The Gendering of Autism – How a few deliberately biased questions turned Autism into a men’s only club.


Since the 1980s, the prevalence of those with Autism associated with science, computing and other hi-tech industries, has once again singled out Autism as being a primarily male condition.

Diagnosing famous scientists, engineers, and computer scientists with autism has become both a parlour game and a cottage industry—Albert Einstein, Paul Dirac, Bill Gates, and Isaac Newton are among the most commonly cited in this category.

Contemporary understandings of geek masculinity have become one of the more common, gendered screens through which autism is now understood.

According to the Extreme Male Brain (EMB) theory of autism, people with autism possess hyper-male brains, therefore the existence and/or severity of their Autism, can be graded based on a scale that views those who are less adept at systemizing tasks (mostly women) as non-autistic, and those with extremely high systemizing abilities (mainly men) as being people with Autism.

To achieve this male and female interests are categorized as either “systemizing” or its opposite, “empathizing,” and then placed along a scale that grades from female to male to the extreme male (or autistic).

In particular, the EMB theory takes the form of a “double hierarchy,” in which an established series (e.g., male-female) forms the basis for a second series (systemizing-empathizing).

In a 2009 study Baron-Cohen co-authored with Bonnie Auyeung, et al., the authors provide these points of evidence for their extreme male brain theory:

-“The typical male brain is heavier than the female brain and individuals with autism have heavier brains than typical males”.

-“The amygdala is also disproportionately large in boys compared to girls … and children with autism have enlarged amygdala”.

Not only does this evidentiary criteria over emphasize differences between those with autism and those without; but it also forces males and females further apart by exaggerating the differences between average women, average men, and autistic people.

Further to this Baron-Cohen claims men are more interested in systemizing tasks, such as engineering, computer programming, and mathematics, or hobbies based on mechanics, construction, and categorizing—metalworking, boat-building, crafting musical instruments, even bird-spotting.

Whereas women tend to enjoy “having supper with friends, advising them on relationship problems, or caring for people or pets, or working for volunteer phone-lines listening to depressed, hurt, needy, or even suicidal anonymous callers”.

Based on these insights, Baron-Cohen devised a series of three tests:

-The systemizing quotient (SQ),

-The empathizing quotient (EQ),

-The Autism quotient (AQ).

All of which reflect his predetermined, gendered notions of male and female-appropriate activities.

On the Systemizing Quotient Test, testers are asked to rank their answers on a Likert scale to such questions as -“If I were buying a car, I would want to obtain specific information about its engine capacity” and “If there was a problem with the electrical wiring in my home, I’d be able to fix it myself”.

Meanwhile, the Empathy Quotient test includes such prompts as “I try to keep up with the current trends and fashions” and “When I talk to people, I tend to talk about their experiences rather than my own”.

In these prompts it can be seen that stereotypically masculine activities are assumed to reflect systemizing, while stereotypically feminine activities are assumed to reflect empathizing.

Clearly, these questions can easily be seen to reflect socialization as well as biology.

Therefore changing even a small number of the questions to more gender-neutral issues could have easily reduced or removed the sex differences found within SQ scores.

In 2006, Sally Wheelwright, Baron-Cohen, and their collaborators published a revised version of the SQ, the SQ-R, which included a wider range of questions about systemizing.

The original SQ, the authors admitted, “Were drawn primarily from traditionally male domains.” For this reason, the SQ-R included “more items that might be relevant to females in the general population,” a feature that would allow the researchers to determine whether men would continue to score higher on the SQ “even with the inclusion of items selected from traditionally female domains” (Wheelwright et al).

Some of the new prompts included “When I have a lot of shopping to do, I like to plan which shops I am going to visit and in what order” and “My clothes are not carefully organised into different types in my wardrobe” (answering “no” on this prompt presumably indicates an SQ type of brain).

The SQ-R successfully shifted the results. In the original SQ, men had a higher mean score on 86 percent of the questions, while women had a higher mean on only 13.2 percent. In the revised version, men scored higher on 68 percent and women on 32 percent—a rather dramatic shift in the sex ratio.

The SQ-R itself demonstrates that these sex differences may largely be an artefact of the testing prompts and the specific mix of questions applied.

One might therefore, hypothesize that the SQ could be revised even further in ways that would more drastically equalize the scores.

This article consists of excerpts form an academic paper written by J. Jack and published in 2011.


25 thoughts on “The Gendering of Autism – How a few deliberately biased questions turned Autism into a men’s only club.

  1. Wow. Isn’t this just like a man? (Just kidding, kinda…). Up until now, I’ve viewed Baron-Cohen’s EMB views as radical or somewhat ‘lunatic fringe’; now, I understand that this view is much more pervasive than I realized… Thank you for this excellent enlightening post!

    1. Thank you so much for the reblog Vivien….. it really does appear that truth is often stranger than fiction….. so glad to be able to expose some of the poor strategies responsible for the lower rates of women being diagnosed with Autism.

  2. What I noticed when I was reading the literature a while back was that they judge validity of a new diagnostic metric by whether or not it preserves the gender bias in autism diagnoses. It is literally hardwired into how autism is studied now.

    And then they turn around and use the gender disparity produced by their gender-biased diagnostic criteria to say that it’s a male disorder.

    I had someone try to imply to me once that I was wrong about how women/girls are often diagnosed late/not at all because of the stereotype that autism is a boys’ disease because gender disparity, and in retrospect, I should have said, “Thank you for proving my point – you take a gender disparity of 4:1 and assume that means girls are being diagnosed just fine because the researched gender disparity is 4:1, rather than questioning whether the criteria themselves are flawed.”

    I dunno, I’m not an expert in the field, but it seems rather flawed to me to say, “This disorder is a predominantly male disorder, which I’ll use to set up diagnostic criteria to maintain the gender disparity, which I in turn use to say it’s a male disorder.” It strikes me as the epitome of circular reasoning.

    1. You are so right…. it is circular reasoning that’s based on what at the time first appeared as a logical fallacy. You’ve expressed your self so well and articulated the issues completely. Thank you so much for taking the time to write this brilliant response.

      1. I’m a chemist and chemical engineering PhD student. So I’m a scientist in a different field. When reading social science out of interest, I often draw analogies to my chemistry training. When I was reading that I was thinking, “You’re begging the question: You’re using the results of diagnostic criteria standardized on the assumption that boys have autism more than girls to justify the assumption that boys have autism more than girls. You can’t justify your assumption with something you based on your assumption!”

  3. There is also another problem this causes. As a female teen with anxiety and depression who liked using computers; I spent a lot of time instant messaging, photo editing and making a website. This was about 10 years ago when most people didn’t use computers/internet regularly. The medical professionals immediately suggested aspergers, because girls weren’t supposed to use computers that much! But guy friends struggling with depression who played computer games much of the day were not accused of being addicted to computers in the way I was.

    My parents and I were advised I should not use the computer everyday, which was awful as it was the only way I was keeping in touch with my friends whilst not able to attend school, so it caused many fights with my parents, which was used as evidence. With the increase of social media use I hope the assumption is not made to quite the same degree, but I do wonder whether the lack of women in tech could be affected by this type of gendered assumptions. I didn’t want to try to learn to program or develop my website, because I was being told these things meant there was something wrong with me and that I was behaving too masculine.

    Telling self-identifying women diagnosed with Aspergers that they have an extreme male brain is I think quite offensive and upsetting to their gender identity. Also when a women behaves in a way which is stereotyped as more masculine is more likely to be told she has Aspergers and has something ‘wrong’ with her, compared to other women with similar symptoms, it is clear how it can play a role in reinforcing gender stereotypes. Since studying in the social sciences gender is seen as at least as much socially constructed as biological, surely the linking of gender in this way is lazy?

    1. I agree that being told my brain is male is not only offensive to my gender identity but also to my right to self-determination.

    2. There’s also the recent evidence that people on the Spectrum experience gender differences at a much higher rate than neurotypical people. We represent a higher percentage of transgender, agender, gender neutral, genderfluid, and other non-binary genders than do neurotypical people.

      I don’t know how that might relate, but the correlation is clear.

  4. When I was assessed for learning disabilities at school each time it came up as unidentifiable. Now I know I was compared to male models of learning problems. At 33 I am finally seeking a diagnosis it feels like relief. I also am dyscalculate and have dyspraxia. Mostly I am looking forward to help from an occupational therapist do I can become more organized for a start.

  5. All Baron-Cohen ever has been to me is a man with rather peculiar notions about autism, empathy, gender and emotional expression and he is hard at work proving all this notions. Never have I been remotely convinced. Now I’m going to take my extreme male hypersensitivity and empathise with animals.

  6. The prompts are particularly problematic if one considers the tendency in autistic people to perceive questions very literally. I am biologically female, and if presented with a question regarding the likelihood that I would repair and electrical problem in my home is nil, since I have a strong fear of electrocution. However, if the question had referenced any other home repair (patching sheetrock, replacing a faucet, fixing the lawn mower), I would have an extremely high likelihood of answering that I would make the repair myself. So I would have been wrongly scored, had I answered the question literally. As an autistic person would, of course, do.

Share with me......

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s