Neonink of Elon Musk he cried making waves the technical side of neural implants, but did not show how to do it use implants. In the meantime, the demonstration of the promise of resources remains in the hands of the students.
This week, locals donated instead an interesting example promise of neural resources. Using another plant, a disabled person was able to write about 90 letters a minute, assuming that he was writing them by hand.
Previous efforts in empowering people with disabilities through the incumbent also include providing the actual keyboard education and allowing them to guide the guide and their ideas. This method is useful but slow, and requires the user’s attention, as the header should monitor the direction of the cursor and determine the time to match the printer. It also requires the user of the time to learn how to manage the system.
But there are other ways to get rid of brain clots and go to the page. Somewhere in our thinking, we have a goal of using someone else, and using a set to follow that goal can work. Unfortunately, this method does not make sense.
Underneath that goal, the choice spreads to the motor cortex, where it is translated into action. Once again, there is a demanding component, where the motor cortex recognizes that it is making a letter (in writing or in writing, for example), which is translated into the muscles that need to take action. These methods make sense, and that’s what the research team is looking for in their new work.
In particular, the researchers placed two devices in the preotor cortex of a disabled person. This area is considered to be instrumental in making travel goals. Holding on to these goals can produce a much clearer identification than holding on to the movement itself, which can be difficult (each move affects multiple muscles) and depends on the situation (where your hand is aligned with the page you are typing, etc.).
With that in mind, the researchers asked the student to consider writing letters on a piece of paper and recording the neural events in which they did so.
Overall, there were about 200 electrodes in the preortor cortex of participants. Not all were teachers of letter writing. But for those who were present, the authors analyzed the main points, which explained the neural notes that differed significantly when considering different letters. Turning the artwork into a two-dimensional image, it was clear that what appears to be written by one person is always connected. And pronounced the same-p and bFor example, or h, n, and r– clusters formed nearby.
(The researchers also asked the student to create posters such as a comma and a question and used> to show the location with a short time.)
Overall, the researchers found that they were able to understand the right person with less than 94% accuracy, but the machine needed to be screened slowly after the neural history was recorded. In order for things to work in real time, the researchers taught neural integration techniques to see if the identification corresponded to each letter.
Although they worked with very few exceptions (only 242 sentences), the system worked very well. The balance between the ideas and the visuals on the screen was about half a second, and the participants were able to create about 90 characters per minute, easily adding to the old record-taking history, which was about 25 characters per minute. The incidence of uncooked errors was about 5%, and the use of a machine as a faulty recorder could reduce the error by up to 1%.
The test was performed with pre-determined sentences. As soon as the procedure was approved, however, the researchers asked the student to write free answers to the questions. Here, the speed dropped slightly (75 characters per minute) and errors increased to 2% after repair, but the machine still worked.
Source link
