Five people view a never-before-seen color called ‘olo’
Humans typically perceive color based on the particular wavelengths of light reaching the retina. Deposit Photos
Get the Popular Science daily newsletter💡
The rainbow may be due for an update. A new, high-tech method for displaying color has allowed five test subjects to see a shade beyond the standard human range. The research, published April 18 in the journal Science Advances, is proof-of-concept for a technique that could allow neuroscientists to probe previously un-answerable questions about visual perception. In time, it might even help color blind people experience the full color spectrum, and enable regularly-sighted people to differentiate between hundreds, thousands, or millions of previously undetectable hues.
“It’s a technological tour de force,” Jay Neitz, a neuroscientist and professor in the department of ophthalmology at the University of Washington who was not involved in the new study, tells Popular Science. “What they’ve been able to do, it almost falls into the realm of science fiction. It’s so amazing– the technology that’s going on here.”
The newly described method and prototype machine is called the Oz Vision System,(a not-so-subtle nod to reaching somewhere over the rainbow. And the new color enabled by Oz is named “olo,” a reference to its theoretical color space coordinates, which are [0,1,0].
What is Color Space?
Color space is the standard way of charting all the many hues visible to humans. It is based on the idea of trichromacy: that most people have three types of photoreceptor cone cells. We have photoreceptors tuned to short wavelengths, medium wavelengths, and long wavelengths– corresponding to blue, green, and red respectively. With these three types of cones, most can discern somewhere around one million different shades within the spectrum of visible light.
A colorspace chart used to calibrate printers. CREDIT: Paulschou at en.wikipedia, CC BY-SA 3.0, via Wikimedia Commons
But there are colors even within the visible spectrum that are impossible to actually perceive for trichromatic humans. That’s because the three types of cone cells overlap in their responses to certain wavelengths of light. The middle (m) photoreceptor response, tuned to green, overlaps with both the long (l) and short (s) photoreceptor responses on either side of the spectrum. There is no wavelength of visible light that naturally stimulates only these m cells in human eyes. So, every time you see green, you’re seeing it mixed with a little bit of something else: yellow or blue from the l or s cells.
Oz enables researchers to circumvent that inherent limitation of human vision. The protocol allows scientists to stimulate individual, pre-selected sets of photoreceptors, including just m-cells, on their own. In response, subjects get a peek at a green (or blue-green, depending on who is describing it) so intense and pure that it’s long been classified as an “imaginary color”.
Welcome to Oz
The first step with Oz is to create a detailed map of an individual person’s retina: classifying every single cell. That personalized map is then used to program an eye-safe laser to deliver a focused beam of light so precise that it can hit just one cell at a time. To achieve this, a computer has to detect and correct for the tiny, but unavoidable movements of a person’s eye in real-time. Stimulating just a single cone cell doesn’t create any perceivable color, so Oz goes a step further and rapidly moves its laser in a zig-zag pattern across a predetermined patch of cells. Oz only sends out its beam when it passes over a target cell. In the case of the newly published study, these target cells were cones classified as m photoreceptors in the mapping stage.
Normally, humans perceive color based on the particular wavelengths of light reaching our retinas and stimulating our photoreceptor cells in a particular ratio and pattern. But with the Oz Vision System, a single wavelength of light can be used to create the perception of innumerable different colors because cells can be so selectively stimulated.
The current prototype includes an array of sensors, laser light sources, mirrors, and photon counters, and combines multiple advances–years in the making–into a single system. “It’s really the culmination of all these technologies that have been developed over decades,” says Sara Patterson, a neuroscientist and assistant professor in ophthalmology at the University of Rochester who was not part of the new research team. “I just think it’s fantastic,” she adds.
The study authors tested this process out with five human subjects, and took multiple steps to verify that what these participants were seeing was truly new. “It’s a very well controlled experiment,” Patterson says. They stimulated the perception of olo against different color backgrounds, with moving overlays, and also directly up against some shades on the edge of (but still within) the normal human color space. In this final type of trial, they asked subjects to use a dial to modify the olo square until it matched the non-imaginary color square. In all cases, the participants had to dilute olo with a significant amount of white light until they reported a match.
The green machine
So, what is this new color like?
“Olo looks like a blue-green color that is just the most saturated blue-green or teal that I’ve ever seen,” Ren Ng, a computer scientist and visual computing expert at UC Berkeley and one of the study co-authors, tells Popular Science.
In addition to being part of the research team, Ng was also among the study test subjects who got to witness olo firsthand. “It’s very nameable. It’s very perceivable,” he says, but it’s simply more intense than any natural color. He compared the experience of olo to the first time he saw a green laser pointer. “I probably would have said in the moment, ‘wow, that’s the greenest green I’ve ever seen,’” but now, olo trumps it.
Hannah Doyle calibrates the scanning laser opthalmoscopy system prior to an Oz Vision experiment. CREDIT: Ren Ng.
To view the color, subjects had to keep very still, with their eyes positioned exactly in place– facilitated in part by a bite-bar. Then, they fixed their eyes on a point in space, while the laser stimulated a square of cells off to the side. With that stimulation, olo appeared in a patch about twice as large as the full moon looks in the sky, says Ng. Blinking would reset the motion correction system, and so olo was only visible for seconds at a time before disappearing and then flashing back. Nonetheless, even that limited experience was “so cool,” says Ng. “I’m just so tickled about it.”
A rainbow of possibilities
He’s even more excited about the future. Olo is evidence that this type of precision photoreceptor activation is possible. Now that the method is proven, there’s much more that might be done.
The research team is currently exploring whether or not Oz can be used to enable color blind people, who are functionally dichromats (i.e. missing one cone type) to temporarily see the full human range. Theoretically, this is possible by artificially classifying a subset of cone cells as the missing type of photoreceptor, and selectively targeting them with laser stimulation, out of sync from the rest of the cells, Ng explains. So far, he says the work is progressing well.
It’s not the first attempt to reverse color blindness. In a landmark 2009 study, Neitz and a team of colleagues used gene therapy to introduce a third type of photoreceptor cell in color blind monkeys. The experiment was a success according to all of their tests, enabling the monkeys to discriminate between objects they previously couldn’t tell apart.
However, monkeys can’t explain their experience to human researchers nor definitively confirm that they’re perceiving colors they couldn’t before. “We don’t really know what they [were] seeing,” Neitz says. But color blind humans offered a similar opportunity through Oz could confirm, one way or the other, if a third photoreceptor leads to normal visual perception or something else. “This is actually a fantasy that I had years ago,” he says, and now it seems within reach.
[ Related: How this computer scientist is rethinking color theory. ]
In the longer term, Ng and his colleagues hope to go even farther. The authors imagine they could eventually use Oz to simulate the experience of tetrachromats: animals (like birds and fish) and exceedingly rare humans that have four types of photoreceptor cells and a 100-fold larger color repertoire. But the technology isn’t quite there yet.
Though Oz is, by all accounts, an impressive achievement– the system is not perfect, notes Gregory Schwartz, a neuroscientist and associate professor at Northwestern University. The research, he says, is “beautiful” and “really exciting.” However, there are still limitations to the tech, which Ng and his co-authors acknowledge and catalog in the study.
Though Oz is a more targeted photoreceptor stimulation method than has even existed for humans before, it is not 100 percent accurate. There is still a significant amount of “light leakage.” About ⅔ of the photons directed by the laser end up captured by non-target cells. “They were quite honest about that in the paper,” says Schwartz. Despite the leakage, he’s convinced that olo is still outside the normal human color space, “but probably not as far outside as they wanted to go.”
Another other major limitation is the size and scalability of the Oz prototype, Schwartz notes. We’re far away from portable glasses or screens that can track eye movements well enough to deliver a full Oz color experience. And the need for a detailed retina map makes adding study subjects a resource-intensive endeavor (that’s why the number of participants was so small–another limitation). But the possibility of hyper-color virtual reality is closer than it’s ever been before.
Generally, in the field of color perception research, scientists go back and forth on the same sorts of questions, using the same sorts of methods, says Patterson– debating things like the neural pathways that allow for color vision, or the relative role of the retina vs. brain. But Oz offers an entry point to a total new realm, she notes.
Already, the fact that all five of the research subjects described the color olo so similarly and that they were all able to perceive it as distinct from normal human color space prompts interesting questions about how flexible or rigid our visual perception is, she explains. Neuroscientists have long been uncertain if humans would even be able to make sense of a new color when presented with one. This adds to the evidence that, in certain contests, our brains can comprehend unfamiliar hues.
“Sometimes, when you push the system way out of its normal operating range, like they’re doing here, then you can really learn new things,” Patterson says. “ I can’t wait to see what’s next.” It’s hard to imagine how colorful it might be.
English
Arabic
French
Spanish
Portuguese
Deutsch
Turkish
Dutch
Italiano
Romaian
Portuguese (Brazil)
Greek