Brain Rhythms in Speech Production

• Applicant: Professor Dr. Joachim Gross
• Subject Area: Human Cognitive and Systems Neuroscience ; Cognitive, Systems and Behavioural Neurobiology
• Term: from 2019 to 2024
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 417044868

In humans, the highly efficient and remarkably fast exchange of information through speech is only possibly because incoming sensory information is processed for understanding but also for generating predictions about the timing and content of upcoming speech. This predictive comprehension is supported by brain areas that are partly shared between speech perception and speech production. Interestingly, this network seems to operate partly in a rhythmic mode - for example, syllables are typically produced at a preferred rate of about five per second (5 Hz). We and others have demonstrated that listener’s brain activity aligns to the syllable rate of speech during listening to continuous speech. Furthermore, this alignment is stronger for intelligible compared to unintelligible speech and seems to be modulated by directed top-down signals from higher-order areas. This relatively new field of speech-brain entrainment has already inspired many studies and significantly improved our understanding of the role of brain oscillations in speech processing. Surprisingly, with only few exceptions, this approach has not been applied to speech production. Here, we propose to fill this gap. We plan to record brain activity with MEG (magnetoencephalopgraphy) while participants produce continuous speech. By simultaneously recording speech with brain activity (and multiple other signals such as respiration and muscle activity) we can identify areas where brain activity is consistently temporally related to speech production (for example the syllable rate at 5 Hz). We will compare this alignment during continuous speech production and speech processing to identify commonalities and differences in the underlying dynamic network. Different experimental conditions are specifically designed to disentangle rhythmic motor processes from more language-related processes. Finally, we will employ a language-learning paradigm to challenge the predictive comprehension (and production) system. This promises novel insights into the spatial localization and dynamic operation of the brain network supporting predictive processes related to speech.In summary, we propose to perform the first comprehensive MEG study of the (temporal) relationship between brain rhythms and continuously produced speech. We expect novel insights that hopefully complement emerging models of brain oscillations in speech perception by adding the ‘speech production’ dimension. Finally, a more comprehensive account of the dynamic speech production system will be important for the understanding and treatment of speech disorders such as stuttering or speech apraxia.

• DFG Programme: Research Grants