For over 50 years, scientists have been intrigued by how we recognize voices. Our recent study explores how emotional prosodies affect this process. Despite differences in prosodies, we found that the overall old-new effect is robust, indicating that our brains consistently distinguish familiar and unfamiliar voices.
Current experimental challenges in speaker identification arise from the commonly used training testing paradigm in the past, which is not suitable for studying detailed variables affecting speaker identity processing. Ensuring that the training is thorough and effective is crucial to address this challenge. Our fundings reveal that while the LPC, or the old-new effect, shows differences between confident and doubtful processes, it remains robust overall.
Our paradigm is well suited for investigating the processing of familiar and unfamiliar objects with internal variations. My lab focuses on voice perception, including how pragmatically marked emotions are expressed and perceived and how vocal information has an impact on listeners. We use EEGs and fMRI to measure these factors with a key focus on social factors such as accent or whether the talker is AI.This study on JoVE supports our future research in these areas.
