture creation among males, and the effect was partially mediated by risk-taking propensity. In a sample of male students, Apicella et al. (2008) found that testosterone levels correlated positively with financial risk-taking preferences in a monetary investment task. Similarly, in a female sample, van Honk et al. (2004) found that a sublingual administration of testosterone elicited riskier and less advantageous financial choices in the Iowa Gambling Task than did a placebo. Recently, Sapienza, Zingales, and Maestripieri (2009) showed that MBA students with higher testosterone levels were more likely to choose a career in finance than a career in a less risky field after graduation. They also reported that financial risk-taking preferences were positively correlated with testosterone levels among female students, but not among males. Zethraeus et al. (2009) found no significant effects of testosterone treatments over four weeks on risk aversion in a sample of postmenopausal women. On the whole, the literature suggests a positive relationship between circulating testosterone levels and risk-taking proclivity, though some studies have yielded either mixed or null effects.

Whereas the direct measurement and manipulation of circulating testosterone has generated valuable insights regarding the ‘activational’ role of testosterone on risk-taking, other studies have focused on developmental, or ‘organizational’ effects. Testosterone plays a critical organizational role in masculinization both prenatally and at puberty. During puberty, testosterone exposure is essential for the suite of masculinizing effects associated with this developmental stage (Archer, 2006, Mazur and Booth, 1998). Prenatal testosterone exposure influences fetal brain organization and future sexually differentiated behaviors (Archer, 2006, Auyeung et al., 2009, Udry, 2000). This exposure also seems to reduce the growth of the second digit relative to the other fingers (Lutchmaya et al., 2004, Manning et al., 1998). As a result, the second-to-fourth digit ratio (2D:4D) has been used as a proxy of both the exposure and sensitivity to prenatal testosterone (Manning, 2002, Manning et al., 2003). This association has spurred considerable interest in 2D:4D, which has been linked to an array of masculine traits including aggression (Bailey & Hurd, 2005), athletic ability (Manning & Hill, 2009), and perceived dominance (Neave, Laing, Fink, & Manning, 2003). Even among females, a lower 2D:4D tends to predict masculine behavioral traits (Brown et al., 2002, Clark, 2004, Paul et al., 2006). However, some masculine proclivities appear to exhibit no robust relationship to 2D:4D (cf. Voracek, Tran, and Dressler’s (2010) meta-analysis on sensation-seeking).

Of particular relevance to the current work, Schwerdtfeger, Heims, and Heer (2010) showed that 2D:4D was negatively correlated to traffic violations, suggesting that highly androgenized males engage in riskier driving behavior. Coates, Gurnell, and Rustichini (2009) found that male traders with lower 2D:4D performed better than men with higher digit ratios. The authors speculate that part of this association could stem from a greater risk-taking proclivity in low 2D:4D individuals. Coates and Page (2009) obtained evidence in support of this relationship by showing a negative correlation between 2D:4D and the level of risk taken by high frequency male traders. Further, Dreber and Hoffman (2007) showed that a lower, more masculine 2D:4D was associated with a greater preference for financial risk in an ethnically homogeneous mixed-sex sample (controlling for sex) in Sweden albeit no such effect was uncovered for a heterogeneous mixed-sex sample in the US. Apicella et al. (2008) also found that financial risk was not significantly correlated with 2D:4D in an ethnically heterogeneous male sample. The authors conjecture that ethnic heterogeneity and the small size of their sample might have made it impossible to detect a significant digit ratio effect. Finally, Sapienza et al. (2009) did not obtain a significant correlation between 2D:4D and financial risk-taking in a mixed-sex sample of students. However, they reported that students with lower 2D:4D were significantly more likely to select a career in finance than in a less risky field. The equivocal findings reported in the three aforementioned studies may in part be due to the use of ethnically heterogeneous samples, as suggested by Apicella et al. (2008). Manning, Churchill, and Peters (2007) showed that amalgamating the data of different ethnic groups can eliminate digit ratio effects, as the correlation between sexual orientation and 2D:4D was found in certain ethnic groups but not in others. The current study addresses this issue by examining associations between digit ratios and risk-taking propensity in a large, ethnically heterogeneous sample and comparing effects among heterogeneous versus homogeneous sub-samples.

Thus far, much of the research investigating the links between digit ratio and risk-taking proclivity has focused solely on risk preferences within a financial context. Risk-taking preferences are assessed via a financially-related measure, subsequent to which the findings are generalized to all domains of risk (i.e. one index of risk is associated equally to all risk-related contexts). While this operationalization of risk preferences is consistent with the domain-general assumptions of both the expected utility framework and prospect theory (Kahneman & Tversky, 1979), more recent research suggests that risk-taking proclivity is a domain-specific phenomenon in which an individual’s risk proclivities are different across domains (Weber, Blais, & Betz, 2002). In other words, an individual may display a strong appetite for financial risk and a strong aversion to risk in other domains such as recreational activities or social situations. Accordingly, in the current study, we explore the links between the digit ratio and domain-specific instantiations of risk.

In sum, while the results are somewhat mixed, it would appear that testosterone has both organizational and activational effects on financial risk-taking. Furthermore, there is a lack of research exploring the link between digit ratios and risk-taking in other domains. The current paper examines if digit ratio is predictive of risk-taking propensity across recreational, financial, social, ethical, and health domains. We propose that lower, more masculine digit ratios are predictive of riskier behaviors across all five domains among both men and women.

Section snippets

Participants

Four hundred and forty-nine students were recruited from classrooms at a Canadian university. Two students had broken fingers and 34 students did not complete the survey, resulting in a final sample size of 413. Participants were 53% male and were aged 17–44 years (mean = 20.9). The sample was ethnically heterogeneous, consisting of 58% Caucasian, 22% Asian, 10% Middle-Eastern, 2% Black, 2% Hispanic, and 6% other, mixed, or unspecified.

Procedure and measures

Participants were asked to fill out a survey containing the

Results

Table 1 displays the digit ratios (2D:4D and rel2) as well as the risk proclivities scores for both sexes. As expected, men exhibited riskier behaviors than women across all five domains (all p values < 0.001, one-tailed) and they had lower 2D:4D and rel2 compared to women (both p values = 0.001, one-tailed).

To control for the potentially confounding effects of sex and ethnic heterogeneity, we performed analyses on the following four sub-samples: male Caucasians (n = 130), female Caucasians (n = 109),

Discussion

Our results suggest that prenatal testosterone exposure has organizational effects on a man’s recreational, financial, and social risk-taking propensity. Contrary to our expectations, there were no significant correlations between digit ratio and risk in the ethical and health domains among men. One explanation for this pattern of results is that, compared to ethical and health risk-taking, recreational, financial, and social risk-taking serve as more honest signals of desirable traits in men.

References (53)

  • A. Schwerdtfeger et al.

    Digit ratio (2D:4D) is associated with traffic violations for male frequent car drivers

    Accident Analysis and Prevention

    (2010)
  • J. van Honk et al.

    Testosterone shifts the balance between sensitivity for punishment and reward in healthy young women

    Psychoneuroendocrinology

    (2004)
  • H. Vermeersch et al.

    The role of testosterone in aggressive and non-aggressive risk-taking in adolescent boys

    Hormones and Behavior

    (2008)
  • M. Voracek et al.

    Lack of correlation between digit ratio (2D:4D) and Baron-Cohen’s “Reading the Mind in the Eyes” test, empathy, systemising, and autism-spectrum quotients in a general population sample

    Personality and Individual Differences

    (2006)
  • M. Voracek et al.

    Digit ratio (2D:4D) and sensation seeking: New data and meta-analysis

    Personality and Individual Differences

    (2010)
  • X.T. Wang et al.

    Life history variables and risk-taking propensity

    Evolution and Human Behavior

    (2009)
  • R.E. White et al.

    Entrepreneurs and evolutionary biology: The relationship between testosterone and new venture creation

    Organizational Behavior and Human Decision Processes

    (2006)
  • M. Wilson et al.

    Competitiveness, risk-taking, and violence: The young male syndrome

    Ethology and Sociobiology

    (1985)
  • B. Auyeung et al.

    Fetal testosterone predicts sexually differentiated childhood behavior in girls and in boys

    Psychological Science

    (2009)
  • S. Baron-Cohen et al.

    The “Reading the Mind in the Eyes” test revised version: A study with normal adults, and adults with Asperger syndrome or high-functioning autism

    Journal of Child Psychology and Psychiatry

    (2001)
  • A. Booth et al.

    Testosterone and men’s health

    Journal of Behavioral Medicine

    (1999)
  • W.M. Brown et al.

    Differences in finger length ratios between self-identified “butch” and “femme” lesbians

    Archives of Sexual Behavior

    (2002)
  • D.M. Buss

    Sex differences in human mate preferences: Evolutionary hypotheses tested in 37 cultures

    Behavioral and Brain Sciences

    (1989)
  • J.P. Byrnes et al.

    Gender differences in risk taking: A meta-analysis

    Psychological Bulletin

    (1999)
  • P.Y. Chen et al.

    Correlation: Parametric and nonparametric measures

    (2002)
  • J.M. Coates et al.

    Second-to-fourth digit ratio predicts success among high-frequency financial traders

    Proceedings of the National Academy of Sciences of the United States of America

    (2009)
  • Cited by (108)

    • Sex hormones and choice under risk

      2023, Journal of Economic Psychology
    • 2D:4D does not predict economic preferences: Evidence from a large, representative sample

      2021, Journal of Economic Behavior and Organization
      Citation Excerpt :

      For example, the first paper on this topic reports a negative correlation between risk taking and left hand 2D:4D but finds no statistically significant correlation with the right hand 2D:4D, and the significant correlation can only be found in one sample and not another (Dreber and Hoffman, 2007). There are other examples of statistically significant negative correlations (e.g. Brañas-Garza and Rustichini, 2011; Garbarino et al., 2011; Stenstrom et al., 2011; Sytsma, 2014; Bönte et al., 2016; Barel, 2019; Brañas-Garza et al., 2018), though many of these only find statistically significant correlations in a subset of the population studied. There are thus also several null results (e.g. Stenstrom et al., 2011; Sytsma, 2014; Bönte et al., 2016; Barel, 2019; Brañas-Garza et al., 2018; Apicella et al., 2008; Sapienza et al., 2009; Aycinena et al., 2014; Schipper, 2014; Drichoutis and Nayga, 2015; Chicaiza-Becerra and Garcia-Molina, 2017; Lima de Miranda et al., 2018; Alonso et al., 2018; Neyse et al., 2020) and also one finding of a statistically significant positive correlation between 2D:4D and risk taking in a female subsample in one of the papers that found a statistically significant negative correlation in another subsample (Brañas-Garza and Rustichini, 2011).

    View all citing articles on Scopus
    View full text