Scientists at the University of California, Berkeley have developed a technology that can directly stimulate individual photoreceptor cells in the human retina, allowing people to perceive colors that have never been seen before. Using a technique they call “Oz,” the scientists created a vivid blue-green hue—described as “unprecedented” in saturation and named “olo”—that lies outside the range of colors normally perceivable by human vision. Their findings, recently published in Science Advances, offer a powerful new tool for studying vision.
The goal of the project was to explore what happens when light is directed not broadly across the eye, but precisely to specific cone cells responsible for color perception. The research builds on earlier work in color theory and retinal imaging. Traditional color display technologies work by mixing red, green, and blue light to stimulate the L (long), M (medium), and S (short) cone cells.
But this always involves some overlap between different types of cones, due to their shared sensitivities. The Oz system attempts something radically different: stimulating just one type of cone—especially the M cone—while avoiding the others entirely. This kind of precise activation is impossible with conventional lighting or screens but becomes possible through finely targeted laser stimulation.
“I teach a visual computing course at UC Berkeley. In preparing for the lectures on color some years ago, I came across Professor Austin Roorda’s work on mapping the cells in the living human retina, and stimulating individual cone cells,” explained Ren Ng, an associate professor in the Department of Electrical Engineering and Computer Sciences.
“From a color theory perspective, it was clear that no light can stimulate only the M cone cells in the retina in normal viewing, so I was filled with curiosity about what it might look like to show someone a square of color where only the M cone cells were stimulated. We teamed up, became friends, and have pursued this research for some years together.”
To conduct the experiment, the researchers first had to map out the retina of each participant in incredible detail. Five subjects, all with normal color vision, underwent adaptive optics scanning and retinal imaging to classify the L, M, and S cones in a small portion of their retina. Then, using an advanced optical system that tracks eye movement and compensates for it in real time, they delivered laser microdoses directly to specific cones within a 0.9° square visual field—roughly the size of a grain of rice held at arm’s length.
This targeting required both high-speed tracking of eye motion and incredibly fine-tuned hardware to adjust the laser beam in real time. Even a slight misfire would stimulate the wrong cone and produce normal colors. But when done accurately, the stimulation of only M cones produced a visual experience beyond what people typically see in nature.
“When I initially wrote to Austin, I guessed that M-only color might look like the ‘greenest green you never saw’ — this is why we called the system to show it Oz Vision, alluding to the brilliant green of the Emerald City in The Wizard of Oz,” Ng told PsyPost.
“But olo looked blue-green. When Oz system performance matured enough (this took years of improvements) to see olo for the first time, the hue surprised me, delighted me, and excited me — because it was such a clear visual change that signified system precision was now close to targeting individual retinal cells at population scale.”
In color matching experiments, participants compared this new color to others and consistently reported that it could not be matched with any natural or artificial light source. They described it as a highly saturated teal or blue-green that could only be approximated by mixing it with white light. This mismatch confirmed that “olo” exists outside the normal human gamut of colors.
The researchers didn’t stop at color patches. They also used the Oz system to display lines and moving dots in image and video form, such as red shapes on an olo background. In these tasks, subjects successfully perceived orientation and motion only when the laser microdoses were delivered accurately. When the targeting was intentionally disrupted, these shapes disappeared or lost their distinct colors, reducing subjects’ performance to guessing. This confirmed that the perceived images were being formed through accurate cell-by-cell stimulation and not from any inherent qualities of the laser itself.
“As we’ve seen in the past few days, the very existence of olo, a hypersaturated teal beyond the human color gamut, has captured the imagination of the public in an enormous way,” Ng said. “That’s natural, because as humans we are very visual animals, and color is right there in the center of our everyday vision — what could feel more natural and complete than our personal sense of color? I hope what people take away from this story is a sense of the delight that can come from scientific inquiry, of expanding our knowledge, and becoming aware of things just beyond what we can see today.”
The technical achievement of the Oz system lies in its ability to program the retina with unprecedented precision. While other methods—such as silent substitution or adaptation effects—have been used to explore cone-specific activation, they lack the sustained control, spatial resolution, or range that Oz provides. By directly stimulating thousands of cones with rapid pulses of light, the system can produce stable and measurable color experiences, not just fleeting impressions.
The study’s results are significant for both scientific and practical reasons. On the scientific side, they offer a new way to understand the limits and capabilities of human vision. Since all previous color perception research has been bounded by the spectral sensitivities of cone cells, Oz opens a path to explore what the brain does when it receives signals that don’t normally occur in nature. It also allows researchers to test how visual perception is constructed from patterns of cone activation, potentially informing models of visual processing at the neural level.
On the practical side, the Oz platform may have future applications in vision science, clinical diagnostics, or even entertainment. For example, it could be used to test or train individuals with color deficiencies by simulating the presence of a missing photopigment. It might even allow for the perception of “imaginary” colors in people who normally experience the world in only two or three dimensions of color. With further refinement, Oz could potentially enhance virtual or augmented reality experiences by expanding the palette of perceivable colors beyond what any display can currently show.
There are limitations to the current system. It only works in a small region of the retina and requires the subject to maintain strict gaze fixation. Expanding the technique to allow free viewing or to stimulate a wider field of vision would require major technological advances in both optics and computing. The method is also limited to laboratory conditions, relying on equipment that is currently too complex and expensive for broader use.
“This is a basic science project, not a technology or product development effort,” Ng noted. “Folks must understand that they will not be seeing olo on their smartphones or televisions anytime soon. Also, Oz Vision is currently a very small display — the size of your thumbnail at arm’s length or 5 times the area of the full moon.”
Still, the proof-of-principle achieved in this study is an important step forward. By showing that it is possible to program the retina at the level of individual cells and evoke perceptual experiences that go beyond natural vision, the researchers have introduced a new class of display technology—one based not on pixels, but on the biology of the eye itself.
“Olo is a visual sign that Oz Vision is now ready for broad scientific work — we can programmably set the activation of thousands of photoreceptors in the retina, as a basic platform for vision science and neuroscience,” Ng explained. “Some immediate goals are to simulate and study the visual perception that occurs in diseases where a large fraction of the cells have died off but patients are visually unaware; and to study the boosting of color dimension, such as boosting color blindness to full color vision, or full color to tetrachromacy.”
“But the long-term question is the scientific mystery underlying perception. Retinal signals differ so markedly from our experience of color vision, and it remains a mystery how the brain makes sense of such signals and produces the color vision that we enjoy every waking moment. The Oz Vision system enables us to scientifically probe this perceptual miracle in unprecedented ways, and hopefully to elucidate the underlying neural mechanisms.”
“I want to spotlight James Fong and Hannah Doyle, the graduate researchers who co-led this work to success over many years, requiring enormous talent and perseverance,” Ng added. “And my good friend and collaborator, Austin Roorda, who has worked for decades on the science and technology of looking into the eye, seeing individual cells in the retina, and stimulating them with mind-boggling precision.”
The study, “Novel color via stimulation of individual photoreceptors at population scale,” was authored by James Fong, Hannah K. Doyle, Congli Wang, Alexandra E. Boehm, Sofie R. Herbeck, Vimal Prabhu Pandiyan, Brian P. Schmidt, Pavan Tiruveedhula, John E. Vanston, William S. Tuten, Ramkumar Sabesan, Austin Roorda, and Ren Ng.