The research has not yet been peer-reviewed, but it could lay the groundwork for much more powerful brain-computer interfaces designed to better assist those who cannot speak or write.

How did the test go?

In the experiment, the researchers used MRI machines to measure changes in blood flow to decipher three subjects’ broader moods or meanings, rather than the activity of individual neurons that were “noisy” and difficult to decipher. He was thinking while listening to podcasts and radio broadcasts for 16 hours.

They used this data to train an algorithm that they said could correlate changes in blood flow to what subjects were listening to at the time.

The results were promising, as the decoder was able to extract “pretty good” value, University of Texas neuroscientist and study co-author Alexander Hutt told The Scientist.

traps

The system has some disadvantages. For example, the decoder would often shuffle who was saying what on radio and podcast recordings. In other words, the algorithm “knows perfectly well what’s going on but doesn’t know who did it.”

The algorithm also failed to take what it learned from the semantics of one participant’s brain scan and apply it to the other’s scan, which is intriguing.

Despite these shortcomings, the decoder was even able to infer the story when participants watched a silent movie, meaning it wasn’t limited to colloquialism. This suggests that these findings may also help us understand the functions of different areas of the brain and how they intersect in our understanding of the world.

Scientists are happy

Other neuroscientists not directly involved in the research were stunned. Sam Nastase, a researcher and professor at Princeton’s Neuroscience Institute, called the research “amazing” and told The Scientist in an interview, “If you have a smart enough modeling system, you can get a really incredible amount of information of this kind.” record.

Yukiyasu Kamitani, a computational neuroscientist at Kyoto University, concurred, saying in an interview with The Scientist that the study “provides a strong foundation for applications.” [інтерфейсу між мозком і комп’ютером]”.