Plate Nº 39 · recorded November 17, 2026
Neuroscience & MindReported finding
Brain Wiring Pattern May Flag Alzheimer's Risk Years Early
About 26 percent of cognitively healthy seniors carry an Alzheimer's-like brain-wiring pattern that predicted faster decline and plaque buildup over five to 15 years, a Science Advances study found.
By Nathan Brooks5 min read951 words
In brief
- About 26 percent of cognitively healthy older adults showed an Alzheimer's-like brain connectivity pattern on MRI.
- The connectome signature distinguished SuperAgers from Alzheimer's patients with 80 percent accuracy (71 percent in an independent cohort).
- Alzheimer's-like subgroups declined faster over five to 15 years of follow-up and accumulated amyloid plaques more quickly.
- The score added predictive value beyond PET biomarkers, demographics, and the APOE risk gene.
- The study was published in Science Advances by researchers at Stanford University School of Medicine.

About one in four cognitively healthy older adults — roughly 26 percent — carries a brain-wiring pattern resembling Alzheimer's disease, and that group declined faster over follow-ups stretching up to 15 years, according to a study published in Science Advances.
The signature, called the Alzheimer's Resilience-Vulnerability Connectome (ARC), was built by contrasting SuperAgers — adults over 75 with memory as sharp as typical middle-aged people — against diagnosed Alzheimer's patients. When researchers applied it to seniors with no cognitive complaints, those with the Alzheimer's-like pattern later dropped faster on memory, attention, and daily-functioning tests, built up amyloid-beta plaques more quickly, and progressed more often to mild cognitive impairment or dementia.
How do you build a brain-wiring signature?
The team, led by Kanhao Zhao and senior author Yu Zhang of Stanford University School of Medicine, analyzed three public brain-aging databases and posted their analysis code online. They used resting-state functional MRI, a non-invasive scan in which a person lies quietly in a magnetic scanner for several minutes while the machine tracks rises and falls in blood oxygen. Regions whose signals fluctuate together count as functionally connected.
Earlier SuperAger studies compared sharp seniors against typical older adults, a mix that can blur signals because some "typical" seniors already harbor silent Alzheimer's pathology. So the researchers anchored their model at the extremes first: 180 SuperAgers versus 110 Alzheimer's patients, both groups with a median age of 80, drawn from the Open Access Series of Imaging Studies and the Harvard Aging Brain Study.
Dividing the brain's outer layer into 100 regions, they measured synchronized activity between every pair of regions and trained a machine-learning model to tell the two groups apart. It reached 80 percent accuracy on held-out data, well above the 62 percent a model would score by always guessing the larger group. Applied unchanged to an independent replication cohort of 64 SuperAgers and 79 Alzheimer's patients from the Alzheimer's Disease Neuroimaging Initiative, it hit 71 percent accuracy.
SuperAgers showed stronger synchrony across frontal and parietal regions supporting attention and self-control. Alzheimer's patients showed heightened connectivity in the right insula, a folded region involved in emotion and bodily awareness, and at the front tips of the temporal lobes.
What did the healthy seniors reveal?
Applied to 818 unimpaired adults in the discovery group and 497 in the replication group, the model split them cleanly: about 26 percent carried the Alzheimer's-like pattern, three-quarters the SuperAger-like one. At the first visit, the two subgroups scored almost identically on thinking tests, and baseline PET scans showed no significant differences in amyloid or tau, the two hallmark Alzheimer's proteins.
Over five to 15 years of follow-up, their paths diverged. The Alzheimer's-like subgroup worsened faster in overall clinical functioning, memory, attention, and executive function, and a larger share progressed to mild cognitive impairment or dementia. The gap held after adjusting for starting age, baseline scores, and amyloid and tau levels — and it appeared even among seniors already amyloid-positive at baseline. Repeated PET scans showed the Alzheimer's-like group accumulated plaques faster, especially in frontal and temporal regions, though tau differences weren't significant.
The score also tracked predicted brain age — with more Alzheimer's-like scores matching a wider gap between biological and calendar age — plus greater apathy, depressive symptoms, and difficulty with everyday tasks like paying bills. Adding the connectome score improved progression predictions beyond demographics, PET biomarkers, and APOE, the best-known Alzheimer's risk gene.
Grouping the same adults by test scores rather than connectivity produced clusters that differed initially but not in later decline — suggesting the wiring pattern captured something tests missed. An exploratory genetics screen of 506 adults, scanning about 3.4 million DNA variations, found no genome-wide-significant hits but three suggestive signals near CLYBL and FRMD6, genes tied in cellular studies to metabolism, inflammation, and nerve-cell structure.
What do outside experts say?
"This study shows that our brain's functional 'wiring diagram' holds early warning signs for dementia that traditional cognitive tests simply miss," Yoshiyuki Hirano, a professor at Chiba University who was not involved, told PsyPost. "By benchmarking brain networks between SuperAgers and Alzheimer's patients, the authors developed a tool that uses a single baseline MRI scan to unmask silent vulnerability in healthy seniors."
Christian Habeck of Columbia University, also unaffiliated, praised the validation strategy despite his usual skepticism of resting-state fMRI. "I was struck by the intuitively appealing approach regarding statistical robustness and held-out validation," Habeck said. "The methodological rigor increases confidence that the reported effects are not merely artifacts of over-fitting."
What are the limits?
The study is observational, so it can't show whether altered network activity drives decline or reflects damage already underway. At 71 percent accuracy in the independent cohort, the signal is reproducible at group level but too overlapping for individual diagnosis. "This framework represents a research tool for statistical risk stratification across populations, not a standalone clinical test indicating that dementia is inevitable," Hirano said.
Habeck added that even a predictive classifier isn't proof of cause: connectivity patterns and cognitive outcomes could both reflect a third underlying process. The signature also used a slightly relaxed SuperAger definition (age 75 rather than 80), drew only on resting-state fMRI without nerve-fiber wiring or vascular markers, and the amyloid link may partly reflect disease stage.
Next steps include validating the score in larger, more diverse cohorts, testing the CLYBL and FRMD6 candidates, and using the signature to recruit high-risk volunteers for preventive drug trials before symptoms begin.
via psypost.org (Source)
More from Nathan Brooks
Show full bio
Market editor covering companies, deals and strategy in general science news at SciBeat.
456 articles
Nearby plates
- Brain scan score linked to later Alzheimer’s-related decline
- Alzheimer's Reshapes Brain Communication Before Symptoms Appear
- Brain Scans Reveal Which Disorders Are Linked to Faster Brain Aging
- New Roadmap Aims to Make Brain Resilience Research More Inclusive
- Brain Structure Changes May Signal Alzheimer's Years Before Plaques