The structural organization of the human brain changes non-linearly over a person’s life, shifting at four distinct ages. A large study identified major transitions in brain network architecture around ages nine, 32, 66, and 83. The research was published in Nature Communications.
The brain is essentially a network of connected regions. The architecture of this network, known as its topology, dictates how well different areas communicate. Researchers measure this topology using mathematical concepts like integration, segregation, and centrality. Different topological structures have different strengths related to cognition and behavior.
Integration describes how easily information travels across the entire brain. A highly integrated network has many short paths connecting distant regions, optimizing it for rapid communication. Segregation refers to how the network divides into specialized local groups. A highly segregated network has dense local connections that support specialized processing tasks, like vision or motor control. Centrality identifies specific regions that act as highly active hubs for information transfer, making the network more resilient to damage.
Past research has linked brain topology to cognitive function and mental health during specific life stages. But the underlying principles of how this organization shifts across an entire human life have remained unmapped. Alexa Mousley, a researcher at the University of Cambridge, wanted to identify if there are specific turning points when the brain enters a new phase of developmental change.
To map these lifespan changes, the researchers gathered brain imaging data from nine different datasets. The combined data included 4,216 participants ranging in age from zero to 90 years old. Because the sample exceeded 2,000 individuals, this qualifies as a large study.
The team used a specific type of magnetic resonance imaging that tracks the movement of water molecules to map the physical wiring of the brain. They then harmonized the data from the different sources to account for variations in scanning equipment. From there, the scientists calculated 12 different metrics to describe the topology of each participant’s brain network. The network densities were strictly controlled to allow for fair comparisons across different ages.
To make sense of this highly detailed data, the team used a mathematical technique to project the network metrics into three-dimensional spaces. This machine learning approach filters out overlapping information to reveal the fundamental mathematical structure of complex data. By tracing the average trajectory of brain development through these spaces, the researchers could pinpoint where the trajectory abruptly changed direction. They defined these spots as turning points.
The analysis revealed four major turning points in the human lifespan. These occur around ages nine, 32, 66, and 83. These four points separate human life into five distinct epochs of brain development, with each epoch featuring its own unique pattern of structural change.
The first epoch spans from birth to age nine. During this childhood phase, the brain’s global integration decreases while local segregation increases. The extent to which neighboring regions connect to each other is the strongest predictor of a child’s age during this period. The end of this epoch coincides roughly with the onset of puberty and a known biological phase where the brain actively eliminates unused neural connections.
The second epoch lasts from age nine to 32. This phase encompasses adolescence and early adulthood. Over these years, the brain network becomes increasingly integrated and less segregated on a global scale. The balance between global efficiency and local specialization becomes the most defining feature of brain development during this time.
The turning point at age 32 represents the largest structural shift in the entire lifespan. It aligns with the known peak of white matter volume, which is the insulated wiring that connects brain regions. Following this peak, the third epoch stretches across three decades of adulthood, from age 32 to 66.
This middle adulthood epoch is a relatively stable period characterized by slower changes in network architecture. During these years, global integration begins to decline while local efficiency increases. Changes in network segregation drive the relationship between age and brain topology during this long phase.
The fourth turning point arrives at age 66, marking the transition into older age. From 66 to 83, the brain network shows a distinct shift toward increasing modularity. Modularity means the network separates into highly interconnected subgroups. The researchers note this pattern suggests a simplification of the brain’s structural network, which corresponds with expected age-related degradation in white matter.
The final epoch covers ages 83 to 90. In this late stage of life, the relationship between age and brain topology is quite weak. The only metric that tracks with age during this period is the centrality of individual nodes, meaning certain localized hubs become increasingly important for connectivity.
The study has some limitations that affect how the results should be interpreted. The data is cross-sectional, meaning it compares different people of different ages rather than following the same individuals over their entire lives. This design makes it impossible to establish causality or temporal dynamics within a single person. It prevents researchers from tracking how an individual’s specific brain topology changes over time.
Additionally, the researchers used fixed network density thresholds for their main analysis to allow for fair comparisons between different ages. While they conducted secondary tests to verify their choices, this thresholding process might obscure some smaller individual differences in total brain connectivity. The analysis also did not separate the data by biological sex, leaving it unknown whether these major turning points happen at different ages for men and women.
Finally, the oldest age group contained just 93 participants, which lowered the statistical power of the analysis for that specific epoch. The associations in this late-aging epoch were mostly not statistically significant. It is also highly possible that the people in their late 80s who participated in these imaging studies are exceptionally healthy compared to their peers. This selection bias could skew the results for the oldest epoch, making their brains look more resilient than average.
The study, “Topological turning points across the human lifespan,” was authored by Alexa Mousley, Richard A. I. Bethlehem, Fang-Cheng Yeh, and Duncan E. Astle.