In silico evaluation of sleep as the hebbian reinforcement of the cortical inhibitory synapses
Abstract
Why do we need to sleep a third of our life? How do inhibitory neurons replenish their efficiencysince they lost it just because they are efficient at forbidding their target neurons to activate? An hypothesis has been formulated that answers both questions at the same time. It states that sleep is the period of time when the cortical inhibitory connections are replenished. This "inhibitory synapses replenishment" hypothesis is built upon Hebb's rule which states that the co-activation of both source and target neurons reinforces any shared synapses, including inhibitory ones. As a matter of fact, the depolarization waves observed during Slow Wave Sleep are the ideal candidates for inhibitory replenishment. The cortex, which represents 80% of the volume of the brain, is divided into about 360 cortical maps. We use an in-silico model of a cortical map in order to test the "inhibitory synapses replenishment" hypothesis. Adjusting the simulation results with clinical data allows to extract a few impportant coefficients such as the rate of inhibitory replenishing associated to slow wave sleep (SWS) and the rate of excitatory long distance replenishing associated to random eyes movement (REM) sleep. Our simulations also show that the best recognition performances require SWS and REM sleeps, in a ratio and a number of cycles per night that conform to clinical observations.
Related
- Memristive synapses and memcapacitive autapses in hippocampal-cortical neural networks: multi-stability, chaos, and applications in image encryption
- WNK-dependent phosphorylation of gephyrin tunes GABA A receptors at inhibitory synapses and modulates anxiety behavior
- Hebbian learning generates individuality in spatial decision-making
- LC–MS-Based Metabolomic Profiling of Excoecaria agallocha Leaf Extract: Integrated Antioxidant, Anti-Inflammatory, Cholinesterase Inhibitory and Neuroprotective Evaluation with AChE-Targeted Molecular Docking
- Constrained Hebbian learning supports efficient representational allocation under structural constraints.
BibTeX
@misc{touzet2026silico,
title = {{In silico evaluation of sleep as the hebbian reinforcement of the cortical inhibitory synapses}},
author = {Claude Touzet},
howpublished = {HAL},
year = {2026},
month = sep,
url = {https://hal.science/hal-05715874v1},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - UNPB
TI - In silico evaluation of sleep as the hebbian reinforcement of the cortical inhibitory synapses
AU - Claude Touzet
PY - 2026
DA - 2026/09/07
PB - HAL
UR - https://hal.science/hal-05715874v1
AB - Why do we need to sleep a third of our life? How do inhibitory neurons replenish their efficiencysince they lost it just because they are efficient at forbidding their target neurons to activate? An hypothesis has been formulated that answers both questions at the same time. It states that sleep is the period of time when the cortical inhibitory connections are replenished. This "inhibitory synapses replenishment" hypothesis is built upon Hebb's rule which states that the co-activation of both source and target neurons reinforces any shared synapses, including inhibitory ones. As a matter of fact, the depolarization waves observed during Slow Wave Sleep are the ideal candidates for inhibitory replenishment. The cortex, which represents 80% of the volume of the brain, is divided into about 360 cortical maps. We use an in-silico model of a cortical map in order to test the "inhibitory synapses replenishment" hypothesis. Adjusting the simulation results with clinical data allows to extract a few impportant coefficients such as the rate of inhibitory replenishing associated to slow wave sleep (SWS) and the rate of excitatory long distance replenishing associated to random eyes movement (REM) sleep. Our simulations also show that the best recognition performances require SWS and REM sleeps, in a ratio and a number of cycles per night that conform to clinical observations.
ER -