You can break down the “cause” of autism to “nurture and nature,” where nature is genetic causes and nurture is environmental causes.
If you want to skim then skim my bolded text, which I’ve also organized as (a), (b), (c) etc… If you want the full Monty, then read my detailed text with original references. This is a very long answer written in the interest of comprehensiveness (and at the risk of being down-voted because of how insanely long this answer is….)
Genetic Basis of Autism
(a) The genetic basis of autism has been studied in twins — and the type of twins. When researchers compared monozygotic twins (“identical” twins) with dizygotic twins (“fraternal” twins), they saw a higher rate of concordance in identical twins.
In other words, if one identical twin has autism, there is a high probability that the other identical twin also has autism. In fraternal twins, the probability is lower but not zero, especially when you expand the autism definition to include the cognitive or language component. [source: Bailey, A., Le Couteur, A., Gottesman, I., Bolton, P., Simonoff, E., Yuzda, E., et al. (1995). Autism as a strongly genetic disorder: Evidence from a British twin study. Psychological Medicine, 25, 63–77]
(b) The familial linkage of autism also supports the genetic basis of autism. Two independent familial studies conducted in 2009 has shown that there is a higher risk of autism in siblings versus the general population. [sources: Geschwind, D. H., & Konopka, G. (2009). Neuroscience in the era of functional genomics and systems biology. Nature, 461, 908–915. Pennington, B. F. (2009). Diagnosing learning disorders: A neuropsy-chological framework. New York: Guilford Press]
(c) The gender dominance (boys are more likely than girls to be diagnosed with autism) of autism further suggests a genetic basis of autism. The general theory is that, like Hemophilia (an X-chromosome-linked disease) is passed from mother to son, autism may also be passed on from mother to son. For a genetics refresher, let me use this designation (a) as the autism gene(s) —
If the mother has the autism genetic variant but does not display classic autism symptoms, then mother is probably:
Mother X X(a)
If the father does not have autism genetic variant, father may look like:
Father X Y
Children with autism in this mating follows this pattern
XX(a) + XY = XX or XY or X(a)X or X(a)Y
It depends on which X chromosome from the mother is paired with the father’s chromosome.
You will have a 25% chance of an offspring with autism, and that offspring will be a boy = X(a)Y.
You will have a 25% chance of an offspring carrying the autism gene but not necessarily EXPRESS the phenotype of autism unless there is a triggering factor (where “nurture” or environmental causes come in), and that offspring will be a girl = X(a)X.
You will also have a 50% chance of a genetically “typical” or “neurotypical” offspring XY son or XX daughter.
Is autism a Y-linked disorder? It would explain why there is a higher rate of boys diagnosed with autism versus girls — but if this is a purely Y-linked disorder then you wouldn’t expect to have any girls diagnosed with autism.
Thus: even with the prevalence of boys being diagnosed with autism versus girls diagnosed with autism, autism may not be a sex-chromosomal-linked condition. We just don’t know, because we don’t even know the whole set of genes that can definitively be assigned to “causing what we can phenotypically diagnose as autism.”
(d) Yet there is also at least 1 study that suggests that autism rate increases when the father’s age increases. [source: Reichenberg, A., Gross, R., Weiser, M., Bresnahan, M., Silverman, J., Harlap, S., et al. (2006). Advancing paternal age and autism. Archives of General Psychiatry, 63, 1026–1032. Full text http://archpsyc.ama-assn.org/cgi/content/full/63/9/1026 ] In particular from this study:
Offspring of men 40 years or older were 5.75 times (95% confidence interval, 2.65-12.46; P<.001) more likely to have ASD compared with offspring of men younger than 30 years, after controlling for year of birth, socioeconomic status, and maternal age. Advancing maternal age showed no association with ASD after adjusting for paternal age. Sensitivity analyses indicated that these findings were not the result of bias due to missing data on maternal age. The note on bias of maternal age is relevant because one can argue that maybe it's due to older mothers -- not older fathers (therefore making this more of the mother's genes at play), and paternal age remains the overriding factor. (e) Scientists have studied various genes that contribute to the spectrum of phenotypes in autism, and these genes are compiled in Chapter 6 of the International Handbook of Autism and Pervasive Developmental Disorders, which you can buy for $559 for the 2011 version [from http://www.springer.com/psychology/child+%26+school+psychology/book/978-1-4419-8064-9 ]. Genes range from NRXN1 (Neurexin 1) and NRXN2 (Neurexin 2) to RELN (Reelin), FOXP2 (Forkhead box P2), CNTNAP2 (Contactin-associated protein-like 2), and PTEN (phosphatase and tensin homolog -- yes, cancer biologists, this is a tumor suppressor gene!), APP (Amyloid beta(A4) precursor protein -- correlates with self-injurious and aggressive behavior). I'm not going to list the entire table here, but you get an idea of the diversity of genes involved and why autism is a "broad spectrum disorder". Environmental Basis of Autism (a) The support for the role of nurture / environmental role in autism is attractive because this is where "hope" is made. Think about it: if I tell you that cancer is wholly genetic, I don't care how well you take care of yourself you're going to die of a certain type of cancer, you're not going to feel very hopeful. But if you believe that there are indeed a strong genetic basis in cancer YET you can also wield a certain span of control over your environment to counter the genes -- you feel more hopeful. I truly think that the subscribing belief to the environmental basis of autism has led to more quackery in the field of autism intervention than if parents wholly subscribe to the primarily genetic basis of autism (but I won't digress on that here.) (b) For one, the genetic basis of autism has led to the discrediting of the "Refrigerator Mother" theory posited by Bruno Bettelheim (http://en.wikipedia.org/wiki/Bruno_Bettelheim). Mothers did not "create" or "cause" autism in their children by being unloving and aloof. (c) For another, "environment" and "nurture" are difficult to be scientific about. There are claims of autism being caused by chemical substances (such as mercury found in vaccines or lead poisoning), food allergies (basis of the gluten-free casein-free diets), food colorings -- for a consumer article on the blurry lines of environmental causes of autism read http://www.nytimes.com/2011/08/09/health/views/09klass.html -- but in order for us critical consumers to accept these claims as "true" we need to have good scientific studies that look at whether these "cause" autism. (d) As an example, the dietary intervention is particularly contentious -- you have parents who claim with utmost sincerity that dietary interventions have been key in their children's improvement in autism symptoms (some go as far as claiming "cure" -- again I won't go there and I won't use that word). They upload videos of their children's "before and after" dietary interventions. But these children also receive interventions that have been proven through research and replicated in many other studies (translation: behavioral interventions like Applied Behavior Analysis as pioneered by the late Lovaas). So how do we know how much is the diet and how much is contributed by ABA? Published peer-reviewed studies of dietary interventions have shown that children with autism who have demonstrated a sensitivity to glutein/casein based foods and children with celiac disease will see improvement in symptoms when on the diet. However, if you exclude these children from the study and look at children without these sensitivities, the diet has not shown to improve outcomes in autism symptoms: J Autism Dev Disord. 2006 Apr;36(3):413-20. The gluten-free, casein-free diet in autism: results of a preliminary double blind clinical trial. Elder JH, Shankar M, Shuster J, Theriaque D, Burns S, Sherrill L. College of Nursing, University of Florida, Gainesville, 32610, USA. firstname.lastname@example.org Abstract This study tested the efficacy of a gluten-free and casein-free (GFCF) diet in treating autism using a randomized, double blind repeated measures crossover design. The sample included 15 children aged 2-16 years with autism spectrum disorder. Data on autistic symptoms and urinary peptide levels were collected in the subjects' homes over the 12 weeks that they were on the diet. Group data indicated no statistically significant findings even though several parents reported improvement in their children. Although preliminary, this study demonstrates how a controlled clinical trial of the GFCF diet can be conducted, and suggests directions for future research. Notable: "Group data indicated no statistically significant findings even though several parents reported improvement in their children." IN CONCLUSION (yes yes I'm getting there)... My opinion is that autism is primarily a genetically-driven neurological disorder spanning a wide spectrum of genes that contribute collectively to a particular "phenotype" (how the symptoms manifest, the degree of severity) in a person. Whether the reason why we see more boys than girls diagnosed with autism may be due to a sex-chromosomal (X or Y) linkage, we may speculate but we cannot tell whether this is due to the "X" chromosome contributing, "Y" chromosome contributing, or X or Y chromosome not contributing. Whatever environmental causes that may "trigger" autism, is likely to trigger a genetic component that otherwise remains dormant, as opposed to "causing de novo" autism that would otherwise not exist on a genetic level. I personally do not believe in children "becoming autistic" because they are eating certain foods, gotten a seizure due to vaccines, or by wearing a certain type of fabric for their clothes. HOWEVER, I do believe that children who have genetic components that can be triggered, may be vulnerable to specific environmental assaults. [in other words, please don't flood me with hate mail, I'm not saying there isn't a link, I'm only saying there may not be a cause.] Finally, just because parents subscribe to genetic bases of autism does not mean there is no hope. There is a genetic basis of major depressive disorders (I come from a family of undiagnosed depressives and I believe I have those genes and I have suffered from depression) -- but this doesn't mean we don't have bona fide, scientifically-based interventions that produce acceptable outcomes in terms of quality of life. I will never claim that I've been "cured" of depression -- I don't believe this is true for myself. But I have an excellent quality of life and I am happy with my life -- even as I live with depression genes that can be triggered by specific environmental assaults. Link out to: Markram's Intense World Theory -- A Unifying Theory of the Neurobiology of Autism http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3010743/