Reading the Mind in the Eyes Test for Autists: What’s Wrong with This Picture?

My partner found this test recently, at a blog called Musings of an Aspie. From the blog :

The “Reading the Mind in the Eyes” is meant to test Theory of Mind (ToM) or the ability to recognize and understand another person’s mental state. It’s supposed to be a more advanced test than “Fear, Anger, Joy“, which tests simple emotion recognition.

[It] contains 36 items with 4 answer choices for each item, increasing the possible range of scores along with the difficulty level. It also contains a balance of male and female photos, a choice between more closely related mental states (i.e. not a choice between opposites like sympathetic/unsympathetic), and is composed entirely of photos representing complex mental states.

In the original study to validate the test, the AS/HFA group scored a mean of 21.9 while the control had a mean of 26.2. However, the AS/HFA group had only 15 participants versus 239 controls. A sample size of 15 is small, especially for study in which participants only have to complete two questionnaires (the AQ and Reading the Mind the Mind in the Eyes. I’m curious why the researchers didn’t make an effort to obtain a larger AS/HFA sample when they had the resources to administer the test to so many controls.

I did the test, and scored 28, meaning I got 8 “wrong.” My wife got the same score. I found the test engaging but also bizarre, for several reasons. Before you read my impressions, however, go and do the test, here. It’s quick and fun, and it’s much better if you have your own experience before reading about mine.

Done that? OK, here are my initial thoughts.

1) Often, none of the words seemed to fit, so I had to take an almost random guess.

2) As my wife pointed out, all the women images are of younger females wearing eye make-up. This gives a more “seductive” quality to their expressions.

3) In some photos the eyes are not clearly visible.

4) The images appear to be old ones selected from different sources, books, portraits, magazines. How do the people who made this test know what the person was experiencing at the time of the photo being taken? Even if the person in the photo was available for comment, how can we be sure they were honest/self-aware enough to describe their feelings accurately?

5) Do people only “have” one feeling at a time?

6) Is tuning into someone’s feelings (empathy) by looking in the eyes wholly a matter of reading visual information? In other words, can reading a photographic image stand in for having the living person in front of us, with all the countless other factors in play (body language, scent, temperature, breathing, etc? I think it’s pretty obvious the answer to this is no.)

How useful is a test of this sort really? Like IQ tests, it seems to betray the bias of its designers, which presumably (and evidently) is an “NT” bias, the bias, for example, that a person always knows what they are feeling/communicating, or that the only ways to communicate feeling are the ones we agree upon socially.

Bottom line: basically it’s nonsense, but it’s nonsense that’s being used to diagnose people and that may even lead to intervention and treatment.

16 thoughts on “Reading the Mind in the Eyes Test for Autists: What’s Wrong with This Picture?

  1. I scored 22. I found your objections very relevant. I also had to take a random guess at times, when none of the choices seemed to fit. And although I often have problems recognising people by their faces, I’ve found that other nuances – the walk, the mannerisms,or even, to get a bit woo, the aura of the living person, clues me in.Just eyes, static in a photograph, is not enough data.
    Terri in Joburg

  2. Well I scored 32 (4 wrong). Although tests like this are clearly pretty rough and ready, the crucial thing isn’t what the person in the photo was thinking or feeling, but what most people looking at the photo would assume they were thinking or feeling. In other words it’s a test of ‘normality’, and whether what you read into someone’s facial expression is pretty much the same as what other people would read into that same facial expression. By that definition, then yes, scoring 26 means you’re pretty normal in that respect. Well done! 🙂


  3. I scored 26. Very relevant comments about the women’s eyes. I was always looking for “horny”. Likely the pictures came from magazine ads.

  4. 25… never liked these kinds of standardized tests. It’s important to point out the BS of this kind of diagnosis. People are being prescribed drugs for mental conditions which have no real, physical tests ( A lot of these drugs make people violent, unstable and insane. The docs get paid to hand out drugs. 50% of americans have prescriptions, so it is a very fragile, scary society westerners live in. I’ve just never been a fan of tests with quantitive scoring systems. I remember in school, I would often ask myself, ‘Do I answer what they want to hear, or do I say what I think?’

  5. I found the mind in the eyes test via a blog about BPD. Supposedly those with BPD have inverse results as compared to those with autism, and my score of 35 would seem to be in keeping with this hypothesis. When answering I aimed to assign an emotion to the photo at first glance – prior to reading the provided options – and am curious about how others approached it.

  6. Hello nah! I like your approach at trying to formulate an emotion to assign to the photo before looking at the answers. I scored a 29 and when I took this, I looked at the face before the answers and tried to figure out how the mouth would naturally move in correspondence to the motion of the eye muscles. I agree with Jasun that this test is poorly designed and that the women are all young, with cosmetics and shaped brows. The poor quality of the photographs themselves also lend a somewhat sinister air to the possible answers. That this is used as a diagnostic tool is unacceptable, in my opinion. Without full body language, this test is worse than useless…it is to judge people.

    I took a somewhat similar test a long time ago, with a large series of full body photographs. The idea was the same, look at the person and judge what they are thinking, or intending to do. It was a test given to students in criminal justice classes who are training to be police officers. They need to know how to “size up” a person instantly and decide if they are a threat, or could be. That test, I didn’t get a single one right, it was so hard.

  7. In other tests I score highly for aspergers, in this test I did well 32, I think the other tests are probably more accurate as I have always been considered a bit eccentric or odd, also the correct choice is often the odd word out eg it is the only positive or negative emotion and that makes it easy to guess because most pictures can easily be seen as either basically happy or not and therefore gives a clue, also if the person is looking at you or not enables many clues to be eliminated, so I think this test does not represent ones ability in real life.

  8. I asked Professor Simon Baron Cohen about where he got the pictures from and received a reply today. “Dear *****. Sorry for the delay in replying. From fashion and other magazines. Best wishes, Simon Bc” so there you have an answer to your point nr 4. being that they didn’t really know what the people in the images really felt at that moment.

  9. I also found this test through a BPD blog. I scored 34. The two I answered incorrectly I ‘overthought’ it, rather than went with my first impression. I didn’t think the expressions were random at all, but perhaps that’s because I have BPD and have been told by therapists that I’m very attuned to the emotions of others.

  10. Hi. I got a 32 and basically had fun. But I agree wholeheartedly with your concerns. This test has serious flaws as you’ve said. All 4 of my incorrect answers had to do with the young women with eye make-up. Lighting is terrible. Old photos. And it baffles me that this could be used scientifically. You are exactly right to ask under what authority these “answers” have been deemed to be truth. Bogus. Fun, but no more.

  11. I just wanted to ensure that you all know that this task is used among a battery of other tests. It is not used on its own for diagnostic purposes, getting a low score does not mean that you fall on the autism spectrum. I think puts it best: “Our tests are posted on our website to enable free access to academic researchers. None of them are diagnostic: No single score on any of our tests or questionnaires indicates that an individual has an Autism Spectrum Condition (ASC). If you are concerned that you, or a friend or relative, may have ASC, please discuss these concerns with your GP or family doctor or ask the National Autistic Society (NAS) or equivalent charity in your country, for advice.”

  12. Pingback: What Is A Daze In Reading Assessments | News Latest Update

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your 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