Delusions of Gender (6 page)

What’s more, tests like the Reading the Mind in the Eyes task and the PONS are not exactly what you would call realistic simulations of everyday mind reading. Trying to penetrate the expression of the
Mona Lisa
, or talking to a time-pressed Muslim woman in full burka might come close to what they assess – but arguably, social interactions more typically involve a stream of rich and changing information from other people (who do not offer multiple-choice options as to what they might be feeling). In the 1990s, William Ickes and his colleagues developed a new empathy test, one Ickes probably rightly claims is ‘the most stringent test’ of a person’s ability to infer the thoughts and feelings of others.
¹⁹
In this empathic accuracy test, two people wait together for an experiment to begin. The experimenter has departed to find a replacement for the projector bulb that has just blown – and in fact, the experiment has already begun. As they sit there and wait, they are unobtrusively filmed and recorded for six minutes. On her return, the experimenter explains the true purpose of the experiment. If both parties are happy to continue, they then view the film clip of their interaction individually, and as they go through the tape they pause it every time they recall having had a specific thought or feeling, and jot down what it was, and whether it was positive, negative, or neutral. Then, in the last part of the experiment, each person watches the tape again, but this time it’s stopped every time the
partner
reported a feeling or thought. The task is to infer what
this was. This can then be compared to what the partner actually reported feeling or thinking at that very moment.

You will probably agree that, of all the tests mentioned so far, this seems to most closely approximate real-world empathising. There are no actors posing expressions, no narrow strips of eyes, no disembodied voices and hands, no carefully choreographed and scripted scenes. Instead, people are interacting in a natural and unscripted way that generates a stream of successive mental states to be inferred from a rich variety of clues. You might expect men to struggle with such a demanding test, but they do not. As Ickes reports in
Everyday Mind Reading
, much to everyone’s surprise, in the first seven studies to use this measure no gender differences were found:

Where was the empathic advantage that we commonly refer to as ‘women’s intuition?’ It wasn’t evident in the interactions of opposite-sex strangers, or in the interactions of heterosexual dating partners, or even in the interactions of recently-married or longer-married dating partners. It wasn’t evident in comparisons of female-female dyads with male-male dyads or of all-female groups with all-male groups. It wasn’t evident in Texas, in North Carolina, or even in New Zealand. Was it nothing more than a cultural myth? A fictitious bit of folklore that was ripe for scientific debunking?

But then, something ‘baffling’ happened.
²⁰
The next three studies, all of which took place four or more years after the first empathic accuracy study,
did
find gender differences. The researchers quickly spotted that there had been a slight change in the form that the viewers used while going through the tape of the interaction. In the new form, for each thought and feeling that they guessed, they had to say how accurate they thought they were. When this version of the form was used womanly intuition existed; when the old form was used, it didn’t.
²¹
Why might this be? Ickes suggested that this small change reminds women that they
should
be empathic, and therefore increases their motivation
on the task. He concludes from his lab’s research that ‘[a]lthough women, on average, do not appear to have more empathic
ability
than men, there is compelling evidence that women will display greater accuracy than men when their empathic
motivation
is engaged by situational cues that remind them that they, as women, are expected to excel at empathy-related tasks.’
²²

If so, then if the experimental situation can instead be designed to motivate
men
, then their empathic performance should also improve. This is exactly what researchers are beginning to find. Kristi Klein and Sara Hodges used an empathic accuracy test in which participants watched a video of a woman talking about her failure to get a high enough score on an exam to get into the graduate school she wanted to attend.
²³
When the feminine nature of the empathic accuracy test was highlighted by asking participants for sympathy ratings before the empathic accuracy test, women scored significantly better than men. But a second group of women and men went through exactly the same procedure but with one vital difference: they were offered money for doing well. Specifically, they earned $2 for every correct answer. This financial incentive levelled the performance of women and men, showing that when it literally ‘pays to understand’ male insensitivity is curiously easily overcome.

You can also improve men’s performance by inviting them to see a greater social value in empathising ability. Cardiff University psychologists presented undergraduate men with a passage titled ‘What Women Want’.
²⁴
The text, complete with bogus references, then went on to explain that contrary to popular opinion ‘non-traditional men who are more in touch with their feminine side’ are regarded as more sexually desirable and interesting by women, not to mention more likely to leave bars and clubs in the company of one. Men who read this passage performed better on the empathic accuracy task than did control men (to whom the test was presented in a nothing-to-do-with-gender fashion) or men who had been told that the experiment was investigating their alleged intuitive inferiority.

Clearly, one’s performance on cognitive empathy tasks involves a combination of motivation and ability. If social expectations can create a motivation gap, could they also be responsible for an ability gap? Women on average score better than men on another social sensitivity test called the Interpersonal Perception Task (IPT). Here, participants watch and hear people acting out unscripted interactions. From the actors’ verbal and nonverbal behaviour, the viewers have to try to work out the nuances of their relationships. For example, from watching a scene between two men and a child, the participant has to work out which man is the child’s father. Recently, psychologists Anne Koenig and Alice Eagly used the IPT to explore the idea that the gender stereotype of women’s superior social skills might furnish women with an unfair advantage.
²⁵
To one group, the test was accurately described as a measure of social sensitivity, or ‘how well people accurately understand the communication of others and the ability to use subtle nonverbal cues in everyday conversations.’ Before the participant took the test, the experimenter casually mentioned that ‘We’ve been using this test for a couple of quarters now. It’s 15 questions long and, not surprisingly, men do worse than women.’ In this group, the men did indeed do slightly worse than the women. But to a second group of participants, the test was described in a more gender-neutral way. It was presented as a measure of complex information processing, or ‘how well people process different kinds of information accurately.’ In this group, the men performed just as well as the women.

The take-home message of these studies is that we can’t separate people’s empathising ability and motivation from the social situation. The salience of cultural expectations about gender and empathising interacts with a mind that knows to which gender it belongs. So what would happen if we could temporarily trick a female mind into thinking it was male? As we saw in the previous chapter, when people take the first-person, ‘I’ perspective of someone else, the stereotypical traits of the other permeate and seep into the perspective-takers’ own self-concept. This merging of
identities can cross gender boundaries. A few years ago, psychologists David Marx and Diederik Stapel asked a group of Dutch undergraduates to write about a day in the life of a student named Paul. Half of the students wrote in the first-person (‘I’) while the other half used the third-person perspective (‘he’). Afterwards, they were asked to rate themselves on technical-analytic skills and emotional sensitivity skills. For the female undergraduates, thinking of themselves as Paul in the ‘I’ perspective altered their self-conceptions. Women who attempted to be Paul living his life actually incorporated his stereotypical male characteristics into their own self-conceptions. They rated themselves as higher on analytic abilities and lower on emotional sensitivity, compared with women who had written a third-person story. In other words, there was ‘a merging between the self and [Paul], such that female participants became more “malelike” as a result.’
²⁶
Indeed, they became so malelike that their self-ratings on these stereotypical traits were statistically indistinguishable from the men’s. For men, there was no such effect of being Paul on their self-concept, presumably because they already
were
a male student.

The participants were also given a battery of emotion sensitivity tests. These problems included ones like recognising facial expressions of emotion, choosing which two more basic emotions make up more complex ones (like optimism), and working out, for example, what emotional state you reach as you become more and more guilty and lose your feeling of self-worth. (Is it
depression, fear, shame
or
compassion
?) Women who had not put themselves in male shoes performed a lot better than the men on this task, getting on average 72 percent of the emotion-sensitivity questions correct, while men’s scores hovered around the 40 percent mark. But women who had just spent a few moments merely imagining themselves to be a man performed every bit as poorly as the real men.

No doubt an intricate interplay between minds and social expectations affects our capacity for affective empathy, too. Research into group-based emotions investigates the idea that
when ‘people are thinking of themselves in terms of a particular group membership – whenever a social rather than personal identity is salient – people’s emotional experiences and reports will be shaped and determined by that group membership.’
²⁷
In a recent study, researchers found that subtly priming a social identity led people to experience group-based emotions that were different from those they experienced when thinking of themselves as an individual. Is it possible that women become more tenderhearted when thinking of themselves as women or mothers rather than as individuals or, say, saleswomen?

We don’t know, but University of Exeter psychologist Michelle Ryan and her colleagues have found that the social identity you are wearing can certainly change the sway of compassionate feeling in resolving moral dilemmas.
²⁸
In the 1980s, Carol Gilligan famously suggested that women and men reason about moral situations in a different way. She suggested that the ‘ethic of justice’ – which privileges abstract principles of justice such as equality, reciprocity and universal rules – is used more by men. In contrast, the ‘ethic of care’ – which takes greater account of the feelings and relationships of those concerned – is used mostly by women. Researchers since have argued that what kind of ethic is used depends a great deal on who the moral dilemma involves: men and women alike are happy to apply abstract universal laws and principles to strangers, but tend to turn to the ethic of care for answers when considering the plight of friends or other intimates.
²⁹
And any remaining gender difference in moral reasoning appears not to be hardwired, because it can be eliminated with a change of social identity. Ryan and colleagues presented students from the Australian National University (ANU) with a moral dilemma: A student from the local TAFE (a nonuniversity institute of tertiary education) urgently needs a book for an assignment due the next day. Without the book, the student will fail. The book is not available at the desperate student’s own library. The ANU students are asked whether they would borrow the book from their own library, on behalf of the TAFE student.

Before being presented with this realistic dilemma, the researchers manipulated which social self was in charge by asking participants to brainstorm ideas for a debate. Then they were given the dilemma to read, and asked to explain the important factors involved and what they would do in that situation. One group was primed with gender stereotypes (they were asked to come up with debating ideas for the claim that men are still real men or that women are not the weaker sex). Within this group, there was clear evidence of gender difference in moral-reasoning style. Women were twice as likely to offer care-based considerations, such as the alleviation of another’s suffering. This might lead us to think that men are less empathic in their approach to moral dilemmas – except that in two other groups, both primed with a student identity, gender made no difference. The second group of students was primed to think of themselves as tertiary students. With this identity in place, the TAFE student was one of them. The last group was primed with only their more exclusive identity as ANU students. (The Australian National University is arguably the highest ranking in the country.) Regardless of sex, tertiary-student-primed students offered more care-based considerations and fewer justice arguments than the ANU-primed students, who had been primed to feel socially distant to the harried TAFE student.

In other words, when we are not thinking of ourselves as ‘male’ or ‘female’, our judgements are the same, and women and men alike are sensitive to the influence of social distance that, rightly or wrongly, pushes moral judgements in one direction or another along the care-justice continuum. But moral reasoning is also sensitive to another social factor – the salience of gender. Thus, the authors argue that ‘it is the salience of gender and gender-related norms, rather than gender per se, that lead to differences between women and men.’ Of course, as they also point out, ‘the social reality is that gender, for most, is a ubiquitous category and is arguably the most salient’.
³⁰