A research team from Baylor College of Medicine in the United States has found that the human brain still possesses complex and advanced language processing abilities even in unconscious states caused by general anesthesia. This achievement, published in the latest issue of the journal Nature, challenges people's previous understanding of the role of consciousness and cognition, and may open up new avenues for understanding memory, language, and brain computer interfaces. During the epilepsy surgery, the team recorded the neural activity of hundreds of neurons in the hippocampus while the patient was receiving general anesthesia. The hippocampus is an important brain region closely related to memory. With the help of Neuropixels probe technology that has not been previously applied in this region, the team has for the first time obtained information on the unconscious processing of sound and language in this brain area. In the initial experiment, the team played repeated tones to patients and interspersed them with different sounds. The results showed that hippocampal neurons are able to recognize these unusual tones, and this recognition ability increases over time. This indicates that even under anesthesia, the brain still possesses certain learning abilities and neural plasticity. Then they played short stories to the patients while recording their neural responses. It was found that the hippocampus exhibited real-time processing ability for language. Neural activity indicates that the brain is able to distinguish parts of speech based on neuronal firing patterns, such as nouns, verbs, and adjectives. What surprised the team even more was that these neural signals could predict the words that were about to appear in the sentence. This' predictive encoding 'is usually thought to only exist in a conscious and focused state, but this time it has been observed in an unconscious state. These findings suggest that advanced cognitive functions such as language comprehension and prediction are not entirely dependent on consciousness itself. On the contrary, the emergence of consciousness may rely more on extensive collaboration between different brain regions rather than the activity of a single brain region. The study also shows that the brain's ability to predict language is similar to the way large language models in current artificial intelligence work. This provides a new perspective for understanding the information processing mechanisms of biological and artificial systems, and is expected to promote the development of technologies such as brain computer interfaces. (Looking into the New Era)