Slashdot Mirror
TV Really Might Cause Autism
Alien54 writes "Cornell University researchers are reporting what appears to be a statistically significant relationship between autism rates and television watching by children under the age of 3. The researchers studied autism incidence in California, Oregon, Pennsylvania, and Washington state. They found that as cable television became common in California and Pennsylvania beginning around 1980, childhood autism rose more in the counties that had cable than in the counties that did not. They further found that in all the Western states, the more time toddlers spent in front of the television, the more likely they were to exhibit symptoms of autism disorders. The Cornell study represents a potential bombshell in the autism debate."
I'd interpret it in the reverse way. That is, the study shows just that children born with autism are more likely to spend time watching TV (knowing the features of autism, this is entirely possible).
And your explanation for why more children being born with autism increases the incidence of cable?
A statistically significant correlation between cable TV rollout and autism could mean one of four things: -
1) Cable TV causes autism. A feasible, if unsupported, hypothesis.
2) Autism creates a demand for cable TV. No obvious mechanism, so not very likely.
3) A third factor links both. I.e. increased adult exposure to TV leads to a greater awareness of autistic spectrum disorders, causing a greater proportion of children with ASD to be recognised as such (this is merely an example of a third factor).
4) The correlation is a product of chance. Statistically significant implies the probability of the correlation being by chance alone is less than, variously, 1 in 20 or 1 in 100. This does not mean it is not a chance event, merely that it is an unlikely one.
No reasonable conclusions can be drawn from this single study. This is not to say it should be ignored, and certainly further studies should be conducted (preferably by pediatricians this time), but this is no justification for some massive brou-ha-ha about brain-rotting TV and the delinquent parents who sit their children in front of it.
Probability of a journalist thinking the above? About 0.0, I'd say.
The MMR vaccine:
http://www.badscience.net/?p=249
Correlation is need to proove causation. But by itself alone it doesn't implie it. Other finding are necessary like models that explains WHY such causality should be expected (What's the biochemistry involved ?). Or see Koch's postulate : expremients that proove that by adding/removing the candidate cause ou can somewhat control the effect.
But on the other hand some mecanism is partially known. Also there are finding pointing to the fact that autism is associated and may be caused by some abnormal brain wiring that already happen in utero, thus debunking the old "it's-the-mother's-relationship-fault" supposed cause. Also if it start that early in the developpement, later exposure to the TV is less likely to be the main causing factor.
And in this case I think it's clearly a case of pure correlation that depend on an additionnal common cause. From what I've understood during my studies (got a degree in medecine) and what I've observed (one of my brothers has autism) : one of the caracteristics of an autist is being much less capable to anticipate or to cope with complex not easily predicted event. In this case the TV is reassuringly predictible : once turned it just plays the show. No complex social interraction required. And also, if wired to a VCR or DVD player, the TV can always play that specific shows that the autist knows and can correctly anticipate, etc...
TV isn't a cause *of* autism. But, the cognitive mecanism that are specific of autism, also happen to find the TV very reassuring.
"Sufficiently advanced satire is indistinguishable from reality." - [Tips: 1DrYakQDKCQ6y52z6QbnkxHXAocMZJE61o ]
Sounds like perhaps you don't have ADD if the drugs didn't work for you. Your sole experience is not scientific proof. I know plenty of people that are helped by the medication (I don't know any kids, just adults). And besides, the meds aren't going to finish your work for you, they're going to help with the mental block that's preventing you from doing so.
And here is another one I want everyone to repeat:
'MIGHT' DOES NOT EQUAL 'DOES'!
'MIGHT' DOES NOT EQUAL 'DOES'!
'MIGHT' DOES NOT EQUAL 'DOES'!
The title of this article is "TV Really Might Cause Autism". Who the hell is reporting that it is the cause? Who is claiming that this statistically significant relationship is equal to a 'randomized double-blind study'? And how exactly do you propose one do a 'randomized double-blind study' on autism anyways? The problem isn't that studies like this that present statistical relationships make the news, it is that people for some reason fail to read words like "might", "may", or "possibly" and instead read a certainity which no one claimed existed.
Mathematics is made of 50 percent formulas, 50 percent proofs, and 50 percent imagination.
Autism is not a binary condition. There is a range of disorders called Autism Spectrum Disorders that range from Asperger's Syndrome to the low-functioning Autism that you cover.
Each of these manifests itself differently. Having an Autism Spectrum Disorder implies some level of impairment in social functioning, but where you are on the spectrum indicates how much desire you have to engage in social interactions at all.
I think the greater knowledge of the spectrum disorders in the medical community has led to the greater incidence of diagnosis. It may also be the case that the incidence of autism is increasing, though I would be surprised if TV was the cause. Simon Baron Cohen (he's Sascha's cousin) has theorized that Autism represents an extremely systematizing mind, and that kind of mind (in my experience) tends to shy away from the disordered chaos of TV and towards systems and obsessions (trains, cars, computers, etc.)
If a diagnosis is provided with the intent of helping the child learn how to cope with social situations better, I don't see anything wrong with it. While I am opposed to the search for a "cure" for a spectrum disorder that also includes a number of functioning, intelligent, and unique individuals, I feel that social interaction strategies can be learned and applied to help children with autism learn to communicate and function better. If our goal is to help each child (or adult) to learn how to accomplish the things they want to do, then I don't see how greater awareness is a problem, even if there is some amount of bandwagoning that goes on.
"I disagree. TV (television) is based on broadcasting, watching DVDs etc isn't."
/.ers who cared to read the pdf in the included link of the story saw that the authors included DVDs as part of TV watching (even if it is or isn't, or is or isn't actually studied properly in their correlation study).
And those
The problem I am having with the way things are going, is that Autism is suggested to be a lifetime handicap, while the different kinds of people that are labeled "autistic" these days may well have a simple development delay -- which can be treated.
I, myself, am known to panic under stress, resulting in somewhat manic-depressive behaviour (I get intensely absorbed in whatever it is I am doing, or I show signs of depression) combined with tension problems. People are trying to see if the label fits me as we speak.
But whenever I go to the library to get information on Autism, I recognize so little about myself there. For instance, I am not rigid at all; I dislike talking about exact topics; I have a vivid imagination.
There may well be an interesting theoretical causality, but it doesn't help me much. I haven't had any advice or help whatsoever from the shrinks -- only tests, tests, tests, and more to come (it takes a very long time to check for autism) -- and the books on the subject don't help me manage myself.
Instead, I have learned a lot from reading about bipolar II, and managed, with much effort, to stabilize my mood so that I can think clearly again.
I insisted on intake that I learned about the underlying cause of my problems, so that I could learn to manage them, so you may say that I have caused my own problem. What frustrates me more, however, is that I have also had to solve it myself with so little help from the experts.
"We can confirm that Debian does *not* ship the version with the trojan horse. Our version predates it." [CA-2002-28]
I don't really think the argument is as much about semantics as you think it is. But you are correct that it comes down to how you define it. Personally I think the act of "watching TV" under debate here is the act of sitting down and watching cable live for an extended period of time. Even if you're sitting down at a particular time to watch a particular show you like.
There is something fundamentally different about the experience when the network dictates when you watch, and dictate what you watch during the commercial breaks. How often do people hang around after the show just because? How many people have a 20 to 30 minutes to kill before the show so they turn the TV on early and watch whatever is on in the mean time?
I think a majority of people who "watch TV" basically just sit down and accept whatever is on. It's a distinctly different activity then pre-recording a show using a DVR or VCR, or watching a file on your PC or on DVD. Some TVs show do generally have worthwhile entertainment that can spark intellectual discussions with friends and family, perhaps even some introspect. But viewing it on the networks schedule you map out a larger portion of your time, you're subject to the massive amount of FUD in advertising (which IMO is where most of the real evil resides), and you're often tempted into watching just another half hour, just another half hour, just another half hour.
When people say "watching TV" they're typically referring to "watching network TV" Anything else is just video entertainment... but that's another argument altogether.
Basically when watching it "live" as you say then you're molding your life around the TV show, as opposed to putting the TV show where it best fits into your life. It fundamentally changes the role TV plays in your life reguardless of what shows you're watching.
Collector's Edition
They take too strong a stance based on what they actually found. The title, of course, is sensational, and in the abstract they claim they've found "strong support" for the idea that TV can trigger autism, which I think is sort of a pointless claim to make until you can compare at risk groups (based on genetic factors, e.g., close relative with autism) to see if more of the TV watchers flip than the non-watchers. But they do specify controlling for increased diagnosis by using time series data, as well as comparing California data to the overall US DOE data to show that autism diagnosis was increasing even before expanded surveillance definitions were employed. They also rely on ATUS data to show that precipitation DOES correlate with more TV watching in youngsters, so to say that their proxy is unvalidated is a bit inaccurate.
In short, it's not the bombshell the media will hail it to be, but neither would it seem to be the flaming heap you say.
They are health economists (well, one of them is, and the other is an economist). Health economists do these kind of large scale public health studies all the time. They are using statistics and economics to investigate issues of public health concern, not looking for medical explanations of causes.
If you look at the paper itself, the title may be provocative, but what they are actually doing is what health economists generally do in their research.
Also - the way they are doing their analysis seems (to me at least) to address the question of correlation vs. causation fairly well, because they are breaking down data county by county and region by region, looking at other intermediating variables and so on. See pages 37-39 where this is addressed fairly well.
You may take issue with their conclusions or their methodology, or the fact that they are examining a hypothesis that is not widely addressed in the medical community about a disease, but don't shit all over it just because they aren't doctors. Sometimes, from the micro perspective of a doctor, it may not be obvious what is causing a disease, whereas somebody trained in statistics who plays with some numbers from the NBER could find out something quite interesting.
...and here we have another M.D. who thinks he knows something about science. I wish medical schools would concentrate less on memorization and more on critical thinking skills, especially with respect to statistical studies.
This is a spectacularly good example of really stupid statistical games.
In actuality, the paper is a good example of the way in which social research can take advantage of natural experiments.
I only skimmed it...
Then why write with such unwarranted authority and in such certain terms about its contents and conclusions?
Aside from the basic methodologic errors (confusing correlation with causation, adopting a highly questionable proxy indicator without validating it, and spending almost no time ruling out confounding factors or tainted data), there remain the dozens of smaller tactical problems...
They made none of the errors you list. I would like to think you might have realized this had you bothered (as I did) to actually read the paper, but based on the evidence of your post, I would be reckless to assume that.
Er, read first, then complain if warranted. The abstract doesn't make it clear, but the data does indeed suggest that they've found a trigger for autism. Very interesting.
Correllation does not equate causation. That's all that needs to be said about this experiment.
And actually, they only found a correlation between precipitation and autism, and cite another study that shows a correlation between precipitation and tv watching.
;)
It's a very tenuous "study". Its basically a 67-page explanation of a hypothesis that someone could do an actual experiment on.
I haven't gotten to reading the data yet to see if they even support their insistence that there is a correlation.
For a great illustration of "correlation isn't causation", I recommend people check out the graph of number of pirates vs. global temperature over at the Flying Spaghetti Monster's page.