To understand the neural mechanisms underlying speech production and stuttering, as a disorder of the speech production system, investigating the details of the sensorimotor system is crucial. The sensorimotor system comprises of three main components: 
(1) sensory feedback (auditory and somatosensory feedback), 
(2) control component (responsible for generating control signals that achieve a desired goal), and 
(3) prediction component (responsible for predicting consequences of issued control signals) 

We are investigating each of these components, separately and in relation with each other, using a combination of behavioral measurements—acoustic and kinematic analyses of speech—and electrophysiological recordings and stimulations—electroencephalography (EEG) and transcranial current stimulation (tCS)—and neuroimaging techniques—functional magnetic resonance imaging (fMRI) and diffusion tensor imaging (DTI).

Acoustic Analyses of Speech

We use acoustic analyses of speech to examine sensitivity, accuracy, and integration of auditory feedback during the production of speech movements. To systematically manipulate specific parameters of the auditory feedback, we employ auditory feedback perturbation techniques. Using such paradigms, we can investigate the integration of auditory feedback for (a) online feedback monitoring, and (b) calibration and maintenance of the speech production system.

For more information see: 



To fully understand sensorimotor processes, we examine sensory prediction as one important aspect of the sensorimotor integration, using EEG technique. We use high-density EEG to measure electrical fluctuations resulting from brain activity in real time. In addition, we use noninvasive transcranial current stimulations to causally modulate specific aspects of the sensorimotor processes. The ultimate goal of this translational research is to develop new clinical treatments for stuttering. 

For more information see:
Kinematic Analyses of Speech

Another approach for understanding the neural mechanism of speech production is to study the kinematic characteristics of speech movements (in combination with speech signals). We use electromagnetic articulography (EMA) to continuously track the position of different speech articulators (tongue, lips, jaw) during speech production. In addition, we study the articulatory movements during auditory feedback perturbations and robotically-induced somatosensory feedback perturbations. 

For more information see: 


As another approach to study the neural substrates underlying speech production, we use fMRI and DTI techniques. These techniques allow us to examine the structure, function, and functional and structural connectivity between different neural substrates of speech production. In addition, we implement the results of these structural and functional analyses in a computational model that will be used to understand the neural mechanisms of speech and stuttering, to make predictions, and to develop hypothesis-driven research questions.