Related Experiment Video
Updated: Oct 14, 2025
05:31
Examining Bilingual Language Control Using the Stroop Task
Published on: February 26, 2020
14.9K
Single-Trial Classification of Disfluent Brain States in Adults Who Stutter
John C Myers1, Farzan Irani2, Edward J Golob3
1Department of Psychology, University of Texas San Antonio, San Antonio, United States.
Conference Proceedings. IEEE International Conference on Systems, Man, and Cybernetics
|November 1, 2021
Summary
Researchers developed a method to detect stuttering by analyzing brain activity before speech. This brain-computer interface (BCI) approach shows promise for real-time stuttering detection and potential therapeutic interventions.
Area of Science:
- Neuroscience
- Speech Science
- Biomedical Engineering
Background:
- Normal speech relies on intricate motor and sensory coordination.
- Stuttering, or persistent developmental stuttering, affects approximately 1% of the global population, often persisting into adulthood.
- Stuttering events frequently occur at the utterance's onset, suggesting pre-speech brain activity differences.
Purpose of the Study:
- To develop a method for classifying brain network states associated with fluent versus stuttered speech on a single trial basis.
- To explore the feasibility of a brain-computer interface (BCI) for early stuttering detection.
Main Methods:
- Electroencephalography (EEG) recorded brain activity before participants who stutter read pseudo-word pairs.
- Independent Component Analysis (ICA) identified neural sources underlying speech preparation.
- Spectral power and coherence data were extracted from salient time windows, analyzed using stepwise linear discriminant analysis (sLDA).
Main Results:
- The sLDA algorithm successfully predicted fluent versus stuttered speech with 81% accuracy across trials in two subjects.
- This demonstrates the potential to differentiate brain states linked to fluent and disfluent speech production.
Conclusions:
- The findings support the feasibility of a BCI system for pre-emptive stuttering detection.
- This technology holds potential for future therapeutic applications in speech disorders.
