Scientists have questioned one of the main theories of memory
For a long time, neuroscientists believed that long-term memory depends primarily on the strength of connections between brain neurons. However, a new study by scientists from the Okinawa Institute of Science and Technology, published in the journal Live Science, showed that even after artificial hibernation, during which more than half of the synapses disappeared in the hippocampus of mice, the animals did not forget the previously learned information. The authors of the work believe that memories can be stored differently than previously thought, and the discovery itself can change the approach to the study of memory and neurodegenerative diseases. For more information, see the Izvestia article.
Artificial hibernation preserved memories contrary to expectations
The study began with a question that has long interested neuroscientists: how does the brain retain memories if its structure is constantly changing. It is known that neural connections do not remain unchanged — they form, disappear and rebuild throughout life. However, until now it was believed that strong synaptic connections are the basis of long-term memory.
To test this hypothesis, scientists from the Okinawa Institute of Science and Technology used an unusual model. They put the mice in a state of artificial hibernation, torpor, in which body temperature and metabolism decrease significantly. This condition is accompanied by large—scale brain restructuring and the loss of a large number of synapses in the hippocampus, an area responsible for learning and memory.
Before the experiment, the mice were taught several tasks. Some animals memorized the location of the exit in the maze, while others learned to avoid places where they had previously received a weak electrical discharge. After coming out of artificial hibernation, the researchers tested the animals' memory again.
The result was unexpected. Despite the fact that more than half of the synaptic connections in the hippocampus had disappeared, the mice still remembered the way through the maze and avoided the dangerous area. In other words, such a large-scale restructuring of the brain did not lead to the loss of already formed memories.
For comparison, the scientists created a control group in which a similar loss of synapses was caused by anesthesia and chemicals. In this case, the animals' memory actually deteriorated. It was the differences between the two groups that led the researchers to assume that during the natural restructuring of the brain, there are mechanisms that protect memories.
Not individual neurons, but the entire network: which theory of memory has been questioned
The results call into question one of the most common memory storage models. According to classical concepts, memories are fixed due to the strengthening of individual synaptic connections between neurons. If a significant part of these connections disappears, the memory should weaken.
However, the study showed a different picture: despite the loss of a large number of synapses, the overall organization of the neural network remained relatively stable. The researchers concluded that long-term memories may depend not so much on individual connections as on the structure of the entire network.
The scientists paid special attention to the so—called synaptic hubs, areas where one neuron connects to several others at once or where groups of dendritic spines are located nearby. It was these structures that were more often preserved after artificial hibernation and could serve as a kind of "framework" around which the brain restored lost connections.
The authors suggest that memory may be stored not in individual neurons or even in specific synapses, but in the overall topological architecture of a neural network. In other words, it is not each connection that is important individually, but how the entire system of brain cell interaction is organized. At the same time, the researchers themselves emphasize that the work does not refute existing ideas about the role of synapses. Rather, it shows that the mechanisms of memory formation and storage may be much more complex than previously thought.
From Alzheimer's disease to artificial intelligence: why this discovery is needed
So far, the findings of the study relate only to laboratory mice, so they cannot be directly transferred to humans. If the hypothesis is confirmed in further research, it may change the approach to the study of Alzheimer's disease and other neurodegenerative diseases in which neural connections are gradually destroyed. Perhaps the key role is played not by the number of preserved synapses, but by the brain's ability to maintain the overall organization of the neural network.
The authors also note that the principles of sustainable organization of neural networks can be useful far beyond the boundaries of neuroscience. In their opinion, understanding how the brain retains information, despite constant restructuring, can help in the development of new methods of data storage and more sustainable artificial intelligence systems.
The study does not provide a definitive answer to the question of how exactly the brain stores memories. However, it shows that memory is probably much more resistant to change than previously thought. If this concept is confirmed in future work, scientists may have to reconsider one of the fundamental principles of modern neuroscience.
Переведено сервисом «Яндекс Переводчик»
