Autism doesn’t have a single cause. It develops from a combination of genetic factors, prenatal environment, and birth circumstances, with genetics playing the largest role. Current CDC data shows about 1 in 31 children (3.2%) are identified with autism by age 8, and it’s diagnosed more than three times as often in boys as in girls.
Understanding what contributes to autism has advanced significantly in recent years. Here’s what the science actually shows.
Genetics Are the Biggest Factor
Genetic factors account for an estimated 40 to 80 percent of autism risk. That’s a wide range because hundreds of genes appear to be involved, and they interact with each other and with the environment in complex ways. There’s no single “autism gene.” Instead, researchers have identified mutations in genes like SHANK3, CHD8, SYNGAP1, and NLGN3, among others, that each contribute a piece of the puzzle. Some of these genes affect how brain cells form connections with each other, while others influence early brain development more broadly.
Most of these genetic changes aren’t inherited from a parent in a straightforward way. Many arise spontaneously during conception or early development. Others are common genetic variants that individually have a tiny effect but collectively shift risk when enough of them are present. This is why two parents with no family history of autism can still have an autistic child, and why the condition can seem to appear “out of nowhere.”
Family Recurrence Rates
If you already have one autistic child, the chance of a younger sibling also being autistic is about 20%, according to data from the Baby Siblings Research Consortium published in Pediatrics. That’s roughly one in five, far higher than the general population rate of about 3%.
The risk isn’t evenly distributed, though. Younger brothers have a 25.3% recurrence rate, while younger sisters have a 13.1% rate. Families that already have more than one autistic child (called “multiplex” families) see even higher odds: about 37% for the next sibling, compared to 21% in families with only one autistic child. And interestingly, when the older autistic sibling is a girl, the recurrence rate for younger siblings jumps to nearly 35%, compared to about 23% when the older sibling is a boy. This likely reflects a higher genetic load in families where girls are affected, since girls generally require more genetic risk factors to develop autism.
What Happens in the Brain
Research from Yale School of Medicine using PET brain scans found that autistic adults have about 17% fewer synapses, the junctions where brain cells communicate, compared to neurotypical adults. Lower synaptic density correlated with more pronounced social communication differences like reduced eye contact and difficulty reading social cues. This suggests that differences in how brain connections form and are maintained play a central role in autism.
During typical development, the brain overproduces connections in early childhood and then prunes away the ones it doesn’t need. In autism, this process appears to go differently, resulting in patterns of connectivity that affect how sensory information, social signals, and other inputs are processed. These aren’t deficits in the conventional sense so much as differences in how the brain is wired.
Prenatal Environment and Exposures
What happens during pregnancy matters. A systematic review of case-control studies found that exposure to environmental pollutants, including heavy metals, pesticides, and air pollution, during pregnancy and early life is a risk factor for autism and other neurodevelopmental conditions. Exposure to particulate matter (PM10, the tiny particles in polluted air) during a child’s first year of life showed a significant association with autism risk in a meta-analysis. The relationship between certain metals and autism appears to follow a non-linear pattern, meaning both very low and very high levels of some metals may increase risk.
Certain medications taken during pregnancy also affect risk. Valproate, an anti-seizure drug, is the clearest example. When taken during the first months of pregnancy, it can disrupt brain development by interfering with gene regulation during a critical window. It affects how brain cells migrate, form connections, and communicate. The risk is well-established enough that valproate carries strong warnings against use during pregnancy.
Antidepressants, particularly SSRIs, have received a lot of attention, but the actual numbers are reassuring. In studies, just over 1% of women taking SSRIs had children with autism, compared to slightly under 1% of women not taking them. That’s a very small difference, and untreated severe depression during pregnancy carries its own risks.
How the Environment Changes Gene Activity
One of the most important discoveries in autism research is that environmental factors don’t just act alongside genetics. They can actually change how genes behave. This field, called epigenetics, studies chemical modifications that turn genes on or off without altering the DNA itself. Things like diet, infections, medications, and chemical exposures can all modify these gene switches.
The gut microbiome appears to play a role in this process. Bacteria in the gut produce short-chain fatty acids that interact directly with the molecular machinery controlling gene expression, particularly through modifications to DNA and the proteins that package it. These interactions have been associated with autism. Environmental exposures can tip the balance in both directions: too much or too little activity of certain gene-regulating enzymes leads to disruption of the same brain development pathways.
Perhaps most striking, some of these epigenetic changes may be heritable. Animal studies have shown that social differences caused by certain chemical exposures can be passed to offspring, suggesting that environmental effects on gene regulation in one generation could influence the next.
Birth Complications and Prematurity
Difficult births don’t cause autism on their own, but several birth-related factors correlate with higher rates. Babies born weighing less than about 5 pounds 8 ounces have roughly double the risk of autism compared to those born at a normal weight. For low-birth-weight girls with intellectual disability, the risk is four times higher. Extreme prematurity, being born before 26 weeks, is associated with significantly more autistic traits and a much higher rate of diagnosed autism compared to full-term birth.
Other factors studied include breech presentation, cesarean delivery, and fetal distress, all broadly categorized as “less than optimal” birth conditions. The challenge with these findings is that the same genetic factors contributing to autism may also contribute to pregnancy complications, making it hard to separate cause from correlation. A difficult birth may be a marker of underlying developmental differences rather than a direct cause.
Vaccines Do Not Cause Autism
This question comes up often enough that it deserves a clear answer. Vaccines do not cause autism. A 2025 WHO analysis drawing on 31 studies published between 2010 and 2025, covering data from multiple countries, confirmed no causal link between childhood vaccines and autism. This includes vaccines containing aluminum adjuvants and those that once contained thimerosal (a mercury-based preservative largely removed from childhood vaccines years ago). A large cohort study using nationwide data on all children born in Denmark between 1997 and 2018 reinforced the same conclusion. The WHO has reaffirmed this finding repeatedly since 2002.
The timing of autism’s most visible signs, which tend to emerge around 18 to 24 months, overlaps with the childhood vaccination schedule. This coincidence likely fueled the original concern, but decades of large-scale research have consistently shown no connection.
Why There’s No Single Answer
Autism arises from many paths converging. A child might have a strong genetic predisposition that would lead to autism regardless of environment. Another might have moderate genetic risk that tips over only with certain prenatal exposures. Still another might develop autism primarily through spontaneous genetic mutations that neither parent carries. The 40 to 80 percent heritability estimate reflects this complexity: genetics set the stage, but the full picture includes prenatal chemistry, birth circumstances, and the interplay between genes and environment that begins before birth and continues through early development.
What’s consistent across the research is that autism begins in the womb, during the earliest stages of brain formation. It is not caused by parenting style, screen time, diet in childhood, or anything a parent does after a child is born. The factors that shape autism are largely biological, largely prenatal, and in most cases, not something any parent could have predicted or prevented.