Files and links (1)
Free to read and download, Open Access
For any speech category there are multiple sources of information (both acoustic and contextual) that are relevant to categorization. Complicating matters further, these sources of information are not always available simultaneously, but present themselves over the course of several hundred milliseconds. These features of spoken language complicate an already difficult task, and raise three important questions: 1) how do listeners weight different cues to the same speech category, 2) how do listeners integrate asynchronous information during speech perception and 3) how do listeners cope with contextual variability. While these questions have been explored, to varying degrees, with adults, there have been very few attempts to explore these questions from a developmental perspective. Furthermore, some of the more complex interactions between these factors remain uncharted territory even in the adult literature. For example, while adult listeners compensate for context when categorizing speech, and utilize acoustic cues as soon as they become available, we still do not know how this process is affected by context.
This dissertation addresses these lingering issues by assessing 7-year-olds', 12-year-olds' and adults' perception of the /s-ʃ/ contrast (one that is influenced by multiple acoustic cues and context) using eye-tracking and the visual world paradigm. This work demonstrates that there is considerable development between 7 and 12 years of age for the /s-ʃ/ contrast in terms of real-time cue integration, cue-weighting and context compensation, and that development likely continues beyond these ages. In addition, the adult work demonstrates, for the first time, a pattern of real-time cue integration in which listeners' (both adult and child) buffer acoustic cues. Finally, several hypotheses are considered that may account for these findings, including the possibility that the unique developmental pattern of fricative perception may play an important role in understanding why adults buffer this contrast, and the implications of buffered speech perception are discussed.