MEG in 75 Adults Shows Connectivity Decouples from Volume Loss

By Priya Raman

A study of 75 healthy adults aged 20 to 86 years found that mean resting-state functional connectivity does not track age-related brain volume loss, a dissociation the authors describe in Clinical EEG and Neuroscience, Volume 57, Issue 5 (September 2026, pages 426-437).

The cohort comprised 39 women and 36 men with a mean age of 59.31 years. Each participant underwent a task-free resting-state magnetoencephalography (MEG, a non-invasive recording of magnetic fields produced by synchronized neuronal populations) recording, alongside anatomical MRI (magnetic resonance imaging, used here to quantify whole-brain tissue volume).

What did the investigators measure?

The team computed two principal variables per subject: whole-brain structural volume from MRI and mean functional connectivity derived from MEG signals during rest. Functional connectivity in this context denotes the statistical dependence between time series recorded from different brain regions while the participant lies awake but unengaged in a task.

The cohort spanned 6.6 decades of adult life, a range wide enough to capture middle-aged plateaus and late-life atrophy. The 39:36 women-to-men ratio kept sex relatively balanced, reducing the chance that a single-sex confound would dominate the comparison.

What does the dissociation mean clinically?

The paper's title frames the central finding as a dissociation between functional connectivity and brain volume loss. Two quantities clinicians often assume move together instead follow separate trajectories across aging. A reader in a monitoring unit or research lab should therefore treat resting-state MEG connectivity as an independent biomarker rather than a surrogate for atrophy.

For epilepsy surgery planning, dementia staging, and other neurophysiology applications that weigh structural imaging against functional readouts, the data argue against equating preserved volume with preserved network behavior. The result aligns with prior magnetoencephalography work showing that oscillatory features can remain stable while gray matter thins.

How should the methodology shape interpretation?

MEG differs from scalp EEG in being largely insensitive to skull and scalp as distorting layers, yielding better source localization for deep and tangential sources. It also differs from intracranial electrocorticography (ECoG, a direct cortical recording used in surgical epilepsy patients), which captures signals from the cortical surface but requires surgical access. The current study used non-invasive, task-free acquisition, extending applicability to participants who cannot perform demanding cognitive tasks.

The available abstract excerpt does not specify which frequency bands the analysts evaluated or whether connectivity derived from sensor-level or source-level data. Readers seeking those details will need the complete text in Clinical EEG and Neuroscience.

What remains unresolved?

A cross-sectional design of 75 participants cannot establish whether the connectivity-volume dissociation holds within individuals over time. Longitudinal follow-up would clarify whether participants whose functional connectivity drops sharply also show accelerated atrophy, or whether the two features diverge as a general rule.

The healthy-volunteer sample also leaves open how disease alters the relationship. Comparable MEG and volumetric protocols in temporal lobe epilepsy, mild cognitive impairment, or Alzheimer's disease would test whether pathological states compress or amplify the dissociation seen here.

The September 2026 publication places the work in a peer-reviewed clinical neurophysiology venue, positioning it for citation by groups developing MEG-based biomarkers of cognitive aging. Whether the dissociation survives in larger multicenter samples and across scanner platforms will determine its eventual role in routine clinical reporting.