Animal brains have somewhat mirror-symmetric neural networks, and asymmetries are hypothesized to be more common in species with higher cognitive abilities. This assumption is based on a well-established hypothesis that more sophisticated neural tasks have the ability to transform mirror-symmetric neural circuits into circuits that exist only on one side of the brain. A mathematical model created by Lus Seoane at the National Center for Biotechnology in Spain has now been used to support this theory. The study's findings may shed light on how cognitively demanding tasks, as well as disease or aging, can change the structure of the brain.
A mirror-symmetric neural network is useful when editing mirror-symmetric body components such as arms and legs. Additionally, having redundant circuits on both sides of the brain can improve computational accuracy and provide a backup in case one of the circuits fails. However, the redundant nature of this type of replication may result in higher energy usage. This trade-off raises a very important question: Does the complexity of the cognitive processes carried out by the neural network affect the ideal level of mirror symmetry?
According to Seoane's hypothesis, a neuronal circuit should always be completely mirror symmetric or completely localized to one side of the brain. More importantly, it shows how an increase in job difficulty can cause a shift between these two arrangements. This change may occur as biological evolution progresses or as the brain changes with age. Although this study focused on mirror symmetry in the brain, the findings apply to any group of duplicated neural circuits, such as those located on the same side of the brain, according to Seoane.
Source: physics aps org
📩 14/09/2023 09:07