Artificial intelligence in neuroscience: a system to turn thoughts into text
Artificial intelligence — While artificial intelligence has recently become a hot topic in technology, it has also become a hot topic in science.
Artificial intelligence is being researched by scientists in a variety of domains.
For example, a peer-reviewed study published in the Monday issue of Nature Neuroscience magazine demonstrated how it may be applied to brain activity.
Scientists developed a noninvasive artificial intelligence system that can convert people’s brain activity into a stream of text, according to the research.
Artificial intelligence & neuroscience
By making large-scale neuroscience datasets more efficient and accurate to examine, Artificial intelligence can benefit neuroscience.
It has the promise of producing more accurate models of neural systems and processes.
Artificial intelligence can also aid in the development of new diagnostic tools and treatments for neurological diseases.
The system
The system is known as a semantic decoder.
It may be beneficial to those who have lost their physical ability to communicate as a result of a stroke, paralysis, or other degenerative illnesses.
The technology was developed by academics at the University of Texas in Austin using a transformer model.
The transformer paradigm is similar to OpenAI’s ChatGPT and Google’s Bard.
Participants in the most recent trial taught the decoder in an fMRI machine by listening to hours of podcasts.
It is also a larger piece of equipment used to monitor brain activity.
The semantic decoder does not require surgical implantation.
Benefits
Artificial intelligence can help neuroscience create techniques for thoughts to become text by using machine learning algorithms to evaluate brain activity patterns associated with language processing.
By analyzing patterns of brain activity, artificial intelligence systems may distinguish certain words or phrases that a person is thinking about and then use this information to generate associated text output.
This technology has the potential to revolutionize communication for individuals who are unable to speak or type, such as those suffering from severe paralysis or communication issues.
More research, however, must be done to enhance the accuracy and reliability in these systems, as well as tackle the ethical and privacy concerns connected to accessing and interpreting people’s thoughts.
Text
The artificial intelligence system creates a stream of text when users are listening to or imagining telling a new tale.
The prepared text may not be an exact transcript, but it was meant to give basic notions or ideas by the researchers.
According to a new release, the trained system creates language that around half of the time closely fits the intended context of the participant’s original thinking.
For example, if a participant in the experiment overheard the words “I don’t have my driver’s license yet,” the response would be “She hasn’t even begun to learn to drive yet.”
Read also: Snapchat and its AI set off alarm among parents
The absence of implants
One of the study’s key researchers, Alexander Huth, stated:
“For a noninvasive method, this is a real leap forward compared to what’s been done before, which is typically single words or short sentences.”
“We’re getting the model to decode continuous language for extended periods of time with complicated ideas.”
The semantic decoder, unlike earlier decoding systems under development, does not require surgical implants, making it noninvasive.
In addition, participants are not forced to utilize just terms from a preset list.
Potential misuse
Concerns regarding the technology’s potential misuse were also addressed by the researchers.
According to the study, decoding only worked when people volunteered to educate the decoder.
The results of individuals who did not use the decoder were unintelligible.
Furthermore, participants who used the decoder but showed resistance produced useless results.
“We take very seriously the concerns that it could be used for bad purposes and have worked to avoid that,” said researcher Jerry Tang.
“We want to make sure people only use these types of technologies when they want to and that it helps them.”
Due to the time required on an fMRI machine, it is only available for use in the laboratory.
The findings, according to the researchers, might be extended to other, more portable brain-imaging methods, such as functional near-infrared spectroscopy (fNIRS).
“fNIRS measures where there’s more or less blood flow in the brain at different points in time, which, it turns out, is exactly the same kind of signal that fMRI is measuring,” said Huth.
“So, our exact kind of approach should translate to fNIRS.”