Cortical and subcortical activity during speech segmentation: Exploration from simultaneous cortex EEG and subthalamic nucleus LFP in Parkinson’s disease

• Applicant: Professor Dr. Michael Barbe
• Subject Area: Human Cognitive and Systems Neuroscience ; Experimental and Theoretical Network Neuroscience ; Clinical Neurology; Neurosurgery and Neuroradiology
• Term: since 2026
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 567518152

The general aim of our project is to refine our understanding of neural bases of speech perception, focusing on speech segmentation by statistical speech learning task as a fundament process of spoken language acquisition. The main scientific objective is to advance fundamental knowledge on the involvement of STN in speech processing. This will be done by means of a highly innovative approach combining the simultaneous exploration of cortical and subcortical activity recorded in people with Parkinson’s disease (PwPD) who underwent a bilateral deep brain stimulation (DBS) within the subthalamic nucleus (STN-DBS). The new research paradigm will provide us the unique opportunity to investigate the understudied role of the basal ganglia in speech processing. Specifically, we will directly collect brain activity during a statistical speech learning task, where PwPD segment a continuous acoustic signal into (pseudo)words. The simultaneous recording of cortical and subcortical responses is possible because the more recent version of DBS generators is capable to record local field potentials (LFP) from the implanted electrodes, becoming sensing DBS. This technological advance allows the recording of STN activity long after the surgery. We will conduct a series of perceptual speech segmentation tasks and analyse task performance, as well as subcortical (STN LFP) and cortical (EEG) recordings. Considering that learning is a key function of the basal ganglia, and that speech segmentation relies on phonological representations and automatization of phoneme chunking that can be affected in Parkinson’s disease, our study appears to be original, innovative and relevant, especially in PwPD where the protocol has never been applied. The core experiment of our project lies upon the performance, and recording of the associated brain activity, during a speech segmentation protocol presented in a statistical learning task. This paradigm evaluates a powerful implicit learning mechanism that allows individuals to segment a continuous acoustic signal into (pseudo)words in the absence of explicit instructions. Additionally, in this context, the role of STN in integrating information from multiple cortical areas remains pivotal. However, understanding the dynamics of speech processing in the basal ganglia, especially the activity coupling between STN and cortical areas, remains elusive. The question on whether the basal ganglia are involved in the learning of speech sequences, such as phoneme sequences forming words, remains open. This is where our project stands. To bridge this gap, we propose to use an integrative approach to record simultaneously STN LFP – with the DBS sensing device – and cortical activity with surface EEG. Our team, which includes experts from different fields, will join forces to address several research questions from an interdisciplinary perspective.

• DFG Programme: Research Grants
• International Connection: France
• Partner Organisation: Agence Nationale de la Recherche / The French National Research Agency
• Co-Investigator: Professorin Dr. Doris Mücke
• Cooperation Partner: Professor Serge Pinto