Measuring neuronal avalanches to inform brain-computer interfaces.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Biomedical Engineering

Background:

• Neuronal avalanches are critical for brain function, reflecting large-scale neural interactions.
• These activity patterns dynamically reconfigure based on cognitive demands.
• Their potential for brain-computer interfaces (BCIs) remains largely unexplored.

Purpose of the Study:

• To investigate if neuronal avalanche dynamics can be leveraged for BCI applications.
• To determine if task-specific information is encoded within avalanche propagation patterns.
• To assess if avalanche topology can enhance BCI decoding performance.

Main Methods:

• Utilized source-reconstructed magneto/electroencephalography (M/EEG) data during resting state and motor imagery tasks.
• Constructed avalanche transition matrices (ATMs) to quantify inter-regional avalanche spread probabilities.
• Analyzed differences in ATM topology between task conditions.

Main Results:

• Identified significant differences in transition probabilities between resting and task states, particularly involving premotor regions.
• Demonstrated that ATM topology allows for superior task-decoding compared to existing BCI methods.
• Confirmed that neuronal avalanches capture task-relevant neural information.

Conclusions:

• Neuronal avalanche dynamics contain interpretable, task-specific information crucial for BCI development.
• ATMs offer a novel, high-performance approach for decoding neural activity in BCIs.
• This research supports the use of neuronal avalanches as a foundational element for next-generation BCIs.