Hacker News new | ask | show | jobs
Does watching television trigger autism? (2007) (nber.org)
38 points by metamuas 1514 days ago
8 comments

Proximity to freeways increases autism risk, study finds [0].

Having a mother who lived within 1,000 feet of a freeway while pregnant doubles a child's odds of having autism [1].

Basically, white noise can cause autism. People are screwing up their kids at an exponential rate [2].

[0]: https://web.archive.org/web/20210609042055/https://www.latim...

[1]: https://www.webmd.com/brain/autism/news/20101217/fast-lane-t...

[2]: https://www.google.com/search?client=firefox-b-1-d&q=childre...

I did a -lot- of research on a house about 1500 feet of a freeway.

It seemed to me nearly all of it was correlative. Which is fine, as it may be true.

But very few corrected for the fact freeways often run through the worst areas in town, which could mean nutrition, drugs, alcohol, lead paint, asbestos, heck about anything else bad about living in older, run down areas.

Your point relies on the causative link not being that having a freeway through a section causes it to decline into being the worst part of town.
Well, there's plenty of rural and 'rich suburban' areas close to freeways. In my county in Florida, they're building brand new 500k dollar houses 100 feet from the freeway. For reference, 500k is high end of house prices here.

Would be interested to see if those show the same.

If not, you could rule out white noise completely. That would then seem to point to other factors, including total pollution, or more socioeconomic things.

Apologies, I should have left my point more open. That is, I don't know that there is a link. Would love to see evidence.
I could see how living next to a major road or having a noisy environment could disturb a child's sleep at night. Perhaps it is the role of noisy cars that disturb the sleep that triggers these changes?

I was a strange child that could only sleep if mother turned on a hair dryer next to me some nights. This is materially different from highways - I briefly got to experience what it was like living next to a noisy road and I would argue it would be more likely that it is the role of noisy cars, screeching tires, horns, and pollution that is at play in these studies, not the white noise of highways. Would children that lived in windy forests with no sound isolation having to listen to the noise of the trees fare the same?

Most autistic people are sensitive to stimuli - sensitivity being nearly completely genetic determined and little to do with environment. I wonder if it this greater sensitivity being triggered by highways and disturbs sleep that impairs brain development.

More likely would be tire particulates.
Don't underestimate noise pollution. It's a very real stressor.

Though, what's more likely is that a child with mild ASD is more obviously autistic under stress, which leads to diagnosis earlier.

Don’t underestimate particulate matter and ozone. Kills millions every year.

But the same theory could apply, as health-induced stress instead of distraction stress. Or both.

Both seems plausible. I just wanted to note that noise pullution is often overlooked and much more impactful than people assume.
What else correlates with living within 1000 feet of a freeway? Population density? Pollutants in the atmosphere? Proximity to Starbucks? Lower socioeconomic status?
Uhm... wouldn't that rather be "more noise/television makes you more likely to notice your child's autism"? An overstimulated autistic person has a much harder time masking their autism. Seems likely to me that a child raised in a quiet environment may not get diagnosed until adulthood, while the same child raised in noise may have a hard time fitting in already. Most of "symptoms of autism" that people notice are actually how autistic people react to distress when they can't mask anymore, and overstimulation definitely causes such distress.

Also, I'd say that modern children programming that's optimized for grabbing as much attention as possible isn't exactly autism friendly; I find older cartoons with their slower pacing much more comfortable to watch.

"White noise causes autism" is a pretty weird conclusion to get from all of that and would require extraordinary evidence to support it.

[2] does not support the idea that white noise causes autism. It's a pretty big leap from "freeway proximity, even in utero, correlates with autism" to "white noise causes autism."
Isn't the womb a fairly noisy place, anyway? From the perspective of a fetus, that is. I remember when our twins were born, white noise was recommended to us by practically everyone because of the similarity to sounds heard in utero.
Odd, I would expect a steady beat to mimic your heart to be a big part of mimicking sounds of the womb. Along with a colored noise. Though, I would suspect pink is better than white.
They're essentially under water and a pregnant mothers noisy internal processes, other than her breathing and her heart beat, are not really following any sort of pattern. And we can't discount the environment that the pregnant mother finds herself in, either, since that noise will also be present to varying degrees.

edit/ Having said that, the "white noise" machines that I've seen, which are available as sleep aides specifically for young children, are not really white noise. There is white noise there but there is also something that sounds like a tumbling drying machine, or a fan, or a hair dryer... Something that a brain could attach itself to as a pattern.

This makes sense. Probably most is marketed as "white noise" because that was a familiar term.
Why do you think white noise causes autism?
Basically, humans are pattern seeking creatures, and children exposed to white noise will try to find patterns in white noise, when there are none, thus leading to the brain maladapting to random patterns. My theory, but I couldn't find a good definition of traffic noise being related to white noise or classified in particular, but who knows.
In argument against your theory - a lot of parents use white noise explicitly to help their babies to sleep. From my own personal experience, most of those kids don’t then suffer from autism.
In argument against your theory - personal experience is not data.
Yet it's at least as valid to me, as a parent, as some random HN commentors pet theory about white noise causing autism.
Is there data showing that white noise _does_ cause autism?
I used to share a room with someone who would fall asleep to TV. In the UK after the last show of the night there would be beep tone for a long time and then it would just be static, no signal.

This lead to me having a recurring nightmare which I also got for a long time into adulthood, seeing the tone as 1/2D line, flowing along nicely, and then the static would screw it up and turn it to knots and I would be unsuccessful in making it turn back to a line.

I have this same reoccurring nightmare and I also grew up in the UK. The line would be smooth, but the violently contort into anxiety inducing sharp triangles. This is the first time I have heard someone echo the same dream.
Nice to hear someone else having the same thing!

In a way perhaps it's kind of an abstract "powerlessness in the face of chaos" nightmare.

Not sure why exactly it went away. But I do remember being conscious of it in my 20s and trying to understand it. I think I tried to understand it consciously and accept some of the powerlessness perhaps. Or maybe later on, it was meaning general confusion and anxiety which was suppressed.

At any rate the nightmare was very scary and awful, as if something evil and wrong was there.

I think air pollution is a much more likely explanation.
Autism seem to be a politicized issue by looking at the replies.

As far as I can tell the paper is nothing but honest truth seeking, you know a study doesn't equal truth, the article itself doesn't say it is a confirmed hypothesis. And while correlation doesn't mean causation, when you see a correlation, the first thing you do is you investigate if it is a causation too. The most logical thing to do and thanks God smart people did that with lead and many other toxic chemicals that affected society in the past.

All it tells me is that the standard for evidence in economics is much lower than the standard of evidence in medicine or sociology. This paper has some problems that run very deep in established methodology, so while it might be good faith, it's also useless.
Sample size: 1.

Conclusion: the standard for evidence in economics is much lower than the standard of evidence in medicine or sociology.

What specific problems are you referring to?
The number of assumptions made without good evidence (neither rain nor having a cable subscription is automatically more TV time - especially the presence of cable says nothing about how parents control their childs media diet and is just an indicator of wealth) and the blindness for alternative explainations for these supposed relationships.

They started from a typical armchair hypothesis and then looked for a way to interpret data to fit that hypothesis. That by itself is not remarkable, but the quality of the interpretation is questionable at best and outright malicious as worst.

They address the connection between rain and TV time in the article:

>The authors first use data from the American Time Use Survey to confirm the link between precipitation and television watching. The results suggest a strong relationship - a child under age 3 watches an average of 27 additional minutes of television on a day with one inch of precipitation (which is equivalent to a day of heavy rain) than on a day with no precipitation.

If you're going to post joke research, please don't do it about autism. It's already bad enough as it is with all the quackery going around.
I don't think it's joke research, I think they chose an effect that might have positive correlation and proved that, then showed at least one environmental factor that could explain it and how you could find correlation there too. They say themselves at the end that they aren't trying to say it's conclusive evidence in any direction, I believe they're just trying to generally indicate that there does seem to be some sort of environmental factor, and that it's likely possible to discover (or at least get set on the path towards) the true cause via this sort of analysis
> I don't think it's joke research

Perhaps not consciously, but I take NBER research on causes of autism as seriously as NIH research on causes of recessions.

Well the stock market succeeding depends on people having irrational faith in the stocks like Tesla, despite making most of their money from Bitcoin than selling cars[0], so that isn't really a valid counterfactual point, since market opinions are based on psychological beliefs.

[0]: https://web.archive.org/web/20210427112905/https://www.wsj.c...

Maybe I'm missing something here but how does the article you linked support "making most of their money from Bitcoin than selling cars"? It says:

- Revenue of $10.4 billion

- Net income reached $438 million

- sold some of the $1.5 billion worth of bitcoin that it purchased in February, contributing $101 million to the bottom line

- sales of regulatory credits to other auto makers to help them meet emissions mandates, which carry a 100% profit margin, reached $518 million

All this tells us is that they sold some of the bitcoin they bought earlier (a tiny fraction of their total income), they're selling a shedload of cars, and that they're spending a ton of money. Which you'd expect given they're building multiple giant-ass factories at the same time.

I'm no crazy Tesla fan (love some of their tech, hate their 'we own your car' attitude) but you can't argue that they don't usually deliver (let's not talk about FSD :P ).

In other words: Hand waving.

This is the kind of "research" that leads to CNN headlines like "TV is turning your child autistic!", and other such scare stories for frightened and confused parents that lead to more bullshit salves and treatments that only harm autistics even more (as if we're not getting shit on enough already).

And if it's not intended as a joke, that's even worse.

Precipitation means the kids stay inside and are more likely to watch television[0]. This was influenced by another paper about the link between precipitation and autism[1]. Why do you think it is joke research? Why did you criticize before looking into it?

[0]: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=989648

[1]: https://jamanetwork.com/journals/jamapediatrics/articlepdf/3...

They're both written by the same people, econ professors that are helplessly out of their field.

They're assuming a lot of things here:

* Rain means children watch televsion, no rain means they do not. This is a harsh oversimplification of how parents work.

* Cable TV was correlated to increased autism diagnosis. But what else is correlated to Cable TV? Wealth. And what's correlated to wealth? Getting your kid diagnosed properly in the country where healthcare is not easily affordable maybe? If you think that's a stretch, well, the entire paper is argued on this sort of thin ice.

* Autism isn't generally considered to have "a trigger", but some circumstances like stress can make it more obvious how a someone with autism is different. Autism is one of these things that aren't objectively measured, but often are diagnosed as a result of how annoyed parents are.

> And what's correlated to wealth?

Delayed family starts.

And what's a well-established contributor to autism?

Paternal age at conception.

N=1, but I grew up without cable TV (only one movie on the weekend), and I'm autistic. very so as a little kid.

my parents were both mildly on the spectrum, I apparently got a double dose. hooray for assortative mating!

They don't assume that, they showed rain correlates with TV watching. Are you making assumptions about what their study does without reading the article that explains it?
> The authors reason that children are likely to watch more television if they live in an area that gets more precipitation

Growing up in part of the California Central Valley where it is intolerably hot outside for a large portion of the year and has near-desert levels of precipitation, and then moving to the Bay Area which has more precipitation and less intolerable heat, I found the reverse to be the case, which given where the research was conducted...

> If that is the case, then a finding that areas with higher levels of precipitation have higher autism rates would be strongly suggestive of a role for television watching as an autism trigger,

“Strongly suggestive” ludicrously overstates the case.

I mean, there's a lot of things more strongly associated with precipitation than television watching.

One of the big problems with studying or talking about autism is assuming that because we have one word for it that autism is just one thing, subject to root cause analysis, and then you have the cause for that one thing. It is very much not like that.
Correlated, not caused.
Corrected, thanks.
If you have the time, please find out what traffic noise is closest to. I tried to find out but couldn't get any further than this wikipedia page[0].

[0]: https://en.wikipedia.org/wiki/Roadway_noise

search: >'traffic noise effects on people'< @DDG: <https://html.duckduckgo.com/html?q='traffic noise effects on...>

Edit: Add;

Health Effects from Noise : https://en.wikipedia.org/wiki/Health_effects_from_noise

This is a hilariously bad line of reasoning and basically a signal for me to take econ researchers less seriously.
"Diagnoses of autism have risen dramatically over the past few decades, from an estimated one in 2500 children thirty years ago to one in 150 today." is one of the first lines of the article. If you are so smart, and only you are so smart, what environmental factor has increased that rapidly?
Maybe because it's easier to get a diagnosis now and parents are more inclined to get their children checked? The medical system is not unchanging over the years.

This is the same kind of thoughtful question as "why are there so many more left-handed/gay/trans people than x years ago" - because social factors have changed visibility of these traits, not because they didn't exist.

We're much better at diagnosing it than we used to be.
Agreed. I have a friend who has been autistic their entire life, but was only acknowledged/diagnosed in the past little while, in their 30's. Back when they were a little kid, it was interpreted as just being a challenging/picky/quirky child, nothing "diagnosis-worthy". If the medical system and society better understood autism back then, it would have been noticed and diagnosed back then. I have heard the same approximate story over and over.
That’s disputed; another point of view is that “autism” is a label and we’ve increased the scope of that label to cover a lot of things which in the past were either given different labels or no label at all. There are a number of voices in the academic/professional community questioning the great and growing breadth of the “autism” label and the flimsiness of the scientific justification for its expansion (people such as Laurent Mottron, Lynn Waterhouse, Allen Frances)
This. I think the problem is simply we're over-diagnosing autism. Basically borderline people would've been considered normal in the past, but are now considered autistic. Frankly, the entirety of symptoms involved in diagnosis sounds subject to a person's opinion of what people's behavior should be vs shouldn't be. Since this line is hard for everyone to agree on, we've come up with degrees of autism, like "high functioning" autism such as Asperger's syndrome. I'd argue, it's only when it's obvious (i.e. the person is largely dependent) while having said symptoms, that they're autistic... Just because someone behave's abnormally socially, has peculiar interests, and/or has delayed abilities (unless extreme), isn't sufficient IMO. Additionally, two different doctors may end up giving different diagnoses.
Autism (and ASD) should really not be seen as this essentialist "yes or no".

There are degrees to which people, considered autistic or not by your favorite arbitrary standard, exhibit autistic traits. And those traits are not strictly inherently better or worse than neurotypical traits.

The standard for consideration as disorder should be the same as with most psychiatric disorders: Not mere presence, but causing issues that need support.

But at the same time, the people that are doing fine with autistic traits are no less autistic.

(Also, Asperger's has ceased existing a while ago)

> Just because someone behave's abnormally socially, has peculiar interests, and/or has delayed abilities (unless extreme), isn't sufficient IMO.

Then it's a good thing that this is absolutely not sufficient to declare someone autistic (even "high functioning", although this kind of categorization is frowned upon these days) even now. Those are just some of possible externally visible symptoms.

I and many other autistic people would strongly disagree. Neurotypical vs ASD isn't some gradient like you seem to think, it's much bigger, clearer wall between us.

I am one of those that you wouldn't diagnose autistic, but in reality me and my thoughts are much much much closer to the ones you would than neurotypicals. I wasn't diagnosed as a kid, partly because autism wasn't diagnosed at least in my country when I was born, partly because even though it was obvious to everyone how different I was my parents tried to pretend I'm just a normal kid. And partly because with time I honed my high quality human emulator to the perfection so at some point I could play that almost-"normal" kid part that they and society wanted from me.

Doesn't mean I didn't constantly feel there's been a mistake, I've been born on wrong planet, in middle of different species that doesn't understand me and which I don't understand. Absolutely nothing about the world made sense, life was hard and strange even if I was extremely good at pretending it isn't. It all changed when I found out and found others like me. Nowadays I'd be diagnosed at two years old, and my life would have been so much easier. I wouldn't have the need to pretend (or mask as we call it) as much, which is a HUGE constant stress. I could understand the world and myself much better, and adjust my life knowing better why and how I'm different. I feel like I could have grown up to be much less of a broken person. There are millions and millions like me, many of whom still don't know and are thinking what is going on in this world. They should be diagnosed as early as possible so they could start enjoying life like I have in much healthier way after finding out.

It's not really surprising that it was - and is - underdiagnosed and we're now finding more cases, as not that long time ago we didn't really diagnose it at all, and after that we only diagnosed the minority of cases who couldn't come up with an acceptable human emulator.

And because the mental image we have of autism is so wrong. I spent decades trying to find people like me, disorders that would explain the things. I gave up, thinking I'm unique. But my mental image was wrong also, so I didn't really give it enough thought - autism was a list of symptoms on wikipedia page and Rainman - that's not me, it's just that my brain and thinking is completely different and nobody understands, I can't even tell anyone because it's impossible to explain. It wasn't until I randomly read an article written by autistic person, and like 1/4 way through it I knew I found my kind - before that nobody understood, and that article could have been word to word written by me, except he knew the reasons behind it.

I don't see how both can't be true.

The diagnosis itself has changed a lot (retiring Aspergers, changing it to "ASD") to account for this though, and I'd generally say that having more awareness of how people deviate from neurotpicallity is a good thing.

If you see it purely as a disorder (and thus a bad thing) maybe you have an interest in continued underdiagnosis, but many people now see autism as different, but not strictly worse or better.

> I don't see how both can't be true.

Laurent Mottron believes that the original narrow view of autism had some genuine scientific value, but DSM-5 ASD has broadened the concept to the point that it has become scientifically meaningless. He presents the idea of a “spectrum” as essentially taking a wrong turn, and he argues the only way to get autism research back on track is to move away from the idea of a “spectrum” and return to a focus on “prototypical” autism

Lynn Waterhouse goes further - she believes that even the original narrower concept of “autism” is a failed scientific hypothesis, and the best way forward is to drop the concept of “autism” entirely (and related concepts such as ASD, Asperger’s and PDD-NOS), and look for alternative constructs to replace it with (her tentative proposal is “phenotypes of neurodevelopmental social impairment”)

Allen Frances has said that (in hindsight) he made a big mistake by agreeing to put Asperger’s in the DSM-IV. He says it was sold to him as a very rare disorder and he was shocked to see how frequently it came to be diagnosed. And he’s said that DSM-5 ASD is “even worse” than DSM-IV Asperger’s. (Like Mottron, unlike Waterhouse, he has no objection to DSM-III/IV autistic disorder; Mottron is less negative on DSM-IV Asperger’s than Frances, but I suspect there is actually less distance between them in practice on that issue than a cursory reading of their public statements might suggest.)

While they don’t agree with each other in all the details, what they all have in common, is a critical attitude towards contemporary mainstream diagnostic practices - maybe I’m misunderstanding you, but your response sounded to me like an attempt to harmonise those criticisms away

> I'd generally say that having more awareness of how people deviate from neurotpicallity is a good thing.

I think “neurotypicality” is a myth. Nobody is “neurotypical”. As the English child psychiatrist Sami Timimi says, “we are all, all humanity, neurodiverse”.

Besides the obvious differences in diagnosis that others point out (my parents specifically had me screened, and both neuropsychologists said no, while every therapist I've seen as an adult says yes), there's some reasonably compelling evidence that assortive mating has a role, and don't less compelling evidence that parents being older at time of conception may play a role.
Why not diagnosis over environment?
While it can be argued for other things like ADHD, I doubt it can be argued for cases of severe autism.
1. The quote says "diagnoses of autism" not "diagnoses of severe autism" so if the definition of autism has expanded over time due to improvements in our understanding of autism (which it has) the diagnosis argument is far more effective.

2. It is easier for children of poorer people to get an autism diagnosis now than it was 50 years ago, so even severe cases might not have shown up in the stats if the child was born to a poor family.

I meant it could explain some of it, but not the full 16x times.

I am talking under the presumption that researchers today are aware are correcting for that fact that there is some selection bias (if that's what I can called) in stats.

The HN headline should probably have a (2007) indication appended.