Summary: Researchers report that people with Alzheimer’s disease exhibit unusually high neural flexibility—the frequent reorganization of functional brain networks—even while at rest. In a large study of older adults, greater neural flexibility in the visual network predicted which cognitively normal participants later developed dementia.
These findings suggest that measures of dynamic brain reorganization could become an early biomarker of Alzheimer’s risk. Although the approach remains experimental, the results also emphasize the brain’s ongoing adaptability in the face of neurodegenerative processes.
Key Facts
- Neural flexibility: People with Alzheimer’s disease show more frequent reorganization of brain networks than cognitively normal adults.
- Predictive value: Higher flexibility in visual networks was associated with later transition from healthy cognition to Alzheimer’s.
- Resilience: The brain retains dynamic reconfiguration capacity even during disease progression, offering potential insight into resilience and compensation.
Source: University of Michigan
Overview: A collaborative study from the University of Michigan and Columbia University examined resting-state brain activity and found that some brain regions in people with Alzheimer’s reorganize their functional roles more often than in people without the disease. In a subset of healthy participants, this increased reshuffling in visual regions foreshadowed later development of dementia.
“The brain is constantly forming and reforming functional networks to allocate resources for different cognitive demands,” said Eleanna Varangis, assistant professor at the U-M School of Kinesiology and the study’s first author. “In Alzheimer’s disease we observed more frequent reassignments of brain regions, which suggests this dynamic behavior could carry information about disease status.”
Because early intervention can preserve independence for people at risk of dementia, researchers are seeking reliable early indicators of Alzheimer’s. Functional brain imaging—especially methods that capture dynamics over time—may offer useful biomarkers that appear before clinical diagnosis.
Funded by the National Institutes of Health and the Brain & Behavior Research Foundation, the study analyzed resting-state functional MRI scans from 862 older adults enrolled in the Alzheimer’s Disease Neuroimaging Initiative. Participants were grouped as cognitively normal (CN, n=461), mild cognitive impairment (MCI, n=294), or Alzheimer’s disease (AD, n=107).
The team measured neural flexibility by tracking how often each brain region changed its community assignment across sliding time windows, then normalized that count by the number of possible changes. They calculated a global neural flexibility metric as well as flexibility values for 12 functional networks. Statistical models examined group differences, and survival analysis tested whether flexibility predicted later dementia conversion among participants who were non-demented at baseline.
Key results showed that global neural flexibility was significantly higher in the AD group than in the CN group (β = 0.002, 95% CI 0.001 to 0.004), and that six specific networks also exhibited elevated flexibility in AD. The MCI group differed from the CN group primarily in the visual network, where flexibility was higher.
Of the 617 participants who were non-demented at baseline, 53 (8.6%) converted to dementia over up to 11 years of follow-up—consistent with national prevalence estimates for older adults. Higher neural flexibility in the visual network was associated with an increased risk of conversion to Alzheimer’s (hazard ratio = 1.323 per 1 SD increase in visual-network flexibility, 95% CI 1.002 to 1.747, p = 0.049), after adjusting for age, sex, and education.
Varangis noted that the visual network finding was somewhat unexpected because sensory areas are typically affected later in Alzheimer’s progression. One interpretation is that these relatively preserved regions may show greater reconfiguration as they compensate for or react to pathology occurring elsewhere in the brain.
“High flexibility is usually considered beneficial for adaptability,” Varangis said. “But when flexibility increases during rest in the context of a disease process, it may indicate that networks are being reorganized in response to malfunctioning regions rather than as a healthy adaptive response.”
She emphasized that neural flexibility assessed from resting-state fMRI is still an experimental research tool and is not ready for clinical diagnosis. Nonetheless, the results point to the value of dynamic brain measures and highlight the brain’s continuing capacity to reorganize, which may reflect resilience or compensatory mechanisms even as cognitive decline progresses.
Co-authors on the study include Jun Liu, Yuqi Miao, Xi Zhu, Yaakov Stern, and Seonjoo Lee from Columbia University. The research received support from NIH grants [R01AG062578 (PI: Lee), K01MH122774] and a Brain & Behavior Research Foundation NARSAD Young Investigator Grant (PI: Zhu).
About this Alzheimer’s disease research news
Author: Laura Bailey
Source: University of Michigan
Contact: Laura Bailey – University of Michigan
Image: The image is credited to Neuroscience News
Original Research: Closed access.
“Neural flexibility is higher in Alzheimer’s disease and predicts Alzheimer’s disease transition” by Eleanna Varangis et al., Journal of Alzheimer’s Disease. DOI: 10.1177/13872877251360025
Abstract
Neural flexibility is higher in Alzheimer’s disease and predicts Alzheimer’s disease transition
Background
Neural flexibility (NF) quantifies dynamic functional connectivity and has been linked to psychiatric conditions. Its role in Alzheimer’s disease (AD) had not been systematically examined before this study.
Objective
The study evaluated whether NF differs across clinical stages of AD and whether NF can predict future conversion to dementia.
Method
Researchers analyzed resting-state fMRI from 862 older adults in the Alzheimer’s Disease Neuroimaging Initiative: 461 cognitively normal (CN), 294 mild cognitive impairment (MCI), and 107 AD. NF for each brain node was defined as the proportion of times a node changed its community assignment across sliding windows. Global NF and 12 network-specific NFs were computed. Group differences were examined with linear mixed models, and survival analysis evaluated NF’s predictive utility for dementia transition.
Results
Global NF and NF in six networks were significantly higher in AD compared with CN (β = 0.002, 95% CI 0.001 to 0.004). NF in the visual network was significantly higher in MCI than CN. Among 617 participants who were non-demented at baseline, 53 (8.6%) converted to dementia during follow-up. Higher visual-network NF was associated with greater risk of AD conversion (HR = 1.323 per 1 SD increase, 95% CI 1.002 to 1.747, p = 0.049), controlling for age, sex, and education.
Conclusions
Resting-state neural flexibility was elevated in AD and predicted later dementia transition, suggesting NF may be a useful biomarker of Alzheimer’s disease. Further validation and mechanistic studies are needed to confirm these findings and to clarify their clinical implications.