2.2.1 The Evolutionary Foundation

About a century and a half ago, Charles Darwin (Reference Darwin1871) proposed the concept of sexual selection to refer to a special type of natural selection. The primary difference between these two evolutionary concepts is as follows: Natural selection involves forces that impinge from outside of a group of organisms that can affect their ability to reproduce (e.g., food availability and predation risks). Sexual selection, on the other hand, occurs between members of sexually reproducing species. Examples of sexually selected traits are preferences and actions taken by members of one sex toward a prospective mate, along with responses to actions by members of the opposite sex. Such preferences and actions can often affect how many offspring each sex is able to leave in subsequent generations (Fisher, Reference Fisher1930; Trivers, Reference Trivers and Campbell1972). For instance, by possessing a trait that enhances a male’s sexual attractiveness to females, the trait is being subjected to sexual selection. Sexually selected traits can be physical, cognitive, or behavioral in nature, but they must all have some underlying genetic foundation (Fromonteil et al., Reference Fromonteil, Marie-Orleach, Winkler and Janicke2023). Examples of sexual selection include both physical and behavioral traits (e.g., size and coordination) that males use to help ward off rival males from mating opportunities, or traits exhibited by females that males find sexually attractive (Daly, Reference Daly2017). Even the care given by males or females to their offspring can be seen as having been subjected to sexual selection (Puts, Reference Puts2016; Wilson et al., Reference Wilson, Miller and Crouse2017).

In Darwinian terms, because of the lengthy gestational period involved in producing an offspring, females should be more cautious than males when it comes to mating. Thus, the sexual selection component of Darwin’s theory can help to explain several sex differences in behavior, not only in humans (Buss, Reference Buss2007; Del Giudice, Reference Del Giudice, Frisby, Redding, O’Donohue and Lilienfeld2023; Geary, Reference Geary2021; Okami & Shackelford, Reference Okami and Shackelford2001; Puts et al., Reference Puts, Jones and DeBruine2012) but in other species as well (Key & Ross, Reference Key and Ross1999; Petersen & Higham, Reference Petersen and Higham2020; Schuett et al., Reference Schuett, Tregenza and Dall2010).

To provide a specific example, consider promiscuity. Theoretically, males who desire sex with numerous sex partners, and at least occasionally act on those desires, are likely to leave more copies of their genes in subsequent generations than males who choose to be sexually exclusive to just a single mating partner (Schmitt, Reference Schmitt2003; Geher & Kaufman, Reference Geher and Kaufman2013; Janicke et al., Reference Janicke, Häderer, Lajeunesse and Anthes2016). This would be especially true for earlier periods in human history before the advent of modern contraception and the widespread availability of abortion. On the other hand, females with equally strong desires for multiple sex partners would rarely have a reproductive advantage over females who are content with having just one sex partner (Buss & Schmitt, Reference Buss and Schmitt2017).

From the perspective of sexual selection, sex differences are rooted in recognizing that, among mammals, only females can gestate offspring. One consequence is that, once a female becomes pregnant, her reproductive potential depends on successfully gestating each fetus she is carries and then nurturing the offspring after birth. Males, however, have fewer constraints on their abilities to reproduce since their investment is limited to the insemination process.

Another example of how sexual selection theory has been used to explain sex differences involves physical strength and the ability to spatially navigate. Both traits are likely to have helped humans track, chase, and throw lethal projectiles at prey (Geary, Reference Geary2021; Kolakowski & Malina, Reference Kolakowski and Malina1974). Theoretically, males would have been favored for these activities more than females because gestating and breastfeeding offspring restricts the time and effort females can devote to effectively tracking and killing wild game (Kordsmeyer et al., Reference Kordsmeyer, Hunt, Puts, Ostner and Penke2018; Marlowe, Reference Marlowe2007; Trivers, Reference Trivers1976).

Overall, Darwin’s (Reference Darwin1871) theory of sexual selection is an extension of his theory of natural section, and has provided a springboard for many hypotheses having to do with how males and females, on average, might differ. In most cases, the theory implies that a substantial number of USDs should exist (Fromonteil et al., Reference Fromonteil, Marie-Orleach, Winkler and Janicke2023). One proponent of sexual selection theory argued as follows: “If a sex difference occurs consistently, despite all the variations in learning and socialization practices that occur across cultures, then … an innate predisposition is probably showing through all the cultural ‘noise’” (Lippa, Reference Lippa2002, p. 116).

2.2.2 Augmenting Sexual Selection with Neurohormonal Variables

Sexual selection theory has achieved impressive success in offering accounts for why a variety of sex differences in various cognitive and behavioral traits might exist (Clutton-Brock & Vincent, Reference Clutton-Brock and Vincent1991; Jones & Ratterman, Reference Jones and Rossiter2009; Lüpold, Manier et al., Reference Lüpold, Manier and Puniamoorthy2016). Nevertheless, this theory is limited from the fact that, because evolved traits of a cognitive or behavioral nature are rarely fossilized, direct evidence of their evolution is all but impossible to obtain.

Given the difficulties involved in directly testing sexual selection theory, critics have pointed out that much of the supportive evidence has involved searching for sex differences in traits that happen to conform to the theory, while ignoring many other sex differences that do not conform (Hankin, Reference Hankin2013). This point was made by Eagly and Wood (Reference Eagly and Wood1999, p. 420) in their criticism of sexual selection theory as follows: “It is far too easy to make up sensible stories about how these differences might be products of sex-differentiated evolved tendencies or the differing placement of women and men in the social structure.”

Scientific theories can often be enhanced by incorporating conceptual components that make them more testable. Of course, adding concepts usually involves trade-offs between simplicity and greater complexity; obviously, all else being equal, simpler theories are better than complex theories. In other words, if adding more conceptual components to a theory gives it greater explanatory power, the resulting increase in complexity can often be justified (Murphy & Medin, Reference Murphy and Medin1985). To illustrate this point, Darwin’s theory was expanded in the early part of the twentieth century with what has come to be known as the Modern Synthesis. This synthesis involved adding genetic concepts to Darwin’s original theory, concepts that were nonexistent during Darwin’s life (Huxley, Reference Huxley1942; Pinker, Reference Pinker2002). Today, scientists can use this gene-enhanced version of Darwin’s theory to help account for many more biological phenomena than was possible with his original theory (Kutschera & Niklas, Reference Kutschera and Niklas2004; Jobling & Tyler-Smith, Reference Jobling and Tyler-Smith2019).

While many genes may play a role in producing sex differences, the most crucial genes reside on the two so-called sex chromosomes, especially the Y chromosome (Charlesworth, Reference Charlesworth1991; Iannuzzi et al., Reference Iannuzzi, Bacalini, Franceschi and Giuliani2023; Roldan & Gomendio, Reference Roldan and Gomendio1999). Over the past several decades, a great deal has been learned about how genes help to guide sexual development, including sexual differentiation of the brain (Jacobs, Reference Jacobs1996; Ryan, Reference Ryan2021). Considerable evidence suggests that the brain is sexually dimorphic both anatomically and functionally in numerous ways (review: Ellis et al., Reference Ellis, Palmer, Hopcroft and Hoskin2024a, pp. 330–474). Most of the differentiation of the brain is the result of both prenatal and postpubertal exposure to various sex hormones, particularly testosterone (Celec et al., Reference Celec, Ostatníková and Hodosy2015; Herbert, Reference Herbert2015; Jašarević et al., Reference Jašarević, Geary and Rosenfeld2012; Negri-Cesi et al., Reference Negri-Cesi, Colciago, Celotti and Motta2004). In other words, many of the genes that result in the production of male-typical amounts of testosterone among males have accumulated mainly (but not exclusively) on the Y chromosome. Since only males carry this chromosome, its genes virtually ensure that males will produce more testosterone than females do (Knickmeyer & Baron-Cohen, Reference Knickmeyer and Baron-Cohen2006). Testosterone is primarily produced outside of the nervous system (especially by the testes). Some externally produced testosterone is able to penetrate the brain, both prenatally and following puberty by passing through the so-called blood–brain barrier (Kopsida et al., Reference Kopsida, Stergiakouli, Lynn, Wilkinson and Davies2009; Savic et al., Reference Savic, Frisen, Manzouri, Nordenstrom and Lindén Hirschberg2017). In this way, testosterone seems to have multiple influences on how organisms (including humans) think and behave.

Overall, due to genes affecting the production of testosterone and the ability of testosterone to affect brain functioning, males and females exhibit a variety of average differences in thought and behavior (e.g., Hooven Reference Hooven2021; Manning & Taylor, Reference Manning and Taylor2001; Martel, Reference Martel2013; Sinervo et al., Reference Sinervo, Miles, Frankino, Klukowski and DeNardo2000). This means that one can add neurohormonal concepts to Darwinian theory. By doing so, his theory can be tested in many ways than was the case prior to the Modern Synthesis and even more after the addition of neurohormonal concepts (Jašarević et al., Reference Jašarević, Geary and Rosenfeld2012).

I have proposed calling this neurohormonal version of Darwin’s gene-enhanced theory the evolutionary neuroandrogenic (ENA) theory (Ellis, Reference Ellis, Walsh and Ellis2003, Reference Ellis2006, Reference Ellis2011). In essence, ENA theory supplements the Modern Synthesis version of Darwinian theory with additional variables having to do with testosterone and brain functioning. And, because the brains of males and females are exposed to differing amounts of testosterone, ENA theory predicts the existence of sex differences in cognitive and behavioral traits.

Overall, ENA theory asserts that the main driver of sex differences in cognition and behavior involves the evolution of hormones (testosterone in particular) that also affects sex differences in traits that make males more attractive to females. As stated by Malisa Hines (2004, p. 65), “[t]he existence of sex differences in behavior implies the existence of sex differences in the brain because the brain provides the basis for all behavior.” I would add that her statement holds true not only for behavior but also for cognitive traits such as emotions, reasoning, interests, preferences, and a variety of physical traits.

To conceptualize average sex differences in exposure to testosterone, readers can refer to Figure 1. This graph provides an estimate of how this exposure varies from conception to the end of life. As one can see, there are major sex differences throughout life in testosterone levels except in the case of childhood, when differences tend to be minor.

Figure 1 Testosterone levels of human males and females from conception through old age.

Figure 1Long description

Testosterone levels begin to peak for males in the first trimester and then exhibit a brief spike in early infancy, and then dramatically rise in the late teens and early 20s. Females, on the other hand, primarily exhibit a moderate rise in testosterone in the late 20s.

Two qualifying comments are worth making regarding this graph: First, most of the permanent sexing of the brain occurs prenatally, whereas most of testosterone’s so-called activational effects occurs following the onset of puberty (Auyeung et al., Reference Auyeung, Lombardo and Baron-Cohen2013).

Second, Figure 1 provides a rough indication of sex differences in the production of testosterone throughout life, including the levels that penetrate the brain (Celec et al., Reference Celec, Ostatníková and Hodosy2015; Gooren, Reference Gooren2007). One can see that by the time males are born, their average exposure to testosterone is much greater than is the case for the average female. Following puberty, male exposure to testosterone also greatly surpasses that of the average female (for a summary of the evidence regarding sex differences in testosterone exposure, see Ellis et al., Reference Ellis, Palmer, Hopcroft and Hoskin2024a, pp. 223–226). Research has indicated that the more the brain is exposed to testosterone, the more it is masculinized (i.e., made more male-typical), and the more masculinized the brain, the more male-typical cognition and behavior will usually be throughout life (Baron-Cohen et al., Reference Baron-Cohen, Lutchmaya and Knickmeyer2006; Pfaf et al., Reference Pfaff, Rubin, Schneider and Head2018).

Scientific evidence regarding the effects of testosterone on brain functioning is extensive (Mehta et al., Reference Mehta, Jones and Josephs2008; Volman et al., Reference Volman, Toni, Verhagen and Roelofs2011; Welker et al., Reference Welker, Gruber and Mehta2015). While investigations are still ongoing, findings have already shown that, when neurological levels of testosterone are high (i.e., in the male-typical range), key reward-sensitive areas of the brain often promote dopamine production. Dopamine is a neurotransmitter that is often associated with feelings of accomplishment and keen motivation to socially compete (Breuning, Reference Breuning2018; Smith et al., Reference Smith, Chein and Steinberg2013). Also, dopamine in the brain often promotes whatever types of thoughts and behaviors individuals are engaged in at the time of exposure (Campbell et al., Reference Campbell, Dreber and Apicella2010; Eisenegger et al., Reference Eisenegger, Kumsta, Naef, Gromoll and Heinrichs2017). Many other neurological changes also occur in response to testosterone, changes that are likely to have numerous influences on cognition and behavior (Davis & Pfaff, Reference Davis and Pfaff2014; Geary, Reference Geary2019; Henriques-Neto et al., Reference Henriques-Neto, Peralta and Marques2023).

According to ENA theory, sex differences in many cognitive and behavioral traits have evolved because they have a tendency to help males and females obtain mating opportunities, thereby contributing to the representation of their genes in future generations. Regarding neurology, sex differences in brain functioning have evolved because many of these differences promote cognitive and behavioral traits that have been sexually selected. It is only a slight over-simplification to say that, for every USD in cognition and behavior that exists, ENA theory predicts that both sexual selection and brain exposure to testosterone are involved (Ellis, Reference Ellis2011).

An important issue to address regarding ENA theory has to do with measuring testosterone exposure. While the theory states that brain exposure is a central feature, in fact, this is nearly impossible to empirically document without risking damage to the brain. Therefore, most researchers who are interested in the effects of testosterone on brain functioning end up relying on peripheral measures (i.e., measures outside of the brain itself such as the amniotic fluid, blood, or saliva). If these measures are obtained before birth, they are known as prenatal levels; if they are assessed at any time thereafter, especially following puberty, they are usually known as circulating levels. Most measures of circulating testosterone are from the blood or saliva. However, testosterone levels fluctuate throughout the day, with the highest and most stable levels occurring soon after awakening, which is when blood or saliva levels are usually obtained (Keevil et al., Reference Keevil, MacDonald and Macdowall2014).

Measuring prenatal levels of testosterone is especially difficult. The most precise way involves sampling the amniotic fluid surrounding the developing fetus (Sarkar et al., Reference Sarkar, Bergman, O’Connor and Glover2008). A few studies have derived samples from the placenta soon after delivery with the assumption that this provides a rough gage of fetal exposure a few days prior to birth (Galiano et al., Reference Galiano, Solazzo and Rabinovici2021). In recent years, a relatively crude method has been established involving the so-called 2D:4D finger length ratio. Basically, the longer an individual’s pointing finger compared to his/her ring finger provides a rough estimate of relatively low prenatal testosterone exposure (Eler, Reference Eler2018; Mikac et al., Reference Mikac, Buško, Sommer and Hildebrandt2016). Finally, some studies have inferred exposure to testosterone among twins based on the sex of their co-twin, often known as the twin testosterone transfer hypothesis (Tapp et al., Reference Tapp, Maybery and Whitehouse2011). If both siblings are boys, both should have testosterone levels in the male-typical range, and, if both siblings are girls, their testosterone levels should be in the female-typical range. However, when twins are of the opposite sex, boys are thought to have slightly lower male-typical exposure to testosterone than singleton males, while girls usually have exposure that is in the high female-typical range compared to singleton females (Miller, Reference Miller1994; Talia et al., Reference Talia, Raja, Bhattacharya and Fowler2020). Examples of these methods for inferring brain exposure to testosterone will be cited as I subject ENA theory to empirical scrutiny.

2.3.1 Social Role Theory

In the case of social role theory, Figure 2 shows that sociocultural factors interact with three basic physiological sex differences (i.e., males being taller and stronger and females being the only sex capable of bearing offspring). Together, these three factors are thought to be responsible for USDs is cognition and behavior. All other cognitive and behavioral sex differences are assumed to be unique and varied from one culture to another.

Figure 2 A flowchart summary of the social role theory of sex differences in cognition and behavior.

Adapted from Eagly & Wood, Reference Eagly and Wood2012, p. 465 and Wood & Eagly, Reference Wood, Eagly and Zanna2012, p. 58.

Figure 2Long description

Represents the main variables responsible for sex differences in cognition and behavior according to social role theory. At the most fundamental level are sociocultural traditions and greater male size and strength, along with the ability of females to bear offspring. Together these factors lead to socially constructed sex roles and to hormonal regulation and self-regulated conformity to sex role expectations and training. The final outcome of these factors involve average sex differences in cognition and behavior, some of which are universal.

According to social role theory, societal sex roles regulate hormones, thereby providing the context in which individual thinking and behavior are impacted (Eagly & Wood, Reference Eagly and Wood2012, p. 465; Wood & Eagly, Reference Wood, Eagly and Zanna2012, p. 58). One can see in Figure 2 that social role theory includes “hormonal regulations” as contributing to cognitive and behavioral sex differences. However, this concept is vague and fails to provide any specifics regarding how “hormonal regulations” impact cognitive and behavioral sex differences.

2.3.2 ENA Theory

The flowchart for ENA theory is presented in Figure 3. It identifies two separate causal categories: one focusing on “why” USDs exist (the evolutionary factors) and the other pertaining to “how” these sex differences are physiologically regulated (the neurochemical factors). In evolutionary terms, sexual selection is seen as driving males and females apart in various ways so that each sex can pass genes on to future generations. Especially in mammals, where females must gestate each offspring (and usually breastfeed each one for substantial time thereafter), they are shown as having been sexually selected for biasing their mate choices in favor of males who appear capable of sustained resource provisioning. Furthermore, this female bias is hypothesized to have favored males with resource provisioning tendencies. In other words, resource provisioning males will have a reproductive advantage over males who do otherwise. Also, once pregnant and breastfeeding, women who provision on their own will usually pass on fewer genes to future generations than women who can rely on resources provided by a mate.

Figure 3 A flowchart summary of the ENA theory of sex differences in cognition and behavior.

Adapted from Ellis, Reference Ellis2011.

Figure 3Long description

Represents the two major components for evolutionary neuroandrogenic (ENA) theory. According to the first component, females have evolved tendencies to bias their mate preferences toward individuals who appear to be loyal and competent long term provisioners of resources. Males, on the other hand, have evolved tendencies to comply with (or sometimes circumvent) these female preferences. Regarding ENA theory’s other component, testosterone is hypothesized to be the main biochemical that has evolved to alter brain functioning in ways that cause males to exhibit cognitive and behavioral traits that help them comply with female preferences for resource procuring capabilities.

The neurohormonal component of ENA theory is represented in the bottom segment of Figure 3. It asserts that genes (especially ones on the Y chromosome) interact to impact male cognition and behavior to help males conform to female mate preferences. Many of these genes regulate the production of sex hormones, especially testosterone.

In the case of females, genes promoting preferences for males with resource provisioning skills will have been favored by sexual selection. Among males, genes that promote abilities and interests to comply with female preferences will be sexually selected. At the present time, ENA theory can still be improved if the precise hormonal regimens involved in these evolved cognitive and behavioral traits can be more precisely identified. Nonetheless, several lines of evidence point toward combinations of testosterone and estrogens (and possibly progesterone levels), all being involved (e.g., De Jonge et al., Reference De Jonge, Eerland and Van De Poll1986; Van Goozen et al., Reference Van Goozen, Wiegant, Endert, Helmond and Van de Poll1997).

One can see in both flowcharts that both theories hypothesize the involvement of hormonal factors in bringing about sex differences in cognition and behavior. However, ENA theory is more specific in this regard in four respects: First, it names the primary hormone (i.e., testosterone). Second, it specifies the direction of the main influence (i.e., greater testosterone produces more male-typical cognition and behavioral traits). Third, it identifies that the timing of testosterone’s influences on cognition and behavior occurs both prenatally and postpubertally. Fourth, ENA theory stipulates that, to impact cognition and behavior, testosterone primarily operates on the brain.

Figure 2 shows that social role theory is nonspecific about hormonal contributions to sex differences in cognition and behavior, the timing of these contributions, or the involvement of the brain in the process. These features of social role theory make it all but impossible to scientifically test regarding any hormonal contributions to sex differences in cognition and behavior. Philosophers of science have long recognized that the best scientific theories are those that make the most predictions of what will eventually be empirically confirmed (Maxwell, Reference Maxwell1974; Popper, Reference Popper2013).

Throughout the remainder of this Element, I will assume that all fifteen variables identified in Table 1 are USDs, although the basic information supporting this assumption is provided in the table (which in turn summarizes the pertinent studies provided in our book). Overall, the primary goal will now be to determine which of the two theories can best account for why these fifteen variables would be culturally universal. The narrative assessment will consider each of the fifteen variables one at a time.

3.1.1 Social Role Theory

To explain USDs in average salaries or other types of income, social role theory can draw on the fact that only females can become pregnant. Thus, at least for women who bear children, they are likely to periodically interrupt their involvement in the paid workforce to devote time to providing childcare, especially if they routinely breastfeed after giving birth. Substantial evidence supports this line of reasoning (Cebrián & Moreno, Reference Cebrián and Moreno2015; de Quinto et al., Reference de Quinto, Hospido and Sanz2021), especially for women who begin having child early in life (Putz & Engelhardt-Woelfler, Reference Putz and Engelhardt-Woelfler2014) and for those living in countries with few paid work-leave accommodations (Karimi, Reference Karimi2014). Furthermore, one study indicated that most of the adverse impact of childbearing on female earnings occurred in the first few years after giving birth, and that the impact was no longer significant after fifteen years of a woman’s last birth (Angelov & Karimi, Reference Angelov and Karimi2012).

Also supporting the view that pregnancy among females is a major cause of sex disparities in wages, a Danish study indicated that, before women give birth, their salaries were statistically equal to those of males. However, after giving birth, the average salaries for males begin to surpass that of females (Kleven et al., Reference Kleven, Landais and Søgaard2019).

Obviously, these pregnancy-related work interruptions are bound to adversely affect career advancement, which is detrimental to salary increases. This “penalties of motherhood” explanation of sex differences in earnings has been criticized as providing only a weak explanation for sex differences in earnings, arguing instead for an explanation rooted in the types of occupations both sexes choose (England, Reference England2005). Nonetheless, it is entirely consistent with sex role theory.

Another study involved persons who never married. It reported that men still earned more than women (Cutler & Harootyan, Reference Cutler, Harootyan, Woodruff and Birren1975). This study also indicated that the types of jobs males and females tend to choose seems to be far more important in explaining sex differences in wages than sex differences in the ability to bear offspring. Later, under the discussion of Variable 14, I will elaborate on the possible importance of sex difference in occupational choices. It has to do with being educated in the physical sciences and engineering.

Proponents of social role theory have also argued that, because males are taller and physically stronger than females, they are often able to achieve greater social dominance (Eagly & Wood, Reference Eagly and Wood2012), one result of which is often higher average earnings (Cheng & Tracy, Reference Cheng, Tracy, Cheng and Tracy2014). One way to independently test this reasoning would involve documenting that, even within each sex, height and/or physical strength are positively correlated with average income. Studies have found a great deal of support for this latter hypothesis, especially regarding height (Judge & Cable, Reference Judge and Cable2004; review: Ellis et al., Reference Ellis, Hoskin and Ratnasingam2018, p. 259).

Overall, social role theory offers explanations for why males in all known societies earn more on average than females. The theory does so by attributing the differences to the fact that pregnancy has suppressing effects on earnings and by noting that males exhibit greater height and strength than females, which promotes greater dominance. These explanations have substantial empirical support, thereby justifying three checkmarks in Table 1 for this theory.

3.1.2 ENA Theory

Both the evolutionary and the neurohormonal components of ENA theory can provide explanations for sex differences in average earnings. Some of its explanations parallel those offered by social role theory and others diverge.

Regarding evolution, ENA theory argues that females have been sexually selected for making mate choices based on indications that a prospective mate is competent at resource provisioning. Without this male ability, the females will usually have to curtail their production of offspring or at least provide less care to them (Ellis, Reference Ellis2001). Theoretically, this would be especially true in societies where little governmental support is given to mothers who are pregnant or with dependent children and nonsupportive fathers. Thus, whether their choices are made consciously or not, women who bias their mate selection in favor of males who appear to be willing and capable provisioners of resources will increase their chances of passing genes (and genes of their mates) on to subsequent generations.

This evolutionary argument can also be applied to males. Presumably, males have been sexually selected for their abilities to comply with female preferences regarding resource provisioning. In other words, males who become competent and reliable resource provisioners will usually pass their genes on to future generations at higher rates than other males. The main qualification to this generalization involves males who use deception or force to obtain copulatory access (Ellis & Hoskin, Reference Ellis and Hoskin2015).

Besides evidence that males on average earn more than females, there is additional evidence supporting this evolutionary reasoning. As will be discussed more regarding Variable 13, studies throughout the world have shown that females prefer mates who are, or appear to be, “good providers” to a greater degree than do males. This suggests that sexual selection for high-earning males is likely to still be ongoing.

Another line of evidence that is consistent with using the sexual selection argument for sex differences in earnings comes from research indicating that males consider monetary compensation for their work as being more important than do females (review: Ellis et al., Reference Ellis, Palmer, Hopcroft and Hoskin2024b, pp. 941–943). In other words, women appear to have greater interests in jobs where the work environment is clean and friendly (review: Ellis et al., Reference Ellis, Palmer, Hopcroft and Hoskin2024b, pp. 937–938), while men focus more on earnings, no matter what the job entails. Overall, the evolutionary component of ENA theory appears to have considerable evidentiary support.

Turning to the physiological aspects of ENA theory, brain exposure to testosterone should be positively correlated with how much individuals earn. Evidence in this regard has been somewhat mixed regarding prenatal testosterone. One study was based on a sample of twins. As already noted, studies have indicated that female fetuses who share the womb with male fetuses usually exhibit at least slightly masculinized behavioral traits when compared to singleton-born females. This is true for studies of both nonhumans (review: Ryan & Vandenbergh, Reference Ryan and Vandenbergh2002) as well as humans (Cohen-Bendahan et al., Reference Cohen-Bendahan, Van De Beek and Berenbaum2005; Miller, Reference Miller1994). Such masculinizing tendencies appear to be the result of females with male co-twins being exposed to higher-than-normal amounts of prenatal testosterone than females who share the womb with other females.

Following this line of reasoning, ENA theory predicts that fetuses sharing the womb with an opposite-sex co-twin should have higher lifetime earnings than those with same-sex co-twins. In other words, when one co-twin is secreting male-typical levels of testosterone during gestation and one is only secreting female-typical levels, the latter is likely to receive above average exposure to testosterone. Some evidence has been reported on this theoretical line of reasoning. One study of earnings by adult twins indicated that male–male co-twins had higher average lifetime earnings than did male–female co-twins. In the case of females, however, the sex of the co-twin was not significantly correlated with lifetime earnings (Gielen et al., Reference Gielen, Holmes and Myers2016). Another study of twins reported that females with a male co-twin earned less than did females with a female co-twin (Bütikofer et al., Reference Bütikofer, Figlio, Karbownik, Kuzawa and Salvanes2019). Overall, the results from these two studies provide little support for ENA theory.

Another study designed to estimate prenatal testosterone exposure used the 2D:4D finger length ratio. The study reported that a significant positive correlation existed between prenatal testosterone exposure (i.e., relatively low 2D:4D ratios) and wages in adulthood among women but not among males (Nye, Yudkevich et al., Reference Nye, Yudkevich, Orel and Kochergina2014). This finding is obviously mixed regarding providing support of ENA theory.

Two other studies of prenatal testosterone exposure looked for evidence that high exposure promotes desires to be self-employed (rather than working for others). Such research is relevant here to the extent that self-employed persons have relatively higher incomes (including profits) than do persons who work for others (Bernhardt, Reference Bernhardt1994; Schneck, Reference Schneck2020). One study of self-employment among men found it to be positively correlated with prenatal testosterone exposure, as indicated by low 2D:4D ratios (Hughes & Kumari, Reference Hughes and Kumari2019). The other study also reported that, among both sexes, persons who were self-employed had 2D:4D ratios indicative of higher prenatal testosterone exposure (Nicolaou et al., Reference Nicolaou, Patel and Wolfe2018).

In addition to prenatal testosterone, ENA theory also predicts that postpubertal testosterone (often called circulating testosterone) should contribute to earnings. Four relevant studies were located. These studies were based on samples of men, and all of them indicated that significant positive correlations were evident (Coates & Herbert, Reference Coates and Herbert2008; Harrison et al., Reference Harrison, Davies, Howe and Hughes2021; Hughes & Kumari, Reference Hughes and Kumari2019; Luoto et al., Reference Luoto and Varella2021). In other words, on average, men with relatively high circulating testosterone levels appear to earn more than men with low circulating levels.

Overall, findings have been somewhat mixed regarding the effects of brain exposure to testosterone on lifetime earnings regarding prenatal testosterone exposure. However, in the case of circulating testosterone levels among adults, the predicted positive correlation has been strongly supported.

How can the apparent positive relationship between testosterone and earnings be explained? Much of the answer seems to involve noting that exposing the brain to high testosterone has been shown to promote competitiveness (Apicella et al., Reference Apicella, Dreber and Gray2011; Simmons & Roney, Reference Simmons and Roney2011; Eisenegger et al., Reference Eisenegger, Kumsta, Naef, Gromoll and Heinrichs2017) as well as risk-taking (Mehta et al., Reference Mehta, Jones and Josephs2008; Volman et al., Reference Volman, Toni, Verhagen and Roelofs2011; Welker et al., Reference Welker, Gruber and Mehta2015). Both personality traits, in turn, have been found to be positively associated with earnings (Hughes & Kumari, Reference Hughes and Kumari2019). More will be said about testosterone and risk-taking when Variable 12 is discussed. Because both the evolutionary component and the postpubertal aspects of the physiological component of ENA theory are strongly supported, it receives three checkmarks for explaining sex differences in average earnings.

3.2.1 Social Role Theory

For many years, proponents of social role theory have argued that sex differences in doing housework reflect cultural traditions (Hoffman, Reference Hoffman1977; Lachance-Grzela & Bouchard, Reference Lachance-Grzela and Bouchard2010; Lennon, Reference Lennon1994) and sex role stereotypes (Eagly et al., Reference Eagly, Wood and Diekman2000). Both explanations lead one to expect that there should be some cultures in which the sexes are equal, or where males actually do more household chores than females. Table 1 shows that, at least among adolescents and adults, this is not the case.

Social role theory assumes that while all behavioral sex differences are culturally learned, biological sex differences (such as physical strength and bearing offspring) can sometimes produce USDs in behavior. Could doing housework be one such example? Perhaps. Based on social role theory, one could argue that because only females can bear offspring and breastfeed them, they are likely to end up spending more time in the home (Wood & Eagly, Reference Wood, Eagly and Zanna2012, p. 59). Thus, because of sex differences associated with childbirth and rearing, females are likely to be more inclined to maintain a clean and orderly indoor environment and would have more opportunities to do so.

While this line of reasoning cannot be considered disproven, it is challenged by other evidence. Specifically, even for individuals living along (Kanazawa, Reference Kanazawa2005, p. 283) and cohabitate with no children (Batalova & Cohen, Reference Batalova and Cohen2002), females still perform more housework than do males. Furthermore, one study of dual-earning couples without children also concluded that women spend more time doing housework than men (Evertsson, Reference Evertsson2014).

Overall, there is no reason to doubt that women perform more household cleaning and maintenance chores than do men throughout the world. From the perspective of social role theory, this universal sex difference can be attributed to long cultural traditions possibly rooted in the fact that, because only females bear children, they feel obliged to fulfill homemaking duties. However, because no specific empirical evidence was found to support this line of reasoning, I assigned social role theory an “x” regarding the universality of this variable.

3.2.2 ENA Theory

Both the evolutionary and neurohormonal components of ENA theory can be seen as addressing the issue of sex differences in performing household chores. Regarding evolution, scientists have noted that, especially among mammals, females can identify their offspring with greater certainty than can males (Buss & Schmitt, Reference Buss and Schmitt1993; Dixson, Reference Dixson2009). As a result, females will have been more strongly favored for providing care to offspring. At least among humans, one expression of caring for offspring would focus on health, among the expressions of which would entail maintaining a clean and hazard-free home environment (Preston, Reference Preston2013).

In accordance with this line of reasoning, while differences are modest, research indicates that females have a keener sense of smell than do males (review: Ellis et al., Reference Ellis, Palmer, Hopcroft and Hoskin2024b, pp. 12–13; Sorokowski et al. Reference Sorokowski, Karwowski and Misiak2019). Also, some research indicated that genetic factors may be partly responsible for sex differences in the ability to detect odors (Hoover, Reference Hoover2010; Majid et al., Reference Majid, Speed, Croijmans and Arshamian2017).

Moreover, a potentially relevant line of evidence comes from outside the human species. It involves a phenomenon known as “nest sanitation behavior” among birds (Blair & Tucker, Reference Blair and Tucker1941; Skutch, Reference Skutch1976, p. 282). In about 20 percent of bird species, breeding pairs of males and females co-parent their chicks (Heinsohn & Double, Reference Heinsohn and Double2004). Besides both parents cooperatively feeding their nestlings, they also rid the nest of debris such as parasites, broken eggshells, and especially fecal sacs produced by the chicks. Researchers have reported that while there may be exceptions, in most co-parenting bird species, females devote more time to keeping their nests clean than do their male partners (review: Guigueno & Sealy, Reference Guigueno and Sealy2012).

Returning to humans, our book identified fourteen studies having to do with sex differences in people’s tendencies to be health conscious. Every one of these fourteen studies concluded that females were more so than males (review: Ellis et al., Reference Ellis, Palmer, Hopcroft and Hoskin2024b, pp. 837–838). Similarly, the vast majority of fifty-one studies of sex differences in the taking of preventive health measures concluded that females surpass males in this regard (review: Ellis et al., Reference Ellis, Palmer, Hopcroft and Hoskin2024c, pp. 92–93). Thus, considerable independent research points toward females being more concerned with cleanliness and health than are males.

Regarding evolution, it is reasonable to suspect that females may have been more strongly favored than males for being health conscious, not only for themselves but also for their offspring. If so, one expression of health consciousness could be a greater desire for household cleanliness among females, ergo greater tendencies to perform household chores.

If evolutionary forces have helped to produced sex differences in traits such as doing more housework, ENA theory hypothesizes that neurohormonal processes should also be involved. Specifically, brain exposure to testosterone should inhibit spending time performing household chores. This is obviously a testable hypothesis, but I found no relevant evidence. Because of the empirical evidence supporting the evolutionary arm of ENA theory but no evidence bearing on the physiological arm, I just assigned two checkmarks to this variable in Table 1.

3.3.1 Social Role Theory

Social role theory is likely to explain sex differences in criminal and delinquent behavior by noting that males have greater physical strength. Physical strength could give males more confidence in winning when engaging in physical aggression (Eagly & Wood, Reference Eagly and Wood2012, p. 465). Of course, this would have no bearing on most property and drug offenses, which also appears to be more prevalent among males than among females (review: Ellis et al., Reference Ellis, Farrington and Hoskin2019, pp. 55–56).

Overall, no independent empirical evidence could be found to support social role theory’s explanation for why most forms of self-reported criminality appears to be more prevalent among males than among females. However, one could argue that throughout the world, antisocial behavior is more likely to be tolerated and/or expected among males than among females (due to men’s greater physical strength). For this reason, and because sex differences in violent forms of crime can be predicted with social role theory (as I will discuss regarding Variable 15 further), I have assigned social role theory one checkmark beside Variable 3 in Table 1.

3.3.2 ENA Theory

ENA theory’s explanation for sex differences in self-reported criminality comes from both of its evolutionary and physiological components (Ellis & Hoskin, Reference Ellis and Hoskin2015). Regarding evolution, the theory makes three assumptions regarding sex differences in criminality (Ellis, Reference Ellis2001). First, females have been sexually selected for preferring mates with the ability to provide resources. Second, males with the greatest likelihood of passing genes onto future generations will have relatively strong tendencies to comply with this female preference (Fales et al., Reference Fales, Frederick and Garcia2016; Waynforth, Reference Waynforth2001; Wiederman, Reference Wiederman1993), and this compliance would have little to do with whether legal or illegal means are used to obtain the resources. Third, males can sometimes pass their genes on to future generations by using force or deception to obtain mates, thereby circumventing the tendencies females may have evolved to choose mates with resources. This latter assumption takes ENA theory into the realm of sexual assault, the commission of which has been shown to be much more common among males than females, although the number of studies fell short of 100 (Ellis et al., Reference Ellis, Palmer, Hopcroft and Hoskin2024c, pp. 526–528).

Turning to the physiological aspects of ENA theory, substantial evidence points toward brain exposure to testosterone as promoting several victimizing forms of criminal behavior. In this regard, numerous studies have indicated that both prenatal and postpubertal testosterone levels are positively correlated with criminality, especially among males (Ellis et al., Reference Ellis, Farrington and Hoskin2019, pp. 353–358). Overall, the neurohormonal aspects of ENA theory appears to have substantial empirical support regarding sex differences in self-reported criminality. For this reason, I assigned it three checkmarks.

3.4.1 Social Role Theory

As noted regarding self-reported offending (Variable 3), social role theory has difficulty explaining why males are universally involved in official criminality than females. One reason for this difficulty involves noting that social role theorists consider the criminal justice system as a male-dominated institution used to control women and other subjugated people (Covington & Bloom, Reference Covington, Bloom and Bloom2003; Eagly, Reference Eagly1987). If so, it is difficult to explain why most of those subjected to arrest, prosecution, and punishment would be males.

Perhaps, one could account for why official criminal behavior is more common among males using social role theory by assuming that the commission of crimes is enhanced by men’s greater size and strength. There is some support for this idea. Specifically, among males, criminals do appear to be more muscular than males in general, especially for violent offending (review: Ellis et al., Reference Ellis, Farrington and Hoskin2019, pp. 341–342). Regarding height, however, findings have been inconsistent regarding any specific within-sex association with criminality (review: Ellis et al., Reference Ellis, Farrington and Hoskin2019, pp. 339–341). Overall, the absence of independent empirical evidence warrants assigning social role theory two checkmarks for Variable 4.

3.4.2 ENA Theory

To explain why males are more likely than females to be involved in the commission of officially measured crime throughout the world, ENA theory can draw on both evolutionary and neurohormonal variables. In evolutionary terms, most victimizing forms of criminality can be seen as being (a) efforts to obtain resources (that males can often use to attract mates), (b) forms of competition with other males for mating opportunities with females, and/or (c) sexual assaults on females directly (Daly, Reference Daly2017; Ellis & Hoskin, Reference Ellis and Hoskin2015). This means that criminality can sometimes help males contribute genes to subsequent generations just as do noncriminal forms of competition for resources and/or mating opportunities.

Regarding neurohormonal factors, considerable evidence indicates that victimizing forms of criminality are positively correlated with both prenatal and postpubertal testosterone levels (review: Ellis et al., Reference Ellis, Farrington and Hoskin2019, pp. 351–355). It is also worth mentioning that risk-taking has been found to be positively correlated with involvement in crime and delinquency (Wilson & Daly, Reference Wilson and Daly1985; review: Ellis et al., Reference Ellis, Farrington and Hoskin2019, pp. 219–221). This comports with evidence that risk-taking is more prevalent among males than females (see Variable 12 for evidence in this regard). Finally, risk-taking appears to be promoted by exposing the brain to testosterone (review: Apicella et al., Reference Apicella, Carré and Dreber2015; Ronay & Von Hippel, Reference Ronay and Von Hippel2010; also see Variable 12). Given the weight of evidence that both prenatal and postpubertal testosterone is related to criminality, I have entered three checkmarks for ENA theory’s explanation of sex differences beside this variable in Table 1.

3.5.1 Social Role Theory

If one assumes that physical strength is associated with involvement in greater physical exercise, social role theory can provide an explanation for why males engage in more physical exercise than females. Supporting this theoretical reasoning, research has found physical strength and physical exercise are positively correlated (Haynes & DeBeliso, Reference Haynes and DeBeliso2019). Furthermore, at least some of the positive association between exercise and strength appears to be due to the fact that exercise promotes physical strength (Buriticá-Marínet al., Reference Buriticá-Marín, Daza-Arana, Jaramillo-Losada, Riascos-Zuñiga and Ordoñez-Mora2023; Rismayanthi et al., Reference Rismayanthi, Kristiyanto and Doewes2022). Unfortunately, social role theory provides no guidance for assessing why sex differences in strength would actually promote sex differences in physical exercise. Therefore, I have provided a single checkmark to this theory’s explanation of sex differences in vigorous exercise.

3.5.2 ENA Theory

The ENA theory offers both an evolutionary and a physiological explanation for why sex differences in physical exercise is found throughout the world. In evolutionary terms, physical exercise is likely to have been favored among males more than females, especially in pre-industrial societies, as a way of enhancing their abilities to obtain resources with which to attract mates and help them rear offspring. More specifically, engaging in frequent exercise would promote men’s success at hunting and tending livestock, as well as engaging in strenuous manual laborers (Pettay et al., Reference Pettay, Helle, Jokela and Lummaa2007; Smith, Reference Smith2004).

Regarding the physiological aspects of the theory, research has shown that when muscles are exposed to testosterone, physical strength is enhanced (Auyeung et al., Reference Auyeung, Lee and Kwok2011; Storer et al., Reference Storer, Woodhouse and Magliano2008). Furthermore, when the brain is exposed to testosterone, physical activity tends to rise (Lightfoot, Reference Lightfoot2008; Vaamonde et al. Reference Vaamonde, Da Silva-Grigoletto, García-Manso, Barrera and Vaamonde-Lemos2012). These findings suggest that when both muscle tissue and nervous tissue are exposed to testosterone, the effects are synergistic. Furthermore, not only does testosterone help to promote physical exertion but physical exertion in turn also promotes testosterone production (Kumagai et al. Reference Kumagai, Yoshikawa and Zempo-Miyaki2018; Riachy et al. Reference Riachy, McKinney and Tuvdendorj2020).

Experiments with laboratory animals have indicated that part of the enhanced physical activity that testosterone promotes is mediated by the release of the neurotransmitter, dopamine (Jardí et al., Reference Jardí, Laurent and Dubois2018). Overall, physical exercise, especially when strenuous, is more common among males than among females, and ENA theory provides empirically supported arguments for explaining these relationships, earning it three checkmarks in Table 1.

3.6.1 Social Role Theory

The explanation offered by social role theory for sex differences in occupational prominence would involve noting that, because males bear no children, they can devote more time to developing and refining their occupational skills. One way to test this line of reasoning would involve comparing occupational prominence for men and women who have never had children. If social role theory is correct, occupational prominence among childless couples should not exhibit significant sex differences. In making such comparisons, it might also be necessary to exclude individuals who attain prominence in sports where physical strength could give males an edge.

Unfortunately, I found no studies pertaining to sex differences in occupational prominence among childless couples. Therefore, I assigned one checkmark to this theory in Table 1.

3.6.2 ENA Theory

To explain sex differences in occupational prominence from the perspective of ENA theory, one can draw on both its evolutionary and its neurohormonal components. In evolutionary terms, ENA theory would focus on female tendencies to prefer mates with resources along with recognizing that occupational prominence is often associated with the procurement of resources (Bereczkei et al. Reference Bereczkei, Voros, Gal and Bernath1997; Ellis Reference Ellis2001; Souza et al., Reference Souza, Conroy-Beam and Buss2016). In other words, while not all prestigious occupations are associated with high incomes, very few fail to garner at least moderate monetary rewards (Garcia-Mainar et al., Reference Garcia-Mainar, Montuenga and García-Martín2018; Magnusson, Reference Magnusson2009).

Regarding the neurohormonal aspects of ENA theory, it predicts that relatively high brain exposure to testosterone should be associated with occupational prestige. Support for this hypothesis comes from studies indicating that status-striving behavior is positively correlated with both prenatal testosterone (Bijleveld & Baalbergen, Reference Bijleveld and Baalbergen2017; Millet & Dewitte, Reference Millet and Dewitte2008) and postpubertal circulating testosterone (Dabbs et al., Reference Dabbs, La Rue and Williams1990; Mazur & Booth, Reference Mazur and Booth1998). Exposure to prenatal testosterone among males also appears to be positively correlated with general competitiveness (Bönte et al., Reference 54Bönte, Procher, Urbig and Voracek2017).

While more research is needed, both the evolutionary and the physiological arms of ENA theory have substantial supportive independent evidence regarding sex differences in occupational prominence. For this reason, I gave ENA theory three checkmarks in Table 1.

3.7.1 Social Role Theory

Using social role theory to explain why males suffer more accidental injuries than females (except possibly for persons over the age of fifty) is difficult. About the only way to do so would involve asserting that physical strength promotes physical activity, which in turn increases the likelihood of accidental injuries. However, as noted earlier (under Variable 5), social role theory provides no way of logically connecting physical strength with physical activity. Granting that social role theory might hypothesize that physical strength promotes physical activity and therefore greater risk of being injured, I have given this theory a single checkmark for explaining sex differences in sustaining accidental injuries.

3.7.2 ENA Theory

Later in the section, Variable 12 will document that males take more risks (or express a greater willingness to do so) than do females. At the present junction, this sex differences can be invoked to help account for sex differences in accidental injuries. From an evolutionary perspective, scientists have argued that risk-taking is more likely to benefit males than females. This is partly due to the contribution that risk-taking can make toward the procurement and retention of resources (Baker & Maner, Reference Baker and Maner2008; Maxfield et al., Reference Maxfield, Shapiro, Gupta and Hass2010). Of course, an unfortunate side effect of risk-taking, at least in physical terms, is an elevated probability of accidental injuries (Turner et al., Reference Turner, McClure and Pirozzo2004).

Turning to the physiological aspect of ENA theory, no research was found directly linking testosterone to accidental injuries. However, as will be documented under Variable 12, substantial evidence points toward a positive correlation between testosterone and risk-taking (Apicella et al., Reference Apicella, Carré and Dreber2015; Dariotis et al., Reference Dariotis, Schen and Granger2016; Votinov et al. Reference Votinov, Knyazeva, Habel, Konrad and Puiu2022; reviews: Kurath & Mata, Reference Kurath and Mata2018). Therefore, ENA theory has considerable support when it comes to explaining why males suffer more accidental injuries than females, leading to its being assigned three checkmarks in Table 1.

3.8.1 Social Role Theory

The only way social role theory appears capable of explaining why males would be more likely than females to hold elective political offices would involve the fact that only females can become pregnant. In this regard, it is reasonable to assume that pregnancy curtails the ability of women to spend time campaigning for elective offices. One way to test this hypothesis would be to determine if women with no children are statistically equivalent to men in their tendencies to hold elected office. No such research was located, so I assigned social role theory a single checkmark regarding its potential for explaining sex differences in tendencies to hold elected political office.

3.8.2 ENA Theory

The ENA theory offers both an evolutionary and physiological accounts for sex differences in elected office holding. Regarding evolution, in many species, females are more likely than males to mate with individuals who display dominant behavior than to those who do not (Weisfeld & Dillon, Reference Weisfeld and Dillon2012, p. 25). This apparent attraction to dominant males probably has to do with the fact that dominance is generally associated with control over territory and the resources it contains (Anderson et al., 2012; Smuts, 1995). If holding political office at least roughly depends on displaying dominance and resource control, it is likely to attract mates, even among humans. More will be discussed in this regard under Variable 13, having to do with mate preferences.

In terms of a physiological explanation, several studies have found evidence that brain exposure to testosterone promotes tendencies to pursue power and to achieve dominance and social status. Regarding prenatal testosterone exposure, two 2D:4D finger length studies indicated that high exposure was associated with status-striving later in life (Neave et al., Reference Neave, Laing, Fink and Manning2003; Moffit & Swanik, Reference Moffit and Swanik2011).

Regarding postpubertal (circulating) testosterone levels, at least five studies of males have found significant positive correlations between testosterone and status-striving behavior (Arnocky et al., Reference Arnocky, Albert, Carré and Ortiz2018; Dabbs & Dabbs, Reference Dabbs and Dabbs2000; Määttänen et al., Reference Määttänen, Jokela and Hintsa2013; Mazur & Booth Reference Mazur and Booth1998; Tremblay, Reference Tremblay1998). Two studies of females came to the same conclusion (Cashdan, Reference Cashdan1995; Grant & France, Reference Grant and France2001). Similar associations have been reported for dominance-striving behavior among nonhuman animals (Helle et al., Reference Helle, Laaksonen, Adamsson, Paranko and Huitu2008; Perret, Reference Perret2018). Overall, if one assumes that serving in an elective office partially reflects a form of status-striving, ENA theory provides an explanation for why males in all societies are more likely to hold elected offices than females. Therefore, I assigned three checkmarks to this variable for ENA theory in Table 1.

3.9.1 Social Role Theory

Nothing in social role theory seems to provide any way to account for why ADHD would be more common among males than females throughout the world. Therefore, it is listed in Table 1 as being silent regarding this sex difference.

3.9.2 ENA Theory

As already noted regarding Variable 5, males appear to have been evolutionarily favored for being more physically active than females (Cordain et al., Reference Cordain, Gotshall and Eaton1998; Jonason, Reference Jonason2007; Wilson et al., Reference Wilson, Daly, Pound, Pfaff, Arnold, Etgen, Fahrbach and Rubin2009). In foraging societies, high activity levels could promote exploratory and predatory behavior. For humans living in more technologically advanced societies, high activity levels could increase the chances of obtaining resources with which to attract mates.

While evolutionary reasoning can provide a possible explanation for why males would surpass females in average activity levels, one is still left wondering why a minority of males would be unusually hyperactive and inattentive. The answer may lie in noting that most evolved traits take on roughly normal curves of distribution, with population means being what is most favored (Parker, Reference Parker1974; Stearns, Reference Stearns1993). Accordingly, one can imagine two overlapping normal curves, one for males and one for females, regarding tendencies to be hyperactive. Because of sexual selection for traits linked to high activity levels, more males would be at the extreme right end of this normal curve. Thus, greater proportions of males than females would be diagnosed as exhibiting ADHD.

In line with this evolutionary reasoning, findings from three recent studies are worth noting. Two of these studies found that children who exhibit at least modest degrees of ADHD were more likely than boys in general to eventually become successful business owners (Lerner et al., Reference Lerner, Verheul and Thurik2019; Moore et al., Reference Moore, McIntyre and Lanivich2021). If we assume that business ownership is more likely to result in above average incomes compared working for wages (see Bernhardt, Reference Bernhardt1994; Schneck, Reference Schneck2020), a positive relationship between entrepreneurship and at least modest forms of ADHD is predictable.

The third study was based on a computer-simulated foraging experiment. It indicated that persons with ADHD ranged further afield than did persons in general. This finding was interpreted as suggesting that ADHD traits may have evolved to promote the use of relatively risky resource-securing strategies (Barack et al., Reference Barack, Ludwig and Parodi2022).

Turning to the possibility that brain exposure to testosterone contributes to sex differences in ADHD, there is a great deal of evidentiary support. In the case of prenatal testosterone, seven studies have reported significant within-sex relationships between 2D:4D and ADHD symptoms (Akkuş & Avşar, Reference Akkuş and Avşar2024; De Bruin et al., Reference de Bruin, Verheij, MA and Ferdinand2006; Martel et al., Reference Martel, Gobrogge, Breedlove and Nigg2008; McFadden et al., Reference McFadden, Westhafer, Pasanen, Carlson and Tucker2005; Roberts & Martel, Reference Martel2013; Stevenson, et al., Reference Stevenson, Everson and Williams2007; Wang et al., Reference Wang, Chou and Chou2017), although one study reported no significant relationship (Lemiere et al., Reference 64Lemiere, Boets and Danckaerts2010).

Regarding postpubertal testosterone, the evidence has been mixed (review: Camara et al., Reference Camara, Padoin and Bolea2022). Such findings are not particularly surprising, however, given that ADHD symptoms are primarily diagnosed prior to the onset of puberty. To summarize, ENA theory seems to provide reasonable explanations for why ADHD is more common among males than among females in all countries ever studied. Nevertheless, because of a few inconsistencies in the current evidence, I have assigned two checkmarks to ENA theory for explaining this sex difference.

3.10.1 Social Role Theory

The only way social role theory seems capable of explaining why conduct disorder is more common among males than females would entail attributing the difference to the fact that males are taller and physically stronger than females. In other words, given that conduct disorder is often associated with physical aggression, persons who are larger and stronger might stereotype themselves (and be stereotyped by others) as being more likely to get away with being physically aggressive in order to achieve their goals. Such an explanation, however, would be weakened by the fact that conduct disorder is usually diagnosed in childhood, years before sex differences in height and physical strength are detectible (Ellis et al., Reference Ellis, Palmer, Hopcroft and Hoskin2024a, pp. 48, 145–147). For this reason, I have given an x (meaning no explanation) to social role theory’s ability to explain sex differences in conduct disorder.

3.10.2 ENA Theory

From an evolutionary standpoint, ENA theory would anchor its explanation for sex differences in conduct disorder in the assumption that males have been reproductively favored for being relatively competitive, aggressive, and prone to take risks and that the underlying neurology is likely due to prenatal exposure to testosterone. Later in life, such sex differences would likely enhance an individual’s tendencies to acquire resources with which to attract mates (Griskevicius et al., Reference Griskevicius, Tybur and Gangestad2009; Wong & Candolin, Reference Wong and Candolin2005). Furthermore, it is worth reiterating the point that evolved traits will often vary around means that have been optimally favored.

As explained earlier regarding ADHD, at a population level, most evolved traits take on roughly normal (bell-shaped) distributions around their optimal reproductive value. This means that some individuals will exhibit these traits in ways that substantially undershoot and overshoot what is being naturally (including sexually) selected (Vercken et al., Reference Vercken, Wellenreuther, Svensson and Mauroy2012). If we assume that the optimal expressions of traits – such as competitiveness, risk-taking, and aggression – is higher for males than for females, we should expect higher proportions of males than females to overshoot what is optimally favored as well. Such reasoning leads one to expect greater proportions of males to exhibit conduct disorders than females.

Regarding the possibility of brain exposure to testosterone promoting conduct disorder, some supportive evidence has appeared in the scientific literature. Specifically, two studies of adolescent males found significant positive correlations between circulating levels of testosterone and conduct disorder (Rowe et al., Reference Rowe, Maughan, Worthman, Costello and Angold2004; Kirillova et al., Reference Kirillova, Vanyukov, Kirisci and Reynolds2008).

In the case of prenatal testosterone, one study indicated that levels were positively correlated with “disruptive behavior” among males, although it found no significant correlation between such behavior and circulating testosterone levels (Wang et al., Reference Wang, Chou and Chou2017). Overall, ENA theory seems to provide a reasonable evolutionary account for why males exhibit conduct disorder to a greater degree than females. More research is needed, however, regarding any testosterone connection. At the present time, I believe that ENA theory warrants just two checkmarks for its explanation of sex differences in conduct disorder.

3.11.1 Social Role Theory

Social role theory can draw on the fact that only females bear offspring to help account for greater males participate in the paid workforce. Thus, at least in the case of females with children, particularly those who breastfeed their infants, the likelihood of interruptions in workforce participation would be enhanced. Supporting this reasoning, substantial research has indicated that females are more intermittent in their paid workforce participation than are males (review: Ellis et al., Reference Ellis, Palmer, Hopcroft and Hoskin2024c, p. 651).

Another way that social role theory can account for sex differences in participation in the paid workforce involves men’s greater physical strength. This sex difference would qualify men for a wider range of jobs than would be available to women. For example, strenuous manual labor, such as work in construction and the manipulation of heavy equipment, would be open to greater proportions of men than women. Research supporting this line of reasoning has come from studies comparing sex differences in occupational diversity, all indicating that jobs filled by males substantially surpass those filled by females (review: Ellis et al., Reference Ellis, Palmer, Hopcroft and Hoskin2024c, p. 674). Overall, social role theory explains sex differences in time spent in the paid workforce rather well, warranting it receiving three checkmarks in Table 1.

3.11.2 ENA Theory

The explanation offered by ENA theory for why males are more likely than females to be involved in paid workforce can be addressed in both evolutionary and neurohormonal terms. Regarding evolution, the theory’s assertion that females bias their mating choices in favor of males with resources means that males who focus greater time and energy on obtaining resources will often have a reproductive edge over other males (Booth et al., Reference Booth, Carver and Granger2000; Ellis, Reference Ellis2001; Pawłowski, Reference Pawłowski2000).

In the case of ENA theory’s neurohormonal component, time spent in the paid workforce should be positively correlated with brain exposure to testosterone. No direct empirical evidence pertaining to this hypothesis was found. As a result, ENA theory receives two checkmarks beside this variable in Table 1.

3.12.1 Social Role Theory

The ability of social role theory to explain why risk-taking behavior is a USD would require assuming that risk-taking is promoted by physical strength. To test this assumption, one could seek to determine if, even within each sex, physical strength and risk-taking are positively correlated. Two studies that were located indicated that such a positive intra-sex correlation exists (Ball et al., Reference Ball, Eckel and Heracleous2010; Cho & Ahn Reference Cho and Ahn2020). Since this evidence is limited to confirming just one hypothesis, I gave two checkmarks to social role theory for explaining sex differences in risk-taking.

3.12.2 ENA Theory

The evolutionary component of ENA theory would offer the following reasoning about risk-taking: Such behavior could have evolved because it can often promote successful resource provisioning, and when males succeed in this regard, they increase their chances of attracting mates and thereby passing genes on to subsequent generations (Kruger et al., Reference Kruger, Wang and Wilke2007). In other words, because females have evolved tendencies to prefer mates with resources (see Variable 13), risk-taking should be more pronounced in males than in females.

Regarding physiological underpinnings for risk-taking, ENA theory asserts that exposing the brain to testosterone should be central. Considerable evidence supports this assertion, both in terms of prenatal and postpubertal testosterone.

Concerning prenatal testosterone, most of the research comes from studies of the 2D:4D finger length ratio. Despite the crudeness of this indicator, I found nine pertinent studies, all of which indicated that high prenatal testosterone exposure is associated with greater risk-taking tendencies later in life (Aycinena et al., Reference Aycinena, Baltaduonis and Rentschler2014; Brañas‐Garza et al., Reference Brañas-Garza, Galizzi and Nieboer2018; Coates et al., Reference Coates, Gurnell and Rustichini2009; Evans & Hampson, Reference Evans and Hampson2014; Finley et al., Reference Finley, Kalwij and Kapteyn2022; Manning, Bundred et al., Reference Manning, Bundred, Newton and Flanagan2003; Ronay & von Hippel, Reference Ronay and Von Hippel2010b; Stenstrom et al., Reference Stenstrom, Saad, Nepomuceno and Mendenhall2011; Vermeersch et al., Reference Vermeersch, T’Sjoen, Kaufman and Vincke2008).

Turning to postpubertal circulating testosterone, which can be measured directly from the blood or saliva, studies also support the hypothesis of a positive correlation between brain exposure to this hormone and people’s tendencies to take risks (Dariotis et al., Reference Dariotis, Schen and Granger2016; Votinov et al., Reference Votinov, Knyazeva, Habel, Konrad and Puiu2022; reviews: Apicella et al., Reference Apicella, Carré and Dreber2015; Kurath & Mata, Reference Kurath and Mata2018). Overall, substantial scientific research supports ENA theory’s explanation for why risk-taking is more common among males than females. For this reason, I gave this theory three checkmarks in Table 1 for explaining sex differences in risk-taking.

3.13.1 Social Role Theory

Social role theory can be used to explain this sex difference in mate preferences by noting that only females are capable of bearing children and, once born, children benefit from receiving substantial parental care (often including breastmilk, which only females can provide). One way to test this explanation for why females are more interested than males are in a mate’s ability to make a living involves females being stereotyped as the primary caregiver to children. Sex role theorists could then attribute the universality of this stereotype to the fact that only females can bear offspring (and breastfeed them after birth). One way to test this rather circular theoretical argument would be to study heterosexual women who intend to have children to those who do not intend to have children. If social role theory is correct, women who intend to remain childless should be less interested in marrying a man with the ability to make a comfortable living than would women intending to have children. Unfortunately, I could find no research bearing on this hypothesis. Therefore. just one checkmark was assigned to social role theory for potentially explaining this universal sex difference.

3.13.2 ENA Theory

The evolutionary component of ENA theory argues that, because only females can bear offspring and breastfeed them during infancy, they will have been favored for preferring mates with the ability to provide for them and their offspring throughout the offspring’s childhood (Bereczkei & Voros, Reference Bereczkei, Voros, Gal and Bernath1997; Buss, Reference Buss1988; Ellis, Reference Ellis2001; Zhu & Chang, Reference Zhu and Chang2019). In other words, under most conditions, women who choose males who seem capable of being a “good provider” will leave more offspring in future generations than women whose mate choices prioritize other mate characteristics (Brooks et al., Reference Brooks, Blake and Fromhage2022).

At the physiological level, ENA theory makes two predicts pertaining to sex differences in preferences for mates. First, it predicts that low brain exposure to testosterone will be associated with preferring mates with resources. No evidence directly bearing on this hypothesis was found.

The second prediction is that individuals with relatively high testosterone exposure will be better at accessing resources than those with low testosterone levels. This should even hold true within each sex. As noted earlier regarding Variable 1, considerable evidence supports this prediction, especially regarding postpubertal testosterone. Specifically, four studies of men have all concluded that income was positively correlated with circulating testosterone levels (Coates & Herbert, Reference Coates and Herbert2008; Harrison et al., Reference Harrison, Davies, Howe and Hughes2021; Hughes & Kumari, Reference Hughes and Kumari2019; Luoto, Krama et al., Reference Luoto and Varella2021). Evidence regarding prenatal testosterone also points toward a positive correlation (Nye et al., Reference Nye, Yudkevich, Orel and Kochergina2014; Gielen et al., Reference Gielen, Holmes and Myers2016). Overall, I have assigned three checkmarks to ENA theory for its ability to account for sex differences in the tendency to prefer mates with resources.

3.14.1 Social Role Theory

According to social role theory, sex differences in occupational interests, and thereby in academic areas of study, are the result of sex role socialization (Eccles, Reference Eccles1994). This idea would lead one to expect considerable cultural variability in the interests expressed by males and females when it comes to choosing college majors. Instead, our review of the literature indicated that, except for chemistry, considerably higher proportions of males select the physical sciences or engineering as majors in all countries that have been sampled. Overall, social role theory offers no explanation for this universal sex difference, leading me to assign it an “x” in Table 1.

3.14.2 ENA Theory

From an evolutionary perspective, female tendencies to prefer mates who can provide them and their offspring with resources should motivate males to take up a wider diversity of occupations that is true for females. Accordingly, ENA theory would predict that males would be more likely females to develop interests in many occupations having nothing to do with rearing offspring or even people in general. Obviously, the physical sciences and engineering often fit such a description. Evidence supporting such reasoning includes research showing that when males are compared to females, males (a) are more interested in things rather than people (Ellis et al., Reference Ellis, Palmer, Hopcroft and Hoskin2024b, pp. 868–871), (b) are better at spatial reasoning, especially regarding mental rotation (Ellis et al., Reference Ellis, Palmer, Hopcroft and Hoskin2024b, pp. 261–267), and (c) have a greater probability of being on the autism spectrum (Ellis et al., Reference Ellis, Palmer, Hopcroft and Hoskin2024b, pp. 656–657).

If ENA theory is correct, these three sex differences evolved to help males to comply with female preferences for competent resource procuring mates. Furthermore, the theory predicts that sex differences greater interest in things than in people are promoted by exposing the brain to male-typical levels of testosterone. Nevertheless, extreme testosterone exposure appears to sometimes result in what is known as the autism spectrum disorder (Baron-Cohen et al., Reference Baron-Cohen, Lutchmaya and Knickmeyer2006). Among the evidence that supports this deduction is that indicating that brain exposure to testosterone promotes interests in things, spatial reasoning ability, and autism (Ellis et al., Reference Ellis, Palmer, Hopcroft and Hoskin2024d, pp. 388–406, 461–463; Greenberg et al., Reference Greenberg, Warrier, Allison and Baron-Cohen2018).

Furthermore, given the relatively high incomes associated with physical science and engineering occupations, ENA theory would hypothesize that individuals entering these professions will often be attractive to females. This is despite the fact that some males end up being exposed to such high levels of testosterone that they risk suffering from the autistic spectrum disorder (Cohen-Bendahan et al. Reference Cohen-Bendahan, Van De Beek and Berenbaum2005; Greenberg et al., Reference Greenberg, Warrier, Allison and Baron-Cohen2018), ADHD, and extreme antisocial disorder (see especially the information on Variable 15 next).

Overall, tending to major in the physical sciences and engineering in college are more common among males than females in all countries studied. However, I was unable to find direct evidence linking testosterone to these tendencies. Therefore, I have given ENA theory just two checkmarks in Table 1.

3.15.1 Social Role Theory

Social role theory has already been discussed regarding two other crime-related variables: self-reported criminality (Variable 3) and official criminality in general (Variable 4). In neither case did social role theory seem to provide anything beyond weak explanations for why these general criminality measures would be more prevalent among males than among females. However, given that nearly all forms of homicide can be considered inherently violent, social role theory can predict that males would be more likely than females to engage in this type of offending.

The explanation would be as follows: Eagly and Wood (Reference Eagly and Wood2012, p. 465) have argued that because of men’s greater strength, they are more confident in their abilities to intimidate rivals and to win in aggressive encounters than are females. Consequently, social role theory predicts that homicide, as well as other violent crimes, should be more prevalent among males.

Supporting this prediction, one study indicated that, among males, muscularity is positively correlated with involvement in violent crime (Maddan et al., Reference Maddan, Walker and Miller2008) although another study reached the opposite conclusion (Beckley et al., Reference Beckley, Kuja-Halkola, Lundholm, Långström and Frisell2014). Overall, since social role theory has at least some empirical support, I have given it two checkmarks beside Variable 15 in Table 1.

3.15.2 ENA Theory

The ENA theory offers both an evolutionary and a physiological explanation for why most murderers are males. As already noted, ENA theory postulates that males have been favored for being more competitive, aggressive, and inclined to take risks, thereby increasing their chances of obtaining resources with which to attract mates (Ellis, Reference Ellis2005). Violence can be seen as a frequent expression of extreme competitiveness, aggression, and risk-taking (Wilson & Daly, Reference Wilson and Daly1985; Villarreal, Reference Villarreal2002). Accordingly, one can predict that males would have evolved all three of these tendencies more than females. While there are some exceptions depending on species and circumstances surrounding the violence, humans appear to share this sex difference with nearly all other mammals (Ellis et al., Reference Ellis, Palmer, Hopcroft and Hoskin2024c, pp. 371–374). In other words, in nearly all species of mammals, males are more violent than females. This suggests that evolutionary reasoning, rather than cultural sex role training, offers a more parsimonious explanation for sex differences in violence, with homicide being an especially extreme expression of violence.

The neurohormonal component of ENA theory asserts that brain exposure to testosterone contributes to physical violence (Ellis & Hoskin, Reference Ellis and Hoskin2015; Johnson & Thayer, Reference Johnson and Thayer2016). Much of this contribution seems to involve testosterone’s ability to alter the brain in ways that enhance an individual’s motivation to be competitive and take risks (Casto & Edwards, Reference Casto and Edwards2016; Ronay & Von Hippel, Reference Ronay and Von Hippel2010). Other testosterone influences seem to promote a lowered sensitivity to the pain caused to others (Vongas & Al Hajj, Reference Vongas and Al Hajj2015; Zhuo et al., Reference Zhuo, Zhang and Fan2022). Regarding the latter point, nearly all studies have indicated that males are less empathetic than females (review: Ellis et al., Reference Ellis, Palmer, Hopcroft and Hoskin2024b, pp. 170–172).

It is also worth mentioning that while social role theory asserts that physical strength helps to promote physical aggression (Eagly & Wood Reference Eagly and Wood2012, p. 465), it does not explain why such a relationship would exist. The ENA theory does offer an explanation. Specifically, it indicates that testosterone not only affects brain functioning but also impacts other parts of the body, including muscularity and physical strength (Bhasin, Woodhouse et al. Reference Bhasin, Woodhouse and Casaburi2001; Sipilä et al., Reference Sipilä, Narici and Kjaer2013). This difference between social role theory and ENA theory means that the latter is more vulnerable to refutation, which is, of course, a good feature in scientific theorizing (Archibald, Reference Archibald1967).

So, is there support for ENA theory regarding testosterone exposure being responsible for both physical strength and physical aggression? At least three lines of evidence point toward an affirmative answer. First, within both sexes, testosterone appears to promote violent behavior (Mhillaj et al., Reference Mhillaj, Morgese and Tucci2015; Pfaff, Reference Pfaff2002; Swerdloff et al., Reference Swerdloff, Wang, Hines and Gorski1992). Second, even within each sex, most research has indicated that both prenatal and postpubertal testosterone levels are positively correlated with aggression and violence (review: Ellis et al., Reference Ellis, Farrington and Hoskin2019, 351–355). Third, regarding murderers specifically, three studies have concluded that they have higher average levels of testosterone than do males in general (Alnourani et al., Reference Alnourani, Osman and Alsheikh2017; Delgado et al., Reference Delgado, Maya-Rosero, Franco, Montoya-Oviedo, Guatibonza and Mockus2020; McAndrew, Reference McAndrew2009). Given the weight of the this evidence, I have given ENA theory three checkmarks in Table 1 for its ability to explain the universal tendency for males to commit more homicides than females.