Speech Perception and Spoken Word Recognition (NIH/NIDCD R01-DC000111)
We are interested in the earliest stages of speech perception at which the initial sensory and perceptual information in the speech signal makes contact with phonological, lexical and indexical knowledge to support spoken word recognition, lexical access and spoken language processing. The proposed research is divided into the following four projects:
Project 1: Spoken Word Recognition and the Mental Lexicon
This project investigates the role of the lexicon in speech perception and spoken word recognition using computational and behavioral techniques. Our goal is to describe how words are represented and organized in lexical memory and how lexical knowledge affects spoken word recognition. Computational analyses of the similarity relations among words will be carried out to identify both local and global constraints on the sound patterns of words. Behavioral studies will be conducted to study the effects of frequency, phonological similarity, and global lexical connectivity relations on spoken word recognition.
Project 2: Variability in Speech Perception
The goal of this project is to learn more about speech variability and the effects that different sources of variability have on speech perception and spoken word recognition. Experiments will be carried out to study perceptual adjustments to different talkers and dialects, as well as the contribution of item-specific episodic information to the perceptual analysis of speech. We will investigate how listeners encode sources of variability in memory, the role of specific instances in perception and how indexical information affects speech perception and spoken word recognition.
Project 3: Perceptual Learning and Adaptation
This project is concerned with perceptual learning and adaptation in speech perception and spoken word recognition. We are interested in how listeners learn to rapidly adapt to novel voices and degraded speech signals. We also want to gain a better understanding of how indexical information about the talker’s voice is perceived and used to facilitate spoken word recognition and the retention of phonological and lexical information in memory.
Project 4: Individual Differences and Working Memory
This project investigates individual differences in speech perception and spoken word recognition in prelingually deaf children with cochlear implants (CIs). The goal of this project is to identify the sources of variability in word recognition performance in this clinical population and describe the factors that affect the speed and efficiency of phonological coding and verbal rehearsal processes. Implicit learning and working memory will be studied using a new methodology based on immediate reproductive memory span. We will also test the hypothesis that the individual differences in spoken word recognition in deaf children with CIs reflect a loss of perceptual distinctiveness resulting from partially-specified phonological representations of speech.