Deep learning in a bilateral brain with hemispheric specialization

The brains of all bilaterally symmetric animals on Earth are are divided into left and right hemispheres. The anatomy and functionality of the hemispheres have a large degree of overlap, but there are asymmetries and they specialize to possess different attributes. Several studies have used computational models to mimic hemispheric asymmetries with a focus on reproducing human data on semantic and visual processing tasks. In this study, we aimed to understand if and how dual hemispheres can improve ML performance at a given task. We propose a bilateral artificial neural network that imitates a lateralization observed in nature: that the left hemisphere specializes in specificity and the right in generalities. We used two ResNet-9 convolutional neural networks with different training objectives and tested it on an image classification task. The bilateral architecture outperformed architectures of similar representational capacity that don't exploit differential specialization. It demonstrated the efficacy of bilateralism and constitutes a principle that could be incorporated into other computational neuroscientific models and used as an inductive bias when designing new AI systems.