Doctoral Dissertation Research: Articulation and Altered Auditory Feedback

University Of California-Berkeley, Berkeley CA

Investigators

Abstract

Individuals vary from one another in how they speak. Speakers listen to themselves while talking, and they constantly use this auditory feedback to correct their speech. This project examines two factors that influence the way that speakers are able to effectively use this auditory feedback: vocal tract shape and native language. The shape of the hard palate, which is an important factor in the relationship between articulation and acoustics, will be used as a measure of individual anatomical variation. Two similar sounds with comparable articulatory variation in American English and Tamil are compared to test the role of native language. In this experiment, participants say words containing the speech sounds of interest, and their auditory feedback is altered in real time by a computer so speakers hear themselves saying something different from what they actually produced. Previous studies (e.g. Houde and Jordan 1998) show that speakers change their articulation in response to altered auditory feedback, but it is not known how. Here, the novel addition of simultaneous ultrasound imaging will reveal how speakers compensate. The results of these studies stand to improve models of how we modify our motor plans while speaking. A complete model of typical speech serves as a baseline for understanding motor control in populations with disordered speech, such as those with apraxia of speech or Parkinson's Disease. This research will also contribute to our understanding of the role of auditory feedback in motor control, which has direct implications for people with hearing loss. Comparing English and Tamil helps to support generalizations about motor control in speech without confining results to English alone. This research also addresses long-standing questions concerning the instigation of language change. Exploring the factors that cause individuals to be fundamentally differently from each other in their speech production may provide an explanation for why some individuals might be more likely than others to contribute to the pool of language variation that results in sound change.