Functional hierarchy of the human neocortex across the lifespan | Nature

Download PDF

Subjects

- Computational biology and bioinformatics

- Computational neuroscience

Abstract

Large-scale gradients of functional connectivity between brain areas organize the human neocortex, linking brain topography to the texture of cognition1,2. In adults, three dominant axes—sensory–association, visual–somatosensory and modulation–representation—run, respectively, from primary sensory to transmodal association areas, from visual to body-centred systems and from control and attention networks to default mode and sensory areas1,2,3,4. These gradients provide a compact description of large-scale cortical hierarchies that underlie distinct modes of information processing. However, how these gradients and their multiscale biological and cognitive correlates evolve across the lifespan is unknown. Here we establish a continuous normative reference of functional organization from birth to 100 years of age, revealing complex, nonlinear developmental trajectories. Gradient architecture is anchored by primary sensory systems in infancy, differentiates along association and control axes during childhood and adolescence and gradually dedifferentiates during ageing. The importance of this functional architecture is corroborated by biology and behaviour: gradient metrics predict cognitive performance across development; structure–function coupling varies by axis and age; and distinct transcriptomic signatures are strongest early in life and weaken with age, consistent with a transient genetic scaffold for gradient architecture. Our lifespan gradients unify diverse research into developmental brain connectivity and provide a shared multimodal reference for future studies.

Main

Understanding how brain network organization changes across the human lifespan is a central and long-standing goal in neuroscience. Previous work has shown that functional connectivity (FC) is organized along smoothly varying cortical ‘gradients’1. These gradients stratify cortical locations according to orthogonal connectivity patterns, capturing smooth variation in fundamental motifs of network architecture. These motifs align with established progressions in information processing2,5, supporting the interpretation of gradients as proxies for large-scale processing hierarchies.

This gradient-based framework exists in a healthy tension with the foundational principle of cortical arealization—the parcellation of the cortex into discrete areas defined by their unique cytoarchitecture and connectivity. Previous work has highlighted this apparent conflict, suggesting that the sharp boundaries between functional systems observed in individuals are at odds with the smooth, continuous nature of population-level gradients6. This debate is reminiscent of historical arguments in physics over light as a wave or a particle, where the resolution lies not in choosing one correct description, but in understanding the context in which each provides value. A key test of any framework’s validity is therefore its explanatory utility: its ability to provide novel insights into brain function and behaviour. The initial discovery of the principal functional gradient, for example, provided a new explanation for the spatial distribution of functional networks1, an insight that had not emerged from purely arealization-based studies. Therefore, a central goal of this study is to directly test the explanatory utility of the lifespan gradient framework by linking its organizational metrics to cognitive and behavioural outcomes, microstructural and morphological metrics and gene ontology data from infancy through to old age. By establishing a normative timeline of gradient development and demonstrating its behavioural relevance, we aim to provide compelling evidence for the value of gradients as a crucial organizing principle of the human brain.

In adults, FC is reliably organized along three dominant gradients that together provide a compact coordinate system for large-scale cortical architecture1,2. These axes are often interpreted as topographic hierarchies2,7,8 that enumerate continuous transitions between processing motifs rather than strict causal or unidirectional pathways. The most well-studied connectivity gradient is the canonical sensory–association (SA) axis. Explaining the most variance in FC, the SA axis is anchored at one end by primary unimodal cortex and at the other end by high-order association cortex coinciding with the default mode network1,2. Similarly, repeated observations have shown that the gradient that explains the second-most variance in FC is an axis spanning between visual and somatosensory cortex (the VS axis), smoothly partitioning cortical locations according to their preferential involvement in modality-specific processing1,2. Finally, many studies have identified a tertiary gradient of FC, often described as a modulation–representation (MR) axis, anchored at opposite ends by cortical regions that are preferentially invo