Why would low acetylcholine during slow-wave sleep matter for memory?
Acetylcholine is a neurotransmitter that helps control the balance between learning new information and recalling old information. During wakefulness and REM sleep, acetylcholine levels are high, which supports encoding new memories. During slow-wave sleep, acetylcholine levels drop, and this change is thought to allow the brain to consolidate and reorganize memories without interference from new sensory input [1][2].
A computational model of neuronal networks shows that when acetylcholine is low, networks shift from asynchronous, high-frequency firing to synchronous patterns of activity. In this low-acetylcholine state, highly connected hub cells that had integrated information during high-acetylcholine states begin to transfer that information to the rest of the network [1]. This state-dependent change in information flow is proposed as a mechanism that could support memory consolidation during sleep, though the model does not specifically address declarative memory transfer from the hippocampus to the cortex.
Does low acetylcholine make the hippocampus replay or retrieve memories?
Another computational model focusing on the hippocampus suggests that low acetylcholine levels favor retrieval over encoding. In simulations, a hippocampal network isolated from external inputs with constant low acetylcholine spontaneously retrieved stored information [2]. This is consistent with the idea that during slow-wave sleep, when acetylcholine is low, the hippocampus may replay or retrieve memories rather than encode new ones.
However, this model does not simulate the transfer of those memories to the cortex, nor does it provide causal evidence that low acetylcholine during slow-wave sleep drives such transfer. It offers early insights into the hippocampal role during states like slow-wave sleep, but the link to declarative memory transfer remains speculative [2].
How do slow waves themselves relate to memory consolidation?
Slow waves in the cortex are known to participate in memory consolidation, but not all slow waves are equal. Research combining recordings and models has identified two types of slow waves: one that occurs in natural slow-wave sleep and another that is ubiquitous under anesthesia. The natural sleep type is more sensitive to external stimuli, which may explain why slow waves in sleep support memory consolidation while those under anesthesia do not [5].
This work also shows that acetylcholine receptors can switch the cortex between these two types of slow waves. By varying the strength of adaptation in excitatory cells using an acetylcholine receptor agonist, researchers demonstrated a neuromodulatory switch between the two slow-wave types [5]. This suggests that acetylcholine levels can shape the dynamics of slow waves in ways that influence memory consolidation, but it does not directly show that low acetylcholine during slow-wave sleep causes the transfer of declarative memories from the hippocampus to the cortex.
What would it take to show that low acetylcholine causes memory transfer?
The evidence so far comes from computational models and indirect experiments. One model proposes a broad circuit involving the medial habenula, interpeduncular nucleus, and median raphe that promotes slow-wave sleep and memory replay, and it lists acetylcholine among several substances that activate this circuit [3]. Another model simulates episodic memory and includes low-acetylcholine states like slow-wave sleep, but it focuses on encoding and retrieval within the hippocampus rather than transfer to the cortex [4].
None of these studies directly manipulate acetylcholine during slow-wave sleep in humans and measure the transfer of declarative memories from the hippocampus to the cortex. To establish causality, researchers would need to selectively alter acetylcholine levels during slow-wave sleep and track memory-related information flow between brain regions. Until then, the idea that low acetylcholine causally supports this transfer remains a plausible hypothesis supported by indirect and modeling evidence, not a proven fact.
Sources used in this answer
Dynamical Mechanism Underlying Scale-Free Network Reorganization in Low Acetylcholine States Corresponding to Slow Wave Sleep
In a computational model, low acetylcholine states corresponding to slow-wave sleep cause a shift in information flow, with hub cells that integrated information under high acetylcholine transferring it to the rest of the network, which may underlie memory consolidation.
Septohippocampal acetylcholine and theta oscillations can modulate memory encoding and retrieval: Insights from a neural masses network
A neural mass model suggests that low acetylcholine levels favor retrieval over encoding in the hippocampus, and a network isolated with constant low acetylcholine spontaneously retrieves stored information, offering insights into slow-wave sleep.
Mechanisms for anesthesia, unawareness, OIRD, sleep and memory replay: MHb→IPN→ PAG + DRN + MRN→claustrum→ cortical slow-waves.
A circuit model proposes that the MHb-IPN-MRN pathway promotes slow-wave sleep and memory replay, and lists acetylcholine among substances that activate this circuit, but does not establish a causal role for acetylcholine in memory transfer.
Acetylcholine-Mediated Neural Mass Model for Episodic Memory
A neural mass model of episodic memory shows that low-acetylcholine states such as slow-wave sleep can be simulated, but it does not address the transfer of declarative memories or provide causal evidence.
Cholinergic switch between two types of slow waves in cerebral cortex
Two types of cortical slow waves exist, with the natural sleep type more sensitive to stimuli and linked to memory consolidation; acetylcholine receptor activation can switch between these types, suggesting cholinergic modulation shapes slow-wave dynamics relevant to memory.