Neuroscientists invented a method that allows to convert brain waves into words.
In fact, our thoughts are located in a region of the brain called superior temporal gyrus, which is a part of the auditory system. It makes it possible for us to elicit sense from the sounds we hear, distinguish words and understand their meaning. This brain region was used by the scientists of the University of California at Berkeley (USA) to decode the inner dialogue of the human mind.
The experiment involved 15 people, and 256 electrodes were introduced into the skull of each of them. The electrodes were used to catch electrical signals, emitted by the brain when thinking, since each word has its own unique set of brain impulses.
Then the subjects were taken to listen to audio recordings of different words, and the devices fixed the brain signals. As a result, it was quite difficult to unravel the chaotic flow of electric impulses arising in the temporal lobe while listening to the audio.
“We have seen in which parts of the brain cortex the activity increased at a time when a person heard a sound of a certain frequency and tried to restore the original sound”, explained Robert Knight, professor of psychology and neuroscience.
The researchers have found the key frequencies that distinguish each speech sound from the others and then, with the help of a computer model, restored the words that “sounded” in the subjects’ head.
Then the volunteers were given a choice of a word set and asked to choose one word and think about it. It turned out that the same computer model allowed to find which word had been chosen. Moreover, it turned out that it was possible even to restore them by converting recorded brain waves back into sound waves.
“Words uttered by the brain were decoded with difficulty, but this way of “reading” the internal monologue of human mind works quite well”, says Dr. Pasley.
The new way of “mind reading” will allow paralyzed patients to recover speech. Once they think what they want to say out loud, their requests will be fulfilled.
However, the developers admit that there is still a lot of work to improve the sound recognition technology and to replace the procedure of the electrode introduction into the skull by something less painful.