CE37 - Neurosciences intégratives et cognitives 2025

The role of beta bursts in sensorimotor predictive coding – Bursts2Predict

Submission summary

Bursts2Predict aims to uncover the role of transient beta bursts in sensorimotor predictive coding, focusing on their contribution to motor control and inference-based movement planning and execution. Predictive coding models propose that the brain continuously generates and updates expectations about sensory inputs, integrating top-down predictions with bottom-up sensory information. While these models have significantly advanced our understanding of sensory processing, extending predictive coding principles to motor control remains a fundamental challenge. The motor cortex exhibits unique connectivity and laminar organization compared to sensory regions, and sensorimotor beta rhythms occur as transient bursts rather than sustained oscillations, raising critical questions about their functional role in predictive processing.
This project will systematically investigate three hypotheses regarding beta bursts: (1) they stabilize predictable sensory inputs and motor states that do not require updates, (2) they encode feedback predictions against which sensory inputs are compared, or (3) different beta bursts play distinct roles in predictive processing depending on their type and temporal dynamics. To address these hypotheses, the project will employ an interdisciplinary approach combining behavioral experiments, electrophysiology, neuroimaging, and computational modeling.
We will employ state-of-the-art neural recording technologies. In humans, high-resolution laminar MEG will be used to infer depth-specific oscillatory dynamics. In macaques, novel multi-area 3D probes will provide chronic, high-density laminar electrophysiological recordings across four key parieto-frontal regions: M1, PMd, 5d, and 7a. These areas are critical for cognitive control and adaptive visuomotor behavior, making them ideal for investigating beta burst contributions to predictive motor control. The MEG approach will complement invasive macaque recordings by establishing the spatial extent of cortical networks supporting inference-based motor control. This cross-species approach will allow us to identify spatiotemporal beta burst states and task variables that modulate behavior across different levels of neural organization.
Our computational framework consists of three key stages: (1) identifying task variables that influence behavior in both species, (2) characterizing beta burst states modulated by these variables, and (3) developing an interspecies computational model linking beta burst characteristics to predictive processing dynamics. The final model will leverage hidden Markov models (HMMs) and other statistical inference techniques to unify data across species and recording modalities, revealing how transient beta bursts contribute to sensorimotor predictive coding.
The project is organized into four interconnected work packages (WPs): WP1 focuses on task optimization and animal training, WP2 on experimental data collection, WP3 on data preprocessing and analysis, and WP4 on data-driven computational modeling on combined human and NHP data. By publicly sharing high-resolution electrophysiological and MEG datasets capturing beta burst dynamics, this project will provide valuable resources for future cross-species and meta-analytic investigations. Ultimately, this research will advance our understanding of sensorimotor predictive coding and could inform translational applications for motor disorders and rehabilitation strategies.

Project coordination

Bjørg Kilavik (Institut des Neurosciences de la Timone)

The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.

Partnership

INT Institut des Neurosciences de la Timone
ISC-MJ CENTRE NATIONAL DE LA RECHERCHE SCIENTIFIQUE
LNCA UNIVERSITÉ STRASBOURG

Help of the ANR 861,488 euros
Beginning and duration of the scientific project: December 2025 - 48 Months

Useful links

Explorez notre base de projets financés

 

 

ANR makes available its datasets on funded projects, click here to find more.

Sign up for the latest news:
Subscribe to our newsletter