Scaling and tuning to criticality in resting-state human magnetoencephalography
New analysis of human brain activity shows robust scaling laws and points to non-invasive E/I balance measurement.
A research team including Irem Topal, Fabrizio Lombardi, and colleagues has published a groundbreaking study in quantitative biology, analyzing human brain activity through magnetoencephalography (MEG). Their work demonstrates that the resting human brain exhibits robust scaling behaviors across different coarse-graining scales, with exponents remarkably close to those previously measured in populations of spiking neurons in animal models. The researchers applied a renormalization group-inspired coarse-graining approach to large-scale electrophysiological recordings, showing that dynamics of neuronal avalanches—scale-free cascades of neural activity—remain invariant under this transformation.
Technical analysis using a non-equilibrium adaptive Ising model, inferred directly from the MEG data, indicates these scaling behaviors emerge when brain networks operate close to criticality. Crucially, the study reveals this critical tuning depends on the precise balance between excitation and inhibition (E/I) within neural circuits. The model successfully reproduces a wide repertoire of resting-state brain dynamics observed in the MEG recordings, providing a computational bridge between microscopic neuronal interactions and macroscopic brain signals.
This research extends previous observations from small-scale animal studies to the whole human brain, suggesting common scaling laws may operate across mammalian species. The most significant practical implication is methodological: the approach opens a pathway toward developing robust, non-invasive techniques for estimating E/I balance. Currently, measuring this fundamental property of neural circuits requires invasive procedures, limiting research and clinical applications. The ability to assess E/I balance non-invasively could transform research into neurological and psychiatric disorders where excitation-inhibition imbalance is implicated, including epilepsy, autism spectrum disorders, and schizophrenia.
- Study of human MEG data shows brain activity follows scaling laws across coarse-graining scales, with exponents matching animal neuron data
- Adaptive Ising model simulations indicate scaling emerges near criticality and depends on excitation/inhibition balance
- Research enables potential non-invasive measurement of E/I balance, a key metric currently requiring invasive methods
Why It Matters
Enables non-invasive measurement of neural excitation/inhibition balance, potentially transforming research into epilepsy, autism, and schizophrenia.