Blind Mice Detected Infrared Light After Scientists Implanted an Artificial Retina

The device may one day restore sight while adding a new visual channel.

Researchers have developed a soft artificial retina that helped blind mice respond to near-infrared light, a type of light normally invisible to mammals.

The device sits on the surface of the biological retina and converts near-infrared light into electrical signals. Those signals stimulate retinal ganglion cells, which can still survive after diseases such as retinitis pigmentosa or macular degeneration destroy the eye’s light-sensing photoreceptors. In tests, the implant triggered retinal activity, visual-cortex responses, and light-guided behavior in mice.

The most remarkable thing was that it also allowed normal mice to respond to infrared light while keeping their natural visible-light responses intact.

The findings are early and limited to animal experiments. But they suggest a path toward retinal implants that do more than restore fragments of lost sight. Perhaps, at some point, some humans will be able to sense infrared light and, why not, ultraviolet or other wavelengths currently outside our visible spectrum.

A Workaround for a Damaged Retina

The retina is a thin layer of tissue at the back of the eye. Its photoreceptor cells detect visible light and convert it into electrical signals (or better said, electrical impulses), which then eventually reach retinal ganglion cells, which in turn send the information to the brain through the optic nerve.

In many blinding diseases, photoreceptors die first. But other retinal neurons, including ganglion cells, can remain partly intact. This is where the new artificial retinas come. Developed by a team of researchers led by Professor Park Jang-ung at Yonsei University, these artificial retinas stimulate the remaining cells directly, rather than attempting to repair dead photoreceptors.

“Many people suffer from blindness due to retinal diseases that cause photoreceptor degeneration,” the researchers wrote in their new study out now in Nature Electronics. “Electrical stimulation of retinal neurons can recreate the action potentials associated with seeing that are generated by these cells.”

How the Device Works

The artificial retina has three main components.

×

Get smarter every day...

Stay ahead with ZME Science and subscribe.

Daily Newsletter
The science you need to know, every weekday.

Weekly Newsletter
A week in science, all in one place. Sends every Sunday.
No spam, ever. Unsubscribe anytime. Review our Privacy Policy.

Thank you! One more thing...

Please check your inbox and confirm your subscription.

The first is an ultrathin filter that blocks visible light from the device’s sensor while allowing near-infrared light to pass. The second is a phototransistor array, a grid of tiny light-sensitive electronic elements that converts near-infrared light into electrical current. The third is a set of three-dimensional liquid-metal micropillar electrodes that deliver the signal to retinal ganglion cells.

This soft electrode design is one of the most important parts of the study. The retina is delicate, curved, and soft. Traditional rigid electrodes could damage the tissue, trigger inflammation, or scar the interface between the device and the eye.

So, the team used tiny liquid metal electrodes made from an alloy of gallium and indium (both are metals that turn liquid at room temperature). These micropillars were about 20 micrometers wide and 60 micrometers tall. Their softness brought them closer to the mechanical properties of eye tissue than solid metal electrodes.

“The liquid metal electrodes enhance proximity to retinal ganglion cells, providing effective charge injection while minimizing tissue damage, owing to their low Young’s modulus,” the authors wrote.

Testing the Retina Outside the Body

The researchers first tested the device on extracted retinas from normal mice and from lab rodents that were engineered to carry a mutation that led to retinal degeneration.

Normal mouse retinas responded strongly to visible blue light. They showed almost no natural response to near-infrared light. So far so good. Mammalian photoreceptors are not built to detect near-infrared wavelengths.

But when the artificial retina was active, near-infrared light triggered retinal ganglion cell activity in both normal and degenerated retinas. The firing rate increased as the near-infrared light grew stronger. This showed that the device could translate an invisible light signal into neural activity.

The effect was surprisingly strong. The signals from device-driven near-infrared stimulation reached levels comparable to the retinal responses produced by visible light in healthy tissue.

Did the Brain Receive the Signal?

Next, the team implanted the device in living mice. They placed it on the epiretinal surface, meaning the inner surface of the retina. They also implanted probes into the primary visual cortex, the brain area that processes visual signals.

RelatedPosts

Novel glasses can help the color-blind perceive more colors
New Urine Test Could Diagnose Eye Disease
How cats see the world (with pictures)
Japanese woman is first recipient of next-generation stem cells

Before implantation, blind mice did not respond meaningfully to visible or near-infrared light. After implantation, near-infrared stimulation produced strong activity in the visual cortex. This showed that the circuit is complete as the infrared-red signal captured by the artificial retina was read by the brain.

In normal mice, brains continued to respond to visible light through their natural photoreceptors. But when the implant operated, their visual cortex also responded to near-infrared light.

Visible light passes through the biological visual pathway, while near-infrared light passes through the artificial retinal pathway. In the experiments, these routes appeared to coexist, working in parallel, rather than interfering.

A Behavioral Test of Infrared Vision

The researchers then asked a more practical question: could the mice use the infrared signal?

They trained water-restricted mice in a task where a light cue predicted water delivery. If a mouse detected the cue, it began licking before the water arrived.

Blind implanted mice showed anticipatory licking in response to near-infrared light. Blind mice without implants did not. In normal mice, the implant allowed responses to both visible and near-infrared cues. The licking response also increased with stronger infrared light.

This does not mean the mice saw infrared the way we humans see green or blue. It means the device created a signal the animals could sense, and which they could then learn and use.

Why Infrared Vision Could Be Useful

A near-infrared prosthesis has one major advantage: it may avoid competing with a patient’s remaining natural vision.

Many people with retinal degeneration retain some sight. They may still detect light, motion, or peripheral shapes. The world for these patients is like one of dark shadows, but it’s still something. A visible-light prosthesis can interfere with this precious remaining ability. An infrared channel could supplement vision instead, especially in low-light environments.

The authors wrote that “the retina could, in the future, be used to create a NIR visual channel in patients with photoreceptor degenerative blindness without interfering with their remaining natural vision.”

In practice, a human patient might rely on an infrared illumination source in darkness, somewhat like night vision routed through the eye rather than a display. The device could also be tuned for other wavelengths by changing its materials and filter design.

“This research suggests the possibility of human enhancement beyond disease treatment,” said Professor Park Jang-ung, “We anticipate its application in various fields such as night surveillance, national defense, medical diagnostics, and brain-machine neural interfaces.”

The Long Road to Human Vision

The safety data are encouraging but preliminary. Human retinal pigment epithelial cells grown on the device showed high viability. In mice, the implanted electrode structures stayed positioned on the retina for six months without obvious inflammation, malignancy, gliosis, or microglial activation.

“The artificial retina device using liquid metal 3D electrodes significantly reduces damage to retinal tissues and ensures precise and stable contact with the retina’s irregular surface compared to traditional hard metal artificial retinas,” said co-author Professor Byeon Suk-ho of Severance Hospital. “This research paves the way for the development of customized artificial retinas for patients with blindness.”

Still, this proof of concept needs a lot of work. Mouse eyes are tiny compared with human eyes. A useful human implant would need many more pixels and finer control. It would need to operate safely for years. It would also need to work in complex real-world lighting, where sunlight and artificial sources create background near-infrared noise.

It’s also an entire mystery how these infrared light signals would be processed in the eventuality that a person will one day be fitted with a similar artificial retina. Would infrared signals appear as flashes, brightness, outlines, or something entirely unfamiliar? Could patients learn to use them for walking, reading signs, or recognizing objects?

For now, the achievement is a first step in the right direction — and perhaps a glimpse of vision beyond the limits nature gave us.

The findings appeared in Nature Electronics.