Science: Images Obtained through "Brain Decoding"

Artificial Intelligence continues to amaze, even as its use becomes increasingly widespread. Japanese scientists have achieved a groundbreaking feat by creating mental images of objects and landscapes generated by human brain activity using artificial intelligence (AI) technology. The research team from the National Institute of Science and Technology for Quantum Physics (QST) and other organizations managed to produce raw images of a leopard, complete with mouth, ears, and a specific spotted pattern, as well as objects like an airplane with red lights and wings, as reported by the Kyodo news agency. The technology, known as "brain decoding," allows the visualization of perceptual content based on brain activity and could be applied in the fields of medicine and social assistance, according to scientists.

During the study, participants were shown 1,200 images of objects and landscapes, and the relationship between their brain signals and the images was analyzed and quantified using functional magnetic resonance imaging. The same images were then fed into a generative AI system to explore their correspondence with brain activity. According to the researchers, this technology could be used in the development of communication devices and to understand the brain mechanisms during hallucinations and dreams. QST researcher Kei Majima stated that people have used microscopes and other devices to see a world that was invisible to the naked eye, but they had not been able to enter another person's mind. Majima mentioned that this is the first time people can look inside someone else's mind. The findings were recently published in the online edition of the scientific journal Neural Networks. This marks a significant first step on a long journey, the destination of which is currently unknown.