Scientists are one step closer to reading people's minds with a new technology that can decipher the internal speech of the wharf brain's mind with nearly 80 percent accuracy.
Some people are unable to speak due to illness or injury, but brain-machine interfaces (BMIs) are devices that can help patients communicate again.
BMI, also known as a "speech decoder," captures the brain activity of internal speech — words that are thought in the brain without having to make any action or sound — and translates them into words.
Until now, it has been difficult to achieve highly accurate results.
Researchers at the California Institute of Technology implanted tiny devices in specific areas of the brains of two participants.
After hearing a verbal or visual cue, participants are first asked to imagine a word in their minds – such as "spoon", "python", or "battlefield".
The researchers decoded the internal speech in real-time and achieved an accuracy rate of up to 79%.
This is carried out by electrodes attached to a part of the brain, the supraoccipital sulcus (supramarginal gyrus), an area that plays an important role in the processing of spoken and written language.
The device is capable of decoding different internal speech strategies, including silently reading words and imagining the objects represented by the words.
The technology works in the same way as other brain-computer interface devices such as Elon Musk's Neuralink.
Unlike this device, neural links convert the electrical signals it receives from the brain into a series of motor controls that can be used to interact with the machine.
However, it has now been revealed that the first human trial of the neural link was nearly interrupted after a patient developed a potentially life-threatening condition shortly after the implantation procedure.
While further research is needed to improve the function of the technique by testing more participants and new words, the authors state that they recommend the supraoccipital sulcus as a promising location for BMI.
In a paper published in the journal Nature Human Behaviour, the scientists wrote: "In this work, an online internal speech BMI achieved significant decoding in two participants...... The online decoder trained only eight repetitions of 1.5 seconds per word, suggesting that only a few minutes of training data per day is needed to obtain meaningful classification accuracy. ”
"This proof-of-concept suggests that the supraoccipital sulcus may be able to represent a larger internal vocabulary."
"Our results may apply to people who can't speak up or can't act at all."