Jul 16 2008

More Evidence that Autism is Genetic

Published by under Uncategorized
Comments: 17

Much of the controversy surrounding autism in recent years is based upon the premise that there is an autism epidemic – that autism rates are climbing. However, the evidence strongly suggests that this is not the case and that autism diagnoses are increasing through a combination of broadened diagnosis and increased surveillance and awareness. However, proponents of various discredited theories, such as those anti-vaccinationists who claim there is a link between vaccines and autism (there isn’t) want to believe there is a true epidemic because that in turn implies an environment factor (like the vaccines they despise).

But the environmental hypothesis of autism, while impossible to completely eliminate, has not been fairing well. On the other hand, the theory that autism is largely (if not entirely) genetic has been very fruitful. Researchers are finding more and more genes that correlate strongly with autism. Also, studies have shown that the risk of autism rises with paternal age – suggesting that aging sperm may play a role. It is also true that as our ability to diagnoses autism more reliably, and based upon earlier and more subtle signs, we are finding that infants show early signs of autism – before they receive most of their vaccines or environment has much of a chance to play a role.

Essentially, multiple independent lines of evidence are converging on the conclusion that autism is dominantly genetic. Well, now there is yet another line of evidence supporting this conclusion

Researchers at Caltech and UNC studied the manner in which those with autism and those without pay attention to the faces of other people and how they discern emotion. Non-autistics generally look to the eyes, while autistics prefer to focus on the mouth.  The researchers emphasize that one way is not necessarily better than the other or “normal” – they are just different.

Their hypothesis was that if autism is genetic than perhaps the parents of children with autism may also share this tendency to favor the mouth over the eyes.  From the press release:

The parents participated in an experiment that measured how they make use of the face to judge emotions. The subjects were shown images depicting facial expressions of emotion that were digitally filtered so that only certain regions of the face were discernible–the left eye, for example, or the mouth. The subjects were then asked to decide as quickly as possible if the emotion depicted was “happy” or “fear.” The part of the face shown, and the size of the revealed area, randomly varied from trial to trial.

An analysis of the subjects’ correct responses revealed that “aloof” parents relied much more heavily on the mouth to recognize emotion than they did on the eyes, as compared to nonaloof parents and, to a greater extent, to a group of parents of children without autism. Prior studies by Adolphs and his colleagues have shown that humans normally evaluate emotions by looking at the eyes–but studies by Adolphs and Piven have shown that individuals with autism do not.

Very interesting.  Perhaps, then, parents have some genes which predispose to autism and bestow some of its features. Of course, the alternate hypothesis is that aloof parents raise aloof children through their parenting style, rather than their genes. In order to eliminate this possibility follow up studies looking at adopted children or separated twins would be helpful. But this hypothesis – that aloof parents (once referred to as “refrigerator moms”) cause autism has already been rejected by autism researchers and is therefore an unlikely explanation. But confirmation would be reassuring, as always.

Promoters of the environmental hypothesis of autism don’t really have much to say about all the evidence for genetic causes. Often they say, correctly, that a genetic cause does not rule out an environment trigger. This is true enough – but the real question is how strong of a genetic influence is there. The evidence so far suggests that it is very strong, leaving little room for environmental factors. Also, no one has established that vaccines, toxins, or other specific environmental factors are truly correlated with autism. The science simply does not support those claims, as is evidenced by the revolving door of claimed environmental trigger. First it was MMR, then thimerosal, now a host of dubious toxins.

Meanwhile a genetic picture (although very complex – autism is not a single entity caused by a single gene) is slowly and consistently emerging.


17 responses so far

17 Responses to “More Evidence that Autism is Genetic”

  1. sjames1958on 16 Jul 2008 at 9:49 am

    “aloof” parents, can someone clue me into what this term means in this context.

  2. Niobeon 16 Jul 2008 at 10:55 am

    People who appear non-emotional, unaffected, detached.

  3. jonny_ehon 16 Jul 2008 at 11:34 am

    If I’m worried that I may be ‘aloof’ does that mean I’m not aloof, since aloof people don’t worry?

  4. fontinalison 16 Jul 2008 at 1:00 pm


    I’ve heard numerous times that improvements in surveillance and diagnoses likely account for the increased rates of autism, but wouldn’t the simple correlation with increasing paternal age suffice as an explanation almost on its own? If true, the simple societal pattern (relative to the past) of delaying the age of marriage, and thus pregnancy and child birth, would cause a corresponding increase in autism rates in any population exhibiting that behavior. Average paternal age likely varies regionally, so wouldn’t an examination of regional incidence rates offer an opportunity to answer the autism/genetic question definitively?

    Or am I missing something?

  5. BAon 16 Jul 2008 at 1:03 pm

    Don’t let the term aloof throw you. Such terms are merely definitions of differences that are borne out in certain behavioral contexts like facial/emotional recognition tasks. Such usage of terms by scientists (e.g., “aloof”, “autistics”) is due to a reliance on classification schemes as a means of gathering reliable evidence and then often using this evidence as an axiom and positing further hypotheses for investigation. When speaking to lay audiences I’ve found it helpful to place the person before the disability (e.g., persons with x) to avoid any possible offensiveness. This is not just political correctness but rather a humane means of talking about afflicted populations (many of whom may be in the audience).

    It is often stated (by the media, lay community and scientific community) that we know very little about the cause of autism. One of the things that is a potential by-product of this humble approach is that we are not putting out just how much we actually know about it. From the twin studies of the 70s through today, there has been no shortage of evidence that genetics underlies autism. It is getting to the point where specific genetic anomalies are being spoken of as accounting for a percentage of all ASDs. On the other hand, that genes are expressed in an environment (inside the skin) means that they are interacting with environmental variables. Granted, autism is most likely not at all associated with vaccination (and one could make a specious argument that thimerosal is associated with lower rates of autism but this is confounded with a general increasing trend in prevalence) but it is also associated with environmental variables like maternal drug use, smoking, disease, and other events. However, such environmental events are pre-natal.

  6. psamathoson 16 Jul 2008 at 4:08 pm

    “the risk of autism rises with paternal age”

    That’s interesting. So autism rates could be actually be increasing, if only because people are having children at a later age than they used to. Hardly something you could call an epidemic, though.

  7. Niels Kjaeron 16 Jul 2008 at 4:14 pm

    I agree to 99.9%!

    What if environment (not the physical but human) is getting more “autistic”? Would that in fact increase or decrease the number of humans being diagnosed as autistic? The answer is according to my opinion important and non-trivial. A more autistic environment would increase the awareness of autism, and at the same time decrease the impact autism has on the individual autist (I have been diagnosed as one).


    My “(C)V”:
    “As an Asperger, I have two special interests: Science and information theory.”
    “As a human, everything interests me.”
    “Please do not ignore the role of other “disorders” like AD(H)D”
    “Language is an important part of the problem and an equally important part of the solution”
    “I “gave up” learning languages and history at the age of 13. I “gave up” mathematics and politics at the age of 19. I “gave up” physics and data analysis at the age of 41. I “gave up” “giving up” at the age of 45″.
    “The uncertainty of a meaurement is as important as the measurement and reflects how the measurement can be used. The uncertainty on the uncertainty of a measurement reflects the maturity and importance of the measurement and is therefore as important as the measurement itself. The uncertainty on the uncertainty on the uncertainty of a measurement is most likely meaningless, but could reflect the arrogance related to the measurement”
    “A consistent theory of Type I, II_1 and II_infinity, and III_0 amenables in von Neumann algebra was completed in 1987 by the proof of Uffe Haagerup that unique type III_0 amenables exist and are stable in compact mathematical structures.”
    ” My rantings can be considered an empirical proof of why language is so important.”
    “In Denmark we normally have only one conspiracy theory: All conspiracy theories are created from other conspiracy theories”
    “The abundance of conspiracy theories reflects the maturity and importance of the society creating the conspiracy theory”
    “The abundance of theories about conspiracy theories is most likely meaningless, but could reflect something related to arrogance”

  8. Fred Cunninghamon 16 Jul 2008 at 5:21 pm

    There was a recent report on Medscape of a link between shampoos containing pyrethrins and autism. The overall effect should be small but should be looked into further.

  9. anniepemaon 16 Jul 2008 at 11:24 pm

    There is a very high average paternal age in the US and Great Britain and Europe. A rise in autism and childhood schizophrenia, which is now called autism is quite expected.

  10. Trying to Stay On Topicon 17 Jul 2008 at 5:25 am

    [...] amnesia) no matter how much scientific evidence (and there has been more recently, concurrent with more evidence that autism is genetic) arises that disputes a [...]

  11. woobegoneon 17 Jul 2008 at 6:01 am

    Hang on, is the association of autism with paternal age because paternal age causes autism (through genetic damage to sperm, presumably) or is it because male carriers of autism genes tend to have children later in life than non-carriers – maybe they’re more work-focussed and less interested in children?

    This is the kind of thing that’s almost impossible to prove, I guess.

  12. daedalus2uon 17 Jul 2008 at 11:30 am

    I think the “genetic damage” hypothesis of the association of advanced paternal age with autism is likely to be not corrrect. If it were correct, then that “genetic damage” would be heritable and cumulative even if the child did not express the autism phenotype. Even a small amount of “damage” per generation would quickly (in an evolutionary sense) become very large.

    Perhaps it is an epigenetic effect related to differential methylation of DNA due to paternal age. That methylation might affect the phenotype of the born individual, but the methylation would then be reset in that born individual’s gametes. Methylation is involved in Rett Syndrome.

  13. Matlatzincaon 19 Jul 2008 at 12:00 am

    I wonder if this study sorted the results out by cultural background. Apparently the Japanese, for example, focus much more on the eyes to discern emotions, whereas in the US we focus more on the mouth. This is why there are different emoticons for the same emotions in both countries.

  14. ebohlmanon 20 Jul 2008 at 2:34 pm

    woobegone: Your hypothesis would be pretty easy to test. If paternal autistic traits led to later childbearing, you’d expect to find a relationship between a father’s age at the birth of his first child and the prevalence of autism among any of his children. If increasing paternal age led to increasing autism prevalence in children, then you’d expect increasing prevalence among later children than earilier children.

    I do agree that it’s irritating to see journalists report a combination of observation and a speculation about that observation’s cause as if they were the same thing. If older fathers are more likely to have autistic kids, the speculation goes toward paternal sperm damage. If later-born sons are slightly more likely to be gay, the speculation goes toward some immune response of the mother to the fetus, even though there’s little difference between the two scenarios (and in the latter case, it may actually be a maternal genetic predisposition to both high fecundity and having gay sons).

  15. Niels Kjaeron 21 Jul 2008 at 2:45 am

    I just noticed that there are no posts on this blog about Oxytocin!

    Is this stuff too old news to be taken seriously by anybody? If yes, then anybody is probably more autistic than I am! The discovery of Oxytocin in 1953 gave a Nobel prize in Chemistry already in 1955, which does not prove anything, but it does indicate that there might be brilliant information to be re-understood.

    I myself just re-discovered the strange and wonderous but old work of Lars Onsager and applied it very succesfully to climate models solving the Navier-Stokes equations in an elegant but abstract way. The work of Lars Onsager was so brilliant that it was beyond the realm of Physics and Chemistry in 1933 and his Ph.D. thesis was rejected because it was said to be “incomplete”. Some mathematicians intervened saying that they would give Lars the Ph.D he deserved in mathematics. In 1935 Lars got his Ph.D. in chemistry.

    Two conclusions: Lars Onsager was probably one of the most autistic human beings ever who learned to speak english. Some students of his even doubted if Lars actually spoke a language that anybody could understand. Never ever ever ever ever reject a theory because it is “incomplete”, unless you understand why it is incomplete! Just state calmly that you don’t understand it.
    During my studies for my master degree in physics in 1988, I got rejected in the same way by a senior physicist when I had made a simple general analytical theory about the behaviour of the Kauffman automata and showed what could be proven and what could not be proven. This senior physicist is still working on the same things he was doing back in 1988. My own Ph.D. in physics was a unoriginal rehash of old ideas officially stolen from other people and was considered brilliant.
    As a physicist I recognize a local symmetry from the above considerations. And local symmetries are always related to locally conserved quantities. Is this the conservation of arrogance or the conservation of empathy I am describing?
    Or am I just rewriting one of the fairy tales of Hans Christian Andersen? If you are interested in Autism try to find out more about the life of Hans Christian Andersen.

    I wouldn’t dream of taking Oxytocin myself as a medicine, but I would seriously consider giving it to a 1 year old autistic child, if the child is desperately trying to learn human empathy/language.

    Any ideas on how to circumvent the blood-brain barrier? Nano-technology with time-released nano-tablets carried through the barrier by nano-carriers? Do we know animal “models” suited for such tests?

    Could we measure a quantitative effect by a applying a fMRI-DIS scan on the region where the mirror neurons are situated? It should be trivial to compare an on/off effect because the lifetime of Oxytocin is only supposed to be 3 minutes.

    Can we do other measurements on other mammals that show autistic traits? Oxytocin is produced in the posterior pituitary gland. Is this similiar to other mammals? Has it been proven that Oxytocin is only produced in mammals?

    Anyone with any information about the entropy map of Oxytocin? Can it be produced easily? What is known about its connections with cholesterol. Are oxytocin neurons related to mirror neurons? Has anyone tested the influence of Oxytocin on IQ? A high measured IQ is a double edged sword in my opinion: Marilyn vos Savant, whos IQ is basically unknown, used her brilliant mind to criticize the original proof of Fermat’s last principle. She is in my opinion right in saying that the proof is not brilliant because the amenableness of von Neumann algebra suggests that there ought to be a simpler proof, but she also fundamentally misunderstood the way modern mathematics actually works. And it took Marilyn more than a year to realize that she was wrong on the second account. I myself never understood the Monty Hall problem to be an interesting problem, I solved it trivially as a kid using a very simple symmetry argument. This shows that you don’t always have to know what you are doing to get good ideas. But I always try to check afterwards that the idea actually works.
    And I still wish to advocate my theorem of IQ being approximately 100 for nearly all human beings.

    Any experts out there? Anyone with contacts to pharmaceutical companies? I have a sister-in-law working at Novo Nordisk, but she is a senior project manager, and not into hardcore research.
    And are Oxytocin and Vasopressin complementary representations of the same thing? Is it the balance of Oxytocin and Vasopressin concentrations which governs my thoughts
    and make me rant like this?

    I’m simply too busy doing other things… Not stressed just busy.
    It would take me many hours to sort things out myself, simply because the medical information is written in a language that I have a hard time understanding. Also, many old news are not available to the public on WWW.

  16. Niels Kjaeron 21 Jul 2008 at 6:56 am

    30 minutes after I finished my last post, a Danish new-tech firm called SimeHealth came out with a press release describing a new cheap analysis technique. It is applied to small amounts of bodily fluids (reminds me of Peter Sellers and von Neumann). It creates a 2-dimensional spectrum using a variable infrared lightsource. My guess is that they use 2 different polarizations to disentangle the 2 components. They then compare the measured spectrum to previously measured spectra for the same human being, and any significant change in the 2-d spectrum will indicate a possible diagnosis of an emerging disease. They have been using the measurement in clinical tests for a while and their method explicitly removes diagnoses that are not statistically significant. They do explicit cross-validation tests in all their work, so it is going to be difficult to prove them wrong. They claim that they now can diagnose emerging diabetes, heart diseases, lung diseases, cancer, arthritis, and dementia with a machine costing 1000US$. They are rather cryptic in their press release saying that they do not want to reveal the underlying algorithm. At the same time googling their names: Agnar Höskuldsson and Lars Sørensen gives all the information I need to reconstruct what they are doing.

    Are they being careless? I don’t think so. And in a short time we will probably have a similar 2-component MR scan of the brain, and then the fun starts…

  17. [...] http://www.theness.com/neurologicablog/?p=336 [...]

Trackback URI | Comments RSS

Leave a Reply

You must be logged in to post a comment.