Grain-of-salt implant records brain signals wirelessly for a year

A mouse's whiskers twitch. Deep in its brain, a device smaller than a grain of salt picks up the electrical pulse and beams it to a nearby computer using nothing but light. No wires. No ports in the skull. No inflammation.

The implant, called MOTE, has just completed a year of recording in mice—roughly half their lifespan—with almost no scarring and no seizures. It's a small proof of concept that could reshape how we monitor and treat the brain.

Why This Matters

Brain implants today face a brutal trade-off. The ones that work best require threading wires through your skull and anchoring them to the bone. The ones that are minimally invasive sit on the brain's surface but pick up weaker signals. Either way, as your brain moves and shifts, the implant stays fixed. Scar tissue builds up. Inflammation creeps in. Within months or years, the device stops working as well.

Scientists have been chasing wireless alternatives for years—using radio waves or ultrasound to power and communicate with implants. But most wireless devices are still too bulky. "Equivalent to a sizable fraction of the mouse brain," as the MOTE researchers put it. For an animal the size of a mouse, that's like carrying a marble in your skull.

MOTE sidesteps this entirely by using light. Red and infrared wavelengths penetrate the scalp and skull with minimal distortion, delivering power directly to a tiny diode inside the implant. The device converts that light into electricity, powers up, records the brain's electrical chatter, and sends it back out as pulses of light—essentially Morse code made of photons.

Inside the implant are 186 transistors arranged into three circuits: one amplifies the brain signals, another recodes them into light pulses, and the third drives the LEDs that transmit the data. The whole package is coated with a custom sheath, built one atomic layer at a time, that protects it from the brain's acidic, corrosive environment.

What the Tests Showed

When researchers implanted MOTE into the barrel cortex of mice—the region that processes whisker sensations—it worked. For a full year, the device reliably detected activity from single neurons and from entire networks firing together as the mice explored their world. When scientists tickled the whiskers, the implant caught it. When the mice behaved naturally, MOTE recorded the neural signature.

None of the mice developed seizures. Scarring around the implant was minimal, even after twelve months. The device could theoretically record from up to six millimeters deep—deep enough to monitor most of a mouse brain, or even organoids, the lab-grown "mini-brains" used in development research.

This is the first time a wireless implant this small has sustained reliable recording for this long without obvious harm. It suggests that the core challenge—keeping a device functional inside the brain without triggering inflammation—might actually be solvable.

Clinical use is still years away. The implant would need to be scaled up for human tissue, tested for safety in larger animals, and validated in actual patients. But the trajectory is clear: brain-computer interfaces are getting smaller, less invasive, and more durable. If MOTE's approach holds, it could eventually be upgraded to record from the spinal cord, heart, or other tissues—opening doors to treatments we haven't yet imagined.