How to Improve Your Eye Health & Offset Vision Loss | Dr. Jeffrey Goldberg

Huberman introduces the guest with a dual credential that matters for everything that follows: Goldberg chairs the Department of Ophthalmology at Stanford University School of Medicine, sees patients every week as an MD, and also holds a PhD and directs his own laboratory. That laboratory works on the mechanisms and cures for glaucoma, retinitis pigmentosa and macular degeneration. Huberman's framing is unusually strong, calling him "one of the world leaders in developing methods to cure blindness" and "intensely knowledgeable about all things related to vision."

The episode is positioned as a near-complete survey of visual and eye health rather than a narrow specialty interview. The promised inventory includes the benefits and drawbacks of corrective lenses, the upside and the cautions of sunlight, behavioral eye exercises with details on frequency and durability of benefit, surgical options such as LASIK with age-related considerations, and even how to clean contact lenses. That breadth is what makes the later shifts between nursery screening, outdoor light, eyelid hygiene and reading glasses coherent rather than scattered.

Huberman also explains why he is asking rather than answering. As a neuroscientist who has worked on the visual system he says he often has partial knowledge: "I sometimes have answers or partial answers, but more often than not, I don't have the answers." He attributes the volume of listener questions about vision to that gap, and explicitly hands the authority to Goldberg.

Goldberg's answer to "when should a child first be examined" is staged by age rather than given as a single schedule. The first checkpoint is universal: every baby gets, or should get, an eye exam, and in the newborn nursery during the first few days the target is narrow. Clinicians are looking for a red reflex, the same phenomenon that produces red eye in a flashed photograph, where light reflects off the retina and comes back out.

The test is crude but consequential. A red reflex is normal and reassuring. Its absence, replaced by a whitish or gray reflex, is the sign that matters, because a small number of diseases can present even in newborns. Goldberg names retinoblastoma as the most concerning and, thankfully, the least common: it is the most common pediatric eye cancer and still quite rare. He notes the exam may be nothing more than a pediatrician flashing a penlight into the baby's eyes.

"It's not typical, as long as that's looking good, to worry about getting an eye exam from there through childhood, maybe early elementary school"

That quote carries a condition, which Goldberg states in the same breath: the low-intensity path holds "unless your baby is presenting with one of a number of features that parents often pick up on." He also calibrates what normal looks like in the first couple of years, noting that infants do not have great visual acuity and that roving eye movements, searching the environment, are expected rather than alarming.

Huberman supplies the mechanistic premise before asking about timing. The brain is highly plastic early, it receives input from both eyes, and it must learn to use a properly aligned pair. His claim is that waiting too long can leave a child unable to recover normal vision, whereas realigning the eyes early allows recovery. He asks how early alignment should be considered.

Goldberg endorses the premise and gives the mechanism in plain terms: the brain "starts ignoring" one eye, failing to fully develop strong connections for the data arriving from it. Cross certain developmental thresholds without correcting that connectivity and the loss can become permanent. This is amblyopia, and it is a wiring problem as much as an eye problem.

The numbers Goldberg offers are explicitly old and explicitly rough. Intervention before age three was considered fully corrective, before six partly corrective, before nine still worth a chance. He then revises the picture with follow-on studies showing that children into their young teens still have a shot at repairing the eye-brain connection, so late detection in the tween or early teen years is not a reason to give up on retraining the weaker eye and realigning the muscles.

He also introduces a benign look-alike that parents and clinicians both have to rule out. Pseudostrabismus is the appearance of turned-in eyes caused by extra skin at the inner corners, so that eyes which are actually straight read as convergent. Distinguishing appearance from actual misalignment is the first step; the second is acting quickly once misalignment is real.

The most interesting admission in this section is about the limits of the model. Goldberg notes that critical period plasticity has been studied most thoroughly in the visual system, and that different components of the eye-brain connection have different critical periods. Even after the amblyopic eye is made to see well and the eyes are realigned, he says many children will never recover something further; the transcript cuts off mid-sentence at that point, so the specific function is not stated in this segment.

Huberman offers his own case as a curbside consult. As a child he swam without goggles, keeping one eye closed and the other opening and closing as it passed in and out of the water, then emerged seeing double. The double vision resolved only after one eye was patched, forcing use of the eye that had been shut, and recovery took roughly a day. What did not recover was depth perception: he describes being the kid the fly ball always hit and says he is better at darts with one eye closed.

He turns the anecdote into a sharper question about dose and about iatrogenic risk. Can a few hours of mismatched input early in life permanently rewire the brain? And if a ten-year-old scratches a cornea and the injured eye is patched, should the healthy eye then be patched in turn once the scratch heals, to restore the balance of input? The underlying question is how many hours or days of asymmetry actually matter.

Goldberg separates the typical from the atypical. The shape of the story is standard amblyopia management: patch the strong eye so the weak eye recovers, without necessarily regaining depth perception. The atypical element is the trigger. For most children, an hour or two of one-eyed viewing, even practiced daily, would not be expected to produce strabismus or amblyopia.

"it'd be very unusual for that to trigger this kind of either strabismus misalignment of the eyes, let alone amblyopia"

His alternative explanation is cumulative rather than acute. Without having examined Huberman before that day, Goldberg guesses at years of intermittent strabismus, an eye turning in or out occasionally, each episode causing a small failure to connect, with the swimming pool simply being the moment the accumulated deficit became visible. He also reassures the general case: for the overwhelming majority of children who get a corneal scratch or wear a patch briefly, this cascade does not follow.

Huberman, 47, situates himself in a transitional generation that had some classroom computers but not many, then contrasts that with children now looking at tablets and phones at close range from an early age. He cites animal experiments in which restricting vision to near distances lengthens the eyeball, so the image lands in front of the retina instead of on it, producing myopia. The question he puts to Goldberg is what the evidence says about a healthy visual environment from birth through adolescence.

Goldberg confirms the scale of the problem and its uneven distribution: myopia is very common, more common in Asian populations, and "called an epidemic in China." At Stanford he sees it in both children and adults, which is why the science of myopia control, meaning how to provide the right environments, has become a practical question rather than an academic one.

He then traces how the hypothesis shifted. For decades, mouse models plus well-designed human cohort studies that asked families how much time children spent reading or at the computer pointed toward near work as the driver. The timing fits, because most myopia progression and eyeball elongation happens in the pre-teen and teen years. Only in the last few years have studies pointed "in a slightly different direction": near activity may not be the whole story, and the kind of light entering the eye may carry more of the effect, since typical indoor lighting is not full-spectrum.

On dose, Goldberg is careful to separate what cohort data suggest from what trials have proven. The cohorts show something like a dose-dependent response, with more time outdoors apparently better, but no known ceiling. The gold standard, a randomized controlled trial that assigns children to zero, one, two or three hours outdoors per day, has not been done. What the existing evidence does support is the contrast between zero and one or two hours; whether five minutes suffices or five hours is superior remains unanswered.