Phonetic learning in adults with and without language impairment

University Of Connecticut Storrs, Storrs-Mansfield CT

Investigators

Linked publications, trials & patents

Abstract

? DESCRIPTION (provided by applicant): Despite decades of debate devoted to the etiology of Specific Language Impairment, the origin of deficit has yet to be identified. Hallmark symptoms include the relative difficulty to acquire grammatical morphemes of low perceptual salience (Leonard, 2014), poor verbal working memory (Ellis-Weismer, 1996), slowed language processing (Friedrich, Weber, & Friederici, 2004), and difficulty identifying speech in noise (Ziegler et al., 2005). Joanisse and Sciedenberg (1998) hypothesize that instability in phonological representations lead to poor verbal working memory, cascading to the impairment in the acquisition of grammar. While this framework accounts for speech processing/perceptual deficits, the origins of phonological deficit remain in debate. Some argue that poor representational quality arises from differences in low-level auditory processing (McArthur & Bishop 2001, Wright et al. 1997), while others (e.g. Ullman & Pierpont, 2005) propose a domain-general implicit learning deficit. A deficit in implicit learning better accounts for the differencs observed in performance measures outside the auditory/linguistic domain (e.g. Lum, Conti-Ramsden, Morgan, & Ullman, 2014); however, how a deficit in implicit learning relates to building phonological representations is not yet clear. Our lab has developed an experimental protocol that examines the time course of encoding of new (nonnative) speech sounds following perceptual training in typical adults (Earle & Myers, in press, submitted). We have argued that post-training sleep plays an important role in building new phonetic categories (see Earle & Myers, 2014, for review); thus, our interest has been not only in the initial phase of acquiring phonetic information, but in how that information is encoded during the initial 24 hours following training. In the current proposal, we extend our work on phonetic learning to individuals with language impairment in order to contribute a memory account of the phonological deficit. We will assess participants' ability to perceive a nonnative contrast before training, immediately aftr training, and after a period of sleep. We will further obtain a rudimentary measure of sleep to determine if individual differences in sleep duration affect post-sleep performance relative to immediate posttest. We also intend to gather electrophysiological measures of sensitivity to the trained sounds, to ensure that differences observed between groups are not dependent on metalinguistic task performance. The proposed research will advance our knowledge of how memory encoding processes contribute to speech sound learning, and further inform the mechanism of breakdown in encoding phonetic features in individuals with language impairment. This account may direct further research into the neurobiological underpinnings of the memory deficits in SLI, with the eventual goal of identifying an appropriate remediation/intervention. The studies outlined in our project aims therefore point to potential contributions to three areas of research: speech perception, memory consolidation, and Specific Language Impairment.