The Science of Hearing, Balance & Accelerated Learning

Andrew Huberman, who introduces himself as a professor of neurobiology and ophthalmology at Stanford School of Medicine, frames the episode around two sensory systems: the auditory system and the balance system, which he calls the vestibular system. His central claim is that these systems interact with the rest of the brain and body and, "used properly," can help people learn information faster, remember it longer, and improve both hearing and balance. He presents this as an area where the cells and mechanisms in the ear and brain are well understood, and says he will cover that biology briefly before moving to protocols.

He also previews several topics: tinnitus, which he describes as a ringing of the ears that is very intrusive for people who have it, along with treatments that work "in some circumstances" and newer emerging treatments; otoacoustic emissions, sounds the ears themselves produce; and advice on what type of music to listen to for faster learning. Within this section of the transcript, only some of these previews are developed. The discussion of tinnitus treatments, balance tools and music does not appear before the transcript breaks off.

Before turning to hearing, Huberman describes a recent peer-reviewed paper in Cell Reports on skill learning. He places it against his earlier advice: skill learning depends on many repetitions, errors matter for learning, and earlier studies showed that a 20-minute shallow nap or a period of doing nothing after learning improves the rate and depth of learning. The new claim is different. He says the data indicate that rest should be inserted within the learning episode itself, and he clarifies that he is not talking about sleeping while learning.

In his description of the design, participants learned sequences of numbers or of piano keys. Using the piano example, he notes he is not a musician. One group practiced a sequence such as G, D, F, E, G continuously. The other practiced for 10 seconds and then rested for 10 seconds, repeating that cycle. During the rests, participants did not look at a phone, focus on anything, or touch the keys; they let their minds drift. He reports that learning rate, skill acquisition and skill retention were "significantly faster" with the inserted rests, and adds that this meant "much, much faster." He says he will give the size of the effect in a moment, but no figure for the size of the learning improvement appears in this section.

The researchers called these effects "micro offline gains." Huberman argues that the brain is not actually offline. According to his account, brain imaging during the task showed the hippocampus, which is involved in memory, and the neocortex, which processes sensory information, active during the 10-second rests in a way that amounted to a 20-fold, temporally compressed replay of the sequence. His illustration is arithmetic: if someone normally completes about five repetitions in 10 seconds, multiplying by 20 means the brain has practiced the sequence about a hundred times during the rest periods. He presents the five-repetition figure as a hypothetical ("let's just say").

So basically, the rehearsal continues while you rest, but at 20 times the speed.

He links the result to the spacing effect, which he attributes to "Ebbington, in 1885," and says it has since been shown across many domains of learning, both cognitive, such as language learning, and physical, such as motor sequences. He points to a review article titled "Parallels between spacing effects during behavioral and cellular learning," which he says connects behavioral skill improvement to underlying changes in neurons. He identifies the newer study as coming from the laboratory of Leonard Cohen, and notes that this is not the late poet and musician of the same name.

His practical takeaway is to keep generating repetitions while inserting 10-second periods of doing nothing, avoiding attention to anything else as much as possible, for example by closing the eyes or simply waiting. He also suggests combining this with a 20-minute nap or a 20-minute decompression period after learning. He is explicit that the combination has not been tested: he thinks the two "could both synergize," but notes "that hasn't been looked at yet."

Huberman states that the podcast is separate from his teaching and research roles at Stanford, describes it as an effort to provide zero-cost-to-consumer science information, and acknowledges the episode's sponsors.

The hearing section starts from the premise that a speaker's voice creates tiny changes in the air, and that the ears and brain can turn those sound waves into meaning. The external ear is technically called the auricle, but more often the pinna. Huberman says its cartilaginous shape is arranged to capture sound in the best way for a person's head size. He also says ears change size across the lifespan and that the rate of change is "a pretty good indication of how fast you're aging," a point he defers until later. That later discussion is not part of this section.

He explains that the shape of the ear amplifies high-frequency sounds, meaning squeakier sounds, while low-frequency sounds include the snoring of his dog, Costello, in the background. He describes sound waves as fluctuations in air, comparable to waves in water. These waves reach the eardrum, a membrane that moves back and forth. Attached to it is a small bony structure he compares to a hammer, which has three parts: the malleus, incus and stapes.

The hammer strikes the cochlea, a coiled, snail-shaped structure in the inner ear where sound is converted into electrical signals the brain can use. Huberman makes an aside that the word is pronounced differently in different parts of the United States. One end of the cochlea is more rigid than the other, and he says this difference matters for separating low-frequency sounds from high-frequency ones. Inside the cochlea are hair cells, named for their shape and unrelated to hair on the head or body. When particular hair cells move, they signal to the brain that a given sound is present; when they do not move, that sound is absent.

He then maps frequency along the cochlea's length. At the rigid base, hair cell movement corresponds to high-frequency sounds, such as a cymbal. At the flexible top, the apex, the cochlea behaves more like a bass drum and responds to low frequencies, with medium frequencies in between. He emphasizes that the pinna, hammer and cochlea are "purely mechanical" and have "no mind of its own"; they only break sound into frequency components, and the brain does the work of making sense of the result. For listeners unfamiliar with frequency, he again uses ripples on a pond: ripples close together correspond to high frequency, and ripples farther apart to low frequency.

Your cochlea essentially acts as a prism.

The prism analogy carries the argument. A prism splits white light into wavelengths, and he mentions the cover of Pink Floyd's "Dark Side of the Moon" as an example. In the same way, the cochlea splits environmental sound into frequencies, and the brain reassembles that information. Hair cells in each of the two cochleae send axons that carry activity patterns into the brain, passing through several stations before reaching areas involved in conscious awareness.

For listeners who want the nomenclature, Huberman lists the pathway in order. He presents it as optional detail and says listeners do not need to remember it.

Spiral ganglion: a ganglion is a clump of neurons. Cochlear nuclei: located in the brain stem, which he places near the neck. Superior olive: one on each side of the brain; he says they look like little olives, with a divot he likens to a pimento. Inferior colliculus: named inferior because it sits below the superior colliculus. Medial geniculate nucleus. Neocortex: where the information is made sense of.

He argues that the large number of stations has a functional reason. Knowing where a sound comes from is more important for survival than knowing what it is, for instance whether something is falling toward you from behind or a car is approaching from the left or right. The auditory and visual systems collaborate to locate things in space: people turn toward a voice on their right, and assume speech comes from a mouth they see moving. He says disruptions in this auditory-visual matching underlie the ventriloquism effect, in which a sound seems to come from somewhere it does not. He promises more on this later, but it is not developed in this section.

For left-right localization, Huberman describes interaural time differences. Neurons in a brain stem station calculate the difference in arrival time of sound at the right and left ears. Simultaneous arrival is read as a source straight ahead; earlier arrival at one ear is read as a source on that side.