Credits: Social Media
Scientists use brain decoding and AI to create the first ever image from human thought
Mind-reading might be a futuristic concept on paper, but scientists in Japan have now generated the first ever mental images purely from human brain activity.
Mind-Reading technology has forever sounded like a distant concept better suited to science fiction movies and TV shows than reality. Now, the process of visualizing images solely through brain activity has taken an enormous step forward thanks to a team of researchers in Japan generating the first-ever mental images from human thought.
Japan scientists create first ever mental images from brain activity
Over the past decade, scientists have been able to take data from human brain activity and generate images based on what a subject was thinking about, aka mental image reading.
However, that process was extremely challenging, and were typically only ever able to reveal images of human faces, letters, or shapes – but never complex pictures such as a specific animal, location, or object.
Now, a team of Japanese scientists at the National Institutes for Quantum Science and Technology and Osaka University have claimed that they have successfully created the worlds first mental images of objects and landscapes, purely from human brain activity.
According to the paper, published in the Neural Networks Journal Volume 170, the scientists were able to use new technology that they dub Brain decoding and generative Artificial Intelligence software to visualize elements of a mental image.
Participants were shown 1,200 sample images of different (but distinctive) animals, objects, and landscapes as their brain activity was being analyzed by functional magnetic resonance imaging (fMRI).
The researchers were then able to train a generative AI to decipher the signals and create digital replicas with remarkable accuracy, being described as a significant milestone for the scientific community.
The experimental results demonstrated the capabilities of our proposed framework in reconstructing both natural images and artificial shapes that were imagined by human participants.
What direction could this research take moving forward?
As you can probably imagine, the potential consequences of research such as this are outstanding especially for the fields of psychology, cognitive science, and neuroscience.
As the team noted, their results not only highlight the efficacy of our proposed framework but also offer a unique tool for investigations into the subjective contents of the brain, encompassing phenomena such as illusions, hallucinations, and dreams.
In line with recent attempts to reconstruct imagined speech, our proposed framework potentially paves the way for mind-reading technology, underscoring the necessity of addressing ethical concerns associated with mental privacy.
The scientists cite two critical questions for future research: 1) Can brain decoders be trained without the subject’s cooperation? and 2) Can imagery spontaneously arising in the mind be accurately reconstructed?
Here, the team believes that with more research, it could be possible to extend this mental image process beyond the 1,200 sample pictures and across decoders who werent specifically trained for the same participant. If both questions could be successfully answered, it would revolutionize the medical and welfare landscape.
Balancing the fine line between science fiction vs science reality; the ultimate goal is that one day, research like this could even lead to friends and family being able to peer inside the minds of patients in medically induced comas, or patients suffering from hallucinations, to see specific mental images.
The entire research paper is available to read online for free.
