A conundrum of the sexes

Are the differences between men and women driven by genes, hormones or patterns of upbringing? Michael Rutter assesses the arguments.

It is evident to everyone that males and females differ in innumerable ways, but it is far from obvious quite why. Over the years, the pendulum has swung to and fro on the supposed importance of inbuilt biological influences and of patterns of rearing. An exclusive emphasis on biology is called into doubt by the major changes in sex differences that have occurred over time. For example, half a century ago only about 10 per cent of medical students were women, now more than half are female. Conversely, exclusively cultural explanations are difficult to sustain in the light of animal evidence. In addition, much research raises questions about the homogeneity of broadly defined behavioural features. Thus, although antisocial behaviour as a whole is much more common in males than females, there is little difference between the sexes in the frequency of domestic violence. By contrast, there is a big sex difference in lifecourse-persistent antisocial behaviour, which is much more common in males, but a smaller sex difference in antisocial behaviour limited to the adolescent age period.

The time is ripe for a clear-headed, critical look at the evidence and Melissa Hines has provided this with her very accessible, clearly written account, focusing especially on the role of sex hormones.

Before turning to research findings, she notes the problem of publication bias (whereby positive findings are likely to be published but negative ones not even mentioned); the situation-specificity of some sex differences; the effects of how psychological features are measured; and the substantial overlap between the two sexes on virtually all psychological attributes. Her summary emphasises the small size of most sex differences, with the individual differences within groups of males or females far greater than the difference between the sexes in the average score on the same psychological trait.

Clearly, she is correct in what she says with respect to the traits that she considers, but her overall conclusions are nevertheless somewhat misleading because she ignores the much larger differences with respect to many forms of mental disorder. Thus, both depression and adolescent eating disorders are substantially more common in women (the sex difference in childhood depression, however, is quite small). Conversely, early onset disorders involving neurodevelopmental impairment - such as autism, attention deficit hyperactivity disorder (ADHD) and developmental disorders of language - are much more common in boys.

Surprisingly, she does not mention Simon Baron-Cohen's controversial hypothesis that autism represents the extreme male brain. Similarly, she does not note the marked variation in sex differences with respect to particular types of scholastic performance. Thus, the Programme for International Student Assessment study showed that the sex differences in mathematics tended to be both small and inconsistent across countries, whereas the substantially better performance of girls on reading was found everywhere, mirroring the marked male excess in the rate of dyslexia.

However, there was an interesting and important twist: although girls were better readers around the world, the size of the sex difference varied markedly across countries. An opportunity to consider the implications (with respect to causal influences) of this mixed pattern of sex differences was missed. Instead, Hines approaches the topic from the perspective of what is known about the biology, rather than which sex differences in psychological functions need to be explained. That is very much her territory and she does an excellent job in providing a balanced account of both the evidence and what it might mean.

She notes the fact that, biologically speaking, sex does not have a single meaning. Thus, there is chromosomal sex, defined according to whether the person has two X chromosomes (in females) or an X and a Y chromosome (in males). Sex chromosome abnormalities complicate the picture because there may be only one X (as in Turner's syndrome) or an extra X or Y. On the whole, of course, chromosomal sex goes along with gonadal sex, meaning the sex organs (ovaries or testes), but this is not always the case. Third, there is hormonal sex, defined according to the pattern of ovarian and testicular hormones, but these too may not go along with either chromosomal or gonadal sex because the same hormones are produced by the adrenal glands and because the hormonal effects depend on particular receptors. As a consequence, the secondary sexual characteristics (such as breasts or facial hair) may be anomalous in relation to chromosomes or gonads. There must also be psychological differentiations according to core gender identity (whether individuals consider themselves male or female), sexual orientation and sexually differentiated patterns of behaviour. With respect to sex hormone effects, the book reminds us that we need to consider separately the possible programming effects of prenatal hormones during sensitive periods on brain structure and function, and the later activating effect of postnatal hormones (such as could occur with the changes that accompany puberty).

Hines describes how the human evidence tends to come from the unusual medical conditions in which there is a disjunction between the various types of biological sex, and where there is a variation in how this is dealt with in terms of sex of rearing and surgical or hormonal reassignment of sex. As she notes, interpretation of the evidence is not straightforward, both because each anomaly tends to have multiple effects and because cultural factors will influence expectations and also responses to anomalies. Animal evidence is in some ways easier to interpret because experimental control is possible, but attention has to be paid to relevant species differences with respect to brain structure, developmental timing and the malleability of psychological functioning.

All of that sounds very complicated and indeed it is. However, Hines does an admirable job pulling the findings together and integrating, so far as possible, the results of diverse research strategies. She notes that the effects of male sex hormones appear to be much stronger than those of female ones. Because females have male hormones produced by the adrenal glands, that applies to them too. Animal studies provide a clear indication that prenatal male sex hormones have a lasting effect on brain development.

The same seems to apply in humans to some extent (as indicated, for example, in brain imaging studies).

Certainly, humans and other animals share the hormonal programming effects on the development of the genitalia, and there are comparable effects on sex-differentiated play patterns. There are also some prenatal hormonal effects on human core gender identity and sexual orientation, although these are far from determinative. In part, this is because the effects of biology are almost always probabilistic rather than determinative, and because nature makes use of what is there. It has been said that biology is a tinkerer and not an engineer. In part, it is because, in both humans and other animals, the brain has been found to be much more malleable in response to post-infancy experiences than was once thought. In part, too, the effects in humans might be expected to be less clear cut because evolution has provided them with the power to conceptualise their experiences and themselves, and to plan what to do with their lives. Hines urges us to adopt a broad perspective, to recognise the importance of biology but also of culture and personal psychological agency and construct. The interplay between these different influences inevitably gives rise to important individual differences. Although the book is strong on biological influences, it is much weaker on the effects of experiences on the organism. Thus, there is no discussion of the possible epigenetic effects of prenatal/perinatal hormones - meaning the effects on gene expression rather than gene sequence.

The book concludes with a thoughtful, but inevitably somewhat inconclusive, account of the clinical implications for how sexual anomalies in humans should be dealt with. The discussion is well informed, and is sensitive to the practical problems as well as to the relevance of individual differences and personal decisions. Hopefully, clinicians and parents alike will be encouraged to think critically and constructively about the dilemmas and how they might be dealt with. More general readers will find much of interest, as well as more that provokes thought, in this well-written, easy-to-understand account of what is known on a most important topic.

Sir Michael Rutter is professor of developmental psychopathology, Institute of Psychiatry, London.