Summary: Researchers find that groups of brain regions that synchronize their activity during memory tasks become smaller and more numerous with age, reflecting reduced cohesion in memory-related brain networks.
Source: PLOS.
Groups of brain regions with coordinated activity remain individual-specific but lose cohesion as we age
New research published in PLOS Computational Biology reports that the patterns of synchronized brain activity—groups of brain regions that fire together during memory tasks—tend to fragment with age. Using functional MRI (fMRI) data and a dynamic network approach, the authors show that younger adults exhibit a few large, cohesive groups of connections spanning much of the brain during memory performance, while older adults show many smaller, more fragmented groups. This change suggests a loss of coordinated functional connectivity in memory-related networks even when no clinical memory impairment is present.
Most studies of brain activity examine averaged measurements across groups of people, which can obscure important individual differences in functional brain networks. To address this, Elizabeth N. Davison and colleagues at Princeton University applied a novel method from dynamic network theory—hypergraph analysis—to fMRI recordings from individuals performing cognitive tasks and during rest. They converted each participant’s fMRI time series into a network of brain regions and the changing functional connections between them. The hypergraph framework then quantified how sets of these connections co-evolve over time, identifying synchronous groups of connections that act together as coherent units of brain dynamics.
The study found two central results. First, the number and organization of synchronous connection groups (hypergraph cardinality) are consistent within an individual across different mental states such as memory tasks, attention tasks, and rest. This stability suggests each person has a characteristic dynamic fingerprint of brain connectivity that persists across contexts. Second, there is substantial variation between individuals: some people have relatively few, large synchronized groups linking broad swaths of the brain, while others have many more, smaller groups.
When focusing specifically on memory tasks, the researchers observed a clear relationship with age. Younger participants tended to exhibit a small number of large, tightly coordinated groups of functional connections that linked many brain regions during memory retrieval and encoding. By contrast, older participants showed an increasing number of smaller groups, indicating that the brain’s functional networks become more fragmented and less globally coordinated with advancing age. This pattern of reduced cohesion in dynamic functional connectivity may represent a general signature of healthy aging that predates overt cognitive impairment.
“This method elegantly captures important differences between individual brains, which are often complex and difficult to describe,” says Elizabeth Davison. “The resulting tools show promise for understanding how individual differences in brain dynamics relate to behavior, cognitive function, and age-related changes in health and disease.”
Future research will explore how individual dynamic connectivity signatures can help distinguish healthy aging from age-related pathological decline, and whether these hypergraph-derived measures can predict cognitive outcomes or serve as biomarkers for early detection of impairment.
Funding: This research received support from multiple sources, including the David and Lucile Packard Foundation; the Institute for Collaborative Biotechnologies; the National Science Foundation Graduate Research Fellowship Program; fellowships such as the Francis Robbins Upton and the Worster Fellowship; the John D. and Catherine T. MacArthur Foundation; the Army Research Laboratory and the Army Research Office; the National Institute of Mental Health; the National Institute of Child Health and Human Development; the Office of Naval Research; and the National Science Foundation. The funders did not influence study design, data collection and analysis, the decision to publish, or manuscript preparation.
Competing interests: The authors declared that no competing interests exist.
Source and original research: Elizabeth N. Davison; original study titled “Individual Differences in Dynamic Functional Brain Connectivity across the Human Lifespan,” published in PLOS Computational Biology. The study applied hypergraph analysis to quantify individual variation in dynamic functional connectivity and found a significant correlation between hypergraph cardinality and age across a wide age range.
Abstract
Individual Differences in Dynamic Functional Brain Connectivity across the Human Lifespan
Individual differences in functional brain networks relate to complex personal traits including age, health, and cognitive ability. While dynamic network theory has been applied to fMRI data to identify properties of brain dynamics, most findings are reported at the group level. In this work, hypergraph analysis—a tool from dynamic network theory—is used to quantify individual differences in brain functional dynamics. A summary metric derived from this method, hypergraph cardinality, was evaluated in two complementary datasets. The first (“multi-task”) included 77 individuals performing multiple cognitive tasks, showing that hypergraph cardinality varies between people but is consistent within individuals across tasks, with a marginal correspondence to age on a memory task. Motivated by this, a second dataset (“age-memory”) with 95 subjects aged 18–75 performing a memory task revealed a significant correlation between hypergraph cardinality and age. These results align with established associations between age and network structure and suggest hypergraph analysis offers a valuable approach for studying dynamic brain networks across the lifespan.