Our days of keeping things to ourselves are long gone. We no longer keep our thoughts to ourselves. It has now been possible for scientists to decode the brain scan of a person’s thoughts with the help of a decoder that uses artificial intelligence (AI). University of Texas is the maker of this invention, which has the goal of being of assistance to people who are unable to speak or communicate for the first time as a means of self-expression.
By using this method, it is possible to decode the language in real-time. In spite of the fact that many might not understand how this method works, this method uses functional magnetic resonance imaging (fMRI) to feed information to the algorithm which then reconstructs arbitrary stimuli that are heard or thought by the person into natural language.
Must Read: Copy and paste the brain?
It was during the course of the experiment that scientists recorded many stories and scanned parts of the brain associated with natural language to get a better understanding of how the decoding algorithm was working. In the future, a decoder will return a summary of what someone listened to in the past.
A current method of reading brain signals is by implanting electrodes into the brains so as to read the signals, but this is not a complete solution. The new model which is the first non-invasive technique, which can analyze the ideas of the patient’s thoughts, is able to analyze the ideas of the patient’s thoughts, however, it cannot decode the words that the patient thinks from one to the other.
Among the many challenges that humans face when managing their brains, Popular Mechanics reported that the brains of humans are made up of smaller pieces that react differently to different kinds of stimuli. As the different thoughts are circulating, they can vary from simple, such as ‘Cat,’ to complex, such as ‘I should pick up the cat, ” that can vary from simple to complex. When utilized in conjunction with other elements from the brain, it forms the form of complex thoughts that can encompass many different concepts such as place, color, size, and many others that together can point to a specific concept. Depending on how the brain is wired, every letter has a different function so when the functions of each of these parts are combined, a person can understand what that person is thinking because the system is so complex.
There is a limitation to the system’s ability to decode the word-for-word what the individual is thinking, but it can provide an idea of what their thoughts exactly are. An MRI machine enables a person to see what they see in a picture to be described in the same way that they see it in the machine.
Must Read: Two Pakistani engineers have developed a prosthetic arm which can be controlled using the brain
There was no doubt that this overall task had not been an easy one according to the researchers. In order to compare the effects of podcast listening on three parts of the brain associated with natural language and those associated with fMRI, a group of people listened to 16 hours of podcasts while a few fMRI scans were recorded.
This study examined three brain regions in order to determine whether the same effects were present in the prefrontal network, the classical language network, or the parietal-temporal-occipital association network
The newly created algorithm was given to the scanner and then the system took this scan recording and transformed it into content based on storytelling.
As presented in the study in BioXiv, the algorithm decoded the story into the following: “For some reason I thought that maybe my wife would come to me and tell me that she misses me if I see her.” This is the decoding algorithm’s interpretation of the story: “Look for a message from my wife saying that she changed her mind and was coming home.”
To date, this system has not been able to transform the word-to-word thinking of an individual, but, rather, to provide a better sense of what that individual is thinking.
