Neuroscientists gain insights into memory formation

The researchers hope to bring more understanding to how memory functions Keystone

Swiss researchers have developed a model that better explains how the human brain creates, stores and recalls memories. 

The model, published in Nature Communications, concerns synaptic plasticity – the technical term for the way the brain’s network of electrical connections between neurons (synapses) is continually updated.

When a connection occurs through a synapse, an electrical signal is sent from one neuron to another. Scientists have long theorised that changes in the speed, frequency and patterns of these signals form the basis of learning in the brain. It seemed that synapses were strengthened when they fired in sync with one another, but this was not enough to successfully simulate memory formation – a piece of the puzzle was still missing.

Now, researchers at the Swiss Federal Institute of Technology in Lausanne (EPFL) have developed an improved model of this process, focusing on “memory assemblies”. The idea is that the building blocks of memory are stored piece-by-piece in collections of neural networks. They can then be selectively activated and combined to form whole memories.

This theory is novel because it suggests that the strength or weakness of a synapse can be indirectly influenced by other synapses in a neural network. It introduces a much more complex angle to the original theory of synaptic plasticity.

The EPFL researchers used the memory assembly concept to develop an algorithm that can simulate memory formation with unprecedented accuracy. With these findings, a better grasp on how memories are formed could lead to an improved understanding of the brain as a whole.