Universal Sleep Decoder: Aligning awake and sleep neural representation across subjects

Decoding memory content from brain activity during sleep has long been a goal in neuroscience. While spontaneous reactivation of memories during sleep in rodents is known to support memory consolidation and offline learning, capturing memory replay in humans is challenging due to the absence of well-annotated sleep datasets and the substantial differences in neural patterns between wakefulness and sleep. To address these challenges, we designed a novel cognitive neuroscience experiment and collected a comprehensive, well-annotated electroencephalography (EEG) dataset from 134 subjects during both wakefulness and sleep. Leveraging this benchmark dataset, we developed the Universal Sleep Decoder (USD) to align neural representations between wakefulness and sleep across subjects and a real-time staging model comparable to offline staging algorithms. Our model achieves up to 23.00 21.15 decoding accuracy on unseen subjects for N2/3 stage and REM stage, which is much higher than the decoding performances using individual sleep data. Furthermore, fine-tuning USD on test subjects enhances decoding accuracy to 29.20 a substantial improvement over the baseline chance of 6.7 and ablation analyses reveal that our design choices, including the use of (i) an additional contrastive objective to integrate awake and sleep neural signals and (ii) a shared encoder to enhance the alignment of awake and sleep neural signals, significantly contribute to these performances. Collectively, our findings and methodologies represent a significant advancement in the field of sleep decoding.