Fast periodic auditory stimulation reveals a robust voice-selective electrophysiological response in the human brain

Several studies have demonstrated the existence of voice preferring regions in the human brain. However, whether this preference is merely driven by low level acoustic properties peculiar of voices, or whether it reflects a higher-level categorical response is still under debate. Here, we propose a new approach to address this question in an objective, direct, fast and automatic way. We combined electroencephalographic recording with a Fast Periodic Auditory Stimulation (FPAS) oddball paradigm to investigate voice categorisation processes. Participants were tested with 3 types of sequences: sequences containing heterogeneous vocal and non-vocal sounds from different categories, sequences created with frequency scrambled sounds to control for frequency content and, finally, sequences containing voices and instruments matched in pitch and harmonicity-to-noise ratio. Our results show robust voice selective brain responses over superior temporal electrodes that cannot be explained by frequency content nor harmonicity typical of voice samples alone. Moreover, our FPAS paradigm allowed us to characterize voice selective responses with a high signal to noise ratio in a very short acquisition time (4 minutes) in the vast majority of individual participants, suggesting that FPAS could be a powerful tool to investigate voice selectivity in children and in clinical populations