Simulated neurons show that learning is easier than you think

An international team, led by UdeM computational neurologist Elif Mueller, simulates how synapses change in the neocortex to better understand how we learn.

Everyone knows that the human brain is extremely complex – but how exactly does one learn? Well, the answer may be much simpler than is usually thought.

An international research team, including the Université de Montréal, has made great strides in accurately simulating synaptic changes in the neocortex that are considered key to learning, opening the door to better understanding of the brain.

The researchers’ study – including an open source model – was published on June 1 Natural communications.

“World of new directions”

“This opens up a world of new research guidelines for how we learn,” said Elif Mueller, assistant professor of research at IVADO at UdeM and CIFAR’s Department of Artificial Intelligence, who leads research at the École’s Blue Brain project. polytechnique fédérale de Lausanne (EPFL), in Switzerland.

Mueller moved to Montreal in 2019 and continues his research at the Laboratory for Biological Education in Architecture, which he founded at the CHU Sainte-Justine Research Center in collaboration with UdeM and Mila, the Institute of Artificial Intelligence in Quebec.

“Neurons are shaped like trees, and synapses are the leaves on their branches,” said Mueller, co-author of the study.

“Previous approaches to model plasticity have ignored this tree structure, but we now have the computational tools to test the idea that synaptic interactions of branches play a key role in guiding learning. in vivo,” he said.

“This has important implications for understanding the mechanisms of neurological disorders such as autism and schizophrenia, but also for developing powerful new AI approaches inspired by neuroscience.

Associates in five countries

Mueller is collaborating with a group of scientists from the EPFL Blue Brain Project, the Université de Paris, the Hebrew University of Jerusalem, the Instituto Cajal (Spain) and Harvard Medical School to come up with a synaptic plasticity model in the neocortex based on limited data. postsynaptic dynamics of calcium.

How it works? It’s complicated – but ultimately simpler than you think.

The brain is made up of billions of neurons that communicate with each other, forming trillions of synapses. These points of connection between neurons are complex molecular machines that are constantly changing as a result of external stimuli and internal dynamics, a process commonly called synaptic plasticity.

In the neocortex, a key area related to the learning of high-level cognitive functions in mammals, pyramidal cells (PCs) account for 80 to 90 percent of neurons and are known to play a major role in learning. Despite their importance, the long-term dynamics of their synaptic changes have been experimentally characterized between only a few types of personal computers and have been shown to be diverse.

Only limited understanding

As a result, there is only a limited understanding of the complex neural circuits they form, especially in the stereotypical cortical layers that dictate how different regions of the neocortex interact. The innovation of Mueller and his colleagues was to use computational modeling to come up with a more comprehensive look at the dynamics of synaptic plasticity driving learning in these neocortical chains.

Comparing their results with the available experimental data, they showed in their study that their model of synaptic plasticity can capture the diverse dynamics of plasticity of the different computers that make up the neocortical chip. And they did so using only one unified set of model parameters, which shows that the rules of plasticity of the neocortex can be shared between pyramidal cell types and thus be predictable.

Most of these plasticity experiments were performed on rodent brain slices IVFwhere the dynamics of calcium driving synaptic transmission and plasticity are significantly altered compared to learning in intact brain in vivo. It is important that the study predicts qualitatively different dynamics of plasticity from the conducted reference experiments IVF. If confirmed by future experiments, the implications for our understanding of plasticity and learning in the brain will be profound, Mueller and his team say.

“The exciting thing about this study is that it’s further confirmation for scientists that we can bridge the gaps in experimental knowledge using a modeling approach to brain research,” said EPFL neurologist Henry Markram, founder and director of the Blue Brain project.

This is open science

“In addition, the model is open source, available on the Zenodo platform,” he added.

“Here we have shared hundreds of plastic pyramidal cell connections of different types. Not only is this the most widely validated model of plasticity to date, but it is also the most comprehensive prediction of the differences between the plasticity observed in a petri dish and intact brain.

This leap is possible thanks to our joint team-scientific approach. In addition, the community can continue it and develop its own versions by modifying or adding it – it is an open science and will accelerate progress.

Help: Chindemi G, Abdellah M, Amsalem O, et al. Calcium-based plasticity model for predicting long-term potentiation and depression in the neocortex. Natural communications. 2022; 13 (1): 3038. doi: 10.1038 / s41467-022-30214-w

This article has been republished from the following materials. Note: The material can be edited for length and content. For more information, please contact the cited source.

Leave a Comment

Your email address will not be published.