Simon Baron-Cohen knighted

[Flappynewyears]

Simon Baron-Cohen has been knighted. I am aghast that someone who's promoted the dangerous ideas that people with autism lack empathy, and have "extreme male brains" has been awarded like this. I would be interested to hear others' thoughts on this.

I didn't read properly either and thought Ali G got a knighthood...good for him! Then I remembered that scene with his thing poking out from his knees...gross...

@AnnaMagnani

I have always scored very highly on the emotions from pictures of eyes tests.

What SBC seems not to have realised is:

Autistic people spend a lot of time learning to fit in - you can learn this shit if you are bright enough
It is even more likely you will learn it if you are a girl - you could actually have fitting in as your special interest FFS if you are that desperate to avoid being bullied

Did he never ask an autistic person? Especially a woman?

I really can't imagine SBC is unaware of this. It is a commonly known approach and description within ASD.

However the damage his wretched eye emotion test score is huge.

We don't all live in the autism academic community and for people first thinking there might be an issue the first thing they will come across is one of those online tests.

I did mine, scored high on male and female brain, didn't know what to make of it and just left it.

It wasn't until years (and many burnouts) later when I watched an ITV programme about a school for autistic girls that DH and I just looked at each other and said 'are you thinking what I'm thinking?'

I still hear stuff like 'autism is far less common in girls' at work every day.

@AnnaMagnani

However the damage his wretched eye emotion test score is huge.

We don't all live in the autism academic community and for people first thinking there might be an issue the first thing they will come across is one of those online tests.

I did mine, scored high on male and female brain, didn't know what to make of it and just left it.

It wasn't until years (and many burnouts) later when I watched an ITV programme about a school for autistic girls that DH and I just looked at each other and said 'are you thinking what I'm thinking?'

I still hear stuff like 'autism is far less common in girls' at work every day.

Well that 'Reading the mind in the Eyes' test I think you're referring to has been extensively used during research on autism and found to be valid and reliable cross-culturally too. (Well different faces to reflect the prevailing ethnicity of the geographic location were used)

But of course the scores are averaged across large populations and individuals will perform differently, and there is a big overlap between the scores of 'neurotypical' participants and those diagnosed with ASD. Just because an individual scores poorly in it doesn't mean they have autism, and just because someone else scores well in it does not mean they automatically don't have autism.

I honestly can't see how it is possible to deduce an "emotion" from a glance at someone's eyes. There is back story, current context, other body language, actual shape of features to consider. Not to mention that it is highly unusual to have one emotion dominant at any one time - there are usual several in the mix. It seems to me highly reductive and again proscriptive devised by someone who doesn't really understand what empathy is, i.e a complex and nuanced listening to and response to another's experience.

I think he has walked back from the 'male brain' terminology in the last few days actually. I'll try to find his tweet

Here we are: Data from 36K autistic people and 600K typical people confirmed all 10 predictions from the Empathizing-Systemizing (E-S) and the Extreme Male Brain (EMB) theories. We retract the EMB terminology as the E-S terminology encompasses this and is more neutral pnas.org/content/pnas/115/48/12152.full.pdf

@JoodyBlue

I honestly can't see how it is possible to deduce an "emotion" from a glance at someone's eyes. There is back story, current context, other body language, actual shape of features to consider. Not to mention that it is highly unusual to have one emotion dominant at any one time - there are usual several in the mix. It seems to me highly reductive and again proscriptive devised by someone who doesn't really understand what empathy is, i.e a complex and nuanced listening to and response to another's experience.

And yet it works! As tested on probably hundreds of thousands of people, if not millions!

Conceptually it does seem pretty reductive I'd agree, and yet it has proven reliability and validity.

But it is measuring ability to infer emotional states quickly from a static photo of eyes - which I guess is just a step in the process of empathy - a step which makes things easier. Just like having good vision makes it easier to read. You may still be able to read without good vision but you may need to work a lot harder to achieve this, with aids etc, or it might take you longer. In the same way empathy is clearly achievable without that first step of recognising in a flash what someone is feeling from their face, but you may need to work harder by asking the right questions, listening to responses etc. Your brain maybe finds it harder to do the shortcut in the same way.

You could always get into nuanced discussions as to what empathy actually is. There's a well known distinction between affective empathy and cognitive empathy, and some researchers think affective empathy is generally better preserved in those diagnosed with ASD.

There are various 'empathy tests' out there as well, I'm sure none them are perfect and their creators likely know this too. It's a complex area.

@suggestionsplease1 thanks for the response. I am struggling to understand the idea that there is a "right answer", I think often many of don't know how to categorise what we are feeling, therefore the idea that someone else can do it for us by glance at our eyes seems odd. Reductive yes. I guess the reliability and validity is deduced from thousands of people arriving at the same answer?

I honestly can't see how it is possible to deduce an "emotion" from a glance at someone's eyes.

The test doesn't ask people to perform a complex analysis of the person's emotions, just to choose which of four named emotions best fits the picture. I've done the test several times as a research participant and I can ensure you that it is perfectly possible to do this with reasonable accuracy.

[quote JoodyBlue]@suggestionsplease1 thanks for the response. I am struggling to understand the idea that there is a "right answer", I think often many of don't know how to categorise what we are feeling, therefore the idea that someone else can do it for us by glance at our eyes seems odd. Reductive yes. I guess the reliability and validity is deduced from thousands of people arriving at the same answer?[/quote]
Yes, they will be looking at statistical significance over large populations.

So I think this test scores out of 36 - say you score highly at 33 and I score poorly at 8 - that doesn't mean I have autism and you don't because we are just 2 individuals and lots of other things could be happening. I may have other visual difficulties which make it hard for me to do well, English may not be my first language and I haven't grasped the true meaning of the emotional descriptors used. And as for you, well you may just have got lucky with your answers and have autism after all!

However, take 10,000 people and if you see that the average score for people with autism is 22, and the average score for people who have been diagnosed with autism is 29, well actually that is a huge difference, will easily reach statistical significance and shows that there is clear difference in that particular ability for those 2 populations. It's a valid measure of a difference in those 2 populations. Re-test those 10,000 people again, and maybe a 3rd time, if they're showing the same results then you have a reliable measure.

Sorry, that should have read ' the average score for people who have NOT been diagnosed with autism is 29'

Thanks for the explanation and for taking the time to write it out - appreciated and @NotDavidTennant - I am not really able to do that I don't think, which is of course why I am making a song and dance out of it. My answer would be random I think. But I am interested in individuals and this experiment is looking at large data sets so I understand that. Interesting conversation regardless.

I honestly can't see how it is possible to deduce an "emotion" from a glance at someone's eyes

It's not! And the photos are not conveying real emotion. I'm not sure how they developed the test, but it seems likely that they validated it by checking what emotions a large number of people identified from each picture. Rather than, say, how the model actually felt at the time the picture was taken, which would be difficult to measure anyway.

So, it's like, we have certain conventions to communicate emotion types, and these conventions are learned. Eyes wide for surprise, for example, eyebrows down for anger, and so on. It's like, with little kids they learn that pictures with 2 pointy ears and whiskers represent "cat" or any comic with eyelashes is a girl, even those those things don't resemble reality. It's symbolic.

Why people with autism tend to do worse on average as a population at recognizing the convention of emotion representation in eye photos is another matter. I don't know why. Does it have anything to do with constructing/recognising real emotion when time, context and real people are present? Maybe, maybe not. My opinion is not because recognising conventions from photos is a different task entirely.

@Brinn yes those were my thoughts following last post. The emotion attributable for the purpose of the test is the one the majority choose. It may be different to the emotion "felt". But that is ok because the test is to find the majority view. But it was stated as the "right" view which is something different. I think it does matter when you are making pronouncements about empathy but I'm not sure why. I was interested to get them both "wrong". I honestly don't think it is an inability to read the shortcut. More an awareness that the short cut leads many places. Also the baby pictures experiment is odd. I thought eyesight was pretty poor in newborns. Will stop nitpicking now.

@Brinn

I honestly can't see how it is possible to deduce an "emotion" from a glance at someone's eyes

It's not! And the photos are not conveying real emotion. I'm not sure how they developed the test, but it seems likely that they validated it by checking what emotions a large number of people identified from each picture. Rather than, say, how the model actually felt at the time the picture was taken, which would be difficult to measure anyway.

So, it's like, we have certain conventions to communicate emotion types, and these conventions are learned. Eyes wide for surprise, for example, eyebrows down for anger, and so on. It's like, with little kids they learn that pictures with 2 pointy ears and whiskers represent "cat" or any comic with eyelashes is a girl, even those those things don't resemble reality. It's symbolic.

Why people with autism tend to do worse on average as a population at recognizing the convention of emotion representation in eye photos is another matter. I don't know why. Does it have anything to do with constructing/recognising real emotion when time, context and real people are present? Maybe, maybe not. My opinion is not because recognising conventions from photos is a different task entirely.

That's an interesting theory. I guess you could only really investigate it if people had no a priori experience of emotions on faces to discount any possibility of learning conventions!

And I guess that is extremely unlikely!

I actually took the test again today to refamiliarize myself with it. I actually found it incredibly hard when approaching it consciously, deliberately- I thought to myself for many of the questions either 'I have absolutely no idea' or 'well it's either 1 or 2 out of 4' much of the time. By question 5 I decided there was no point in thinking too hard and I went with my gut more. Despite finding it (consciously) impossible, and just clicking intuitively I still scored 31 out of 36, which is pretty high for that test. Retest 8 hours later, questions had been randomized, I scored 30.

I guess what I'm trying to say anecdotally - sample size - me! ...Is that I think it would be really hard to doggedly learn the 'correct' or 'conventional' response from real life experience to correspond to these test items.

socialintelligence.labinthewild.org/mite/

Take a look at them - you would probably need 500 written words at least to try to distinguish one pic from the other to another person...''uh...there's 2 eyes, and i think they're looking slightly downward, and to the side a bit, and uhhh....the eyebrows look quite long and heavy...and uh....there's 2 ....no, well 4 ...if you count the little ones...wrinkles at the side of the right eye,....and uhh...they maybe extend....4 mm? Except for the one on the bottom, it's maybe 8 mm..?...and it goes straight to start off with and then tapers down....maybe around the 5th mm? And the whites of the eyes....they're maybe, uhhh, well for the left eye, on the right of the pupil and a bit to the up as well...I guess the white of the eye is maybe between 120% and 160% of the size of the pupil of the eye...?" The right eye, well it's harder to tell..... it's at an angle more...maybe 80% of the size of the pupil...but like I say, that's distorted because of the angle the viewer is presented with... ...uhhhh...and the eyelashes ...uhh"

Gut instinct isn't consciously measuring any of that, it's just shouting out loud 'OMG that person is so sarcastic'!!!

This isn't a Simpsons cartoon where people can look at 2 eyes and 2 eyebrows and say 'Homer happy', 'Marge pissed off'...this test items comprise surprisingly complex visual stimuli asking people to identify really elaborate emotions - and people are going with gut instinct and the 'neurotypicals' are agreeing with each other! And that's the fascinating thing!

I wonder if researchers have tested other primates with similar test stimuli (of their own species). That would be fascinating.

Gut instinct isn't consciously measuring any of that, it's just shouting out loud 'OMG that person is so sarcastic'!!!

A lot of things are hard to explain though: how to dress stylishly, how to recognize a kind of music, or detect sarcasm etc., they are easier to do than explain. Those are still learned skills though. Knowledge of social conventions is not necessarily conscious knowledge. I just meant that it's not clear what skills are being tested by the eye task, but it is probably not empathy or the ability to detect real emotion, as JoodyBlue said.

@suggestionsplease1 thanks for the share. I did take it and scored 24 which I find really interesting. You may surmise from my responses here that I regard myself as empathetic and so the results are a surprise to me - it is hard not to take it personally :) Moreover the title of the test is not neutral, it describes "social intelligence". So I have obviously taken umbrage!! But self effacement aside it is interesting that there can be such a discrepancy between one's view of oneself and a test like this perhaps? I am not sure what to make of it really. Obviously I am trying to argue with it. But I think in real world terms many people do describe me as empathetic, it is one of my "strengths", but perhaps I have been kidding myself. I am also aware that this post is me, me, me. But how else does one respond on an individual basis......

I tried it. I got 29, which is average. For me, I was thinking, what does the test designer think the right answer is? That's what I always think when I do tests! For example, there was one with a heavily made up women's eyes looking sultry and I picked "desire" as the right answer even though I have no reason to think that the model was actually feeling desire. But I think the test designer was trying to convey that. But maybe that's one I got wrong.