How Meta researchers are using AI to understand human cognition

Meta AI researchers have achieved significant breakthroughs in decoding language from brain activity and understanding how the brain processes language, working in collaboration with the Basque Center on Cognition, Brain and Language.

Key breakthroughs: Two major advances in brain-computer interface technology and neuroscience understanding have emerged from Meta’s research efforts.

• Researchers successfully decoded up to 80% of characters from non-invasive brain recordings, enabling the reconstruction of complete sentences from brain signals
• A new AI model was developed to understand how the brain transforms thoughts into sequential word patterns
• The research utilized MEG and EEG technology to record brain activity from 35 healthy volunteers

Technical approach: The research combines advanced neuroimaging techniques with artificial intelligence to capture and interpret brain activity during language production.

• MEG and EEG devices were used to measure magnetic and electric fields generated by neuronal activity
• The system takes 1,000 brain activity snapshots per second to track thought-to-word transformation
• Researchers identified a ‘dynamic neural code’ that shows how the brain chains successive representations while maintaining them over time

Clinical implications: This research opens new possibilities for medical applications, particularly in helping those with communication disabilities.

• The technology could potentially help millions of people affected by brain lesions that impact communication
• Current approaches require invasive surgical interventions, while this new method is non-invasive
• Several challenges remain, including performance optimization and the need for controlled environments
• Further research is needed to validate the approach with patients suffering from brain injuries

Broader research impact: Meta is fostering collaboration and supporting continued research in this field.

• A $2.2 million donation to the Rothschild Foundation Hospital has been announced
• Partnerships with European research institutions including NeuroSpin, Inria, ENS-PSL, and CNRS continue to expand
• Meta’s open-source AI models have enabled other breakthroughs in medical applications, including fetal heart defect detection and endoscopy analysis

Future implications: While these advances represent significant progress in brain-computer interfaces and understanding human cognition, practical implementation faces several hurdles that will require continued research and development to overcome.

• The technology’s current accuracy and environmental requirements limit immediate clinical applications
• Success with healthy volunteers must be translated to effectiveness with patients suffering from various neurological conditions
• The research provides valuable insights into human language processing that could inform future AI development