Tech-Enabled Telepathy Moves Closer to Reality With Latest Breakthrough

Tech-Enabled Telepathy Moves Closer to Reality With Latest Breakthrough

Scientists are inching closer to making telepathy a reality. In a new study published this week, researchers claim to have created a device that can read and translate a person’s internal speech. The findings only show modest success so far, however, and there are still many hurdles remaining before such devices can have practical applications.

In recent years, scientists have advanced technology that reads and interprets the complex brain signals that allow us to talk to another person—tech that has made it possible for people with speech difficulties to regain some amount of communication ability. Much of this work so far has involved translating a person’s partially vocalized or mimed speech into words or audio. But scientists at the California Institute of Technology appear to have broken new ground in this emerging field, creating a brain–machine interface (BMI) device that can translate internal speech, at least on a rudimentary level.

The researchers recruited two people with spinal cord injuries to take part in their study. Both had electrodes implanted in their supramarginal gyrus, a brain region that previous research had suggested was important to the formation of internal speech.

Over the course of three days, the volunteers were trained to imagine saying a series of six words (battlefield, cowboy, python, spoon, swimming, and telephone) and two nonsensical terms (nifzig and bindip) as their brains were monitored. The initial measurements were then fed into a computer model that tried to decode and interpret the volunteers’ brain signals as they thought about saying these words during a subsequent session conducted in real time.

As hoped for, the researchers did find unique brain activity in the supramarginal gyrus when the volunteers were internalizing their speech compared to vocalizing it, supporting the idea that it plays a major role in the process. Overall, their model was 79% accurate at predicting the first subject’s internal speech and 23% accurate at predicting the second subject’s speech.

“This work represents a proof-of-concept for a high-performance internal speech BMI,” the authors wrote in their paper, published Monday in Nature Medicine.

Obviously this research is only an early illustration of this technology’s potential. Given the widely varied results between the two volunteers, it’s also clear that scientists have much more to learn about how our brains work to produce internal speech. And we’re likely a long way off from using these BMI devices to efficiently translate the thoughts of a person with no external communication ability, such as people with locked-in syndrome, which would be an especially profound application for these individuals and their loved ones.

Aside from its medical uses, learning how to decipher the brain signals that underline internal speech and thoughts could lead to more radical possibilities in the future. Other researchers have created interfaces that allow people’s brains to communicate with one another from afar, for instance. So pairing these various kinds of brain-machine technologies could make a form of mind-reading someday feasible.

Still, these findings do mark an important step forward, and the researchers are already working on further improvements. They next hope to find out if their tech can reliably tell apart the individual letters of the alphabet.

“We could maybe have an internal speech speller, which would then really help patients to spell words,” study co-author Sarah Wandelt told Nature News.

The Cheapest NBN 50 Plans

It’s the most popular NBN speed in Australia for a reason. Here are the cheapest plans available.

At Gizmodo, we independently select and write about stuff we love and think you'll like too. We have affiliate and advertising partnerships, which means we may collect a share of sales or other compensation from the links on this page. BTW – prices are accurate and items in stock at the time of posting.