Using Complex Networks in the Hearing Sciences

Abstract

In this Point of View, we review a number of recent discoveries from the emerging, interdisciplinary field of Network Science, which uses graph theoretic techniques to understand complex systems. In the network science approach, nodes represent entities in a system, and connections are placed between nodes that are related to each other to form a web-like network. We discuss several studies that demonstrate how the micro-, meso-, and macro-level structure of a network of phonological word-forms influence spoken word recognition in listeners with normal hearing and in listeners with hearing loss. Given the discoveries made possible by this new approach and the influence of several complex network measures on spoken word recognition performance we argue that speech recognition measures—originally developed in the late 1940s and routinely used in clinical audiometry—should be revised to reflect our current understanding of spoken word recognition. We also discuss other ways in which the tools of network science can be used in Speech and Hearing Sciences and Audiology more broadly.