Anemone's Research Emporium

Critique of "The Essential Difference"

Anemone Cerridwen
June 27, 2008

The book:
The Essential Difference: The truth about the male and female brain, by Simon Baron-Cohen (2003). Basic Books, NY.

Baron-Cohen proposes that there are biological differences in male and female brains, leading to average differences in personality and cognition. Fair enough. There's a lot of research on that (e.g. Halpern, 2000; Kimura, 1999). Females on average tend to have better language skills while males tend to have higher visual-spatial skills, although females do better on memory for spatial location and males do better on verbal analogies (Halpern, 2000).

Instead of building on this research, however, Baron-Cohen describes the typical male brain as being hard-wired for systems and the female brain for empathy:

The female brain is predominantly hard-wired for empathy. The male brain is predominantly hard-wired for understanding and building systems. [p.1]

This is where things start to get problematic. Baron-Cohen names the two brain types male vs female or systemizing vs empathizing. He defines empathizing as:

Empathizing is the drive to identify another person's emotions and thoughts, and to respond to them with an appropriate emotion. Empathizing does not entail just the cold calculation of what someone else thinks and feels (or what is sometimes called mind reading). Psychopaths can do that much. Empathizing occurs when we feel an appropriate emotional reaction, an emotion triggered by the other person's emotion, and it is done in order to understand another person, to predict their behavior, and to connect or resonate with them emotionally. [p.2]

and systemizing as:

Systemizing is the drive to analyze, explore, and construct a system. The systemizer intuitively figures out how things work, or extracts the underlying rules that govern the behaviour of a system. This is done in order to understand and predict the system, or to invent a new one. [p.3]

This sounds suspiciously like "men think and women feel", rather than "men are good at thinking in some ways and women in others". Fortunately for Baron-Cohen, his description of empathizing sounds more like relationship skills in general, and the ability to read nonverbal communication (something women are relatively good at - Halpern, 2000) and to communicate well, not just a tendency to react empathically to others. Similarly, he seems to be including spatial skills (also known to be higher in males than in females, though, among males, higher in less masculine males than in more masculine males - Halpern, 2000) and a fondness for technology in the concept of systemizing. Whether his two categories correspond to or correlate with the cognitive abilities males and females are already known to excel at is not clear. A quick glance suggests that "systemizing" is probably closer to visual-spatial skills than "empathizing" is to language skills.

At this point I should point out that neither of the scales Baron-Cohen is using to measure these concepts has been put through a factor analysis or external validation to see if they in fact measure what they are supposed to or not. The best test would be to combine the two scales and see if a combined factor analysis indicates that there are two main concepts here, and the two he claims to be measuring, rather than more than two, or one big one and a bunch of smaller ones, or whatever. In addition, external validation of the two scales to see if their scores correlate with other scales that measure the same or similar concepts would also ensure that the scales measure what they say they do. This process would probably help considerably in making sure the basic concept is clearly defined and stated.

Back to the main idea. While Baron-Cohen talks about systemizing versus empathizing, I think it would be more accurate to say that what he is saying is that men are, on average, more interested in and better at technology (things), and women are, on average, more interested in and better at relationships (people), and that this is a biologically based, lifelong tendency.

I have no quarrel with the men-like-technology/women-like-relationships dichotomy. It is not as sophisticated as the research on sex differences in cognition that is already out there, but it may still be a useful model (depending on the use it's put to). The problem I have here is what he calls it. First of all, calling it male and female (and not even "male" and "female") reinforces simplistic dichotomies, and is unfair to those who are atypical for their sex. Yes, he says in the text that these are average differences, not absolute ones. But the labels speak pretty loudly, too, and they suggest more absolute differences.

Second, when he doesn't call them male and female, he calls them systemizing and empathizing. While these labels allow men and women to be either or both, and are thus more accurate, I don't think these are accurate labels, either, as explained above.

An example from a description of a mathematician with Asperger Syndrome:

Here was a man who could fathom any mathematical problem you could throw at him, but who was unable to work out the basics of friendship or how to have a phone conversation. Was there every a more dramatic example of dissociation between empathizing and systemizing? [p.158]

It is hard for me to see how being able to have a phone conversation or not is related to empathy as people would normally define it. (More on that when I dissect the Empathy Quotient scale in another critique.) I think Baron-Cohen would be better off using the terms technology and relationships. And actually, this particular example points even more closely to known sex differences in cognition (visual-spatial versus language skills) than the rest of his discussion so far.

The main problem with Baron-Cohen's model of "male" versus "female" minds, and probably the main reason for it in the first place, is that he seems to be using autistic minds as the basis for his theory. It is true that unusual minds can point researchers to fundamental rules of brain and mind organization. For example, the existence of savant abilities (found in both autistic and non-autistic individuals, with both below normal and normal intelligence, but most obvious in individuals with below normal intelligence) indicates that there are distinct modules of ability (e.g. mathematical, musical) that are independent of general intelligence. So if autistic people had unusual abilities or lack of abilities, it would make sense to study these to see if they have something to say about minds and brains in general.

However, very little is known about what autism actually is, and stereotypes of the technologically-minded autist with little or no empathy may not be particularly accurate in the first place. As it is, right now males are diagnosed with autism far more often than females, and while it may well be true that autism is more common in males than females, it is also possible at the same time that females are underdiagnosed relative to males. This happened with respect to ADHD. At one time the stereotype of ADHD was more masculine than it was now, with an emphasis on hyperactivity. Then researchers noticed that there were many females with an inattentive form of attention disorder. Now there are still more males than females diagnosed with ADHD, but the ratio is much less extreme than before, and in the meantime the diagnostic criteria have been adjusted to take into account that ADHD often looks different in females.

If you assume that autistic people have "male" brains, then you will disproportionately diagnose people who fit the male-brain stereotype, mostly males and females who are less girly. Then when you study this population, you will find that they do in fact have "male" brains. Of course they do - that's how they were selected in the first place. It is probable that autistic girls and women do not fit the stereotype as well as boys and men do, and are being underdiagnosed. In another ten or twenty years, the face of autism may change as more girls and women (and "atypical" males) are diagnosed, and the stereotype may no longer hold true.

Even when we have a more accurate picture of what autism is, and what it looks like, we may not learn much more about major brain "types", since it's actually possible that autism has little or nothing to do with personality or cognitive domains. Given how very little is known about autism, especially in adults, people need to be careful just how much they assume at this point. I don't even know if anyone has checked to see how autistic people do on standardized visual-spatial and verbal-language tests to see how they compare with average populations. If autistic people generally do well on visual-spatial tasks, that's interesting, and suggests right hemisphere dominance. If they also do relatively poorly at verbal-language tasks, but no worse than the average male, that's also interesting, but not that exciting. If autistic people show specific verbal or language deficits on standard tests, that would be different. I'd get excited about that.

I would recommend that Baron-Cohen do a rigourous analysis of his Systemizing and Empathizing Quotients to see what in fact they are measuring, and how closely they are related to already known sex differences in cognition. He will have to make a very strong case to establish that his systemizing and empathizing are distinct from existing visual-spatial and language abilities, and that they add something significant to the body of information on sex differences that is already available. So far he hasn't done that.

References

Simon Baron-Cohen (2003). The essential difference: The truth about the male and female brain. Basic Books, NY.

Halpern, Diane, 2000. Sex differences in cognitive abilities. Third edition. Lawrence Erlbaum Associates, Mahwah, New Jersey.

Kimura, Doreen, 1999. Sex and cognition. MIT Press, Cambridge, MA.