The brain is a complex network that contains billions of neurons, each communicating with thousands of others through their synapses (connections). However, the neuron actually collects its many synaptic incoming signals through several extremely long-branched "arms" called dendritic trees.
With new theoretical results and experiments on neuronal cultures, a group of scientists headed by Prof. Ido Kanter of the Department of Physics and the multidisciplinary brain research center Gonda (Goldsmith) at Bar-Ilan University has shown that the central assumption for Almost 70 years that learning occurs only in synapses is wrong.
In an article published today in the journal Scientific Reports researchers contradict conventional wisdom to show that learning is actually performed by multiple dendrites, similar to the slow learning mechanism currently attributed to synapses becomes.
"The newly discovered process of learning in the dendrites is much faster than in the old scenario, suggesting that learning takes place only in the synapses In this new dendritic learning process there are some adaptive parameters per neuron, in comparison to Thousands of small and sensitive in the synaptic learning scenario, "said Prof. Kanter, whose research team includes Shira Sardi, Roni Vardi, Anton Sheinin, Amir Goldental and Herut Uzan.
The newly proposed learning scenario suggests that learning takes place in some dendrites that are much closer to the neuron than in the previous one. "Does it make sense to measure the quality of the air we breathe over many tiny, remote satellite sensors at the height of a skyscraper, or by using one or more sensors near the nose? Similarly, it is more efficient for the neuron his incoming signals close to his arithmetic unit, the neuron, "says Kanter. Hebb's theory has been so deeply rooted in the scientific world for 70 years that no one has proposed such a different approach. In addition, synapses and dendrites are connected in series with the neuron, so that the exact localized location of the learning process appeared irrelevant.
Another important finding of the study is that weak synapses, previously thought to be insignificant, although they play the majority of our brain play an important role in the dynamics of our brain. They induce oscillations of the learning parameters rather than driving them to unrealistic fixed extremes, as suggested in the current synaptic learning scenario.
The new learning scenario occurs in different parts of the brain and therefore requires a reassessment of current treatments for brain dysfunction. Therefore, the popular phrase "neurons firing a wire together" that summarizes the 70 year old hypothesis of Donald Hebb needs to be reformulated. In addition, the learning mechanism forms the basis for advanced advanced machine learning and deep learning achievements. Changing the learning paradigm opens new horizons for different types of deep learning algorithms and artificial intelligence based applications that mimic our brain functions, but with advanced features and much faster.
When published, the paper will be available online at www.nature.com/articles/s41598-018-23471-7.
Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of the news posted on EurekAlert! by contributors or for the use of any information by the EurekAlert system.