A provocative new study argues that the same evolutionary forces that supercharged the human brain may also have nudged autism rates upward. It is a theory about trade-offs, not villains or heroes, and it asks us to look at neurodiversity through the long lens of evolution.
A fast-evolving neuron, a slower-growing brain
In Molecular Biology and Evolution, researchers analyzed single-nucleus RNA sequencing data from multiple mammalian species to compare cell types across the brain. One group of neurons stood out: the abundant layer 2/3 intratelencephalic (L2/3 IT) neurons in the outer cortex, which relay information across brain regions involved in higher cognition.
Along the human lineage, these neurons appear to have evolved unusually quickly. The same analysis found striking changes in genes associated with autism, suggesting they were shaped by natural selection specific to humans. The authors propose a plausible link: these genomic shifts may have slowed certain aspects of postnatal brain development and boosted capacities like language—features that define us and, in some cases, intersect with autism traits.
Autism is common enough to matter at a population level and diverse enough to defy simple narratives. About one in 31 children in the United States has been identified with autism spectrum disorder, and global estimates hover around one in 100. The new work does not claim that autism is inevitable; it suggests that the routes our brains took to become uniquely human may have increased the probability of autistic neurodevelopment.
Evolution rarely gives without taking
Trade-offs are a hallmark of evolution. Classic examples come from immunity: gene variants that help resist malaria can, in double dose, cause sickle cell disease. Selection can favor traits that confer average benefits while creating risks at the extremes or in certain combinations.
The human brain is a story of trade-offs writ large. Compared with other apes, our brains develop for longer outside the womb, our childhoods stretch across years of learning, and our cortices carry dense networks for language and abstract thought. The new study’s focus on L2/3 IT neurons—key players in those long-range cortical conversations—fits that picture. If selection shaped these cells to favor prolonged development and linguistic prowess, it may also have nudged some developmental pathways toward autistic profiles.
“Our results suggest that some of the same genetic changes that make the human brain unique also made humans more neurodiverse,” said lead author Alexander L. Starr.
That framing matters. It does not cast autism as a mistake; it places it within the spectrum of outcomes produced by the same evolutionary processes that made us who we are.
Inside the spectrum: what biology shows
Biologically, autism is not one thing. Studies have documented early brain overgrowth in some autistic children, especially in frontal and temporal lobes that support language and social cognition. Across development, patterns can shift from higher long-range connectivity in infancy to reduced global coordination in adulthood, alongside pockets of local overconnectivity.
Researchers have also observed atypical lateralization, with some autistic individuals leaning more on left-hemisphere processing—good for detail, challenging for big-picture integration like facial configuration. At the molecular level, dozens of genes tied to synapse formation, circuit pruning, and neuronal migration are implicated, and their effects can be modulated by the environment.
None of this points to a single cause or a single trajectory. Rather, it paints autism as a set of developmental pathways through the brain’s vast landscape of possibilities, influenced by genetics, timing, and experience. In that light, an evolutionary nudge in specific neuron types could influence the statistical odds of these pathways without determining any one person’s fate.
The headline problem: price, by-product, or product?
“Autism may be the price of human intelligence” is a sticky headline, but it lands hard on many families and autistic adults. Words like price imply a deficit, as if people are a cost we pay for genius. The science describes something subtler: a by-product of the same changes that enabled language, long childhoods, and flexible thought.
Advocates in the neurodiversity movement have long argued that autistic minds contribute uniquely to society—often as deep specialists, meticulous analysts, or pattern seekers—while also facing environments that are not built for their sensory and social needs. The new study does not contradict that; it helps explain why a species wired for complex communication and learning might also express a wide range of ways to think and feel.
There is an ethical guardrail here, too. Evolutionary explanations are not value judgments and they are not blueprints for selection. History shows the dangers of treating genetic variation as something to engineer away. In practical terms, the lesson is not to eliminate difference, but to understand it well enough to support people better.
What this changes—and what it does not
The study advances a testable idea: a specific neuron class in the human cortex evolved rapidly, autism-linked genes were part of that change, and both may relate to our species’ prolonged brain development and language capacity. It reframes autism prevalence as an evolutionary feature of human neurodevelopment rather than a modern anomaly.
But it does not provide a clinical test, a single mechanism, or a cure. Science will need replications across more species and brain regions, developmental studies that trace these neurons from infancy, and integrative work that links genetic signatures to lived behavior. In the meantime, the most immediate implications are social and educational.
- Design environments that accommodate sensory differences and reduce overload, especially in classrooms and workplaces.
- Invest in early supports that scaffold communication and social learning without forcing conformity at the expense of well-being.
- Fund longitudinal research that connects cell-type biology to developmental timing, language acquisition, and outcomes across the spectrum.
- Avoid deterministic narratives; emphasize that evolution shapes probabilities, not destinies.
Reading the brain for what it is: human
If the human cortex carries the signatures of a long evolutionary gamble—more years to wire up, more talk between far-flung regions, more room for variation—then autism may be one of the ways that gamble plays out. That does not make autism a flaw or a superpower. It makes it part of being human.
The new findings ask us to expand our imagination about intelligence and development. They tell a story in which the same currents that carried language ashore also carved channels for many kinds of minds. The task now is to build a world where those minds can learn, work, and belong.
