Study examines how post-stroke aphasia disrupts fluent speech

The research, published in the Oct. 15 issue of the journal NeuroImage, could help advance future treatment approaches for individuals with post-stroke aphasia.

Led by Dr. Roozbeh Behroozmand, associate professor of speech, language, and hearing in the School of Behavioral and Brain Sciences, the researchers recorded brain signals in stroke survivors with aphasia to understand how lesions in neural networks of the left hemisphere interfere with the patients’ ability to process auditory feedback — a critical component in producing accurate, fluent speech.

“The auditory system is essential in identifying errors as we speak,” said Behroozmand, the corresponding author of the study. “We control our speech production based on what we’re hearing from our own verbal output to maintain fluent communication.”

Aphasia affects 20% to 40% of stroke survivors, impairing speech production, comprehension, reading and writing without diminishing a person’s intellectual ability.

“Clinical symptoms of aphasia are diverse, ranging from nonfluent speech to difficulty understanding spoken language,” Behroozmand said. “Some individuals can comprehend speech but not produce it; some lose words for naming objects; some have issues with repeating words. These outcomes all represent errors in the sensory and motor systems of the brain for monitoring, detecting errors and generating corrective motor commands to the speech muscles — the physical movements that will produce words and sentences.”

Clinical intervention for post-stroke aphasia patients has traditionally focused on improving speech movement outcomes. Behroozmand explained that such treatment only addresses one of three major components of the brain’s effort to correct speech errors.

“Detecting the error is the first step: Your brain anticipates what you intend to say before you overtly produce words, and then you hear your own feedback to determine if the right words were produced,” he said. “If that is not the case, the second step is to process the errors you hear and translate that into corrective motor commands that adjust muscle movements to reach speech targets. The last step is execution of those physical modifications to ensure corrected words are produced.

“Our findings underscore the importance of the auditory feedback processing step in identifying and correcting speech errors.”

Researchers studied sensory-motor integration in speech production using auditory feedback perturbation, which is a technique that externally alters participants’ own speech and feeds the manipulated version back to them via headphones in real time.

“Using this technique, we artificially generate a speech error signal for the brain, which is interpreted as a disparity between intended and actual speech output when patients are talking,” Behroozmand said. “Then, we measure how this mismatch leads to an adjustment in speech output to probe the integrity of sensory and motor systems in healthy individuals and identify the source of neural deficits in patients with post-stroke aphasia.

“In our study, we found that participants in the aphasia group showed significantly reduced event-related potential activity — an indicator of weakened neural responses — compared with the neurotypical speakers. These patterns point to disruptions across the sensory-motor brain networks responsible for detecting and correcting speech errors.”

The study confirmed that post-stroke aphasia individuals have a diminished ability to recognize their own speech errors in the auditory feedback and to make real-time adjustments to maintain fluency.

In further research, investigators in Behroozmand’s Speech Neuroscience Lab will use electroencephalograms and neuroimaging from different parts of the brain to understand better how stroke affects the interaction between the sensory and motor systems during speech production.

“It could be that both the hearing system and the motor system are intact, but the integration between the two breaks down,” Behroozmand said. “Stroke patients with aphasia could exhibit failures at any of these levels, depending on the location and size of the lesion in the brain and how much gray matter or white matter tracts have been damaged.”

Dr. Vahid Nejati, a research associate in the Speech Neuroscience Lab at UT Dallas, is the lead author of the study. Dr. Ayoub Daliri, associate professor of speech and hearing science at Arizona State University, also contributed to the project.

This research was supported by grants (R01DC018523, K01DC015831, R01DC019905, R01DC020162) from the National Institute on Deafness and Other Communication Disorders, an organization within the National Institutes of Health.