Don't miss the end (pp. 41-43), where he calculates that the net economic benefit of mating between "systematizers" as very positive, modeling the resulting higher wages compared with the cost of autism (as 3-SD systematizing behavior) and Asperger's syndrome (as a 2.5-sigma trait).
In this model, net economic result is still very positive, even if the lifetime economic cost for a case of autism is $5 million, and the presence of Asperger's cancels out any benefit in earning (not likely!)
This is a nice result for the ever-growing industry of online and offline forums for assortative mating, particularly in this field of interest.
So here's the upshot for economic development organizations: support your local geek conventions!
There's an interesting literature on the relationships between IQ and economics on both individual and national levels (I've compiled some links in http://lesswrong.com/lw/7e1/rationality_quotes_september_201... ), but one of the key factors making it difficult is that while it's easy to observe correlations between IQ measured at some point and later income, it's not easy to figure out the additional wealth created!
Much of IQ's economic benefits seem to be related to zero-sum or positional games (only a few thousand students can be let into Harvard each year). However, an additional marginal case of autism is not zero-sum where one case of autism saves you $5m on, I dunno, Alzheimer's somewhere. It just makes everyone $5m worse off.
So, it's possible that most of that 'net benefit' is only the 'net benefit' as considered for the specific person, and a net loss on a society-wide scale (because while that person may be internalizing the benefit of hundreds of thousands of dollars, they're not externalizing any of the cost by paying for that extra autistic kid a few houses over, whose treatment is going to be subsidized to some degree by other people as well).
Calculating the society-wide gains and losses is much harder, and it seems he didn't even try.
and most of the people he credits for helping him on the paper are other economists. He cites psychologist Simon Baron-Cohen a lot, but Baron-Cohen's view of autism is not yet the mainstream view among autism researchers.
Other economists are examining behavior genetics issues, and one group of economists has published a paper updating their fellow social scientists on issues of human behavorial genetics that are little understood among most social scientists.
Chabris, C. F., Hebert, B. M., Benjamin, D. J., Beauchamp, J., Cesarini, D., van der Loos, M., ... & Laibson, D. (2012). Most reported genetic associations with general intelligence are probably false positives. Psychological Science.
"At the time most of the results we attempted to replicate were obtained, candidate-gene studies of complex traits were commonplace in medical genetics research. Such studies are now rarely published in leading journals. Our results add IQ to the list of phenotypes that must be approached with great caution when considering published molecular genetic associations. In our view, excitement over the value of behavioral and molecular genetic studies in the social sciences should be tempered—as it has been in the medical sciences—by a recognition that, for complex phenotypes, individual common genetic variants of the sort assayed by SNP microarrays are likely to have very small effects.
"Associations of candidate genes with psychological traits and other traits studied in the social sciences should be viewed as tentative until they have been replicated in multiple large samples. Failing to exercise such caution may hamper scientific progress by allowing for the proliferation of potentially false results, which may then influence the research agendas of scientists who do not realize that the associations they take as a starting point for their efforts may not be real. And the dissemination of false results to the public may lead to incorrect perceptions about the state of knowledge in the field, especially knowledge concerning genetic variants that have been described as 'genes for' traits on the basis of unintentionally inflated estimates of effect size and statistical significance."
So my caution here would be that an unpublished working paper like this (a common form of preliminary sharing of speculative results in the discipline of economics) is very unlikely to be the definitive word on this interesting topic.
Interesting that a highly transient society has managed to apparently reduce genetic diversity to an observable degree in such a short span of time.
Having recently rewatched Gattaca and falling in the Asperger range of ASD, I can't help but wonder if studies of this nature will nudge us closer to passive forms of eugenics. We're certainly within the realm of feasibility at this point...
Relevant to this issue is that the more brain you devote to analytic thinking, the less you have for empathetic thinking, according to the findings described here: Empathy Represses Analytic Thought, and Vice Versa: Brain Physiology Limits Simultaneous Use of Both Networks http://www.sciencedaily.com/releases/2012/10/121030161416.ht...
He has limited his data to the USA; but should we not also see increases in Europe and UK as well, given the similar shifts in society and the manner in which people choose their spouses?
If there were other factors that were found in the USA but not found in other countries (whatever that may be, environmental or otherwise) then there would be statistically significant divergence.
Both my wife and I lean toward the academic. I went to MIT, and we have an ASD son. This research is fascinating. And I think the Time article on it (http://www.time.com/time/magazine/article/0,9171,2089358,00....) is a good starting place. A while back the "Einstein Syndrome" book did a light treatment on this topic but for speech delays. Ultimately, I think the whole concept of an autism spectrum is helpful and harmful, because severe kiddos are so different from high-functioning (e.g. nerds) ones.
I might be missing something, but doesn't assortative mating in itself guarantee that certain hereditable characteristics would be more likely to flourish?
Also, I thought it was funny that one of the researchers cited (Simon Baron-Cohen) is a cousin of the Baron-Cohen you've probably heard from before, especially since I've long sensed a whiff of spectrum disorder in the whole Ali G/Borat schtick.
I have an autistic son (and my second might have something similar but much milder) and I suppose I would meet the "systemitizing" criterion (my wife wouldn't, but that doesn't really argue against the conclusion of the paper). Of course, anecdote != data, etc.
"Correlation does not imply causation" is not a meaningful reaction to an article like this. Yet, unfortunately, that is the first response many will have to reading this.
This is more about a model than a study, and the model involves causation. Data fitting the model, therefore, implies causation! Not, of course, "implies" in the strict logical sense; it's possible that there is another model that lacks the causal link and also fits the data. But rather, "implies" in the sense that it's reasonably strong evidence - the more confident we are in the model, the more confident we are that there is causation involved.
No, but correlation correlates with causation because causation causes correlation. Correlation between A and doesn't tell you if its (A->B) or (B->A) or (C->A and C->B)[1]. And robust correlation pretty much does imply that causation is happening somewhere. So then you have to look at the plausibility of each of those to try to figure out which it is. And since (assortative mating) -> (reinforced traits) has a clear and reasonable mechanism it has a serious leg up here.
[1] Interestingly (A->C and B->C) doesn't cause correlation between A and B despite causing correlation between A and C and B and C. Which is how you can figure out causation from enough correlation datapoints.
Sometime in the future it will become clear that one of the causes of autism is traumatic experiences during labor and within the first few weeks of life. And the reason why boys are 4 times more likely than girls to be autistic is because of the trauma caused by circumcision.
I can't point at any scientific studies. This is purely speculation based on years of research and observation on my end. I don't have autistic children, but I know people who do.
Yeah, so, ideas someone just makes up one day, which directly contradict all existing scientific observations, tend not to become clear in a few years.
[+] [-] zach|13 years ago|reply
In this model, net economic result is still very positive, even if the lifetime economic cost for a case of autism is $5 million, and the presence of Asperger's cancels out any benefit in earning (not likely!)
This is a nice result for the ever-growing industry of online and offline forums for assortative mating, particularly in this field of interest.
So here's the upshot for economic development organizations: support your local geek conventions!
[+] [-] gwern|13 years ago|reply
There's an interesting literature on the relationships between IQ and economics on both individual and national levels (I've compiled some links in http://lesswrong.com/lw/7e1/rationality_quotes_september_201... ), but one of the key factors making it difficult is that while it's easy to observe correlations between IQ measured at some point and later income, it's not easy to figure out the additional wealth created!
Much of IQ's economic benefits seem to be related to zero-sum or positional games (only a few thousand students can be let into Harvard each year). However, an additional marginal case of autism is not zero-sum where one case of autism saves you $5m on, I dunno, Alzheimer's somewhere. It just makes everyone $5m worse off.
So, it's possible that most of that 'net benefit' is only the 'net benefit' as considered for the specific person, and a net loss on a society-wide scale (because while that person may be internalizing the benefit of hundreds of thousands of dollars, they're not externalizing any of the cost by paying for that extra autistic kid a few houses over, whose treatment is going to be subsidized to some degree by other people as well).
Calculating the society-wide gains and losses is much harder, and it seems he didn't even try.
[+] [-] tokenadult|13 years ago|reply
http://nymag.com/news/media/51015/index3.html
and most of the people he credits for helping him on the paper are other economists. He cites psychologist Simon Baron-Cohen a lot, but Baron-Cohen's view of autism is not yet the mainstream view among autism researchers.
Other economists are examining behavior genetics issues, and one group of economists has published a paper updating their fellow social scientists on issues of human behavorial genetics that are little understood among most social scientists.
Chabris, C. F., Hebert, B. M., Benjamin, D. J., Beauchamp, J., Cesarini, D., van der Loos, M., ... & Laibson, D. (2012). Most reported genetic associations with general intelligence are probably false positives. Psychological Science.
http://coglab.wjh.harvard.edu/~cfc/Chabris2012a-FalsePositiv...
"At the time most of the results we attempted to replicate were obtained, candidate-gene studies of complex traits were commonplace in medical genetics research. Such studies are now rarely published in leading journals. Our results add IQ to the list of phenotypes that must be approached with great caution when considering published molecular genetic associations. In our view, excitement over the value of behavioral and molecular genetic studies in the social sciences should be tempered—as it has been in the medical sciences—by a recognition that, for complex phenotypes, individual common genetic variants of the sort assayed by SNP microarrays are likely to have very small effects.
"Associations of candidate genes with psychological traits and other traits studied in the social sciences should be viewed as tentative until they have been replicated in multiple large samples. Failing to exercise such caution may hamper scientific progress by allowing for the proliferation of potentially false results, which may then influence the research agendas of scientists who do not realize that the associations they take as a starting point for their efforts may not be real. And the dissemination of false results to the public may lead to incorrect perceptions about the state of knowledge in the field, especially knowledge concerning genetic variants that have been described as 'genes for' traits on the basis of unintentionally inflated estimates of effect size and statistical significance."
So my caution here would be that an unpublished working paper like this (a common form of preliminary sharing of speculative results in the discipline of economics) is very unlikely to be the definitive word on this interesting topic.
[+] [-] ewheeler|13 years ago|reply
[+] [-] squarecat|13 years ago|reply
Having recently rewatched Gattaca and falling in the Asperger range of ASD, I can't help but wonder if studies of this nature will nudge us closer to passive forms of eugenics. We're certainly within the realm of feasibility at this point...
[+] [-] Symmetry|13 years ago|reply
[+] [-] felicopter|13 years ago|reply
[+] [-] patrickgzill|13 years ago|reply
If there were other factors that were found in the USA but not found in other countries (whatever that may be, environmental or otherwise) then there would be statistically significant divergence.
[+] [-] _hiss|13 years ago|reply
http://www.time.com/time/magazine/article/0,9171,2089358-3,0...
[+] [-] mistermcgruff|13 years ago|reply
[+] [-] aes256|13 years ago|reply
Teehee. Interesting paper.
[+] [-] hanleybrand|13 years ago|reply
Also, I thought it was funny that one of the researchers cited (Simon Baron-Cohen) is a cousin of the Baron-Cohen you've probably heard from before, especially since I've long sensed a whiff of spectrum disorder in the whole Ali G/Borat schtick.
[+] [-] mpyne|13 years ago|reply
I have an autistic son (and my second might have something similar but much milder) and I suppose I would meet the "systemitizing" criterion (my wife wouldn't, but that doesn't really argue against the conclusion of the paper). Of course, anecdote != data, etc.
[+] [-] jrl|13 years ago|reply
[+] [-] brador|13 years ago|reply
[+] [-] dllthomas|13 years ago|reply
This is more about a model than a study, and the model involves causation. Data fitting the model, therefore, implies causation! Not, of course, "implies" in the strict logical sense; it's possible that there is another model that lacks the causal link and also fits the data. But rather, "implies" in the sense that it's reasonably strong evidence - the more confident we are in the model, the more confident we are that there is causation involved.
[+] [-] Symmetry|13 years ago|reply
[1] Interestingly (A->C and B->C) doesn't cause correlation between A and B despite causing correlation between A and C and B and C. Which is how you can figure out causation from enough correlation datapoints.
[+] [-] herbivore|13 years ago|reply
I can't point at any scientific studies. This is purely speculation based on years of research and observation on my end. I don't have autistic children, but I know people who do.
[+] [-] prodigal_erik|13 years ago|reply
[+] [-] Evbn|13 years ago|reply