Groundlessly Boasting Mouse Hibernation Study At Least Reveals the Short Lifetimes of Synapses and Dendritic Spines

 The recent paper "Artificial hibernation reveals synaptic engram architecture associated with memory retention" is another neuroscience paper making a groundless boast in its title. The authors did nothing at all to reveal a "synaptic engram architecture associated with memory retention."  But this weird study involving putting mice into a kind of artificial hibernation does at least reveal something that helps to debunk all claims of synaptic memory storage. 

The final version of the paper is behind a paywall. But when I search for the paper on Google Scholar, it gives me a preprint as the corresponding .pdf file. The preprint is by the same authors, and has an almost identical title, a title of "Artificial hibernation uncovers distinct synaptic engram architecture for memory retention." So I can use that preprint to analyze the research that went on here.  My use of "the paper" below refers to the preprint. 

The paper is one that has many untrue or groundless statements that reflect the author's dogmas and "improperly jumping to conclusions" thinking. Any statements that the authors make using the words "engrams" or "representations" are groundless. These days cognitive neuroscientists are misusing those words very abundantly. Being guilty of "see what you are hoping to see" pareidolia, neuroscientists are frequently claiming without any good warrant to have seen "representations" of this or that in the brain. It's a "Jesus in my toast" kind of affair.  Similarly, whenever neuroscientists use the term "engram" these days, they are making groundless claims in which some cells or synapses are called "engrams," despite a lack of any decent evidence that any such thing as memory storage has occurred in such cells. "Engram" is a word referring to a claimed place of memory storage in the brain. 

The paper is guilty of very bad methodological sins. The study group sizes were ridiculous, consisting of way-too-small study group sizes such as only three mice, only four mice, only five mice, and a study group size imprecisely described as "6-10 mice for each group."  No study like this should be taken seriously as experimental evidence unless it used a study group size of at least 15 or 20 animals per study group. The authors have used the worthless "freezing behavior" method of trying to judge memory recall, which means that nothing they claim about experimentally demonstrating memory retention should be regarded as reliable. Read my post here for why such a method is worthless as a technique for measuring whether a rodent recalled anything. 

But we do get something of value from the paper: some observations of how unstable and short-lived are synapses and dendritic spines. These are observations which help to discredit the popular but groundless dogma that memories are stored in synapses of the brain, which tend to protrude from dendritic spines. 

A dendritic spine is a tiny protrusion from one of the dendrites of a neuron. The diagram below shows a neuron in the top half of the diagram. Some dendritic spines are shown in the bottom half of the visual. The bottom half of the visual is a closeup of the red-circled part in the top of the diagram. 

Neuroscientists have often speculated that dendritic spines could be some type of memory storage system, although such a speculation is silly. Dendritic spines no more resemble a memory storage system than do the little twigs on the branch of a tree. And while memories can last decades, dendritic spines are very unstable, typically having a lifetime of only weeks or months.  

We read this (the "QIH" refers to a roughly 48-hour period of artificial hibernation produced by the experimenters):

"We also evaluated the similarity of the locations of all dendritic  spines before and after QIH. The similarity in spine locations between Day 1 and 3, with QIH in between, was 0.91 on average, which was not significantly different from the similarity between Day 3 and 7 without QIH (0.88)."

So in a mere two days or four days, the dendritic spines on the mice lost about 9% of their previous appearance. You can extrapolate from that to get a dendritic spine average lifetime of only several weeks, which is consistent with previous observations by others reported here. Because synapses are typically attached to dendritic spines, anything we learn about the short lifetime of dendritic spines is something also telling us about the short lifetime of synapses. 

The paper also tells us something about the lifetime of synapses. It states this, referring to a hibernation period of only two days. 

"In the current study, we used artificial hibernation in mice to induce extreme downscaling of the neuronal activity and dendritic structures in the hippocampus. During this hypothermic and hypometabolic state, hippocampal neurons show a ~70% reduction in firing rate and the elimination of more than 50% of synapses." 

This is a devastating answer for any one believing that memories are stored in synapses. Extrapolating from such a rate of decay, we would guess that synapses have an average lifetime of only a few days or maybe a few weeks. 

It has long been known that synapses are built from proteins that have very short average lifetimes of only a few weeks.   Richard Huganir led a study that was specifically dedicated to trying to find long-lived proteins in synapses. None were found.  The scientific paper found no such thing. Quite to the contrary, the paper found the following:

• Studying thousands of brain proteins, the study found that virtually all proteins in brains are very short-lived, with half-lives of less than a week.
  
• Table 2 of the paper gives specific half-life estimates for the most long-lasting brain proteins, and in this table only 10 out of thousands of brain proteins had half-lives of 10 days or longer.
  
• Of the proteins whose half-life is estimated in Table 2, only one of them has a half-life of longer than 30 days, that protein having a half-life of only 32 days.
  
• A graph in the paper indicates that none of the synapse proteins had a half-life of more than 35 days.
  

Below is a graph from the Huganir paper. It shows that the study found that virtually all proteins in synapses are very short-lived.

Below is another graph from the same paper. It shows that the study found that virtually all proteins in synapses are very short-lived.

Judging from these graphs, none of the proteins found had a half-life of longer than 35 days, and only a few had a half-life of more than 14 days.

The 2025 paper here states this:

"We measured the lifetime of the major excitatory synaptic scaffold protein PSD-95 over a range of spatial scales from brain regions to single synapses. PSD-95–HT lifetimes ranged from 11 to 14 days, depending on the brain region (Fig. 2f,g)."

So if synapses are made from proteins with short lifetimes of only a few days or weeks, we should not be surprised that a researcher would find that about 48 hours would produce a 50% reduction in synapses. Most synapses are attached to dendritic spines. So the previous result about the short lifetime of dendritic spines reinforces the conclusion that synapses have short lifetimes. 

The paper here states, "Experiments indicate in absence of activity average life times ranging from minutes for immature synapses to two months for mature ones with large weights."  The paper suggests that even larger synapses last only two months.  A 2025 paper states that  "the synaptic turnover rate is as high as 1% per day in the visual cortex." Such a rate of turnover is equivalent to a synaptic lifetime of only about a year. Another paper  states that in  hippocampal CA1 cells the synapses have an  estimated lifetime of only 1–2 weeks.

The lifespan of synapses is of very great relevance to whether there is any credibility in the claim that synapses store memories. Old humans can remember very well many things they learned and experienced 50 years ago. But if synapses have short lifetimes, they cannot possibly be a storage place of knowledge humans learned 50 years ago,  and synapses cannot be a storage place of memories humans had 50 years ago.  If synapses have short lifetimes, the synaptic theory of memory is untenable. 

The issue that synapses are made of proteins with short lifetimes has been known for decades. So why on Earth do so many neuroscientists keep claiming that memories are stored in synapses? It's simply that they have no other alternative theory of brain-stored memories that is any better.  Nothing in the brain looks anything like a system for storing, preserving or retrieving learned information or experiences an organism had.