January 31, 2006
Assortative Mating By Systematizers Raising Incidence Of Autism?

Logical organized minds with a knack for pattern recognition preferentially marrying each other in a process called "assortative mating" may be the cause of a rise in the incidence of autism.

Highly analytical couples, such as scientists, may be more likely to produce children with autism, an expert has argued.

Professor Simon Baron-Cohen, of the University of Cambridge, said the phenomenon might help explain the recent rise in diagnoses.

He believes the genes which make some analytical may also impair their social and communication skills.

Perhaps male scientists and computer programmers ought to marry literature professors, art professors, and lawyers? Organized and disorganized people should hook up? Systematic types should mate with chaotic types? Not sure what this would do the divorce rate. But it might lower the incidence of autism. Though it might come at the expense of lowering the frequency people born with the ability to be extremely talented scientists and engineers. Or maybe it would create managers better balanced between having technical and communications skills and therefore make industry more productive over all.

Parents of autistic kids tend to have autistic tendencies themselves.

According to a survey of 1,000 members of the National Autistic Society, fathers and grandfathers of children with autistic spectrum conditions are twice as likely to work in a systemizing profession such as engineering.

Students in the natural sciences have a higher number of relatives with autism than do students in the humanities, and mathematicians have a higher rate of autistic spectrum conditions compared with the general population.

The theory that both parents of children with autism are strong systemizers is also evident from a study that shows both mothers and fathers score above average on a questionnaire that measures autistic traits.

The idea here is that people who hate to socialize who marry like minds are at risk of having kids who get even stronger doses of genes that push them in that direction with the extreme being autism.

Baron-Cohen is not the first person I've seen put forward this idea. Some argue that Silicon Valley has a disproportionate incidence of autism and that this is happening because male and female techies are hooking up there in workplaces and having babies.

IF this theory is correct (and it seems very plausible) then it adds further evidence to the argument that technology is causing big changes in selective pressures on genes which code for cognitive function. Evolution did not stop tens of thousands of years ago. Selective pressures can cause changes in mating practices that cause changes in distributions of genetic variations for brain genes and this can happen in a relatively short period of time.

Share |      Randall Parker, 2006 January 31 09:22 PM  Brain Evolution

Jim said at February 1, 2006 8:26 AM:

the assertion in your post sure makes intuitive sense to me! phd engineering students have a lot of borderline autistic characteristics. fortunately, many of the guys (and yeah, they're mostly guys) really hit it off when paired with the right 'verbal-type' women. very complimentary.

Jake said at February 1, 2006 8:40 AM:

There was a very good article on this subject in Wired a few years ago. It said that autistic type people had trouble earning a living years back. They became hermits or worked in universities. Most of these men and women never married.

Then computers arrived. Now these people could earn big money programming and doing other computer work. With that money, they became more attractive to the opposite sex. More important it threw autistic type men and women together in the same office and propinquity did its magic. Ordinarily, autistic type men and women would never marry each other because they had no way to meet.

Randall Parker said at February 1, 2006 5:14 PM:


Yes, I believe all that has happened. Soon I expect genetic counseling on autism genes for potential mates. People are going to want to get optimal balances of analytical and social skills genes.

Penny said at February 1, 2006 5:15 PM:

Well, this would certainly explain my family. Father: scientist with strong asperger's traits, mother: historian and very analytical, 6 kids, 2 diagnosed aspies, 2 (IMO) undiagnosed aspies, one who is, erm, mentally unwell, and one normal one who thinks we are all anti-social freaks. Resulting grandkids (of my parents): of the 5 who have reached adulthood, 3 are engineers. Maybe I'll start a (secret) family study, testing the grandkids and what not. Might be interesting to someone down the line.

Patrick said at February 2, 2006 3:26 AM:

Add another factor: the internet, dating sites and email allow people who can barely talk to the other sex a chance to meet.

I speak as someone who discovered internet dating after finishing my PhD in Engineering, and is now married to an accountant I met online.

Mark Plus said at February 2, 2006 9:40 PM:

Just speculating: The current scientific recommendations for building up a "cogntive reserve" to protect the brain against the effects of aging -- working challenging puzzles like Sudoku, reading difficult books, studying foreign languages and so forth -- resemble what people with autistic neurologies tend to do without prompting. Do we see some selection pressure in favor of autism because people need to keep their brains functioning over longer life spans than their ancestors could expect? Or do autistic people maintain their cognitive abilities better than other people at comparable ages?

Rhonda said at February 6, 2006 3:50 PM:

The theory of "assortive mating" does not explain an increase in the rate of autism diagnosis. If assortive mating is a cause of autism what is driving the increase rate of autism? An increase in assortive mating?

Bob Badour said at February 6, 2006 4:56 PM:


Microsoft is increasing assortative mating. Actually, Microsoft, Oracle, Google and all the other employers of high-tech geeks that either congregate them in large campuses or near other employers of high-tech geeks.

Randall Parker said at February 6, 2006 5:22 PM:


Yes, an increase in assortative mating is theorized to be driving an increase in autism. The increase in assortative mating is being driven by high tech companies bringing like minds together.

Allison said at February 6, 2006 8:30 PM:

Are you kidding? You think there are a lot of women engineers working at MS, Oracle, Intel, or Google? Uh, no. There aren't. There are few women scientists. There are few women engineers, even in the bay area. There are virtually no WHITE women engineers and scientists in the bay area; there are significantly more Indian and Northern Asian women. Are the incidence of autism up in those communities?

I find it hard to believe that "assortative mating" is increasing. Find this hidden cache of highly analytical women and then show me where IT'S increasing, and then perhaps we'll talk.

some data:

"Women make up 46 percent of the U.S. work force and 48.6 percent of the college-degreed work force, but only 9.48 percent of the work force in science and engineering, according to the National Science Foundation...Agilent says 19 percent of its engineers are women. "
from:http://sanjose.bizjournals.com/bizwomen/sanjose/content/story.html?id=898710 (date Feb 12 2004)

"About 35 percent of workers at HP worldwide are female. " from
--given how many must be in HR, Finance, Sales, Admin, how many do you think that leaves in the sciences/engineering?

"And while 20% of engineering degrees are earned by women in the U.S., only 10.6% of American engineers are women2. Further, the percentage of women graduating with computer science degrees has decreased 25% since 1983."

One in ten is about right. Anecdotally, I've never known a white women scientist or engineer other than myself at any of the SiliValley companies I worked in or my husband worked in, and each had 100-200 people in them. The total number of women engineers/scientists in those places was less than a dozen, give or take. I doubt you can find any way to show statistical significance of more "analytic couples" existing now than ever before, whatever that would mean.

Randall Parker said at February 6, 2006 9:38 PM:


In Santa Barbara I work with several white and Indian women engineers and computer programmers. The married ones are all married to engineers and computer programmers. Most met each other either on the job or in engineering and programming classes.

Your 10% figure misses a basic point: Autism is rare. Even if most women in Silicon Valley aren't engineers, scientists, or computer programmers those women who are in those occupations are far more likely to be married to men who are as compared to women with the same native abilities would have been 50 years ago.

Natural selection works at the margin. To substantially increase the rate of autism it is not required that every carrier of genes that cause a tendency toward autism marry another person who also carries those genes. All that is required is that more marry fellow carriers than was the case previously.

Allison said at February 7, 2006 5:52 AM:

---Even if most women in Silicon Valley aren't engineers, scientists, or computer programmers those women who are in those occupations are far more likely to be married to men who are as compared to women with the same native abilities would have been 50 years ago.

you have no data to support this whatsoever. You don't know that the same analytic women weren't married to analytics in the past--why wouldn't they have been? The percentage of analytic women is small in the first place. Are you suggesting that they were old maids, or just married jocks, so to speak? In all likelihood, there aren't more women at the margin. You are suggesting that a) there are more analytic women that there were 50 years ago, and b) they couldn't find analytic men. Both of these statements are without support.

It's possible that more male analytic types are creating offspring on the margin, now, as more analytic men are able to find women AT ALL, but you have no reason to believe either a) or b) above.

Kathleen Fasanella said at February 7, 2006 10:41 AM:

I'd have to quarrel with the idea that it is strictly "analytical" women as defined by profession. Until recently, women weren't encouraged to enter those fields so the numbers of "analytical" women are artificially low. Why not judge women by interests and hobbies? I've found a striking number of women who sew -who have autism traits. Sewing is very technical; it's mechanical.

And besides, most of these new cases of autism are on the higher functioning end of the spectrum. As someone who is intimately familiar with autism, I disagree with the implication that the increase is negative. Who else will have the intellectual capacity and the focused narrow interests to pursue the solutions to the myriad of problems societies face? People seem to conveniently forget the giftedness inherent to autism, as well as the existing -unlabeled or undiagnosed- autistics who've made a dramatic impact in the quality of all our lives. For example, are there *any* nobel prize winners who weren't/aren't autistic? Even among peace price winners, there's some debate. Consider Ghandi for one.

Bob Badour said at February 7, 2006 10:29 PM:


When I was going through highschool in the early 1980's, it was politically correct to encourage women to enter mathematics and science professions. Everything imaginable was done to encourage them.

Women accounted for no more that 10% of my electrical engineering class. At the same time, men accounted for no more than 10% of the pharmacy class. This 'recently' you talk about when women were discouraged from engineering professions was well over half a career ago and many of the people I went to school with have kids in their teens.

I suspect without the strong encouragement to enter engineering, women would have accounted for 0% of the class.

Kathleen and Allison,

What are your hypotheses that account for the increase in autism and the clustering of cases in silicon valley?


You ask: "Why wouldn't analytic women have married analytic men in the past?" Because analytic men are rare and because individual men are less important reproductively than individual women. Further, analytic women are likely very analytic regarding mate choice and would likely choose a mate able to support her and her children. I suspect geeky analytic men had less worth in past economies.

However, silicon valley changed a lot of that. Analytic men are much less rare there and have significant affluence.

Suppose 2% of the male population are geeky analytic types. The sex ratio in engineering and computer science suggests only 0.2% of the female population would be geeky analytic types. Thus without assortative mating, 1 in 25,000 marriages would combine geeky analytic types. While I just ballparked those numbers, they seem reasonable given entry requirements for engineering undergraduate programs.

Suppose in silicon valley the clustering of geeky analytic types means 5% of the male population are geeky analytic types and 0.5% of the female population are geeky analytic types. Without any additional assortative mating, 1 in 4,000 marriages would combine geeky analytic types with a potential 6-fold increase in autism.

However, the general population in Silicon Valley does not represent the work environment. At some high tech companies, 50% of the men and perhaps 5% of the women are geeky analytic types. Among those who meet their spouse at work, 1 in 40 marriages would combine geeky analytic types with a 600-fold increase in autism.

Note that this achieves a 600-fold increase even before accounting for any increase in geeky male reproductive fitness. And while my numbers might not match the actual numbers, I suggest the actual numbers will account for a significant increase and clustering of autism in Silicon Valley. And very likely Research Triangle Park and Redmond too.

Kathy Howell said at April 29, 2009 5:42 PM:

As a Speech Language Patholgist working predominantly with newly diagnosed children with autism and their families, I have definitely noticed a significant majority of parents ( mainly fathers) working in the computer/high tech industry. many of these parents have aspergers-like traits. Also, the vast majority of my clients ( 75%) are either East Indian or southeast asian demographically. I have often wondered at the possible significance of these uneven demographics. This explanantion by Baron-Cohen seems very plausible and something I can attest to.

almost anonymous said at January 4, 2011 6:08 PM:

Increasing the incidence of Autism is a good thing. Suspected Autistics have been at the top of their fields in science, the arts, and engineering for as far back as recorded history goes. Autistic people are also kinder than NT. The only thing wrong with autistic people are the prejudices NT have against them, and no one has yet suggested resolving bigotry by getting rid of the targets of bigotry before and not been laughed at.

jennifer edwards said at April 19, 2011 2:17 PM:

Has anyone looked into this being linked to computer use - because while theories of genetic links in tech-laden areas apply to silicon valley, it wouldn't necessarily account for the widespread rise of incidence of this disorder globally (I'm sorry, but as an auntie of a child with autism, i do not believe it is just bigotry that is a problem here, it cuts to the very nature of human experience when one can't empathize with another (not to mention a very challenging anxious way to experience life) . .. .where as computer usage (waves emitted?) might -

CureisComing said at June 29, 2011 4:52 AM:

Let people marry who they want.

Maybe there's a reason rates of autism are going up.

Humanity's natural direction is to become more and more analytical. Scientists have estimated that the average IQs of people even as recently as the middle ages would have been in the mentally retarded range if they were alive today.

Paradoxically the more humans gain the ability to analyze the less use people have for intuition. The predictable end result is this:

Once humanity has completely decoded social interaction in all its intricacy and can print it in books autistics will no longer have significant trouble with social interaction. Eventually science can be able to teach us how to socially observe with analysis to the point of getting everyone better than what average is now. Some point in the future every human child will just be taught social skills because our knowledge of social interaction will exceed even the most NT person's ability to intuit social knowledge. On top of that corrupt people will have less chance of taking control in a society where the entire social science and all interaction within it is completely understood by all.

So the cure is coming, it's not a pill, it's not a surgery, it's not genetic engineering, it's information. And the best thing is it will leave all the analytical ability intact(whereas any chemical solution is likely to be aimed at 'normal' even with sacrificing analytical ability). And as we analyze how people think we'll even be able to enhance everyone's analytical abilities. So someday we'll be able to teach NTs autistic stuff and autistic people NT stuff just as easily, because science will have decoded how our minds work in that much detail.

Marcy said at December 22, 2011 2:12 PM:

"Perhaps male scientists and computer programmers ought to marry literature professors, art professors, and lawyers?"
WHY do you ONLY mention MALE scientists and computer programmers?
Why add the word MALE in this sentence?
Why not make it, FEMALE scientists and computer programmers?

A bit gendered don't you think?

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright