Autism is not something you catch or develop from a single cause. It is a neurodevelopmental condition that begins before birth, shaped primarily by genetics and, to a lesser extent, by conditions in the prenatal environment. Twin studies estimate that 64 to 91% of autism risk comes from genetic factors, making it one of the most heritable neurodevelopmental conditions known.
Because autism emerges from a combination of genetic makeup and early brain development, there is no single answer to “how do you get it.” But researchers have identified the major contributors with increasing clarity.
Genetics Play the Largest Role
The strongest driver of autism is your DNA. Hundreds of genes contribute to autism risk, and they work in two broad ways. Some genetic changes are inherited, passed from parent to child like eye color or height. Others are “de novo” mutations, meaning they appear for the first time in a child and weren’t present in either parent’s DNA. De novo mutations tend to have a stronger individual effect, while inherited variants each carry a smaller influence but can add up across many genes.
This is why autism often runs in families without following a simple pattern. A parent might carry a collection of risk-related gene variants without being autistic themselves, then pass enough of them to a child to cross a threshold. In other cases, a spontaneous mutation during egg or sperm formation or early embryonic development creates a significant shift on its own. The inherited variants tend to affect how cells are organized and how nerve cells communicate, while de novo mutations often disrupt broader developmental pathways.
Sibling and Family Risk
If a child has an older autistic sibling, their chance of also being autistic is about 20%, according to a large multinational study from the Baby Siblings Research Consortium. That is roughly ten times higher than the general population rate. The risk increases further if there are multiple older autistic siblings or if the older sibling is female (since girls are diagnosed less often, a female child who does meet diagnostic criteria may signal a stronger genetic load in the family).
Male children are more likely to develop autism in these families, consistent with the broader pattern: boys are diagnosed about three to four times more often than girls, though researchers increasingly recognize that girls may be underdiagnosed due to differences in how their traits present.
What Happens in the Brain
One of the clearest biological differences in autistic brains involves synaptic pruning, the process by which the brain trims excess connections between nerve cells during childhood. Postmortem studies have found a higher density of excitatory synapses in autistic brains, suggesting that the pruning process doesn’t remove as many connections as it typically would.
Research in animal models points to a specific cellular pathway (driven by a signaling system called mTOR) that, when overactive, impairs the brain’s ability to clean up unneeded synapses. This surplus of connections changes how different brain regions communicate with each other, creating patterns of hyperconnectivity. These altered connection patterns appear to influence sensory processing, repetitive behaviors, and social communication, the core features of autism.
Prenatal Factors That Increase Risk
While genetics account for the majority of autism risk, conditions during pregnancy can also play a role. These don’t “cause” autism on their own but can increase the likelihood, particularly in someone already genetically predisposed.
Maternal immune activation. When a pregnant person’s immune system mounts a strong response, whether from infection or inflammation, the signaling molecules involved can cross the placenta and affect fetal brain development. Immune proteins that are normally helpful in fighting illness also play roles in how fetal neurons grow, migrate, and form connections. Animal studies show that strong immune activation during pregnancy can reduce certain types of brain cells and alter brain wiring in ways that mirror features of autism.
Certain medications. Valproate, an anti-seizure medication, is one of the most clearly identified prenatal risk factors. Children exposed to valproate in the womb have an absolute autism risk of about 4.4%, with a nearly threefold increase compared to unexposed children, even after accounting for the mother’s epilepsy. Other common anti-seizure medications studied alongside valproate did not show the same increased risk.
Air pollution. A large Danish study following over 850,000 children found that higher prenatal exposure to traffic-related air pollutants, particularly nitrogen dioxide and black carbon, was associated with a modest but consistent increase in autism diagnoses. The associations held even when researchers controlled for other pollutants and socioeconomic factors. The effect sizes are small on an individual level but meaningful at a population scale given how widespread air pollution exposure is.
Premature birth. Babies born before 37 weeks have about a 40% higher chance of an autism diagnosis compared to those born full-term. The risk rises sharply with earlier delivery: extremely preterm babies (born at 22 to 27 weeks) are nearly four times more likely to be diagnosed. Being born small for gestational age also carries increased risk, though prematurity itself is the stronger factor.
Parental Age
Older fathers have a measurably higher chance of having an autistic child. Men aged 50 and older are about 2.2 times more likely to have an autistic child compared to men under 30, after adjusting for the mother’s age. The risk increases steadily with each decade of paternal age, likely because sperm accumulate more spontaneous genetic mutations over time. Maternal age has also been associated with increased risk in some studies, though the paternal age effect is more consistently documented.
Vaccines Do Not Cause Autism
The claim that vaccines cause autism originated from a 1998 study that was later retracted due to fraud. Since then, extensive research has examined the question. The MMR vaccine specifically has been studied repeatedly, and reviews by the Institute of Medicine and the Agency for Healthcare Research and Quality have found no association between MMR vaccination and autism. This finding has held across multiple large population studies involving millions of children.
Why There Is No Single Cause
Autism arises from a convergence of factors, most of them genetic, that alter how the brain develops before and shortly after birth. No single gene, no single exposure, and no single event explains it. For many autistic people, genetics alone are sufficient. For others, a genetic predisposition combines with prenatal environmental factors to shift brain development along an autistic trajectory.
This also means autism cannot be prevented in any targeted way. It is not the result of anything a parent did or didn’t do during pregnancy or childhood. It is a difference in neurodevelopment with deep biological roots, present from the earliest stages of brain formation.