Scientist Develop First Mind-Reading Helmet That Can Write What You Think

Researchers at the University of Technology Sydney's GrapheneX-UTS Human-centric Artificial Intelligence Centre have created a portable, non-invasive device that can translate silent thoughts into text, a first for the world. This invention can help people who suffer from illnesses or injuries that prevent them from speaking, such as paralysis or a stroke, to communicate with others. It might also make smooth communication possible between humans and machines, which would be useful for controlling robots or bionic arms.

According to the press release, participants in the study read texts silently while donning a cap that used an electroencephalogram (EEG) to capture electrical activity in their brains through their scalps. The moment the sensors detect that a participant is thinking, they begin to function and send them a random text. It records their brain waves while they mentally read the sentences, and then it translates those waves into a different text that looks quite similar to the original. Further, EEG recordings use a smart artificial intelligence model called DeWave to transform the results and recordings into text, giving users a unique mind-reading experience.

According to a video shared by the university, in one of the tests, a participant was asked to think "Good afternoon! I hope you're doing well. "I'll start with a cappuccino, please, with an extra shot of espresso." A screen displayed the AI at work turning the brainwaves into a written response after a short while. The result was, "Afternoon! You well? Cappuccino, Xtra shot. Espresso." 

CT Lin, Director of the GrapheneX-UTS HAI Centre, who led the research said, "This research represents a pioneering effort in translating raw EEG waves directly into language, marking a significant breakthrough in the field. It is the first to incorporate discrete encoding techniques in the brain-to-text translation process, introducing an innovative approach to neural decoding. The integration with large language models is also opening new frontiers in neuroscience and AI."

Overall, certain statements were more difficult than others, but they still got a 40 per cent success rate out of the 29 participants evaluated. "The model is more adept at matching verbs than nouns. However, when it comes to nouns, we saw a tendency towards synonymous pairs rather than precise translations, such as 'the man' instead of 'the author'. We think this is because when the brain processes these words, semantically similar words might produce similar brain wave patterns. Despite the challenges, our model yields meaningful results, aligning keywords and forming similar sentence structures," study author Yiqun Duan said.

The use of EEG signals received through a cap rather than electrodes implanted in the brain results in a noisier output. However, in terms of EEG translation, the study demonstrated a performance that exceeded previous benchmarks.

On the other hand, the scientists also stated that this advancement is non-invasive, relatively inexpensive and easily transportable as compared to Elon Musk's Neuralink, brain-chip startup. Recently, it received approval from an independent review board to begin recruitment for the first human trial of its brain implant for paralysis patients. The study will use a robot to surgically place a brain-computer interface (BCI) implant in a region of the brain that controls the intention to move, Neuralink said, adding that its initial goal is to enable people to control a computer cursor or keyboard using their thoughts alone.

The billionaire has grand ambitions for Neuralink, saying it would facilitate speedy surgical insertions of its chip devices to treat conditions like obesity, autism, depression and schizophrenia.