# Neural Modeling and Imaging of Speech

> **NIH NIH R01** · BOSTON UNIVERSITY (CHARLES RIVER CAMPUS) · 2020 · $350,625

## Abstract

PROJECT SUMMARY
The overall goal of this study is to improve our understanding of the neural mechanisms that underlie speech
production and their breakdown in voice disorders. Our goals will be achieved through a tight coupling of
computational modeling and functional magnetic resonance imaging (fMRI) studies organized around the
Directions Into Velocities of Articulators (DIVA) model of speech motor control. The goal of Aim 1 is to develop
a general computational framework for quantitatively comparing cognitive models with neuroimaging. Software
will be developed to enable (i) the generation of whole-brain activity patterns from any model of brain function
whose components are specified by a computational load function (e.g., the number of items in working
memory) and a standard brain volume template location, and (ii) direct quantitative comparison between
simulated brain activity and observed activity measured by a given imaging modality (e.g., fMRI, positron
emission tomography, electroencephalography, and electrocorticography). DIVA and two alternative speech
production models will be implemented in this framework and used to simulate whole-brain fMRI activity for a
variety of speech tasks. The framework allows quantitative comparisons between models for fitting
neuroimaging data, marking a significant improvement over the qualitative comparative methods that currently
prevail. The goal of Aim 2 is to characterize the brain mechanisms underlying somatosensory feedback control
of phonation. fMRI will be used to measure brain responses while neurotypical participants speak short
sentences. On a subset of sentences, a somatosensory perturbation will be applied to the larynx. Perturbed
utterances will be contrasted with unperturbed utterances to highlight the brain network responsible for
counteracting the perturbation. DIVA-based hypotheses regarding this network will be tested by comparing
observed brain activity to that resulting from simulations of the task using the framework developed in Aim 1.
The goal of Aim 3 is to investigate somatosensory feedback control of phonation in individuals with the voice
disorder adductor spasmodic dysphonia (ADSD). Prior findings suggest ADSD is due to hyper-function of the
somatosensory feedback control system. This hypothesis will be tested using the larynx perturbation fMRI
protocol of Aim 2 applied to individuals with ADSD. Measured brain activity from ADSD and neurologically
normal participant groups will be contrasted, and computer simulations of normal and impaired versions of the
DIVA model will be compared to the experimental results. Successful completion of our aims will move
neuroscience from qualitative descriptions to quantitatively testable accounts of the neural processes
responsible for speech and other cognitive tasks. In the longer term, the improved understanding of brain
mechanisms underlying normal speech and their breakdowns in disorders such as ADSD will pave the way for
improved ass...

## Key facts

- **NIH application ID:** 9982930
- **Project number:** 5R01DC002852-25
- **Recipient organization:** BOSTON UNIVERSITY (CHARLES RIVER CAMPUS)
- **Principal Investigator:** FRANK H GUENTHER
- **Activity code:** R01 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2020
- **Award amount:** $350,625
- **Award type:** 5
- **Project period:** 1996-02-01 → 2022-07-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/9982930

## Citation

> US National Institutes of Health, RePORTER application 9982930, Neural Modeling and Imaging of Speech (5R01DC002852-25). Retrieved via AI Analytics 2026-05-23 from https://api.ai-analytics.org/grant/nih/9982930. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*