Scientists have discovered how our brain understands the meaning of words by studying individual brain cells. They recorded the activity of neurons in the left prefrontal cortex while people listened to various sentences and stories. They found that certain neurons react to specific word meanings and can tell apart words from non-words.

These neurons don’t just remember words as static images but instead respond dynamically based on the context of the sentence. For example, the neuron might react differently to the word "rose" in the sentences "He picked the rose" versus "He finally rose". This means that our brain uses a fine-tuned and context-dependent process to understand language.

The researchers identified nine broad categories of word meanings, such as actions, objects, and animals. Each category activated different sets of neurons. They also showed that neurons are organized in a way that they can group similar meanings together, like "rain" and "clouds" being close, and different meanings further apart, like "clouds" and "dad".

By training computer models with the neuron activity data, the scientists could predict which category a word belonged to even if it was part of a new story. This demonstrates that the brain's method of processing word meanings is both generalizable and robust.

Overall, this study helps us understand the detailed workings of language comprehension at the cellular level, revealing how the brain dynamically represents meanings in real-time during speech. This could have implications for developing better communication aids and understanding language disorders.