Neural Modeling and Imaging of Speech

Boston University (Charles River Campus), Boston MA

Investigators

Linked publications & trials

Abstract

[unreadable] DESCRIPTION (provided by applicant): The primary goal of this research project is the continued development, testing, and refinement of a computational neural model describing the neural processes underlying speech. In the new budget period, we propose to address several important issues regarding auditory feedback control in speech production, both at the prosodic and segmental levels, utilizing tightly integrated combination of neural network modeling, neuroimaging, and psychophysics. The proposed research is organized into three inter-related thrusts. (1) Control of prosodic aspects of speech. This thrust combines neural modeling with psychophysical and functional magnetic resonance imaging (fMRI) experiments involving real-time auditory perturbation of prosodic cues during speech to investigate the neural mechanisms responsible for the control of word- and phrase-level prosody. (2) Representation of speech sounds in auditory cortical areas. In this thrust, we propose fMRI experiments and modeling work designed to further our understanding of the representation of speech sounds in the auditory cortical areas. This issue is central to our speech production model, which utilizes an auditory reference frame to store speech sound targets for production. In particular, we will investigate auditory vs. phonetic representations and talker-specific vs. talker-independent representations in the auditory cortical areas. (3) Integration of feedforward and feedback commands in different speaking conditions. This thrust involves an fMRI experiment and modeling project to investigate hypotheses concerning the integration of feedforward and feedback control mechanisms in different speaking conditions (i.e., fast, normal, clear, and stressed speech). In each thrust, computer simulations of the model will be compared to the psychophysical and fMRI experimental results to test the model's account of the neural mechanisms underlying speech production. Relevance. We believe these studies will provide significant advances in our understanding of the neural bases of speech, including a detailed functional description of each of the brain regions involved in speech production. This in turn will allow the more complete characterization of communication disorders that involve the model's circuitry, such as apraxia of speech, stuttering, and ataxic dysarthria. In the long run, we believe this improved understanding will lead to better diagnosis and treatment of these disorders. [unreadable] [unreadable] [unreadable] [unreadable]