In this article we will discuss about:- 1. Introduction to Sex Differences in Human Physiology 2. Basic Concepts of Differences 3. Models Influences 4. Hormonal Influences 5. Normal Variability in Hormones 6. Sex Differences in Human Behaviour 7. Hormones 8. Aggression and Personality Characteristics of Humans.
Contents
1. Introduction to Sex Differences in Human Physiology:
Sexual differentiation is the process of development of the differences between males and females from an undifferentiated zygote (fertilized egg). As male and female individuals develop from zygotes into fetuses, into infants, children, adolescents, and eventually into adults, sex and gender differences at many levels develop- genes, chromosomes, gonads, hormones, anatomy, and psyche.
Sex differences range from nearly absolute to simply statistical. Sex-dichotomous differences are developments which are wholly characteristic of one sex only. Examples of sex-dichotomous differences include aspects of the sex- specific genital organs such as ovaries, a uterus or a phallic urethra.
In contrast, sex-dimorphic differences are matters of degree (e.g., size of phallus). Some of these (e.g., stature, behaviours) are mainly statistical, with much overlap between male and female populations.
Nevertheless, even the sex-dichotomous differences are not absolute in the human population, and there are individuals who are exceptions (e.g., males with a uterus, or females with an XY karyotype), or who exhibit biological and/or behavioural characteristics of both sexes.
Sex differences may be induced by specific genes, by hormones, by anatomy, or by social learning. Some of the differences are entirely physical (e.g., presence of a uterus) and some differences are just as obviously purely a matter of social learning and custom (e.g., relative hair length). Many differences, though, such as gender identity, appear to be influenced by both biological and social factors (“nature” and “nurture”).
The early stages of human differentiation appear to be quite similar to the same biological processes in other mammals and the interaction of genes, hormones and body structures is fairly well understood.
In the first weeks of life, a fetus has no anatomic or hormonal sex, and only a karyotype distinguishes male from female. Specific genes induce gonadal differences, which produce hormonal differences, which cause anatomic differences, leading to psychological and behavioural differences, some of which are innate and some induced by the social environment.
The various ways that genes, hormones, and upbringing affect different human behaviours and mental traits are difficult to test experimentally and charged with political conflict.
2. Basic Concepts of Differentiation:
In general, gonadal hormones have two types of influences on brain and behaviour, termed organizational and activational. Organizational influences typically occur early in life, usually during critical periods of development, and they are permanent. The critical period concept implies that the hormone must be present at a specific time to exert its effect.
However, despite the brevity of its presence, its effect persists across the life span and cannot be reversed by subsequent hormone withdrawal. For instance, if a genetically female (XX) rat is injected with a single dose of testosterone shortly after birth, as an adult she will no longer be able to show female-typical sexual behaviours.
These early, permanent effects of hormones are thought to occur because of developmental influences on brain organization. Hence, they were termed organizational effects. Activational influences of hormones differ from organizational influences in that they occur later in life, typically in adulthood, and are reversible.
The hormone has an effect as long as it is present, but the effect wanes or disappears when the hormone is withdrawn. An example would be the changes in sexual receptivity that are shown over the estrous cycle in rodents.
Both organizational and activational effects of hormones can involve structural changes in the brain. They are distinguished from one another by their time of occurrence and by their permanence.
Masculinization and Feminization:
In common parlance, masculine means like a typical male and feminine means like a typical female. Accordingly, masculinization would mean becoming more, like the typical male whereas feminization would mean becoming more like the typical female. However, in research on sexual differentiation, these terms have technical meanings that differ somewhat from this colloquial usage.
Masculinization refers to movement along a continuum from none of a characteristic that is more common (or larger) in males than in females to the greatest possible amount of the characteristic. The most typical example of a characteristic that can be masculinized is the mounting behaviour exhibited by adult male animals when they encounter a sexually receptive female.
Feminization refers to movement along a continuum from none of a characteristic that is more common (or larger) in females than in males to the greatest possible amount of the characteristic.
A typical example of a characteristic that can be feminized is the lordosis posture (arching the back and deflecting the tail) that a receptive female rodent shows when mounted by a male animal.
Sex and Gender Differences:
Sex differences and gender differences are characteristics that differ on the average for males and females. In addition, some researchers studying sexual differentiation use the term sexual dimorphisms to refer to sex differences or gender differences. Hence, although this term technically means two forms it will sometimes be used, like sex differences and gender differences, to refer to characteristics that differ on average for the sexes.
3. Models Influences on Sex Differences:
1. The Classic Model:
The classic model of hormonal influences on sexual differentiation posits that testicular hormones cause masculine-typical development, while their absence causes feminine-typical development.
For instance, exposure of XX rodents to testosterone during critical periods of prenatal or neonatal development results in adult animals who are able to show masculine-typical sexual behaviour (e.g., mounting) but not feminine- typical sexual behaviour (e.g., lordosis).
Similarly, castration of XY animals early in life produces adults who show reduced masculine-typical behaviour and increased feminine-typical behaviour. The same hormone treatments also produce permanent changes in the mammalian brain.
Perhaps the best known example involves a sub-region of the anterior hypothalamic-preoptic area called the sexually dimorphic nucleus of the preoptic area (SDN-POA). This nucleus is several folds larger in male than female rats.
However, treating genetic females with testicular hormones during early life permanently enlarges the nucleus while withdrawing these hormones from developing males permanently reduces it. In contrast to the dramatic effects produced by manipulating testicular hormones, removal of the ovaries at comparable early stages of development generally has little or no impact on a female animal’s subsequent ability to show masculine-typical or feminine- typical behaviour or on the SDN-POA.
In addition, treating genetic female animals with estrogen generally produces results that are the opposite of what would be predicted if ovarian hormones had feminizing influences. It promotes masculine-typical behavioural and brain development, including development of the SDN-POA, and impairs feminine-typical development.
This is understood to occur because testosterone is normally converted within the brain to estrogen before interacting with receptors to produce masculine-typical development, at least in regard to many brain regions and behaviours in rodents.
In general, the classic influences of testicular hormones occur during critical periods of prenatal and neonatal development. The critical periods vary somewhat from one species to another, but appear to correspond to times when hormones are higher in developing males than females.
In the rat, which is the most studied animal model of these hormonal effects, this occurs from about the 17th to the 19th day of an approximately 21 day gestational period and from the first to the 10th day of postnatal life. Within this overall critical period there are separate periods when specific sexually differentiated characteristics are most sensitive to hormonal influences.
For instance, mounting is sensitive to hormonal influences somewhat later than lordosis. As we know that, this type of evidence has led to conceptualization of masculinization and feminization as separate processes. As a consequence, sexual differentiation is viewed as involving at least two dimensions.
As an animal becomes more masculine it does not necessarily become more feminine, as would be the case if sexual differentiation involved only a single dimension with masculine at one end and feminine at the other.
In fact, by timing hormonal manipulations to hit or avoid periods during which specific behaviours differentiate, animals can be created which are both masculinized and femininized (i.e., can show both mounting and lordosis) or demasculinized and defeminized (i.e., can show neither mounting nor lordosis), as well as those which are conventionally masculine (i.e., can show mounting but not lordosis) or conventionally feminine (i.e., can show lordosis but not mounting).
2. The Gradient Model:
Hormones influence not only behavioural differences between the sexes, but also differences within each sex. This is because hormonal influences are graded; when a greater quantity of hormone is administered a more dramatic change in behaviour occurs. Naturally occurring variations in hormones also appear to relate to variations in behaviour within each sex.
Female rodents show substantial individual variability in masculine-typical behaviours. For instance, some females mount other females and some do not. Female rats exposed to blood that has contacted male littermates (because of their position relative to male siblings in utero) show more mounting as adults than those who are not so-positioned.
Studies in other rodents, including mice and gerbils, have produced similar results. This is thought to occur because females positioned near males are exposed to some testicular hormones in blood from their male siblings. In support of this interpretation, gerbil fetuses positioned between two males have higher levels of testosterone than those positioned between two females.
Because of these graded influences of hormones, a modification of the classic mode of hormonal influences, called the gradient model, has been proposed. In this model, not only do testicular hormones cause differences between normal male and female animals, but also small amounts of hormones produce movement along the male and female gradients within each sex.
Active Feminization:
Although the classic model of hormonal influence clearly applies to much of mammalian development, it may not explain all of sexual differentiation. Because the model suggests that ovarian hormones are not needed for feminine- typical development, it has sometimes been referred to as the passive feminization model and it has been argued that passive feminization may not apply to all sexually differentiated characteristics. The best-known formulation of the active feminization hypothesis cites two types of supporting evidence.
The first type of evidence cited in support of active feminization involves movement along a dimension of masculinization from no masculinization to the small amount of masculinization seen in the average female. For instance, the SDN-POA exists in both male and female rats, but is larger in males.
This is because both males and females have some estrogen, the hormone that causes masculinization of the nucleus, although males have substantially more. If females are treated during early development with tamoxifen, a substance that blocks the action of estrogen, the SDN-POA is reduced in size.
This has been referred to as defeminization of the SDN-POA. However, this is true only in the colloquial sense of the term, feminization. In the scientific terminology of sexual differentiation, reduction of a characteristic, like the SDN- POA, that is larger in males is demasculinization from the point that the average female reaches on a continuum of no masculinization to full masculinizatiort.
Thus, this is evidence that the classic effects of hormones are graded, not evidence of a separate process of active feminization in the sense of ovarian hormones promoting the development of female-typical characteristics.
The second type of evidence involves the promotion of female-typical characteristics by ovarian hormones, and thus does reflect a novel process to that proposed in the classic model or gradient model of hormone influences. For instance, the presence of ovarian hormones near the time of puberty may permanently enhance some aspects of feminine- typical sexual behaviour in the rat.
Ovarian hormones might also promote feminine-typical development of some structural characteristics in the cerebral cortex, including asymmetries in cortical thickness and the size of the corpus callosum in rodents. Like the effects on feminine-typical sexual behaviour, these feminizing effects of estrogen on cortical development appear to occur somewhat later in life than the classic influences of testicular hormones.
The feminizing influences of ovarian hormones have not been studied as extensively as the masculinizing influences of testicular hormones. However, assuming they prove to be reliable, they suggest that there is a critical period when ovarian hormones actively feminize some characteristics, at least in rodents.
This critical period would appear to occur somewhat later in life than the critical period for the effects of testosterone, and it would be interesting to know if there is a corresponding period when estrogen levels are higher in the female brain than the male brain.
Additional research also is needed to demonstrate that the effects of ovarian hormones on development are truly organizational in the sense of being irreversible, and that they occur only during a particular critical developmental period.
Finally, conceptualization of sexual differentiation has proceeded from a one-dimensional continuum with masculine at one end and feminine at the other, to a two-dimensional space, defined by separate masculine and feminine axes. However, more than two dimensions are probably needed.
The terms, masculinization and feminization were originally used to separate the influences of hormones on different characteristics. Masculinization referred to increases in male-typical characteristics, usually mounting, and feminization referred to increases in female-typical characteristics, usually lordosis, but also sometimes gonadotropin regulation. The necessity for separating these characteristics followed from evidence that they could each be influenced independently by hormones.
The terms masculinization and feminization have been adopted for use in reference to other behaviours as well as brain regions, with masculinization referring to those more common or larger in males, and feminization referring to those more common or larger in females.
These are unlikely to all become sexually differentiated under the influences of the same factors or via the same mechanisms. In fact, there is evidence that in primates they do not. This is also likely to be true for the myriad neural and behavioural characteristics that are more common or larger in females.
A multidimensional conceptualization is probably needed to provide a complete understanding of sexual differentiation, wherein each behaviour or brain region that is susceptible to hormone influence is conceptualized on a separate dimension.
Such a conceptualization would also allow for different sexually differentiated characteristics to conform to different models of hormone action, such as a gradient version of the classic model for some characteristics and a model involving active feminization for others.
4. Hormonal Influences:
It is generally unethical to manipulate hormone levels during human development for experimental purposes. Therefore, true experiments similar to those conducted in animals are not possible. However, some naturally occurring situations have shed light on the relevance of animal models to human neural and behavioural sexual differentiation.
These include genetic and other syndromes involving hormonal abnormality, instances where women have been prescribed hormones during pregnancy and situations where normal variability in hormones during development has been related to subsequent behaviour.
Evidence from genetic syndromes and situations where women have been prescribed hormones during pregnancy indicate that, as in other mammals, differentiation of the human internal and external genitalia follows processes consistent with the classic model of hormonal influence.
High levels of testicular hormones promote masculine- typical development, whereas, in the absence of these hormones, feminine structures appear. Although both internal and external genital structures are influenced by the presence or absence of testicular hormones, some details of the mechanisms for differentiation of the internal genitalia differ from those for the external genitalia.
In the case of the internal genitalia, XX and XY fetuses both begin with Mullerian ducts as well as Wolffian ducts. A portion of the Y chromosome directs the gonads to differentiate into testes, and by week 8 of gestation, almost all XY fetuses have functioning testes.
One hormone produced by the testes, Mullerian inhibiting factor (MIF), then causes the Mullerian ducts to regress, while a second hormone, testosterone, causes the Wolffian ducts to develop into the masculine internal genitalia. In contrast, in XX fetuses, the gonads differentiate into ovaries instead of testes. Then, in the absence of testicular hormones, the Wolffian ducts regress and the Mullerian ducts develop into the feminine internal genitalia (uterus, fallopian tubes, and the upper portion of the vagina).
In contrast to the internal genitalia, where two sets of structures are initially present in both XX and XY organisms and one regresses, the external genitalia begin as one set of structures, identical in both XX and XY fetuses. In the presence of testicular androgens, particularly dihydrotestosterone (DHT), these structures become penis and scrotum.
In the absence of these hormones, the same structures become clitoris, labia, and the lower portion of the vagina. Thus, although both the internal and external genitalia differentiate under the influence of testicular hormones, the processes differ in that for the internal genitalia, both XX and XY fetuses begin with two sets of structures, one of which is lost, whereas for the external genitalia, both XX and XY fetuses begin with the same single set of structures that then develops differently depending on the hormone environment.
Thus, although testicular hormones are important for both internal and external structures and operate in accord with the classic model of hormone action, the specific mechanisms involved differ. In addition, the specific hormones involved in differentiation of the two sets of structures differ, with T and MIF influencing internal structures and DHT bearing primary responsibility for external structures.
These processes of physical sexual differentiation have been established in experimental studies involving hormone manipulations in nonhuman mammals, as well as by observing the consequences of abnormal hormone environments for human genital development. Essentially identical mechanisms govern sexual differentiation of the internal and external genitalia in humans as in other mammals.
XX individuals exposed to high levels of adrenal androgens, because of genetic problems, or to high levels of androgenic progestin, because their mothers were prescribed these hormones during pregnancy, are born with masculinized external genitalia. However, because they were not exposed to MIF, their internal genital structures remain female.
Similarly, XY individuals who produce normal levels of testicular hormones, but whose cells cannot respond to androgen because of a genetic defect, are born with female-appearing external genitalia, but lack female internal genitalia because these have been inhibited by MIF from their testes.
Hormone Administration during Pregnancy:
With the exception of ablatio penis, the syndromes involve genetic abnormalities or other disorders intrinsic to the individual, and they can have continued manifestations across the life span, independent of the prenatal hormonal abnormalities associated with them.
In contrast to these endogenous causes of gonadal hormone abnormality, there are exogenous causes. In these situations, the hormonal abnormality is limited in time and is less likely to be accompanied by nonhormonal symptoms, such, as are often associated with genetic disorders. Exogenous causes of hormone exposure include situations where women have been prescribed hormones during pregnancy, usually for medical reasons.
Hormones that have been prescribed during pregnancy include estrogens and progestins. Historically, the most commonly prescribed hormone was the synthetic estrogen, diethylstilbestrol (DES). DES was prescribed to millions of women in the United States from the late 1940s to the early 1960s.
It was mistakenly thought to provide protection against miscarriage and was prescribed to women with a history of miscarriage, with threatened miscarriage and, in some cases, even as a routine precaution. Double-blind, placebo-controlled, studies eventually demonstrated that DES did not protect against miscarriage. It was finally removed from use in the United States when it was associated with an increased risk of vaginal and cervical adenocarcinoma in the early 1970s.
In non-human mammals, DES and other estrogens promote masculine-typical brain and behavioural development when administered prenatally or neonatally. These masculinizing and defeminizing effects of estrogen were originally considered paradoxical.
However, it is now well-established that testosterone from the testes is converted to estradiol within the brain, and, in normal male animals, estradiol acting through neural estrogen receptors is responsible for many aspects of masculine-typical neural and behavioural development.
Although most of the evidence supporting these masculinizing and defeminizing influences of estrogen has come from studies of rats and other rodents, there is some evidence supporting similar masculinizing influences of estrogen in rhesus macaques. Thus, if DES or other estrogens influence human sexual differentiation, they would be hypothesized to have masculinizing or defeminizing effects on developing females.
Progestins, the second main group of hormones that have been prescribed to pregnant women, are of two general types, progestational and androgenic. Progestational progestins interfere with the actions of androgens, whereas androgenic progestins mimic the actions of androgen. Thus, these two types of progestins would be predicted to have opposing effects, the first impairing masculine-typical development and the second promoting it.
5. Normal Variability in Hormones:
A third approach to studying hormonal influences on human sexual differentiation has related normal variability in hormones to behaviour. This has involved four types of studies. One has measured hormones in umbilical cord blood at birth and related these to childhood behaviour.
A second has measured hormones in amniotic fluid during gestation and related these to subsequent behaviour. A third has measured maternal hormone levels during pregnancy and related these to psychological traits in adult female offspring. And a fourth has compared same sex versus opposite sex, dizygotic, twins.
This last type of study is based on animal research finding that female fetuses located adjacent to male siblings in utero, show more masculine- typical behaviour as adults. In the animal studies, offspring can be cross-fostered to eliminate the postnatal impact of the sibling sex ratio on sexual differentiation of behaviour. This is not possible in humans and so these studies of human twins present interpretational difficulties.
6. Sex Differences in Human Behaviour:
Animal models suggest that behaviours that show sex differences are susceptible to hormonal influences, whereas those that do not are not. Thus, only behaviours that show sex differences would be hypothesized to be influenced by the early hormone environment.
The question of which behaviours these are has itself been debated, and the study of sex differences presents unique difficulties. One is that researchers bring their own preconceptions, or sex stereotypes, to their work. Because expectation’s can influence perceptions of outcomes or even actual outcomes, this can distort results.
A second problem is that finding differences between groups is generally viewed as more exciting, and is easier to publish, than finding no differences. This general problem is exacerbated for research on sex differences, since sex can be easily measured and is often routinely analyzed.
Because statistical decision rules result in a certain percentage of false positive results (5% with a set at .05), there is a high probability that spurious results of sex differences will be published. The group of behaviours thus identified also Corresponds well to the behaviours that have been studied in individuals exposed to atypical hormone environments during development.
The conclusion that behaviour shows a sex difference does not necessarily mean that males and females are dramatically different. Typically it means that when groups of men and women or boys and girls are compared, the groups show average differences. The size of these average differences varies.
Where possible, results of metaanalytic studies, which combine data from many studies to get reliable estimates of the sizes of group differences, will be cited to estimate the size of specific sex differences in human behaviour. In the behavioural sciences, d values for group differences of 0.8 or greater are termed large, those around 0.5 are termed moderate, and those around 0.2, are termed small. Below this point, sex differences are considered negligible. To put the size of behavioural sex differences into a familiar context, the sex difference in height has a d value of 2.0.
i. Core Gender Identity:
Core gender identity, or the sense of self as male or female, shows a sex difference. The vast majority of XY individuals think of themselves as boys or men and the vast majority of XX individuals think of themselves as girls or women. However, even this most basic aspect of sexual identity is not always consistent with genetic sex or with the physical appearance of a person as male or female.
In adults, the incidence of psychological identity as the other sex is approximately 1 in 20,000 to 1 in 30,000 in genetic males and 1 in 50,000 to 100,000 in genetic females in North America and Western Europe. Although gender identity disorder can also occur in children, estimates of its incidence in childhood are not available.
ii. Sexual Orientation:
Sexual orientation also shows a sex difference. The great majority of males are sexually attracted to and erotically interested in females, whereas for the great majority of females sexual attraction and erotic interest is focused on males. Again, this is not universal.
Data from Kinsey et al. suggested that about 10% of men and about 5% of women are bisexual or homosexual. More recent studies provide lower estimates, at least for males having homosexual experience, ranging from 2 to 6% in the United States, France, and Great Britain.
7. Hormones:
Having outlined the theoretical basis for predicting relationships between the early hormone environment and human behaviour, and having described the sources that can provide information and the behaviours that might be influenced by hormones, the question of whether hormones influence sexual differentiation of human behaviour can now be addressed.
i. Core Gender Identity:
There is some evidence that gonadal hormones influence the development of core gender identity. Women exposed to high levels of androgen during early development, because of CAH, appear to be at increased risk for developing the symptoms of gender identity disorder or for changing sex from female to male.
One study found that of 53 XX CAH patients seen at one clinic during a defined time period, one had been diagnosed with trans-sexualism and was now living as a man, despite assignment and rearing in the female sex. Gender identity disorder was estimated to occur in one in 30,400 cases of XX individuals in the general population, resulting in odds of 1 in 608 that this coincidence of gender dysphoria and CAH was a chance happening.
Studies of children also suggest that girls with CAH express reduced satisfaction with being a girl. In one study, 7 of 15 girls with CAH said they were content to be or preferred to be a girl, compared to 14 of 15 controls. In a second study, 6 of 17 girls with CAH said that, if given a choice, they might have chosen to be a boy or would have been undecided as to whether to be a boy or a girl compared to 1 of 17 unaffected sisters of CAH children.
However, in both studies, severe gender dysphoria was reported to be rare or nonexistent in CAH girls. A third study found that 2 of 18 girls with CAH in one clinic population met the diagnostic criteria for gender identity disorder of childhood, as did 5 of 29 children raised as girls who had been exposed to high levels of androgen prenatally because of other intersex conditions, including PAIS, cloacal exstrophy, and true hermaphroditism.
XY individuals whose genitalia appear female at birth, because of CAIS, are assigned and reared as females and have not been reported to wish to change sex as adults. Published reports regarding their gender identity conclude that they are content with the female sex of assignment. This suggests that lack of stimulation by androgen, at least when combined with an unambiguously female sex of rearing, produces a female core gender identity. A second X chromosome is apparently not needed, nor is ovaries.
A final group of syndromes that has been studied in hopes of elucidating the role of hormones in the development of core gender identity are the deficiencies in enzymes needed to produce androgen. Imperato-McGinley et al. reported on 18 individuals with 5aR deficiency who lived in an isolated community in the Dominican Republic.
These XY individuals had been born with undervirilized genitalia, were assigned and reared as females, and were reportedly content in the female role as children. However, following physical virilization at puberty, 17 of the 18 lived as males, and had adopted a masculine core, gender identity.
This outcome was interpreted to support a role for androgen in the development of masculine gender identity. The report was surprising, because it contradicted the point of view which prevailed at the time; that is, that the sex of rearing determines gender identity. It also raises questions from a hormonal standpoint in that the presumed critical period for hormonal influences on gender identity formation is prenatal or neonatal, certainly prior to puberty.
ii. Sexual Orientation:
The two cases where boys were reassigned as girls, following surgical damage to the penis, differed not only in outcomes for core gender identity, but also in outcomes for sexual orientation. The child for whom the damage occurred at the age of 8 months had erotic interest exclusively in women as an adult, whereas the child for whom the damage occurred at the age of 2 months was bisexual.
Although they differ in that, one individual was bisexual and the other interested exclusively in female erotic partners, both cases suggest that early exposure to masculine-typical levels of testicular hormones influences sexual orientation away from the primary or exclusive erotic interest in men that is usually seen in females.
Data on women with CAH also suggest a hormonal contribution to sexual orientation. Even when treated early in life to control postnatal hormone levels and surgically feminized and reared as girls, females with CAH are more likely than women in general to be bisexual or homosexual.
In one study, 7 of 30 women with CAH were noncommittal about their sexual orientation, 12 rated themselves as heterosexual, and the remaining 11 as bisexual or homosexual. This differed from controls who had either CAIS or Rokitansky syndrome, all 27 of whom were willing to reveal their sexual orientation, with 25 indicating that they were exclusively heterosexual.
A second study compared 34 girls and women with CAH to 14 of their unaffected sisters. Looking at those old enough to have had sexual experiences, none of the controls and 44% of the CAH women desired or had experienced sexual relations with other women. On inventories of sexual orientation, the CAH women scored higher on homosexual interest, lower on heterosexual interest, and lower on general sexual interest than did their sisters.
A third study, compared 30 women with CAH to 15 unaffected female relatives, and found reduced heterosexual activity and reduced general sexual activity in the CAH group. A fourth study again found reduced sexual activity in CAH women compared to matched controls, as well as reduced experience in relationships.
However, self-report of homosexuality appeared similar in the two groups. This contrasts with findings from the other studies, in that it does not find evidence of a relationship between early androgen exposure and sexual orientation. This could reflect methodological differences.
This study, unlike the others, included women with late onset CAH whose hormonal disorder was likely to have begun after any critical period for hormonal influences on sexual differentiation. In addition, it assessed whether CAH women called themselves homosexual and lived with a female partner, whereas the other studies included information on desires and fantasies, as well as actual behaviour, in assessing sexual orientation.
iii. Childhood Play:
The clearest evidence of hormonal influences on sexual differentiation of human behaviour has come from studies of childhood play. Girls exposed to high levels of androgens prenatally, because of CAH, show increased interest in masculine-typical toys, activities, and playmates and reduced interest in feminine-typical toys, activities, and playmates.
These findings have been reported based on interviews with and questionnaire data from CAH girls and their mothers, as well as on direct observation of toy choices in a playroom. In the observational study, girls with CAH spent more time playing with toys typically preferred by boys (e.g., vehicles) and less time playing with toys typically preferred by girls (e.g., dolls) than did the unaffected relatives of CAH children.
Similar results have been reported on questionnaire and interview measures in the United States, Canada, and Europe. CAH girls have been found to differ from matched controls as well as from sisters and female cousins who do not have CAH. Behaviours that appear to be influenced include playmate preferences, and rough, active play, as well as toy preferences.
Convergent evidence suggesting that the effect is caused by androgen exposure, rather than other aspects of CAH, comes from studies of girls whose mothers were prescribed hormones during pregnancy for medical reasons.
Girls whose mothers took androgenic progestins have been found to show masculinized play behaviour similar to that seen in CAH girls, and girls whose mothers were prescribed the anti-androgen, medroxyprogesterone acetate (MPA), have been found to show less masculine- typical and more feminine-typical play.
The effects of MPA appear to be smaller than those seen following androgen exposure, perhaps because there is limited scope for further feminization or demasculinization in genetic females.
Prenatal exposure of females to DES does not appear to influence juvenile play behaviour or sex-typed interests. Interviews and questionnaires have been used to assess childhood activities retrospectively in 60 women exposed to DES, and in a variety of control groups. No consistent differences have been seen.
Thus, it seems likely that hormonal influences on the development of childhood play behaviour are exerted directly by androgen, rather than following conversion to estrogen. Alternatively, it might be suggested that the genital virilization at birth in CAH girls and in girls exposed to androgenic progestins played a role in their behaviour, for example, by altering parental encouragement of sex-typical play.
However, the impact of MPA on sex-typical play argues against this interpretation, since MPA would not produce noticeable alterations in the external genitalia of females. In addition, parents are instructed by medical personnel to treat their CAH daughters as they would any other girl, and their responses to interview and questionnaire items suggest that they do.
iv. Cognition:
Early reports concluded that individuals exposed to androgen prenatally had elevated intelligence. The intelligence test scores of individuals with CAH or exposed prenatally to androgenic progestin were found to be substantially higher than the population norm.
Subsequently, prenatal exposure to natural progesterone was also claimed to enhance intelligence, based on evidence of high levels of academic achievement in the offspring of mothers from a clinic that used progesterone to treat difficult pregnancies. Because natural progesterone acts as an anti-androgen, these results would appear to contradict those reported for CAH and other sources of androgen exposure.
Eventually, it was concluded that the enhanced intelligence in the androgen- exposed individuals and the academic excellence in the progesterone-exposed offspring related to factors other than hormones. When CAH patients were compared to their relatives or to controls matched for demographic background.
The original elevation seen in comparison to the norm for the population was probably caused by selection biases; individuals with higher intelligence were more likely to enroll in the research project. In regard to prenatal progesterone exposure, inappropriate statistical analyses contributed to the apparent academic enhancement, and re-analyses of the original data as well as subsequent research found no evidence of academic enhancement.
Other studies of individuals exposed to a variety of progestin, to estrogen and progestin, or to DES also have found no differences in measures of general intelligence in comparison to relative controls. The lack of an influence of androgens, progestin, or estrogens on general intelligence is not surprising, since general intelligence does not show a sex difference, and so would not be expected to relate to gonadal hormones.
Although general intelligence does not show a sex difference, some specific aspects of cognitive performance do, and researchers have also asked whether hormones influence these specific cognitive abilities. Two studies have reported that women exposed to high levels of androgen prenatally because of CAH show enhanced visuospatial abilities.
One study compared 17 girls with CAH to 13 female relatives of children with CAH. The CAH group showed better performance on a three-dimensional mental rotations task, on a two-dimensional mental rotations task, and on a spatial visualization task. The first two tasks typically show sex differences, the third does not.
The second study found that seven girls with CAH performed better on a fourth task than did six unaffected sisters of CAH children. However, a third study, using the same task or a two-dimensional, rotational task, found no difference between 17 girls with CAH and 11 unaffected sisters of CAH children. Other studies have generally found no differences between CAH girls and controls on the block design subtest of the Wechsler scales or on the Embedded Figures Test.
In these same studies, boys with CAH have generally been found to perform similarly to relative and matched controls on measures of visuospatial abilities. However, one study reported that five boys with CAH performed worse than four unaffected male relatives of CAH children on a spatial task.
Thus, data on visuospatial abilities following prenatal androgen exposure owing to CAH are inconclusive. There are some reports of alterations on measures that show substantial sex differences (mental rotations tasks) as well as on some measures that do not. In addition, findings of altered performance have not been replicated consistently. Small samples and the use in some studies of measures that show small to negligible sex differences may underlie the inconsistent findings.
v. Interest in Parenting:
As we know that, girls with CAH show reduced interest in toys typically preferred by girls, including dolls. Interest in dolls, particularly infant dolls has not been examined separately, however. Thus, it cannot be assumed that the reduced interest in toys preferred by girls, or even specifically in dolls, reflects a reduced interest in parenting or nurturance.
The studies suggest that CAH girls are less interested in babysitting and other aspects of child care, including plans to have children of their own. The reduced interest in child care may be more apparent in CAH girls older than 16.
One limitation of these studies is their reliance on a few questions, or even a single question, from an interview, in some cases conducted by someone who knew which respondents had CAH and which did not. A more extensive questionnaire assessment of 23 girls and 16 boys with CAH compared to 12 female and 22 male relatives produced equivocal results.
The questionnaire was completed by parents of the children and showed the expected sex difference among controls, with girls showing greater interest in infants than boys. The CAH girls also received lower scores than control girls on the questionnaire, but when items related to doll play were removed, the difference from female controls was no longer significant (p > .05, one-tailed). Also, the CAH girls scored lower on interest in pets, although control girls and boys did not differ on this measure. Boys with CAH did not differ from male relatives in regard to interest in either infants or pets.
As with other sex-typed behaviours, the CAH girls’ and their parents’ knowledge of their condition and their intersex genitalia could contribute to apparent psychological differences. Because of this, researchers have also investigated interests related to parenting and nurturance in women exposed prenatally to the synthetic estrogen, DES, since DES does not masculinize the external genitalia.
As we know that, DES-exposed women have been reported to be more likely to be bisexual or homosexual than other women. However, their interest in parenting appears unaltered. The same researchers who studied sexual orientation also administered questionnaires that provided retrospective assessments of behaviour.
An initial study suggested some reduction in interest in parenting, but this was just one of many variables assessed, and none of the others differed for DES-exposed women and controls. This raises the possibility that it was a chance finding.
Further research suggested that the initial finding had indeed been a chance occurrence, since it could not be replicated in two subsequent studies. Thus, some data suggest that early exposure to higher than normal levels of androgen prenatally decrease interest in parenting in females, but findings are far from consistent. As with other sex-typed behaviours, methodological issues, such as small samples and reliance on rudimentary measures, prevent definitive conclusions.
8. Aggression and Personality Characteristics of Humans:
Several studies have used questionnaires to assess aggressive response tendencies and other personality characteristics following atypical hormone exposure. One study found enhanced aggressive response tendencies in both boys and girls who had been exposed prenatally to androgenic progestin.
Seventeen girls and eight boys (ages 6 to 18 years) whose mothers had been prescribed hormones during pregnancy were compared to their 17 sisters and eight brothers born from pregnancies where hormones were not prescribed.
The hormone-exposed children gave responses suggesting that they would be more likely to react to provocation with physical aggression. In this study possible confounding influences of genital virilization were not a problem, because the hormone-exposed children had normal- appearing genitalia at birth.
Personality, including aggressive response tendencies, has also been investigated in individuals with CAH. In one study females with CAH in the age range 17-34 years gave more masculine-typical responses than matched controls on measures of Indirect Aggression and Detachment, two of eight personality measures that showed sex differences in controls in the study.
Similar differences were not seen in CAH males compared to controls. The first sample also showed the expected sex differences in 13 female versus 11 male controls, whereas sex differences in the second sample of 5 female and 10 male controls were not significant.
A sample of children, ranging in age from almost 3 to almost 13 years, showed the expected sex difference in 10 female versus 20 male controls, but no difference between 20 CAH girls and the 10 female controls. There were no significant differences in any of the three samples between males with CAH and male controls.