Sleep restores an optimal computational regime in cortical networks

Area of Science:

• Neuroscience
• Computational Neuroscience
• Sleep Science

Background:

• Sleep's homeostatic functions are critical for brain health, but the precise neural set point remains elusive.
• Slow-wave activity (SWA) reflects sleep pressure but doesn't fully explain the necessity of sleep.
• Understanding the computational basis of sleep's restorative function is a key neuroscientific challenge.

Purpose of the Study:

• To investigate whether criticality, a state of optimal information processing in neural networks, serves as the computational set point for sleep.
• To determine if sleep actively restores critical dynamics disrupted during waking.

Main Methods:

• Continuous cortical neuron activity recording in freely behaving rats over 10-14 days.
• Analysis of neural activity to assess network dynamics and criticality.
• Correlation of criticality deviations with sleep/wake behavior and other neural measures.

Main Results:

• Waking experience progressively disrupts network criticality in a context-dependent manner.
• Sleep effectively restores these critical dynamics, indicating a restorative function.
• Deviations from criticality more accurately predict subsequent sleep/wake behavior than SWA or other neural metrics.

Conclusions:

• Criticality represents the computational set point that sleep tunes.
• The perturbation and recovery of criticality constitute a network homeostatic mechanism.
• This finding supports the core restorative function of sleep at a computational level.