Please take your gender bias out of your science

A meme has been going around a couple of the social networks I’m on. I picked it up on LiveJournal, but I’m told it’s also been making the rounds on Facebook. It’s called “Reading the Mind in the Eyes” and is allegedly a test used in diagnosing Asperger’s syndrome. According to the website introduction, it first appeared in Simon Baron-Cohen’s book “The Essential Difference” in 2003. In a nutshell, the test asks you to look at a picture of someone’s eyes and determine their emotional state – you get four words to pick from.
Before you read on, go and do the test. It shouldn’t take you more than 15 minutes, and you might even spot what I’m about to write about. Before I write on, it’s probably a good idea to point out that I haven’t read Simon Baron-Cohen’s book and thus have no idea whether the online test is a faithful reproduction of the one in the book, or just an imitation. So I am writing purely about the online test.
The thing that struck me about the test was the shocking gender bias it displayed. Yes, half the pictures were of men’s eyes, and the other half of women’s, but that’s about as far as balanced coverage goes. The age range of male models used was much greater. We saw young eyes, but we also saw pictures where the wrinkles and eyebrows were threatening to eat the eyes. Heck, men were event allowed *gasp* to have asymmetric eyes! Compare and contrast to the range of images we got for women’s eyes: not a single wrinkle to be seen in 18 images. Without exception, all of the women had their eyebrows plucked into perfectly attractive shapes – no stray hair to be seen. The vast majority were made up to accentuate the eyes and look attractive. There were maybe one or two photos of women without make-up.
Now think about the moods and emotions that the male and female eyes expressed. Did you notice that men were despondent, decisive, insisting, while women got to be fantasising, playful, and flirtatious? Not only were the types of emotions strongly gendered, but women were even limited in the number of emotions they could display. Every single man had an emotion of his own (19 in total, counting the introductory example); yet two women each had to be fantasising, preoccupied and interested, as apparently the creator of the test could not imagine 18 different emotions for women to display.
Now, as I said, I don’t know if this specific version of the test is the one which appears in Simon Baron-Cohen’s book. But if it is, I would be very worried. The underpinning theory of that book, you see, is that there are significant gender differences in how we think, with women more likely to empathise and men more likely to systemise. Baron-Cohen even labels his different ways of thinking the “male brain” and the “female brain”. I would be seriously concerned indeed if it turned out that someone was making such generalisations who did not have sufficient imagination to come up with 18 different emotions which women can exhibit.
I would appreciate it if someone could set me straight and tell me that the test in the book is more representative of real women, but just in case it isn’t, I would like to reassure both my readers and Mr. Baron-Cohen that women can be angry too.

6 thoughts on “Please take your gender bias out of your science

  1. sinclair_furie

    Yeah, I definitely noticed that. About halfway through the test, I thought, why are all the women hot and flirty?

    Reply
  2. Andrew Ducker

    I just went through and made a note of all of the answers:
    Men: Panicked, upset, insisting, worried, uneasy, despondent, cautious, regretful, anticipating, accusing, thoughtful, friendly, defiant, pensive, hostile, serious, concerned, suspicious
    Women: Playful, desire, fantasizing, preoccupied, sceptical, contemplative, doubtful, decisive, tentative, fantasizing, preoccupied, interested, cautious, interested, reflective, flirtatious, confident, distrustful, nervous,

    Reply
  3. Eliza

    …I’ve done that test and I didn’t even notice that. Ouch. Redid and noted answers – exactly the same as you. At first I was wondering how to tell whether the older/un-made-up faces were female, but given that counting only super-pretty as female gives us a fifty-fifty split, I’ll make the assumption.
    Stumbled over while looking for a response from Baron-Cohen to the book ‘Delusions of Gender’ (someone said there is one, but I’m beginning to doubt it). Thought I might plug the book, considering your post – it’s about the shitty science behind claims that men and women have vastly different brains. Well-written, thorough, and includes some well-aimed shots at Baron-Cohen.

    Reply
  4. Milena Popova

    Thanks for the tip – I’ll have a look. You do raise an interesting question which is this: Faced with photos of people’s eyes where the people are roughly the same age and there is no make-up involved, would we be able to tell if they’re male or female? I’d quite like to run that experiment actually.

    Reply

Leave a Reply

Your email address will not be published. Required fields are marked *