Skip to main content

Almost Ten Years On: Why are we still talking about The Essential Difference?

Simon Baron-Cohen’s book, The Essential Difference: The Truth About The Male And Female Brain (2003), is almost a decade old now, but his thesis keeps popping up in various places. For example, in a recent (and truly delightful) book on neuroscience and religion, Why Religion is Natural and Science is Not (2011), Robert McCauley uses Baron-Cohen’s work to suggest that researchers looking for “hyper-empathetic” subjects might want to check out the local convent.

Baron-Cohen’s main argument is that, on average, men and women have different cognitive strengths and weaknesses: men are more adept at “systematizing” and less adept at “empathizing,” while women are more adept at “empathizing” and less adept at “systematizing.” He goes on to argue that people with autism have “hyper-male” brains (in other words, they are especially good at systemizing and particularly poor at empathizing). According to Baron-Cohen, these differences in cognitive abilities are likely to be the result of genetic differences (both in the case of men and women and in the case of people with autism and people without autism).

I will leave it to others with more knowledge about autism to continue to debate Baron-Cohen’s thesis that autism is characterized by an aptitude for systematizing and by deficits in empathizing abilities, and will focus instead on his argument about sex/gender difference. His argument that these differences are primarily genetic or “hardwired” has been ably critiqued by Lesley Rogers. On average, men and women may have different ways of thinking, but these differences are undoubtedly influenced by gender role socialization (in addition to genetic factors). In societies in which women were denied access to education and were given primary responsibility for childcare, differences in patterns of thinking were most likely exaggerated. In theory, we could further reduce differences (or even reverse the averages) if we sent boys to schools in which only language and social etiquette skills were taught and if we sent girls to schools in which only math, science, and motor skills were taught.[1]



I see an even bigger problem
with Baron-Cohen’s argument in how he uses the constructs “empathizing” and “systematizing.”
When looking at averages, researchers have found small sex/gender differences in
cognitive, motor, and personality characteristics. Yet, these differences have
been found in very specific domains (i.e. on average, men may be better at some
types of visual-spatial processing while women may be better at some aspects of
language processing and some aspects of empathizing).[2] Baron-Cohen chooses to amalgamate
all of these different domains under the two constructs of empathizing and
systematizing; thereby suggesting that a number of small differences in degree should
be seen as two differences in kind (empathic vs. non-empathic and systematizing
vs. non-systematizing). By arguing that more men are good systematizers and
poor empathizers, and more women are good empathizers and poor systematizers,
he thereby suggests that male and female brains (on average) don’t differ from
each other by degree in some domains but in fact constitute two different


And although Baron-Cohen
denies it, his definition of what counts as “systematizing” seems highly
reliant on gender stereotypes. When talking about boys and men, he sees their
interest in memorizing sports statistics as evidence of their systematizing
ability. He also sees their preoccupation with gaining social status through
force as evidence for their lack of empathizing abilities and as
evidence for their abundance of systematizing abilities because, in this case,
“social hierarchies are systems” (36). Yet, when talking about girls and women,
he interprets their preoccupation with gaining social status through covert
actions as evidence of their empathizing abilities but he does not see this as
evidence of their systematizing abilities (in this case, social hierarchies are
apparently not systems?). Elsewhere, he argues that girls are better at gaining
access to a group of strangers because girls are “more likely to stand and
watch for awhile in order to find out what is going on, and then try to fit in
with the ongoing activity” (42). Again, this is interpreted as evidence for
empathizing abilities but not as evidence for systematizing abilities. Why
wouldn’t we interpret this, sans evidence to the contrary, as support for the
idea that girls are both empathic and systematic?


It seems to me that
Baron-Cohen simply takes everything that girls and women do as evidence for
their empathizing and “theory of mind” abilities but not as evidence for
their systematizing abilities, while taking everything that boys and men do as
evidence for their systematizing abilities and as evidence for their
lack of empathizing and theory of mind abilities. In this way, Baron-Cohen is
simply drawing on the centuries-old association of masculinity with rationality
and femininity with emotionality. For me, this raises the question: why do so
many scholars, both within cognitive science and within those fields that
connect cognitive science to the social sciences and humanities, remain so
enamored with Baron-Cohen’s thesis?



Want to cite this post?

Gupta, K. (2012). Almost Ten Years On: Why are we still talking about The Essential Difference? The Neuroethics Blog. Retrieved on
, from

[1] For an example of how these differences might be attenuated through learning, see the article“Playing an Action Video Game Reduces Gender Differences in Spatial Cognition” by Feng, Spence, and Pratt (2008).

[2] For a recent review of sex/gender differences in cognitive, motor, and personality characteristics, see Hine 2010. Many studies have tested sex/gender differences in empathy using different measures and have found statistically significant differences between men and women (for a review of sex/gender differences in empathy, see Schulte-Rüther et al 2008). A number of studies have found that on average, men are better at tasks such as targeting, mental rotation, and visuospatial processing (for a recent study of sex/gender differences in targeting abilities, see Moreno-Briseño et al 2010).


  1. I know Kristina is leaving it aside for the most part, but the thing that has always intrigued me about this study is the connection between this (hyper-essentialized) version of masculinity and autism.

    If we think in terms of the overlaps between the representation of gender and the representation of mental illness, there are definitely gendered stereotypes about empathy which reinforce the boundaries of both femaleness and "normal" social interaction. For example, given a man and a woman who display the same level of inability to navigate social situations, the woman is more likely to be seen as "odd," or even placed higher on the autism scale, while the man's activity is more likely be seen as normal "maleness."

    Does anyone who has more information about autism have comments on this?


  2. Hi Cyd,

    Just to note males are diagnosed as having autism at rates much higher than females (sources I have seen put the ratio between 8:1 and 4:1). The high ratio is likely due to two facts: (1) more males exhibit autism-diagnosing behaviors and (2) given the same behaviors exhibited in males and females, males are more likely to be diagnosed with autism (females are actually more likely to be diagnosed as having OCD or a more general learning disorder, given the same behaviors).

    But back to the bigger question: Is autism really characterized by decreased empathy (and if so, is the lack of empathy in autism caused by an exaggeration of typical male mechanisms — whatever that even means)?

    Well, if you define empathy "as the drive to identify another's mental states and respond with an appropriate emotion" (as Baron-Cohen define empathy), then, yes, it seems as if people with autism are going to have empathy deficits. But I see this definition as including two components: (1) The drive to recognize cognitive and affective states of others and (2) the ability to emphatically respond to the cognitive and affective states of others. As you can tell from my parsing of their definition, I think empathy has more to do with an ability to respond with an appropriate emotion, given the recognition of the cognitive and affective states of another.

    In spite of people's with autism decreased ability to recognize the cognitive and affective states of others, they do tend to show appropriate empathetic responses if they apprehend the cognitive and affective states of others. (And Baron-Cohen even admit as much.) What this says to me is that people with autism have deficits in their ability to recognize the cognitive and affective states of others. Not that they have deficits in an ability to emphasize, per se.

    And this is what sets the empathetic abilities of those with autism apart from psychopaths. People with autism generally do not have deficits in their abilities to emphasize given that they recognize the cognitive and affective states of others. Psychopaths, on the other hand, still show a lack of empathy even when they clearly recognize the cognitive and affective states of others. Any definition of empathy that puts psychopaths and people with autism in "the same boat" empathy-wise is probably a misleading definition … And interestingly Baron-Cohen recognize this problem, and they admit that even though on their definition psychopaths and people with autism come as both being low in empathy, people with autism do not show an empathy-profile similar to the empathy-profile of psychopaths.

    As for my parenthetical autism-empathy-male-mechanism question, I won't touch that (for multiple reasons), except to say that psychopaths are actually low on empathy and are (generally) high on systemizing …


  3. Hi Jason,

    Thanks for clarifying that empathy requires two components – I have been thinking about this as well, because I think Baron-Cohen often collapses "theory of mind” and “empathy.” He takes evidence of ToM as evidence for empathy and he takes evidence of lack of empathy as evidence for lack of ToM. I find his discussion of what happens at summer camp to be particularly revealing. Basically, he suggests that a dominant boy will pick on a socially awkward boy in front of the other boys in order to establish dominance, whereas dominant girls will use covert tactics such as gossip to establish dominance. Baron-Cohen takes this as evidence for the boys’ deficits in ToM and empathy and as evidence for the girls’ strengths in ToM and empathy. But, if you read what he actually says, I think both the boys and the girls are displaying strength in ToM and both the boys and the girls are displaying a lack of empathy (in their willingness to ostracize and pick on socially awkward members). The only difference seems to be that the boys are overt about their lack of empathy (perhaps because there are no social sanctions for this: “boys will be boys”), but the girls are worried about appearing non-empathic and so try to hide their behavior (perhaps because there are social sanctions for girls who appear non-empathic: “girls are sugar and spice and everything nice”). This suggests that both boys and girls at summer camp are more similar to psychopaths than people with autism by Baron-Cohen’s own definitions!



  4. Hi Jason and Kristina,
    Yes- you are quite right about the gender disparity in psychiatric diagnosis (also true for anti-social personality disorder, at the rate of something like 3:1)

    What I actually had in mind, but was not terribly clear about, was the way that these diseases are imagined to function, and the way that combines with gender stereotypes.

    As a (perhaps facile) example: If we take the representation of autism as a deficit in recognizing emotions (rather than one of formulating an appropriate response) and combine it with masculinity, doesn't that give us the good old stereotype that men are terrible at recognizing their wives (and their own) emotions?

    In this system, the problem that Kristina is noting with Baron-Cohen's overzealous categorizing of gendered behaviors is made even more problematic (and in a sense, proves itself to be true) because it is woven into and then reproduced by social systems of gendering. Does that make sense?

    Now- lack of empathy, psychopathy, and maleness- that's an entirely different, and more troubling, set of systems. I want to think more before I say any further about that.


Post a Comment

Emory Neuroethics on Facebook