Skip to Main Content

PALO ALTO, Calif. — In the name of science, Andrew Huberman has gone diving 40 feet underwater with great white sharks. He’s gone mountain climbing without ropes or harnesses, traversing some sections where one slip would have sent him plummeting 650 feet.

Now, the Stanford neuroscientist is embarking on a different kind of daring quest, testing an intriguing but unproven hypothesis: that virtual reality could be used to preserve or even restore vision for patients with glaucoma.

Academics and companies all over the world are betting on virtual reality to help patients with conditions like anxiety, depression, PTSD, and ADHD. Huberman believes in the potential of the technology for vision issues, too, but he speaks about it less like an evangelist than someone who’s discovered a useful tool. He’s also harnessing it in pioneering ways.


In the case of his glaucoma clinical trial, Huberman is asking patients to gaze at flashing white dots in the hopes that they can trigger the firing of neurons that connect the eye to the brain and coax them to regenerate.

He’s also using virtual reality to study the physiology of anxiety and fear. He has recruited volunteers to watch virtual reality footage of scenarios they may never encounter in real life: Being attacked by a dog. Climbing more than 250 feet up a tree. And swimming with sharks.

Huberman, 42, traveled to many of the shoots, putting himself in personal danger to collect the footage of the sharks. An intense and earnest man, he said he’s not interested in the extreme for its own sake, but rather “for how it can inform the typical person — and mostly how it can be used to relieve suffering.”


“When the curtain goes down for me, I don’t want to look back on my career and say: ‘Oh, we did all this nice work in mice,’” Huberman told STAT in a recent interview in his lab at Stanford. “I decided about six years ago that, unless I made a deliberate attempt to create a tool to cure blindness, a deliberate attempt to alleviate pathologic anxiety, it wasn’t going to happen the way it could happen.”

“I hope I don’t die trying, but if I do, I do,” Huberman said. “A bad thing about being dead is that the research might halt.”

Andrew Huberman 03
Optics equipment used to take high-resolution images of the retinas of glaucoma patients in Huberman’s clinical trial. Laura Morton for STAT

In his quest to relieve suffering, Huberman picked a formidable target in glaucoma, a disease in which the neurons that relay information from the eye to the brain suffer damage, resulting in a loss of vision. Many patients take prescription eye drops to slow or prevent their vision loss, but there’s no proven way to reverse damage already done.

Huberman’s new clinical trial grew out of an experiment in visually impaired mice. In that study — detailed in a 2016 paper published in Nature Neuroscience — he and his team used gene therapy (to activate a signaling pathway involved in stimulating growth) and visual stimulation (in the form of moving black and white bars that the mice were forced to look at) to try to coax the neurons connecting a mouse’s eye to its brain to regenerate. Fewer than 5 percent of the neurons, called retinal ganglion cells, grew back. But that was enough to help the mice see better, measured by whether they ran away from perceived threats.

The visual improvement was greatest when Huberman and his team used both gene therapy and visual stimulation. Their clinical trial is not testing the former component of the mouse experiment, but rather a more sophisticated version of the moving stripes.

The trial has so far enrolled two glaucoma patients, a 17-year-old woman and a 76-year-old man. A third patient is expected to join soon, with still more being screened. The goal is to eventually enroll 200 glaucoma patients who have lost some vision but are not completely blind. (Huberman has no hope that the experimental treatment will work in people who can’t see at all.)

When patients sign up, Huberman’s team measures their vision and maps the damage to their retinal ganglion cells. “If you have a hole in your neural retina that gives you a blank spot just X number of degrees, or this position off, the center of your visual field, we want to know that,” Huberman said. That information allows the programmers on his team to customize the visual reality experience for each patient, placing flashing white spots in specific locations in their field of vision.

To keep things interesting, the virtual reality experience involves more than white dots. When patients put on special headsets, they’re transported into an art gallery with empty frames on the walls. They can move their eyes or their head to explore the gallery, but the point of it all is the visual stimulation: those flashing white dots, which dance across the screen for periods of one to three minutes at different sizes and speeds.

Courtesy Huberman Lab

When patients complete the task, they get rewarded with a great work of art that appears in one of the empty frames. Among them: Vincent van Gogh’s iconic landscape “The Starry Night” and Rene Magritte’s surrealist portrait of “The Son of Man,” clad in a suit with a green apple obscuring his face.

Delivering this visual stimulation using virtual reality solves a stubborn problem in the field. Researchers have long tried to regenerate patients’ neurons by having them stare at some kind of stimuli on computers or TV screens. But it hasn’t worked. One theory why: Patients’ eyes tend to wander, and when that happens there’s no way to precisely target the visual stimulation to the damaged regions of patients’ eyes. With virtual reality, “you control their entire world and you know exactly where their eyes are,” Huberman said.

Patients get sent home with a virtual reality headset and are instructed to use it five days a week for 30 minutes at a time, while sitting down. Huberman and his team haven’t decided how long they’ll ask the patients to keep doing the visual exercise, but for now they’re bringing them in every six weeks to track changes in their vision.

“When the curtain goes down for me, I don’t want to look back on my career and say: ‘Oh, we did all this nice work in mice.’”

Andrew Huberman, Stanford

Three independent vision specialists consulted by STAT said that while Huberman’s approach is innovative, there are plenty of reasons why it might not work.

Dr. Anne Coleman, a glaucoma specialist at UCLA School of Medicine said that the experimental treatment doesn’t pose a safety risk, but she hopes that patients considering the trial will keep in mind that it’s still early and unproven so that it does not bring them “false hope.”

For now, the trial lacks a control group of patients who would not get the visual stimulation, an element generally considered the gold standard in medical research. (Huberman said he’s considering adding a control group later on but wants to see more data first.) The trial is also open both to patients who take medication to relieve high pressure in their eyes and those who do not, which could potentially confound the results. (Huberman said he expects nearly everyone in the trial in the trial to be on eye drops, since they’re the standard treatment.)

The trial also comes with another important caveat: What works in mice rarely works in people. While the mouse study is interesting, “translating that in humans who have to physically go out and interact with the world, I think, is a whole other level,” said Lotfi Merabet, a clinician-scientist at Massachusetts Eye and Ear who specializes in vision rehabilitation.

Huberman speaks candidly about potential weaknesses in his hypothesis: While the neurons could regenerate all the way up to the brain in tiny mice, the much greater distance in humans could prove too far to traverse. And the neurons involved in vision may not have the same capacity to regenerate in adult patients as they do during development.

There’s also the matter of money.

The Glaucoma Research Foundation is providing initial seed funding for Huberman’s trial, but that won’t be enough. Huberman is scrambling to raise money from other foundations, private donors, and government grants.

Thomas Brunner, president and CEO of the Glaucoma Research Foundation, said he’s pleased to see the experimental treatment moving from mice to humans at a quicker pace than usual for patients who are desperate to stop the ravages of glaucoma.

“The idea of testing early and then refining is much more appealing to me than people who are a little more afraid to take a risk,” Brunner said. Huberman, he said, is “a risk-taker, which is something I admire.”

Andrew Huberman 04
Huberman (right) and videographer Morne Hardenberg gather virtual reality footage of great white sharks off the coast of Guadalupe Island in 2017. Courtesy Michael Muller

Huberman has a Ph.D. in neuroscience from the University of California, Davis, and an associate professorship at Stanford, where he moved his lab in 2016 after five years at the University of California, San Diego.

He also has a taste for the extreme.

In his free time and on his own dime, Huberman has soaked in an ice bath for 10 minutes. He’s training right now for an ocean swim tens of miles long. He’s also training for an ocean dive — without scuba gear — in which he plans to hold his breath. (The depth has yet to be determined, but he said a good goal would be 70 feet underwater.)

“I realize I’m a little unusual in the kinds of eclectic things that I do for fun,” Huberman conceded.

Huberman grew up doing martial arts, so he’s comfortable in scenarios where there’s a physical threat, he said. And, no, he does not think you should try any of this at home or elsewhere without training and supervision from professionals.

He’s driven by a sense of adventure, sure, but he also does it because he’s interested in understanding fear — and wants to find teachable interventions to help people overcome it. For example, he tried out a nasal breathing technique when he went climbing without protective equipment in the Pyrenees mountain range in Spain; he navigated through sections where, as he put it, “basically one slip and you’re dead.”

Huberman said he has always been curious about why some people find a situation tolerable or pleasurable — while others find it terrifying.

He’s not so interested in people who are naturally comfortable in extreme or stressful scenarios. His real interest, he said, is in someone like himself — “the person who’s not comfortable doing it but learns how to be. That’s information you can transfer.”

Courtesy Michael Muller

Huberman’s fascination with the extreme has informed his virtual reality research.

Plenty of researchers use the technology in clinical experiments. But what sets Huberman apart is the lengths he’s gone to gather his virtual reality footage.

For his study on anxiety using virtual reality to evoke fear, he initially thought about pulling from existing libraries of scary stock footage. But they weren’t right for his purposes: The clips were too short; he needed something at least 10 minutes long. And they usually opened right away with a frightening experience, instead of opening with a calming scene from everyday life — necessary for his team to collect baseline measurements.

So Huberman decided to collect his own virtual reality footage, using special 360-degree cameras.

To evoke the claustrophobia of being trapped in an elevator, he and his team captured footage in what he fondly refers to as the “dungeon-like” elevator in his lab building at Stanford. They recruited other occupants of the building to serve as extras, and the building’s maintenance team helped them halt the elevator abruptly.

To elicit a fear of heights, they asked a professional tree-trimmer to wear a camera on his body while climbing more than 250 feet up a tree in the region east of the San Francisco Bay.

And to simulate the frightening experience of being attacked by a dog, they went to nearby Redwood City, Calif., to solicit the help of a professional dog trainer and his 120-pound pitbull. With the camera running, the trainer gave the dog a cue to attack his arm and then cried out as if in pain. (In fact, the trainer was wearing a special dog-bite-resistant sleeve to protect himself from injury.)

Gathering all this footage has proved to be remarkably cheap, Huberman said. The lab’s many collaborators have helped them at little to no cost or allowed them to piggyback onto other trips, all in the name of research. (Other costs of the fear study, such as the virtual reality headsets, are being paid for using internal funds, Huberman said.)

The study, which is not a formal clinical trial, so far has recruited 85 volunteers, with the goal of eventually signing up 250 people, some of whom have diagnosed anxiety. The most effective recruitment method has not been the placement of traditional fliers tacked to a telephone pole, but rather the lab’s Instagram page.

Andrew Huberman 05
One of Huberman’s flyers recruiting volunteers for a study using virtual reality to evoke fear. Laura Morton for STAT

When most volunteers watch the footage — all while having their heart rate, sweating, breathing, pupil size, and body posture measured — they will most likely have no idea of all that went into capturing it.

In the case of the shark footage, Huberman teamed up with an A-list photographer named Michael Muller, who has a passion for photographing sharks in the wild. Their team has twice boarded a research vessel for a 22-hour voyage out into the Pacific Ocean to spend a few days out on the crystal-clear water off the coast of Guadalupe Island.

Visiting tourists or researchers often go underwater in protective metal cages to observe the hundreds of great white sharks that congregate in the area. But Huberman said he managed to get permits from the Mexican government so that he and his team could exit the cage and swim with the sharks for his research project.

The experience was “incredibly calming,” said Huberman, who underwent training on how to stay safe.

Huberman speaks fondly about the sharks. But he also calls them “really diabolic little guys,” too, before quickly correcting himself: At about 3,000 pounds and 15 feet long, they’re “not so little guys.”

Huberman said that, if he returns to Guadalupe Island, he wants to run a field study measuring his collaborators’ respiration, pupil size, and heart rate while they swim among the sharks.

And although he hasn’t yet figured out how to do it, he said he wants to collect perhaps his most extreme virtual reality footage yet: a simulation of the experience of being eaten by a shark.

Comments are closed.