What might be causing the female brain to empathize at a superior level, and the male brain to systemize at a superior level? Most people are likely to assume that such sex differences are due to a mix of cultural and biological factors. There are, of course, more extreme claims which propose that the sex differences can be reduced to just one of these factors. If these extreme theories were true, this would be important since it would mean that the other set of factors was ruled out.
As I pointed out in Chapter 1, it would be politically attractive if the cultural theory were true. If it really were the case that cultural factors (such as sexism in education or the workplace, or differential child-rearing styles, or the media, or the toy industry) made little boys and girls turn out differently, it would at least mean that we could try to eliminate these differences through new social or educational programs. So just how persuasive is the cultural theory?
If you are shown a videotape in which a child appears upset, and you are told the child is male, you are more likely to label the child’s emotion as anger. If you are told the child in the video is female, you are more likely to label the child’s emotion as fear. This is an example of what has come to be known as the Baby X experiment. One conclusion to draw from this is that we are not very good at judging if a baby is male or female. But that is not the important conclusion, and sex recognition becomes easier as the person we are observing gets older. The more important conclusion is that, despite believing that we are perceiving people in an unbiased fashion, we must be unwittingly carrying a set of sexist biases. These could be biases we have picked up from the way society stereotypes the sexes, or simply associations we have formed in relation to each sex. Whatever their origin, such biases may lead us to react differently to a person, depending on what we believe their sex to be.
The implication of this is that the psychological sex differences discussed previously could be just the result of gender expectations by adults of children throughout their upbringing. I will look at parenting shortly, to test this idea. The other implication is that sex differences could be the result of the subjective biases of the researchers who make the observations in the experiments. It is therefore imperative for those conducting studies in this area to attempt to remain “blind” to the sex of the participants.
In practice, this is very hard. In the Cambridge newborn study, the observers made strenuous efforts to remain blind to the babies’ sex. They asked the mothers who consented to participate in the study while they were in the hospital, on the first day of their new baby’s life, not to tell the researchers the sex of their baby, so as to avoid being influenced by this information when they were filming or testing the baby. Mothers agreed to withhold this information until after the filming and data recording was complete.
In the vast majority of cases, this information did not leak out. In a few cases it was hard to remain completely blind because the mother’s hospital bed would be surrounded with cards from well-wishers saying things such as “Congratulations! It’s a boy!” or the baby would be dressed in pink or blue pajamas and blankets. Nevertheless, the experimenters ensured that they filmed only the baby’s face, to record where the baby’s eyes were looking when he or she was shown the two types of objects (a human face and a mechanical mobile). By the time the judges came to analyze each tape, none of the potential cues (such as the color of baby’s clothing) were visible, and they could not tell if they were looking at the face of a male or female baby. And yet the sex difference in looking preference was still found; cultural stereotyping cannot explain the result of this experiment.1
The Baby X experiments have not always produced consistent results, but let’s assume that it really is true that our expectations of the sexes shape how we interact with boys and girls, or what opportunities we offer them. Despite the influence of such biases, there are examples of wellcontrolled studies suggesting that this is not necessarily the only cause of observed sex differences in behavior.2
Could differences in parenting toward boys and girls cause the observed sex differences in children? Could these differences be linked to the fact that mothers are frequently the primary caregiver? This is a valid consideration since, in all societies studied, children (on average) spend more time with their mother than with their father. In some cultures (such as those in Kenya or Japan), children spend about three times as much time with their mother, while in others (such as those in India), children spend more than ten times as much time with their mother. Are mothers doing something to make young girls better empathizers and young boys better systemizers? Or are fathers to blame, treating their sons and daughters differently?3
Boys certainly receive more punishments, prohibitions, and threats from parents, as forms of control. Parents tend to forbid their sons from doing something, and tell them what will happen if they transgress those orders far more frequently than they tell their daughters. Do the following examples sound familiar?
“Max, come back here!”
“Max, stop!”
“Max, don’t DO that!”
“No! I’ve told you before!”
“Max, do what I say or you’ll be in trouble!”
“Max, if you do that ONE more time, I’ll take away your remotecontrolled jeep for a week!”
Of course, girls are spoken to in an admonitory way, too, but it is interesting that in most cultures this style of speaking is more often directed toward boys. Is it a sign of sex-typing (in other words, the different treatment of boys and girls for purely cultural reasons)? If so, what on earth are parents doing?
One view is that parents hold in their mind some notion that boys are wilder or greater risk-takers and therefore need more restrictions. Are boys living up to such assumptions? Have parents unconsciously encouraged their sons to take more risks, or to push back the boundaries more often, and then find themselves having to police their sons more often when their child goes too far? There is, of course, an opposite view. It may be that boys are poorer empathizers than girls, which leads them to be less socially compliant, less skilled at picking up the social cues of boundaries, and requiring more frequent disciplining.
There is some evidence for both views. For example, in one study, fathers were videotaped in the waiting room with their one-year-old children, a good set-up for a naturalistic experiment. The investigators found that fathers reprimanded their sons twice as often as they reprimanded their daughters. But this was not without reason: the boys tried to touch forbidden things more often. Girls seemed to pick up on subtle cues—for example, their father looking somewhat disapproving. Such “social referencing”—looking at a parent’s face to detect whether something is permitted or not—was enough for many girls to get the message of what was, or was not, allowed.
You can imagine these subtle signals in the parent’s face: the father’s eyebrows narrow a touch; or maybe his eyes shoot a quick stern stare at the child, as he wanders further than he is allowed to, or tries to take something that is not appropriate; or the father’s lips purse just slightly into a silent “shhh,” to signal that the child is being a bit too noisy. While girls look up more frequently to pick up these signals, and are more accurate at decoding them, boys may miss them more often. This may be because they are not checking back to see them, or because the signals are being seen but are unread or ignored. In this study boys seemed to get the message that they were doing something that was not allowed only when an explicit verbal expression of disapproval was given.
However, this does not prove that boys are inherently less empathic than girls. The subject of this book is essential differences in the minds of men and women, but perhaps the observed differences in empathizing in this chapter do not reflect any differences in “essence.” There is evidence that we have encouraged young boys to be less emotional and more independent by giving them messages such as these:
“Who’s a tough boy? Well done!”
“Ooh, you’re strong! Look at those muscles! You can do it!”
“Want to try climbing a bit higher? You’re so big and brave!”
“Don’t cry. Boys don’t cry!”
“Stop being babyish! None of the other boys are being clingy! It’s time for me to go now.”
Parents do use phrases like these more often toward their sons. Fathers in particular are less sympathetic to their sons showing emotional dependency.
Such findings are, of course, open to multiple interpretations. Parents might be discouraging boys from showing emotion, thus socializing them into different sex roles; or they might be more tolerant of emotional dependency and clinginess in their daughters, compared to their sons; or they might be giving their sons a description of themselves that portrays them as independent explorers more often than they do their daughters. Yet another interpretation is that girls are better at self-control (we will see some evidence for this in the next chapter), so they do not need to be told to control their emotional outbursts or to inhibit their impulses as often. Young boys are therefore judged by the norms set by young girls, and are thus told more often not to cry or not to touch anything they want. So this difference in parental behavior toward the two sexes could actually be driven by an essential difference between males and females.
In summary, one can see parents’ differential behavior as either a cause of observed differences in children, or as a consequence of them.4
Although there is some evidence for the two sexes being treated differently by their parents in terms of rough-and-tumble play, with boys being given more opportunities and invitations to horse around and to be more physical than girls, systematic studies reveal more similarities than differences in how the two sexes are handled. For example, parents do not differ in how they treat their sons and daughters in terms of the amount of warmth, responsiveness, talking, or restrictions that they provide. Parents also do not differ in how they handle their sons and daughters when it comes to how much encouragement they give. This suggests that sex-typing may not exist as strongly as many people suspect. There is even evidence that, if anything, parents devote more energy toward encouraging their sons to be more empathic. For example, mothers devote more time to copying their baby’s facial expressions if the baby is a boy than if it is a girl. It is difficult to correlate such evidence with a female superiority in empathy resulting from parenting styles.5
Mothers do, however, speak to their toddler daughters with more emotion words, compared with the number of emotion words they use with their sons. There are also differences in the type of emotion words that mothers use when talking to their toddler sons or daughters, and they tend to discuss positive emotions more with their daughters. Look at these examples of mothers talking to daughters:
“Oh, that was so kind of you to have made that for me.”
“Maybe she didn’t want to play with you because she thinks you’re best friends with Sally now? She might be feeling a bit jealous. Why don’t you invite her round on the weekend and make her feel a bit special as your friend?”
Mothers also use more “other-oriented” talk with their daughters when they have done something wrong, compared to how they talk to their sons. For example, they might say to daughters:
“How do you think she might have felt?”
“Imagine if you were the one who had gone off with a new best friend. You might also feel like her.”
“Do you think she’d like this as a present? It’d be nice to make her feel really special after her illness.”
Such differences in style of speech could lead girls to develop better empathy than boys. However, Maccoby makes the important point that this may not be the result of mothers’ unwitting sexism. Rather, it may reflect what mothers perceive their particular child is capable of understanding.6
We saw in Chapter 4 that young girls are more likely to play with dolls, and young boys are more likely to play with toy vehicles, or with mechanical toys, or with building blocks such as Legos. Could this be because they have somehow picked up gender stereotypes of what are considered to be “boy toys” or “girl toys”? Certainly, these messages are out there. Consider fathers who say, “I’m not having my son playing with dolls. He’ll get laughed at at school for being a sissy.” Or consider the toy industry that fills children’s television with constant advertisements. In these advertisements, Spider-Man toys are depicted as the ideal present for a child (who is always male) and Barbie dolls are depicted as the most desirable present for a child (who invariably is female). And as the toy business knows, nagging-power by children (“Oh, can’t I have one of those, Dad? All the other children have got one.”) is a powerful economic force in consumer activity. It is hard for parents to keep saying “No” to their child. And now consider how the influences from parents or the media are strengthened because their older sibling or other children in their peer group have been exposed to the same messages. Sibling rivalry, and the strong need to belong to the peer group, to fit in, might be some of the pressures making children believe that they need these sex-typical toys.
This may well be happening by school age, when the peer group gives signals of approval or disapproval of a child’s interests. But the idea that such social influences determine toy choices is unlikely. Here’s why. If you ask two-year-olds which toys are for boys and which for girls, they will not be able to tell you. Children do not yet know the gender stereotypes; they are as likely to suggest a toy car or a doll for a girl or a boy. Yet at this age they already show the sex-typical toy preferences through the toys that they themselves choose to play with. This suggests that their toy preferences predate their gender stereotypes. The latter cannot be causing the former, not unless you believe in backward-causation, in which case we are not living on the same planet.
In addition, by four years of age children are able to distinguish between their own preferences and those of others. Consider the little girl in one study who had been left with a male-typical toy (a truck) to play with, and who said, “My mommy would want me to play with this, but I don’t want to.” This strongly suggests that children are making choices that are not simply the result of what their parents want for them, but reflect other factors.7
Could the sex differences in toy choice and behavior be the result of early imitation of same-sex peers, or of adults? There is no doubt that imitation plays a powerful role by school age. For example, even in cultures where there is no school, older children’s play tends to involve enacting the roles of the same-sex parent. But as an account of the origins of these sex differences in toddlers, this is insufficient. This is because children younger than school age do not consistently imitate someone of the same sex more than they imitate someone of the opposite sex. Imitation of peers cannot be the relevant factor determining early toy preferences.
Psychologist Eleanor Maccoby makes the important point that when children choose to play with a peer of the same sex, and thereby play with toys that the same-sex peer is playing with, this is unlikely to be due to imitation of adults. This is because, when you stop to think about it, adults interact with the opposite sex all the time. Certainly, there is no evidence that parents encourage their children to avoid the opposite sex. Yet this is what children are choosing to do.8
Leaving aside imitation as an explanation of how sex differences arise, what about the obvious explanation of practice? Take the male superiority in motor skills (such as hitting targets). The common-sense view is that it is due to the greater amount of practice males get, through games such as darts and other male-typical sports. Practice may partly explain why the sexes diverge in their skills with age, but problematic for this view of the origin of these sex differences is the fact that the male superiority in throwing accuracy is present even in children as young as two years old. At this age, we can safely assume there has been little difference in opportunities to practice, and yet the two-year-old boys clearly outstrip the girls. At older ages, these sex differences are found even when the experimenter obtains information about each individual’s sports history and implements controls for this.9
Another explanation of sex differences is the perpetuation of “gender roles.” The basic idea is that we hold different beliefs about our genders that affect our behavior and interests. For example, women are believed to be more “communal” (more selfless, more concerned for others), while men are believed to be more “agentive” (more self-assertive, self-expansive, with a stronger urge to master). These beliefs about gender are held to arise from men’s and women’s different social and economic roles: women’s greater involvement in domestic and childcare activities gives rise to them being more communal, and men’s greater involvement in paid employment gives rise to them being more self-assertive.
Certainly, there is plenty of evidence for the existence of these gender roles. Given a choice, more men choose to work in “dominance-oriented” occupations (i.e., those emphasizing social hierarchies and the control over others), while more women choose to work in “dominance-attenuating” jobs (i.e., working in a team of equals with others, and/or working with disadvantaged people).10
It may be that we hold gender-role beliefs (such as men have a stronger urge to master), but where do such beliefs come from? The systemizing theory would say that men, on average, have a stronger drive to understand systems as thoroughly as possible, whether this is a physical technique to master, or a new computer to understand. But why this is present is a wideopen question.
How can we test if it is our gender-role beliefs that cause the observed sex differences (in behavior, emotion, interests, and skills) or if they arise for Darwinian reasons? The Darwinian David Geary argues that more males are agentive because males depend for their reproductive success on a drive to establish social dominance. For the gender-roles theory to work, it would have to disprove such Darwinian factors. This has not yet been done. In fact, the gender-roles theory runs into specific problems quite quickly. For example, it fails to explain striking similarities across very different societies. Thus, of 122 societies studied, weapon-making (a clear example of a systemizing skill) is an exclusively male activity in 121 of these. (In the one society where women work in weapon-making, the women help the men rather than being involved exclusively.) This must be more than a coincidence.11
To summarize, there is some support for cultural determinism. A clear example can be found in the different ways that parents speak to their sons and their daughters, something that could contribute to the differences we observe in the development of empathy. But some of the sex differences are present so early (at birth) that it is hard to see how culture could be the sole cause. In addition, some parents try to do everything to avoid such cultural influences on their child. They buy their sons dolls, and their daughters toy trucks, only to find that the child still chooses to play with sextypical toys. And even if we think that this simply shows that peer and media influences are stronger than parental ones, this still seems insufficient to explain all the observed sex differences, because of their early onset—before the media or the peer group have taken root in the young child’s mind.
For these and other reasons, it seems possible that the development of sex differences in behavior are due to factors other than, and additional to, the cultural ones. Biological factors are the only other candidates. Let’s have a look at what these could be.