Neural Criticality: How Sleep Calibrates Brain Circuits to Prevent Neurodegeneration

WashU biologist Keith Hengen and physicist Ralf Wessel delivered the first direct evidence for that mechanism in 2024. A new five-year, $2.7 million NIH grant with Stanford's Luis de Lecea now extends the framework into Alzheimer's disease.

The Criticality Lens

Hengen's group treats the brain as a dynamical system under constant load. Wakefulness nudges networks away from criticality. Sleep restores it. The 2024 study used chronic electrophysiology to track the same neurons across thousands of sleep-wake cycles. Result: sleep functions as a calibration routine, not a passive recharge.

For circuit researchers, the operational takeaway is direct. A network that drifts off criticality loses computational power. Quantify the drift and you have a measurable biomarker. Hengen's framing — running a car in the wrong gear until the engine fails — maps cleanly onto the experimental logic. Degradation is slow, but the early signal is detectable.

The Alzheimer's Probe

The NIH award funds a long-duration study in mice engineered for high Alzheimer's risk. The team will record neural activity across disease progression and correlate three variables: sleep architecture, criticality state, and cognitive performance. According to Hengen, breakdown in criticality appears to unify at least four molecularly distinct neurodegenerative diseases. Different upstream pathways. Same downstream circuit failure.

If the framework holds, the clinical window opens before symptoms. Predict vulnerability from subtle shifts in circuit dynamics. Intervene early enough to matter.

A Second Recording Site

Hengen has also launched a parallel project with Aya Takeoka at the RIKEN Center for Brain Science, funded through the Human Frontier Science Program. The question: does the spinal cord cycle through a sleep-like state? If yes, the criticality framework generalizes beyond the brain. If no, the brain-specific reset is itself a finding worth isolating.

The methodological hook mirrors the WashU work. Months-long recordings from identified neurons. Cross-condition comparisons. No shortcuts through population averages.

What to Track

• Criticality drift in Alzheimer's-risk mice: which circuit metrics break first, and at what disease stage.
• Sleep-intervention outcomes: whether targeted modulation pulls criticality back toward baseline in declining circuits.
• Spinal-cord recordings from Takeoka's lab: first datasets will determine whether non-brain sleep exists as a measurable state.