Natalya Kaganovich - US grants

DESCRIPTION (provided by applicant): Audiovisual Processing in Specific Language Impairment. A growing body of literature strongly suggests that the ability to combine auditory and visual information plays a significant role in normal language acquisition. However, whether a breakdown in the merging of these senses may contribute to the development of language disorders, such as specific language impairment (SLI), is unknown. So far, only a few studies have evaluated audiovisual processing in SLI. All of them used the sensory conflict paradigm - the so-called McGurk illusion - in which the sound of pa is dubbed onto the visual articulation of ka typically leading to the perception of ta. These studies have reported a reduced susceptibility to this illusion in children with SLI and in adults with a history of SLI, concludin that their audiovisual integration is impaired. However, the inability to resolve a sensory conflic between auditory and visual modalities during syllable perception does not presuppose an impairment in audiovisual integration during other stages of linguistic processing, such as early sensory encoding of speech sounds or later lexical perception. Furthermore, how an impairment in audiovisual processing interacts with speech-language deficits of children with SLI is unknown. This project is designed to advance our understanding of audiovisual processing in SLI in multiple ways. First, we determine the effectiveness of audiovisual processing in SLI at the acoustic, phonemic, and lexical levels of speech perception by drawing on the strength of the event-related brain potentials (ERP) method, whose excellent temporal resolution makes it uniquely suited for evaluating the stage of information processing at which audiovisual perception falters in SLI. We also add more ecologically valid stimuli - syllables without a sensory conflict and words - to our paradigms in order to examine audiovisual processing in response to more realistic speech stimuli. Lastly, by evaluating correlations between, on the one hand, behavioral and ERP indices of audiovisual integration and, on the other hand, a range of cognitive and language skills in children with SLI, we strive to identify those cognitive and linguistic deficits of SLI that are most likely to be affected by audiovisual perception difficultis in this group.

Project Summary ? Audiovisual processing in developmental language disorder Emerging developmental studies show that visual speech cues facilitate the acquisition of important building blocks of language, such as phonemes and words, and shape the development of both speech production and speech perception. Because visual speech cues influence multiple aspects of typical language acquisition, and speech perception more generally, we ask whether impairment in the processing of visual articulatory movements and/or difficulty in integrating such movements with concurrent auditory speech may underlie some of the deficits observed in children with developmental language disorder (DLD). DLD is characterized by difficulties in acquiring language in the absence of an easily identifiable cause, such as hearing impairment, frank neurological disorders, or low non-verbal intelligence. It affects approximately 7% of preschool children in the US and is typically a life-long condition. Earlier studies suggest that at least some aspects of audiovisual processing are impaired in DLD, but the causes remain unknown. In this project, we propose to examine in children with DLD and their typically developing (TD) peers a set of mechanisms that are essential for normal audiovisual processing. These mechanisms include the following: (1) sensitivity to temporal relationships between events in auditory and visual modalities; (2) sensory encoding of speech-relevant visual features, namely shape and motion; (3) allocation of attention to features of complex visual objects; and (4) knowledge of auditory/articulatory correspondences for speech sounds. Examining these mechanisms in the same groups of children with DLD and TD will provide information about both the hetero/homogeneity of audiovisual impairments in DLD and the effects of impairment in just one vs. several mechanisms on audiovisual language use. We combine behavioral measures with ERP and EEG data analyses in order to map the nature of neural processes engaged by each mechanism. Understanding the causes of audiovisual processing difficulties in DLD has significant clinical implications. Audiovisual integration normally facilitates some of the linguistic skills that are impaired in DLD, such as phonological and lexical processing and speech-in-noise perception. Hence, identifying the mechanisms that prevent children with DLD from using visual speech cues as well as their TD peers do and rehabilitating them could potentially lead to improvements in these linguistic skills. Additionally, comparing the neural processes underlying audiovisual integration in children with typical and atypical language ability provides a unique window into the development of the neural networks connecting auditory and visual modalities and their impact on normal and disordered language function.