This content originally appeared on DEV Community and was authored by Promise Omoigui
Can you picture the idea of someone being able to eavesdrop on your thoughts? It sounds impossible right? That's what I used to believe until I learned otherwise.
Recently, scientists at The University of Texas have managed to translate a person's brain activity into text by utilizing AI and MRI scans. Specifically, they were able to decode brain signals while the subject was either listening to or imagining telling a story.
It's important to understand that the system does not provide a literal translation of the thoughts, as that would be quite unfeasible. Rather, it conveys the essence or general meaning of the imagined scenario.
The study which is published in the journal Nature Neuroscience could prove to be tremendously beneficial for individuals who are mentally aware but unable to verbalize their thoughts, for instance those who have suffered a stroke.
According to the statement, the project was spearheaded by Jerry Tang, a computer science doctoral candidate, and Alex Huth, an assistant professor of computer science and neuroscience at UT Austin. The research incorporates a transformer model similar to those employed by Open AI's ChatGPT and Google's Bard, to some extent. Unlike other language decoding methods, this technique does not necessitate the use of surgical implants. It's worth noting that this does not imply that thoughts can be read without consent; the participant must actively collaborate with the researcher.
"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." - Huth said in a statement.
How Does This Works?
The process of the technology involves subjects being instructed to listen to several hours of podcasts while being scanned in a functional MRI machine. The act of listening to these podcasts triggered specific brain activity, which was then decoded by the machine, generating corresponding text. As previously mentioned, the decoder does not produce an exact word-for-word transcription; instead, it generates text that conveys the intended meaning of the original words. To illustrate, if a participant listening to a speaker said, "I don't have my driver's license yet," the machine decoded it as "She has not begun to learn to drive yet."
In addition, the participants were instructed to watch four brief videos without sound while inside the scanner. The semantic decoder used their brain activity to describe certain occurrences from the videos.
How Practical Is This System
At present, the system is dependent on an fMRI machine, thereby restricting its use to laboratory settings. However, the researchers suggest that this technology could be adapted for use with other brain-imaging systems, such as functional near-infrared spectroscopy (fNIRS), which are more portable.
"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," Huth said. "So, our exact kind of approach should translate to fNIRS," although, he noted, the resolution with fNIRS would be lower.
The researchers are cognizant of the possibility of malicious individuals misusing the technology. Which is why Tang said "We take very seriously the concerns that it could be used for bad purposes and have worked to avoid that. We want to make sure people only use these types of technologies when they want to and that it helps them."
Study Abstract
A brain–computer interface that decodes continuous language from non-invasive recordings would have many scientific and practical applications. Currently, however, non-invasive language decoders can only identify stimuli from among a small set of words or phrases. Here we introduce a non-invasive decoder that reconstructs continuous language from cortical semantic representations recorded using functional magnetic resonance imaging (fMRI). Given novel brain recordings, this decoder generates intelligible word sequences that recover the meaning of perceived speech, imagined speech and even silent videos, demonstrating that a single decoder can be applied to a range of tasks. We tested the decoder across cortex and found that continuous language can be separately decoded from multiple regions. As brain–computer interfaces should respect mental privacy, we tested whether successful decoding requires subject cooperation and found that subject cooperation is required both to train and to apply the decoder. Our findings demonstrate the viability of non-invasive language brain–computer interfaces.
This content originally appeared on DEV Community and was authored by Promise Omoigui
Promise Omoigui | Sciencx (2023-05-03T07:45:01+00:00) Are we achieving yet another unimaginable feat with AI?. Retrieved from https://www.scien.cx/2023/05/03/are-we-achieving-yet-another-unimaginable-feat-with-ai/
Please log in to upload a file.
There are no updates yet.
Click the Upload button above to add an update.