A team at National Yang Ming Chiao Tung University has created a light-
controlled chip that simulates brain functions, according to research
published in Small. The chip’s metal-based structure mimics learning and
memory processes, converting short-term memory into long-term memory
through the application of light pulses and voltage control.
The chip also has sensing abilities, enabling applications in autonomous
vehicle vision, medical imaging, and robotics. A prototype module mimics the
retina by detecting brightness and colors, retaining learned information even
after light is removed. Tests on an artificial neural network showed high
accuracy in recognizing digits and images, even under distortion.
The project follows earlier research from the university’s Biomedical
Engineering Department, which in 2023 developed a graphene-based chip for
nerve monitoring and brain stimulation in collaboration with Abbott.
