The primary aims of this research project are (1) to test and refine the GODIVA neural network model of the brain interactions underlying learning and production of speech sound sequences, and (2) to develop and experimentally test a neurocomputational model of stuttering based on this model. Four functional magnetic resonance imaging (fMRI) experiments, two involving persons with fluent speech (PFS) and two involving both PFS and persons who stutter (PWS) are proposed for the new funding period to pursue these aims. The studies focus on several left hemisphere frontal cortical regions known to be involved in speech motor sequencing, including the lateral prefrontal cortex, ventral premotor cortex (vPMC), supplementary motor area (SMA), and preSMA, along with associated subcortical structures (basal ganglia, cerebellum, and thalamus). Each region will be modeled mathematically with equations governing neuronal activities, and interactions between the regions will be modeled with equations governing synaptic strengths. The resulting models will be implemented in computer software and simulated on the same tasks as speakers in associated functional magnetic resonance imaging (fMRI) experiments. Primary Aim 1 is addressed by three studies designed to identify the neural representations used in the brain regions involved in speech sequence production. In Study 1, the GODIVA model will be modified to exhibit the property of repetition suppression, a phenomenon exploited in fMRI studies to investigate the """"""""units"""""""" represented in different brain regions. Study 2 is an fMRI repetition suppression study (fMRI-RS) aimed at identifying the brain regions that process syllabic frames independent of phonetic content. Study 3 is an fMRI-RS study aimed at identifying the brain regions that treat consonant clusters as single units. Primary Aim 2 is addressed by three studies designed to investigate speech motor sequence learning and production in PWS. Study 4, an fMRI study involving learning of new speech sound, is aimed at determining the neural correlates of impaired speech motor sequence learning in PWS. Study 5 is an fMRI study investigating the neural circuitry underlying the fluency-inducing rhythm effect in which PWS produce syllables in time with a rhythmic external stimulus. In Study 6, a neurologically impaired version of the GODIVA model will be developed to simulate the tasks performed in studies 4 and 5. The integrated approach of computational neural modeling and functional brain imaging will provide a clearer, more mechanistic account of the neural processes underlying speech production in normal speakers and individuals with stuttering. This improved understanding will accelerate the development of treatments for stuttering and other disorders of speech sound sequencing.
This project will improve our understanding of how the brain controls speech production and how this control is disrupted in persons who stutter. The outcome of this research is expected to significantly accelerate the development of treatments for stuttering, which would have a profound impact on millions of people affected by this disorder.
Showing the most recent 10 out of 26 publications
