Discovering the Neuroanatomical Correlates of Music with Machine Learning

Abstract

Music is ubiquitous in our lives yet unique to humans. Over the past decades, a growing body of literature has revealed the neural and computational underpinnings of music processing including not only sensory perception (e.g., pitch, rhythm, and timbre) but also local/non-local structural processing (e.g., melody and harmony). This chapter reviews the neural correlates of unsupervised learning with regard to the computational and neuroanatomical architectures of music processing. Further, we offer a novel theoretical perspective on the brain’s unsupervised learning machinery that considers computational and neurobiological constraints, highlighting the connections between neuroscience and machine learning.