Decoding Continuous Character-based Language from Non-invasive Brain Recordings

TL;DR

This paper introduces a novel non-invasive brain decoding method using fMRI and deep learning to reconstruct continuous language, achieving cross-subject semantic decoding from single trials, advancing brain-computer interfaces.

Contribution

A new deep learning framework for decoding continuous language from single-trial non-invasive brain recordings, capable of cross-subject semantic reconstruction.

Findings

Decodes continuous language with high fidelity

Operates effectively across different subjects

Outperforms existing decoders in cross-subject tasks

Abstract

Deciphering natural language from brain activity through non-invasive devices remains a formidable challenge. Previous non-invasive decoders either require multiple experiments with identical stimuli to pinpoint cortical regions and enhance signal-to-noise ratios in brain activity, or they are limited to discerning basic linguistic elements such as letters and words. We propose a novel approach to decoding continuous language from single-trial non-invasive fMRI recordings, in which a three-dimensional convolutional network augmented with information bottleneck is developed to automatically identify responsive voxels to stimuli, and a character-based decoder is designed for the semantic reconstruction of continuous language characterized by inherent character structures. The resulting decoder can produce intelligible textual sequences that faithfully capture the meaning of perceived speech both within and across subjects, while existing decoders exhibit significantly inferior performance in cross-subject contexts. The ability to decode continuous language from single trials across subjects demonstrates the promising applications of non-invasive language brain-computer interfaces in both healthcare and neuroscience.