ENIGMA: A Unified Lightweight EEG-to-Image Model for Multi-Subject Visual Decoding

To be practical for real-life applications, models for reconstructing seen images from human brain activity must be effective on affordable scanning hardware, small enough to run locally on accessible computing resources, and easily and consistently deployable across multiple subjects in downstream tasks. To directly address these current limitations, we introduce ENIGMA, a multi-subject electroencephalography (EEG)-to-Image decoding model that reconstructs seen images from EEG recordings and achieves state-of-the-art (SOTA) performance on the research-grade THINGS-EEG2 and consumer-grade AllJoined-1.6M benchmarks. ENIGMA boasts a simpler architecture and has ~120x fewer parameters than previous SOTA methods, integrating a set of subject-specific encoder layers with a subject-unified spatio-temporal backbone to map raw EEG signals to a rich visual latent space. We evaluate our approach using a broad suite of image reconstruction metrics that have been standardized in the adjacent field of fMRI-to-Image research, and we describe the first EEG-to-Image study to conduct extensive behavioral evaluations of our reconstructions using human raters. Our simple and robust architecture provides significant performance improvements across both research-grade and consumer-grade EEG hardware, and provides a substantial boost in cross-subject decoding alignment. Finally, we provide extensive ablations to determine the architectural choices most responsible for our performance gains in both single and multi-subject cases across multiple benchmark datasets. Collectively our work provides a substantial step towards the development of practical brain-computer interface applications.