A paper published on 5 September 2025 in Molecular Biology and Evolution proposes that the same evolutionary forces that made human cognition unique also increased the prevalence of autism. Researchers from Stanford University examined over 1 million neurons using single-nucleus RNA sequencing across multiple species and cortical regions.
How Human Brain Evolution Made Autism More Common
Background
Autism spectrum disorder is significantly more prevalent in humans than in other mammals and other species. Approximately 1 in 31 children in the United States is identified with the condition. Data from the World Health Organization indicate that 1 in 100 children globally are diagnosed with the disorder. Nonhuman primates rarely exhibit autism like syndromes.
Postdoctoral researcher Alexander L. Starr and biology professor Hunter B. Fraser were motivated by two big questions in evolutionary neuroscience. The first centers on why different neuron types evolve at different speeds. The second question involves reiterating why autism spectrum disorder is prevalent in the human species but is far less reported in other primates.
The two first hypothesized that abundance constrains evolution. This means that the more abundant a cell type is, the harder it is for its gene regulation to drift, because too many connections depend on it. Moreover, in building their case, they also underscored the fact that some traits impacted in autism, like language and social communication, are unique among humans.
Hence, to understand not only the cause of autism but also its prevalence, Starr and Fraser intended to connect human brain evolution with neurodevelopmental diversity. They did this using large datasets from six mammalian species, including human, chimpanzee, and other primate brains, to categorize different neuron subtypes and compare gene expression profiles.
They then applied single-nucleus RNA sequencing to identify patterns of transcription across species and determined how neuronal abundance influenced levels of molecular conservation or divergence over evolutionary time. Transcriptome divergence across cortical regions was measured to test whether abundance correlated with evolutionary constraint.
Key Findings
• Abundance Predicts Conservation: Neuron types with higher abundance had greater conservation of gene expression across species. This indicates that the more numerous a neuron type is, the more strongly its molecular programs are constrained by evolutionary pressures. This principle held across cortical regions and datasets.
• Exception in Human L2/3 IT Neurons: Human layer 2 and layer 3 intratelencephalic neurons evolved unusually fast compared to chimpanzees and other species, despite being abundant. These neurons are critical for long-range communication across the cortex and play major roles in higher-order cognition and language processing.
• Downregulation of Autism Genes: The same neurons in humans exhibited dramatic downregulation of genes associated with autism. These were consistently less expressed in humans relative to chimpanzees. The pattern was significant. This suggests that these expression changes had functional relevance rather than being neutral drift.
• Specific Polygenic Positive Selection: Analyses showed evidence of polygenic positive selection shaping these down-regulated autism-associated genes. Natural selection favored small coordinated changes across many genes, lowering their expression in human L2 and L3 IT neurons. The changes, therefore, appear adaptive rather than random.
• Further Cortical Organoid Validation: Human-chimpanzee hybrid cortical organoids confirmed that human alleles of autism-related genes were expressed at lower levels than chimpanzee alleles. These differences originated from changes near the genes. Variability of expression was also lower in humans, further supporting selective pressure.
Implications
Starr and Fraser proposed that downregulation of autism genes supported prolonged postnatal brain development and enhanced connectivity required for language. However, this also increased neuronal sensitivity to disruption, making humans more vulnerable to autism. The tradeoff illustrates how cognitive benefits came with developmental risks.
The findings from the investigation suggest that the prevalence of autism spectrum disorder is not solely pathological but reflects evolutionary outcomes. The same genetic changes that shaped human intelligence and creativity also contributed to neurodiversity. Autism can therefore be seen as an intrinsic byproduct of human brain evolution rather than a disorder alone.
Hence, based on the above, the study links human brain evolution and neurodevelopmental conditions. It indicates that autism prevalence may stem from positive selection acting on abundant neurons crucial for cognition. By connecting evolutionary history with modern diversity, the study reframes autism as deeply intertwined with humanity itself.
FURTHER READING AND REFERENCE
- Starr, A. L., and Fraser, H. B. 2025. “A General Principle of Neuronal Evolution Reveals a Human-Accelerated Neuron Type Potentially Underlying the High Prevalence of Autism in Humans.” Molecular Biology and Evolution. 42(9). DOI: 1093/molbev/msaf189