Science Fair: The Makings and Consequences of Autism
An occasional column about science news pertaining to autism
Compiled by Robert B. Waltz
Three years ago, the Diagnostic and Statistical Manual folded together "autism" and "Asperger's Syndrome" to create "autism spectrum disorder." We know that, diagnostically, the change was appropriate, but there was a question about whether it was socially wise -- that the prejudice against "autism" might be stronger than that against "Asperger's Syndrome," leading to even more discrimination against those on the spectrum. A recent study seems to indicate that this is not so, although the presence or absence of a label makes a difference in treatment.
Studies have shown that autism is strongly associated with genetic factors -- in at least 80% of cases, the victims of autism have hereditary risk factors. But genes aren't everything -- the very twin studies that show that autism is largely due to genes show that it isn't entirely due to genes. That leaves environmental factors. We don't have much to go on in that regard, but if you are interested, here is a good Slate article on what is(n't) known.
Commentary: A few random thoughts on this: First, regarding the question of whether it is bacterial or viral infections which increase the rate of autism, my guess is that inflammation plays a role somehow -- it's the mother's body's response, not the particular pathogen that matters. Second, based on the one pair of identical twins I've known (where the left-hander is much more clearly autistic than the right-hander), I suspect hormonal gradients (which contribute to handedness in twins) play a role. Third, I have observed that a particular sub-population of people with autistic traits (a sub-group that I recognize although it's hard to explain) consists entirely of people born between August 14 and January 6 -- in different years, but in those months. The odds against that being coincidence, given the sample size, are on the order of 500:1 against. We also know that schizophrenia is seasonal -- and a different season; the "autism season" (which has been found in other studies) ends about the time the "schizophrenia season" begins. This might tie in with infection rates, e.g., but the point is, there are lots and lots of environmental influences to investigate.
Oxytocin to the Rescue?
Since autism is considered a disorder of social interaction, it is probably little wonder that oxytocin, the so-called "social hormone" (or "trust hormone" or "love hormone") is often suggested as a potential treatment. Many studies have investigated its use, with mixed results. Now comes one that suggests that, in children less than eight, it has definite benefits.
And here is a story about how oxytocin might do what it does -- and a medical gimmick to take advantage, but also with a warning.
Commentary: As usual, there are some cautions about the study in the first article: it is just barely large enough to be statistically meaningful (meaning that the percentages of response could still be pretty far off), and there was no control group. This one in fact should properly have two control groups -- one of people with autism who weren't given oxytocin and one of neurotypicals, ideally of neurotypicals who were given oxytocin, although that has ethical problems. Nor does there seem to have been an objective rating method. But these are, frankly, nitpicks. My real concern is the long-term effects. Long-term as in, "what happens when these oxytocin-using kids reach puberty?" A classic mistake made by those with autism is the inability to tell a friend from a boy/girlfriend, and now you're giving them the strongest bonding hormone we know of? Just being better at bonding doesn't make you better at knowing when you've made a big fat fool of yourself! And indications are that oxytocin makes break-ups even more painful than they would otherwise be. Plus there are indications that, as a feel-good chemical, oxytocin is addictive. I would not be at all surprised if this research holds up -- I thought, long before this came out, that many of my problems are related to oxytocin metabolism defects. But I would strongly urge parents not to try this at home until a lot more is known. What works in five-year-olds could well be a disaster in fifteen-year-olds.
A study in the respected journal Pediatrics finds higher rates of obesity in children with autism.
Commentary: The data in this article isn't very complete. The difference in rate of being overweight is relatively small; without knowing the sample size, we can't tell if it's statistically significant. But I suspect it is -- and I suspect it has at least two causes. One is related to sensory sensations and routine: A kid who eats only pasta or Frosted Flakes with strawberry milk is going to get fat. The other has to do with exercise. A baby who doesn't understand how to crawl won't get much exercise; a pre-teen who gets teased a lot will avoid others and not engage in playground games; a teenager with lousy coordination will get less playing time in school sports. We already have therapies for the eating problems. But perhaps it's time to pay more attention to developing ways for children with autism to get exercise that are interesting and don't cause them to have social problems.
A study out of Cambridge University confirms that scientists tend to have more autistic traits than "the rest of us."
Commentary: The basic assertion here -- that scientists have a lot of autistic traits -- is not new; this is confirmation, not new work. (Which doesn't mean doing the study is bad; confirmation is a key element of science!) My concerns are, first, that the paper supports a false impression of autism -- although almost all scientists have a lot of autistic traits, by no means are all people with autism scientifically inclined. Second, and even more worrisome, is the dependence on Simon Baron-Cohen's work. He invented the Autism Quotient, which is fine, but he also invented the "Extreme Male Brain" hypothesis. There is too much risk of confirmation bias here -- of course Baron-Cohen's test will detect the traits he considers core to autism, which are not the same as those which form the official definition. Speaking personally, I have been fortunate enough to know many women who are brilliant at scientific work, and who have significant autistic traits but without showing any signs at all of "extreme male brain" or excessive testosterone (e.g. they aren't particularly aggressive or competitive). I really wish the "autism linked with science" and "autism linked with male-ness" hypotheses were un-linked, because the only link between them is Baron-Cohen's hypothesis.... The last thing we need is one more excuse for keeping women out of the sciences!
If it matters, my Autism Quotient score (I took the test twice a year or so apart) is 36/50. A score of 32 or more is strongly indicative of autism -- but then, I'm autistic and trained in the sciences. So why do I score high? Autism or science or...?
Incidentally, not all personality tests show those in the sciences as being very different. A recent study of computer science types (also a highly autistic group) showed that their scores on the "five factor model" that is becoming a psychological standard reveals them to be more introverted than the general population but not as showing any other particular psychological peculiarities.
Author bio: AuSM member Robert B. Waltz was diagnosed with autism in 2012. He earned his B.A. in physics and mathematics from Hamline University in 1985. He is the author of three books on folklore, the editor of the online folk music database The Traditional Ballad Index, and recently has been informally studying the biology of autism.