We pursue several lines of research in the Speech & Language Lab. In one line of work, we investigate the structure of the speech plan in child and adult language via measurement of speech acoustics and kinematics. In a second line of work, we use individual difference measures to investigate how language and nonlanguage subsystems interact to support the development of hierarchically organized speech. In a third line of work, we investigate the relative weightings of speech factors that contribute to the perception of disorder and the social import of these weightings. Our research is supported in part by the National institute of Child Health and Human Development.
Project Title: Speech Rhythm Acquisition
Source: National Institute of Child Health and Human Development (R01 HD087452), 2017—21.
PI: Melissa A. Redford [w/ Christina Gildersleeve-Neumann (subcontract PI), & Eric Vatikiotis-Bateson (consultant)]
Atypical speech rhythms, like those observed in children with childhood apraxia of speech, autism spectrum disorders, and stuttering, limit social interactions and can impede normal development and cognitive functioning. The long-term goal of the proposed research is to provide a developmentally sensitive model of language production that explains the emergence of typical and atypical speech rhythms. Current models of production are based on sophisticated descriptions of adult language. Influences from slowly developing speech motor, language, and other cognitive skills on speech plan representations and production processes have rarely been investigated. The paucity of research on such influences leads to fundamental gaps in understanding typical and atypical rhythm development, including understanding why young school-aged children do not chunk grammatical words with content words in connected speech to the same extent as adults. This chunking is important because it yields the rhythms we associate with prosodic words in adult language. The proposed research will investigate age-related changes in grammatical word production in typical development to explain prosodic word acquisition. The working hypothesis is that, once lexical stress is acquired, adult-like English rhythm production depends on a meaning-based chunking process at the intonational phrase level. The chunks evolve over developmental time into the supralexical production units we hear as prosodic words. We will test this hypothesis against well-defined alternatives using long-distance coarticulation to index production units. The influence of specific linguistic and cognitive factors on the development of long-distance coarticulation will also be investigated. The results will inform theory and provide new insights into why speech rhythm is atypical in children with speech planning deficits. The research will also provide important normative data on the extent and strength of long-distance coarticulation in school-aged children’s speech, and fundamental information about language and speech interactions in typical development. Finally, it will determine which cues most influence the perception of immature speech rhythm, with implications for clinical intervention.
Project Title: Acquisition of Temporal Patterns in Child Speech and Language
Source: National Institute of Child Health and Human Development (R01 HD061458), 2009—15 (w/ NCE).
PI: Melissa A. Redford [w/ Laura Dilley (subcontract PI), & Doris Payne (co-Investigator)]
The long-term research goal of this project was to model the structures and strategies that underlie the acquisition of suprasegmentals in the context of a developing system. The project laid the foundation for this long-term goal by characterizing a full range of prosodic patterns in 5- to 11-year-old English-speaking children’s speech, with an emphasis on temporal patterns. We sought to identify the rate at which different patterns are acquired in a relatively understudied age group whose language is functional and complex but prosodically immature. We also sought to identify interactions between timing, intonation, and syntax in children’s production of prosody, and the language and nonlanguage factors that explain why children’s prosodic speech patterns are different from those of adults. Our working hypothesis was that children’s rhythmic speech is increasingly modulated over time to highlight conceptual and structural aspects of the message for listeners, which is a major function of adult prosody. This developmental process was predicted to unfold as children overcome motoric constraints and become more sensitive to their listeners’ needs. Moreover, the speech plan that underlies the production of suprasegmental patterns was predicted to evolve with developing memory and syntactic abilities.