Parental Bias and Autism

There is no such thing as ‘low functioning’, please do not use functioning labels when speaking about fellow members of our autistic community, especially our younger neurokin. It’s dehumanising and inaccurate.

People harmfully and wrongly misconstrue some autistic people as ‘low functioning’ if they have co-occurring needs such as apraxia, learning disability or epilepsy etc. No human deserves to be defined and identified using functioning labels, our worth is not defined by how much we can contribute to society.

Please use the term ‘support needs’ instead and list what specific needs the autistic person you are talking about has. 

Please read about the harm that ensues from using functioning labels to describe our autistic community:

https://www.autisticality.co.uk/functioning-labels

Support needs would be inaccurate. I’m taking about intelligence, in practical terms low vs high IQ. If you prefer we can use low IQ autism instead of low functioning.

Intelligence can be measured in many different ways, the Intelligence Quotient is not accurate or representative of all humans, particularly those who are multiply neurodivergent.

there is a statistical mean when it comes to what is quantifyable. And often (but not always) values clustor around the mean. This is definatly true for IQ tests of neurotypicals. (not so true for autistic IQ tests where you get the double hump effect)

Well by definition you can only mesure what you can messure. You have to mesure IQ by difrences that can be messured. You can't messure inteligence (or anything else) using things that can't be messured.

How does IQ work? Is it normally weighted where the average appears at one value and then you look at how you differ from the average. What sort of distribution is it?

It's literally called a normal distribution or bell curve. https://en.wikipedia.org/wiki/Intelligence_quotient IQ is calibrated so that the average iq is 100 and the standard deviation is about 15 iq points. roughtly 2/3 of people have IQs between 85 and 115.

Also with the IQ test of autistic children versus ordinary children. That means it only includes the autistic children that get diagnosed, compared to the ones that don't. That might have a difference on the result.

That's fair. But since high iq is likely to make it easyer to mask it's likely those discrepencies, if they exist, are mostly present in the normal to high IQ inderviduals. Maybe there are some highly inteligent undiagnosed autistic people who test as having normal IQ because the test isn't well calibrated for them. But that doesn't really hurt my definition since I'm defining 'high iq autistic person' as someone with and IQ over 85. Which is 5/6th of the population for neurotypicals.

Peter said:

ordinary children.

You mean non autistic children, there is no such thing as normal it is a societal construct.

We have lost sight of the forest for the trees. Could it be that parents of high intelligence autistics are more likely to be autistic so are less likely to ask for help? They have spent more time masking and have found ways to deal with their problems that they don't really want to confron their difference in themselves. They have grown up hearing the idea that they must try to fit in and so they might try to enforce that on their children rather then seeking out a different opinion. They are just tired of all this masking they have to do and they probably have more stuff on their plate.

I think you are right. IQ doesn't measure intelligence it measures a bunch of different factors that lead to high intelligence. Which can be useful. Bottom down thinking is extremely useful, it has created the devices we are using to communicate. It has created the modern world. But you also lose sight of the forest for the trees in the bottom up approach. The problem is it is never enough. You are always trying to break down problems into smaller and smaller parts. How much money is spent on modern physics trying to break down the universe into ever smaller components? You need a balance between the two. Bottom-up to fill in the details, but also top-down as a sanity check, to make sure you don't get lost and are on the right path.

Doesn't the fact you can't measure IQ by looking at noticeable differences feel a bit strange to you?

How does IQ work? Is it normally weighted where the average appears at one value and then you look at how you differ from the average. What sort of distribution is it?

Also with the IQ test of autistic children versus ordinary children. That means it only includes the autistic children that get diagnosed, compared to the ones that don't. That might have a difference on the result. You might not be able to say autistic children have a different IQ then ordinary children, but only the diagnosed autistics. Does that make sense? Is that accounted for? There were a lot of strange words in the link and I found it very confusing. Shouldn't it be simpler. Can't they explain it in simple terms. I understand they are probably using it to communicate with other similar scientists who have knowledge of that but it seems so strange that research is inaccessible to anybody else.

I got a bit over the top in my previous message. Sorry.

I mean what kind of noticeable difference (based on IQ) would you like to measure? If it’s job success then that presupposes you assume that intelligent people always get good jobs which I think we both agree isn’t true?

anyway I do not think they are trying to measure autism via IQ tests in that paper. They are comparing the IQ tests of autistic children versus ordinary children.

Well the first thing I will say is that an IQ of 125 is at least one standard deviation above mean making it quite high. IQ gets less accurate to the further you get away from 100 in either direction. Once you get beyond 2 standard deviations The number becomes pretty meaningless beyond simply being very high or very low. It’s difficult to calibrate and test the reliability of IQ tests at those sorts of high and low IQ levels because there are so few people to test them on. All they really tell you is that someone’s IQ is very high or very low.

secondly I am not from moment suggesting that people with low IQ not be allowed to take exams, and I’ve said previously that IQ can be plastic can improve with education. However if I was planning a package of special educational needs for a student and I knew they had a very low IQ it would influence the way I put that together differently than if they had a very high IQ.

you know I’ve yet to meet or hear about a brilliant programer with an IQ of 60, but I’m open to being surprised.

I’m regards to top down thinking. I think the most practical purposes people use a mix of top down and bottom up thinking, or at least they should. But it works better if the top down thinking is used as a shortcut to help Focus on the bottom up thinking on the most important elements.

as you can say models can be wrong and sometimes very wrong. That’s why they have to be validated against experimental data. One mathematician one said all models are wrong about some models are useful. Newtons theory of gravity was wrong but it will get you to the moon and back. Einstein special theory of relativity is is pretty much all you need for GPS satellites but it won’t explain the orbital precession of mercury. That doesn’t mean bottom-up thinking was the wrong way to think about gravity.

bottom-up models can be chaotic and that doesn’t nullify their usefulness either. Weather forecasting being the obvious example. Or over even longer timeframes climate models which can’t predict the weather on a particular day but will tell you about the climate in a year some time in the future.

The sad truth is economists are just very very bad at maths even when they’re pretending not to be. I once sat in a fourth-year Masters economics lecture on stock market analysis where a student raised his hand and asked what is a ‘sine wave.’ For historical reasons we economists tend to get ahead have often bbeen those who have a good verbal / people skills rather than those who are good at maths.

I am loving this, arguing with you refining my argument. The problem with your approach is that you cannot include everything. In all mathematical models you have to make assumptions otherwise it would become too complex to deal with. I took a module introduction to mathematical finance and whilst it wasn't the really advanced I felt like the entire module was a scam. It was all about modelling the random variations of the stock market but not once did the lecturer say that stocks are portions of a business, and if the business does well the stock goes up and if the business goes badly the stock goes down. They were too concerned with the modelling of randomness and ignored the fact there is an underlying business behind the stocks. Do you know who sees stocks as pieces of a business instead of just numbers that go up and down, Warren Buffet the most successful investor of all time and probably autistic. They are trying to model a problem which doesn't really need modelling, reducing it down to numbers and probabilities instead of seeing them as businesses. I think chaos theory has a similar perspective where you can't include everything in a model. I agree that the bottom up approach is useful, it is very good for learning things, especially in maths, first you learn the definitions then you learn the theorems and then you learn the proofs and then you apply to problems but it isn't perfect. You will always miss something. Both a top down and bottom up approach are useful in certain situations. I may be wrong though.

Also I don't agree with just taking someones IQ as proxy for exams, I would rather just see how they did on the exams. If you believe someone will do badly on exams just because they have a low IQ then you are not going to put any effort into teaching them properly and giving them good revision methods. So they do end up doing badly on exams. It is a self-fulfilling prophecy. But if you instead took a top-down approach looking at their strengths and their weaknesses figuring out how they learn best then implemented those strategies then they would end up doing much better and have a much more fulfilling life. I don't like your argument because you are just reducing someone down to a number and taking that as their ability to do well on exams, as their intelligence. I don't think there is a need for it. I would rather just see if someone did well on exams. This is the same problem with aptitude tests, what aptitude tests, I would rather hire someone that has a proven ability that they can learn through exams then scoring high marks on your superficially created tests. People aren't numbers. We are far more complex and varied and unique then can easily be defined by a number. Why would you want to? That is what makes us special. There will only ever be one of us. Ever. Why would you just try and sum us up with a number or a label? Oh because this person has a higher IQ they are more likely to be a better programmer so lets just look at people for high IQ's to become a programmer. That is very dangerous thinking. Our uniqueness should be celebrated. If there was a computer programmer with 60 IQ then they would probably have a completely different way of thinking about it which might make them a much better programmer, because they have had to struggle and figure there way around things. Find tricks to deal with problems they have encountered. Which is far more important then intelligence.

Also I may be misremembering but in the book neurotribes there was this bit where autistic children where treated like animals because they didn't grow up to speak, because they were never taught to speak because to them as they were just autistic there was no use trying to teach them to speak. Why bother, autistic people are just animals. So being autistic was a good proxy for not being able to speak. This was used to justify their abuse. That is not ok. It's not ok to just take IQ as a proxy for exams because if you do that then that could potentially lead down a dark road where people are valued only in terms of IQ score and people are ranked based on IQ score. In some specific cases it might be useful but not as a general measurement for intelligence. It's just a number but it shouldn't define a persons whole life. Richard Feynmann supposedely only had 125 IQ and yet he was still one of the greatest physicists of his age. IQ doesn't really matter. Does knowing someones IQ really improve their lives? I think it might only make things worse, like pigeonholing them into certain categories. If you never try, how will you know what is possible.

Got a bit extreme there. Sorry.

Then you have a different understanding of reductionism than I do. When I think about reductionism I think about reducing all problems to their most basic elements. Building an understanding of systems up from the most fundamental and well understood units. That doesn’t mean you ignore interactions between different units. Emergent behaviour comes from complex interactions between very simple elements and very simple rules. But you build your understanding of that emergent behaviour from the bottom up.

that is a power of reductionism, The power of bottom-up thinking.

that said. if by saying that intelligence is more of a spectrum multifaceted thing then I understand your assertion that people with the same IQ can have quite different intelligence. however IQ is still useful. people with significantly different IQs typically have significant differences in intelligence. I would go further in saying in autistic people IQ is a good proxy for the ability to superficially mask. IQ is a good proxy for an autistic persons ability to pass exams. IQ is probably a good proxy for an autistic persons ability to do the task oriented aspects of a job.

people with higher IQs are more likely to be able to do more complex jobs. someone with an IQ of 100 has a much better chance of making it as a computer programmer than someone with an IQ of 60.

How are they useful? I feel like IQ is just a way to bin people into categories

I don’t feel they are useful, I completely agree with you!

No I have not read the book, thanks for the recommendation!

How are they useful? I feel like IQ is just a way to bin people into categories but how they get binned into those categories is different so two people with the same IQ don't have the same intelligence. Also not all measurements are useful. It is better to just take an individualised approach like with most things. Have you read the book the one straw revolution about this japanese farmer who got amazing yields without listening to all of the advice of modern science, science is all about measurements, and it is very reductionist, where you say this thing has this effect if you ignore everything else but the problem is you ignore everythin else. You don't view it as a whole, but instead if you viewed it as a system you might get better results. I don't know.

I understand and respect what you mean, but I don’t agree with this:

Peter said:

It is precisely because IQ tests are limiting that they are useful.

It is precisely because IQ tests are limiting that they are useful. They are specifically designed to exclude all kinds of factors. factors like social norms and memorised topical knowledge. They are meant to be a measurement of the ability to learn and reason and solve technical problems. It’s precisely because they exclude social elements that actually very useful for autistic people. Because it can give an impression of what a person can achieve if society isn’t in their way.

Slow Clap. Well played my friend. Well played.

Also, I don't know what most of those words mean but to me it seems like they are using an intelligence test to diagnose autism but how well does the intelligence test actually correlate to intelligence not just for autism but in the wider population is there a noticeable difference between someone who score 10 points higher on a test then another person. To me it seems like an IQ test doesn't really test intelligence. It tests your ability to perform well on a specific test if that makes sense. I thought I read somewhere about a study that looked at people with high IQ when they were kids and when they were adults and they weren't noticeably different in some way from the population. I may be misremembering.

No I don’t want to cite any research, all I want to say is that intelligence can be measured in many different ways and the IQ test is limiting. I mean this because it does not suit the way many people’s brains work and also once you measure someone’s intelligence immediate unhelpful assumptions are made about them.

You can’t value someone’s intelligence by looking at them. Also life skills are much more important and useful for lots of people.

Your wish is my command 

www.tandfonline.com/.../15374416.2020.1846543

Sorry, don't mean to be rude but I am going to uno reverse you and say can you cite sources stating the measurement is meaningful. If you can I would love to read them.

Iq tests may be less accurate than for neuro typical people but that doesn’t mean that it is meaningless. Unless maybe you can cite some source stating that the measurement  is meaningless?

Iq tests may be less accurate than for neuro typical people but that doesn’t mean that it is meaningless. Unless maybe you can cite some source stating that the measurement  is meaningless?

there is a statistical mean when it comes to what is quantifyable. And often (but not always) values clustor around the mean. This is definatly true for IQ tests of neurotypicals. (not so true for autistic IQ tests where you get the double hump effect)

Well by definition you can only mesure what you can messure. You have to mesure IQ by difrences that can be messured. You can't messure inteligence (or anything else) using things that can't be messured.

How does IQ work? Is it normally weighted where the average appears at one value and then you look at how you differ from the average. What sort of distribution is it?

It's literally called a normal distribution or bell curve. https://en.wikipedia.org/wiki/Intelligence_quotient IQ is calibrated so that the average iq is 100 and the standard deviation is about 15 iq points. roughtly 2/3 of people have IQs between 85 and 115.

Also with the IQ test of autistic children versus ordinary children. That means it only includes the autistic children that get diagnosed, compared to the ones that don't. That might have a difference on the result.

That's fair. But since high iq is likely to make it easyer to mask it's likely those discrepencies, if they exist, are mostly present in the normal to high IQ inderviduals. Maybe there are some highly inteligent undiagnosed autistic people who test as having normal IQ because the test isn't well calibrated for them. But that doesn't really hurt my definition since I'm defining 'high iq autistic person' as someone with and IQ over 85. Which is 5/6th of the population for neurotypicals.

Peter said:

ordinary children.

You mean non autistic children, there is no such thing as normal it is a societal construct.

We have lost sight of the forest for the trees. Could it be that parents of high intelligence autistics are more likely to be autistic so are less likely to ask for help? They have spent more time masking and have found ways to deal with their problems that they don't really want to confron their difference in themselves. They have grown up hearing the idea that they must try to fit in and so they might try to enforce that on their children rather then seeking out a different opinion. They are just tired of all this masking they have to do and they probably have more stuff on their plate.

I think you are right. IQ doesn't measure intelligence it measures a bunch of different factors that lead to high intelligence. Which can be useful. Bottom down thinking is extremely useful, it has created the devices we are using to communicate. It has created the modern world. But you also lose sight of the forest for the trees in the bottom up approach. The problem is it is never enough. You are always trying to break down problems into smaller and smaller parts. How much money is spent on modern physics trying to break down the universe into ever smaller components? You need a balance between the two. Bottom-up to fill in the details, but also top-down as a sanity check, to make sure you don't get lost and are on the right path.

Doesn't the fact you can't measure IQ by looking at noticeable differences feel a bit strange to you?

How does IQ work? Is it normally weighted where the average appears at one value and then you look at how you differ from the average. What sort of distribution is it?

Also with the IQ test of autistic children versus ordinary children. That means it only includes the autistic children that get diagnosed, compared to the ones that don't. That might have a difference on the result. You might not be able to say autistic children have a different IQ then ordinary children, but only the diagnosed autistics. Does that make sense? Is that accounted for? There were a lot of strange words in the link and I found it very confusing. Shouldn't it be simpler. Can't they explain it in simple terms. I understand they are probably using it to communicate with other similar scientists who have knowledge of that but it seems so strange that research is inaccessible to anybody else.

I got a bit over the top in my previous message. Sorry.

I mean what kind of noticeable difference (based on IQ) would you like to measure? If it’s job success then that presupposes you assume that intelligent people always get good jobs which I think we both agree isn’t true?

anyway I do not think they are trying to measure autism via IQ tests in that paper. They are comparing the IQ tests of autistic children versus ordinary children.

Well the first thing I will say is that an IQ of 125 is at least one standard deviation above mean making it quite high. IQ gets less accurate to the further you get away from 100 in either direction. Once you get beyond 2 standard deviations The number becomes pretty meaningless beyond simply being very high or very low. It’s difficult to calibrate and test the reliability of IQ tests at those sorts of high and low IQ levels because there are so few people to test them on. All they really tell you is that someone’s IQ is very high or very low.

secondly I am not from moment suggesting that people with low IQ not be allowed to take exams, and I’ve said previously that IQ can be plastic can improve with education. However if I was planning a package of special educational needs for a student and I knew they had a very low IQ it would influence the way I put that together differently than if they had a very high IQ.

you know I’ve yet to meet or hear about a brilliant programer with an IQ of 60, but I’m open to being surprised.

I’m regards to top down thinking. I think the most practical purposes people use a mix of top down and bottom up thinking, or at least they should. But it works better if the top down thinking is used as a shortcut to help Focus on the bottom up thinking on the most important elements.

as you can say models can be wrong and sometimes very wrong. That’s why they have to be validated against experimental data. One mathematician one said all models are wrong about some models are useful. Newtons theory of gravity was wrong but it will get you to the moon and back. Einstein special theory of relativity is is pretty much all you need for GPS satellites but it won’t explain the orbital precession of mercury. That doesn’t mean bottom-up thinking was the wrong way to think about gravity.

bottom-up models can be chaotic and that doesn’t nullify their usefulness either. Weather forecasting being the obvious example. Or over even longer timeframes climate models which can’t predict the weather on a particular day but will tell you about the climate in a year some time in the future.

The sad truth is economists are just very very bad at maths even when they’re pretending not to be. I once sat in a fourth-year Masters economics lecture on stock market analysis where a student raised his hand and asked what is a ‘sine wave.’ For historical reasons we economists tend to get ahead have often bbeen those who have a good verbal / people skills rather than those who are good at maths.

I am loving this, arguing with you refining my argument. The problem with your approach is that you cannot include everything. In all mathematical models you have to make assumptions otherwise it would become too complex to deal with. I took a module introduction to mathematical finance and whilst it wasn't the really advanced I felt like the entire module was a scam. It was all about modelling the random variations of the stock market but not once did the lecturer say that stocks are portions of a business, and if the business does well the stock goes up and if the business goes badly the stock goes down. They were too concerned with the modelling of randomness and ignored the fact there is an underlying business behind the stocks. Do you know who sees stocks as pieces of a business instead of just numbers that go up and down, Warren Buffet the most successful investor of all time and probably autistic. They are trying to model a problem which doesn't really need modelling, reducing it down to numbers and probabilities instead of seeing them as businesses. I think chaos theory has a similar perspective where you can't include everything in a model. I agree that the bottom up approach is useful, it is very good for learning things, especially in maths, first you learn the definitions then you learn the theorems and then you learn the proofs and then you apply to problems but it isn't perfect. You will always miss something. Both a top down and bottom up approach are useful in certain situations. I may be wrong though.

Also I don't agree with just taking someones IQ as proxy for exams, I would rather just see how they did on the exams. If you believe someone will do badly on exams just because they have a low IQ then you are not going to put any effort into teaching them properly and giving them good revision methods. So they do end up doing badly on exams. It is a self-fulfilling prophecy. But if you instead took a top-down approach looking at their strengths and their weaknesses figuring out how they learn best then implemented those strategies then they would end up doing much better and have a much more fulfilling life. I don't like your argument because you are just reducing someone down to a number and taking that as their ability to do well on exams, as their intelligence. I don't think there is a need for it. I would rather just see if someone did well on exams. This is the same problem with aptitude tests, what aptitude tests, I would rather hire someone that has a proven ability that they can learn through exams then scoring high marks on your superficially created tests. People aren't numbers. We are far more complex and varied and unique then can easily be defined by a number. Why would you want to? That is what makes us special. There will only ever be one of us. Ever. Why would you just try and sum us up with a number or a label? Oh because this person has a higher IQ they are more likely to be a better programmer so lets just look at people for high IQ's to become a programmer. That is very dangerous thinking. Our uniqueness should be celebrated. If there was a computer programmer with 60 IQ then they would probably have a completely different way of thinking about it which might make them a much better programmer, because they have had to struggle and figure there way around things. Find tricks to deal with problems they have encountered. Which is far more important then intelligence.

Also I may be misremembering but in the book neurotribes there was this bit where autistic children where treated like animals because they didn't grow up to speak, because they were never taught to speak because to them as they were just autistic there was no use trying to teach them to speak. Why bother, autistic people are just animals. So being autistic was a good proxy for not being able to speak. This was used to justify their abuse. That is not ok. It's not ok to just take IQ as a proxy for exams because if you do that then that could potentially lead down a dark road where people are valued only in terms of IQ score and people are ranked based on IQ score. In some specific cases it might be useful but not as a general measurement for intelligence. It's just a number but it shouldn't define a persons whole life. Richard Feynmann supposedely only had 125 IQ and yet he was still one of the greatest physicists of his age. IQ doesn't really matter. Does knowing someones IQ really improve their lives? I think it might only make things worse, like pigeonholing them into certain categories. If you never try, how will you know what is possible.

Got a bit extreme there. Sorry.

Then you have a different understanding of reductionism than I do. When I think about reductionism I think about reducing all problems to their most basic elements. Building an understanding of systems up from the most fundamental and well understood units. That doesn’t mean you ignore interactions between different units. Emergent behaviour comes from complex interactions between very simple elements and very simple rules. But you build your understanding of that emergent behaviour from the bottom up.

that is a power of reductionism, The power of bottom-up thinking.

that said. if by saying that intelligence is more of a spectrum multifaceted thing then I understand your assertion that people with the same IQ can have quite different intelligence. however IQ is still useful. people with significantly different IQs typically have significant differences in intelligence. I would go further in saying in autistic people IQ is a good proxy for the ability to superficially mask. IQ is a good proxy for an autistic persons ability to pass exams. IQ is probably a good proxy for an autistic persons ability to do the task oriented aspects of a job.

people with higher IQs are more likely to be able to do more complex jobs. someone with an IQ of 100 has a much better chance of making it as a computer programmer than someone with an IQ of 60.

How are they useful? I feel like IQ is just a way to bin people into categories

I don’t feel they are useful, I completely agree with you!

No I have not read the book, thanks for the recommendation!

How are they useful? I feel like IQ is just a way to bin people into categories but how they get binned into those categories is different so two people with the same IQ don't have the same intelligence. Also not all measurements are useful. It is better to just take an individualised approach like with most things. Have you read the book the one straw revolution about this japanese farmer who got amazing yields without listening to all of the advice of modern science, science is all about measurements, and it is very reductionist, where you say this thing has this effect if you ignore everything else but the problem is you ignore everythin else. You don't view it as a whole, but instead if you viewed it as a system you might get better results. I don't know.

I understand and respect what you mean, but I don’t agree with this:

Peter said:

It is precisely because IQ tests are limiting that they are useful.

It is precisely because IQ tests are limiting that they are useful. They are specifically designed to exclude all kinds of factors. factors like social norms and memorised topical knowledge. They are meant to be a measurement of the ability to learn and reason and solve technical problems. It’s precisely because they exclude social elements that actually very useful for autistic people. Because it can give an impression of what a person can achieve if society isn’t in their way.

Slow Clap. Well played my friend. Well played.

Also, I don't know what most of those words mean but to me it seems like they are using an intelligence test to diagnose autism but how well does the intelligence test actually correlate to intelligence not just for autism but in the wider population is there a noticeable difference between someone who score 10 points higher on a test then another person. To me it seems like an IQ test doesn't really test intelligence. It tests your ability to perform well on a specific test if that makes sense. I thought I read somewhere about a study that looked at people with high IQ when they were kids and when they were adults and they weren't noticeably different in some way from the population. I may be misremembering.

No I don’t want to cite any research, all I want to say is that intelligence can be measured in many different ways and the IQ test is limiting. I mean this because it does not suit the way many people’s brains work and also once you measure someone’s intelligence immediate unhelpful assumptions are made about them.

You can’t value someone’s intelligence by looking at them. Also life skills are much more important and useful for lots of people.

Your wish is my command 

www.tandfonline.com/.../15374416.2020.1846543

Sorry, don't mean to be rude but I am going to uno reverse you and say can you cite sources stating the measurement is meaningful. If you can I would love to read them.