An ECoG-Based BCI Based on Auditory Attention to Natural Speech

Abstract

People affected by severe neuro-degenerative diseases (e.g., late-stage amyotrophic lateral sclerosis (ALS) or locked-in syndrome) eventually lose all mus-cular control and are no longer able to gesture or speak. For this population, an audi-tory BCI is one of only a few remaining means of communication. All currently used auditory BCIs require a relatively artificial mapping between a stimulus and a com-munication output. This mapping is cumbersome to learn and use. Recent studies suggest electrocorticographic (ECoG) signals in the gamma band (i.e., 70–170 Hz) can be used to infer the identity of auditory speech stimuli, effectively removing the need to learn such an artificial mapping. However, BCI systems that use this phys-iological mechanism for communication purposes have not yet been described. In this study, we explore this possibility by implementing a BCI2000-based real-time system that uses ECoG signals to identify the attended speaker.