How Meditation Works & Science-Based Effective Meditations

Huberman opens by dividing the episode into two questions that are easy to conflate: what happens in the brain and body during practice, and what a practice leaves behind afterwards. He states the second half plainly — meditation has been shown to alleviate symptoms of depression, and it can be used to enhance focus and other states useful for work. The framing matters because it commits him to mechanism rather than testimonial, and it sets up the episode's real thesis: the state you enter is not the deliverable; the trait you export is.

He then dismantles the stock image before using it. The lotus position, the closed eyes, the focus on the so-called third eye center behind the forehead — he describes all of it, concedes the anatomy with a joke ("There's no third eye there, at least there shouldn't be."), and then widens the category: body scans that track contact with the sitting surface, and walking meditations done with eyes open. The point of the demolition is not to debunk the stereotype but to establish that the visual package people associate with meditation is one option among many, and that the options differ in what they do.

From that, he draws a selection rule rather than a dosage rule. If you want to change your default mood, sharpen focus, improve sleep, or improve cognitive or physical performance, meditation is described as powerful — but only if you "pick the right meditation practice", one chosen for feasibility and for the specific goal rather than for generic virtue. This is the organizing claim the rest of the episode has to earn.

The counterintuitive claim he plants early, and returns to later: meditation inverts the endurance-sport model. In running, getting better means running longer — first a 5K, then a 10K, then a marathon. He explicitly rejects the transfer: "With meditation, it's actually quite the opposite." As the ability to drop into a given brain state improves and shifts from state toward trait, the required dose falls rather than rises. He attributes the state-versus-trait framing to a book he names only later, and he closes the segment by promising a menu of practices plus guidance on modifying them when busy or sleep-deprived, including a form of meditation he says can reduce sleep need while preserving cognitive and physical ability. That last claim is asserted as an upcoming topic, not supported here.

Huberman grounds the topic autobiographically: a book handed to him in high school, during what he calls his wild years, which he identifies as Jon Kabat-Zinn's "Wherever You Go, There You Are". The detail is not decoration. It dates the popularization to the late 1980s and establishes Kabat-Zinn as one of the first — he is careful to say not the only — to bring mindfulness practice to a general American audience. It also sets up the timing problem that explains why the science lagged the practice by decades.

The lag is attributed to instrumentation, not to indifference. Until functional imaging — MRI and fMRI — became a common laboratory and hospital tool, there was little mechanistic purchase on meditation, even though cultures outside the United States had long held it to be useful. Studies existed but were few through the late 1980s, then picked up in the 1990s. The causal story he tells is straightforward: you cannot observe a brain state without a device that images a brain while it is active rather than only mapping its structure.

He also flags a definitional trap he never fully resolves. Any discussion of meditation is a discussion of states of mind, and any discussion of states of mind invokes consciousness — a word he calls dangerous in any format because it has no single standard operational definition. Rather than define it, he notes that the term travels with adjacent subcultures, which is why meditation and psychedelics were "close cousins" in the 1960s and 1970s conversation about consciousness.

The split, as he narrates it, was administrative before it was intellectual. Timothy Leary and others at Harvard were interested in both psychedelics and meditation, but their advocacy of LSD — lysergic acid diethylamide — was anti-establishment and involved encouraging students to take it; they were pushed out. He points to a book in the show notes for the fuller account. The consequence was a divorce: the psychedelics conversation and the meditation conversation separated, and Kabat-Zinn's books in the late 1980s and early 1990s deliberately rebuilt meditation as a standalone practice for calmness, stress regulation and sleep.

He is explicit that academia did not immediately follow. It took years for institutions including Harvard and Stanford to fund meditation research, and the questions that mattered most were the least convenient: which areas are involved, how the body changes, and — most importantly — what changes after the session ends and the person returns to ordinary life. When imaging arrived, it arrived awkwardly: subjects meditated inside a magnet, in a tube. The payoff was a long list of brain changes, and, when paired with subjective reports of happiness and sleep plus objective measures such as hormones and markers of inflammation, a dozen or more benefits attributed to regular practice, some obtainable from quite short sessions.

The adoption story is corporate as much as clinical: Bay Area technology companies, investment firms and others began hiring meditation trainers or running courses in the 2000s, which is part of why meditation now reads as universally beneficial. His warning is that this consensus is shallow — most people treat meditation as one thing. The controlling analogy is exercise: weight training, running and high-intensity interval training all count as exercise and all produce different results depending on what you do and how often. Meditation, he argues, is the same kind of word, and this batch's whole architecture rests on that analogy.

Huberman refuses to deliver a bare nomenclature list, and his justification is the load-bearing one for the episode: the names are useless unless you know what the regions do, and knowing what they do is useful only because you "actually can manipulate those brain areas" — turning activity up or down through specific elements of practice. That claim is what converts neuroanatomy into a tool, and it is also the claim most in need of evidence, which he supplies later in the form of practice-dependent shifts rather than direct manipulation.

The first node is the prefrontal cortex, described physically as the front bumper of the head just behind the bone, present on both sides and connected across the midline but doing different things on each side. He narrows to the left dorsolateral prefrontal cortex, unpacking the words as he goes: "Dorsal means up, lateral means to the side." This is a deliberately low-tech anatomical instruction — hand on the left side of the head, moved slightly toward the midline — so that the listener can locate the region without a diagram.

The evidence for what that region does is lesion and stimulation work in animals and, he notes, in humans. Its function is stated as interpretation: controlling and making sense of bodily senses, reading emotional and bodily signals, and then deciding on the basis of that reading. He also fixes a convention that keeps the rest of the episode readable — from here on, "prefrontal cortex" means left dorsolateral prefrontal cortex unless stated otherwise.

The second node is the anterior cingulate cortex, shortened to ACC, and its job is defined by context rather than by absolute value. It tracks breathing rate and heart rate, but more importantly it tracks whether those numbers fit the situation. His example is precise: running uphill with a fast heart is unlikely to alarm you, because the reading matches the circumstance; walking calmly when the heart suddenly races is read as pathological or inappropriate. The ACC does not report a heart rate, it reports a mismatch.

He then lists the ACC's inputs to show how wide its sampling is: the amygdala, described as the threat-detection center and the structure people usually call the fear center; the heart; the gut, including how distended or empty it is; the lungs and related structures for breathing rate; and the skin surface, where an itch, a pain or a mosquito bite would register. The insula, spelled out as I-N-S-U-L-A and described as having multiple parts, is the third node, doing overlapping interpretive work on internal signals while also integrating what is happening outside.

The insula's contribution is what makes the triad explanatory rather than merely anatomical. It can certify that a pounding heart on a steep hill is appropriate — "I don't have to be worried. I don't have to be scared. I might want to slow down, but this makes sense." And it can fail to certify the opposite case: sitting in a calm room while the stomach turns, breathing accelerates and panic rises, because the internal shift has no external correlate. Anxiety is thereby modeled as a correspondence error between inside and outside, not as a surplus of sensation.

He calls the result a "neural conversation": left dorsolateral prefrontal cortex, anterior cingulate cortex and insula continuously negotiating what is going on inside, how it feels, what is being thought, and whether the felt state warrants the current behavior. He explicitly releases the listener from memorizing the names and keeps only the functional claim — that this negotiation runs whenever you are awake, comparing internal state against external circumstance. That is a deliberately modest ask, and it is the premise the entire interoception argument later depends on.

The human addition is time. Because we hold memories, present awareness and anticipation, the mismatch can be understood rather than merely suffered: a dinner-table thought about tomorrow's exam degrades the current bodily state, and the degradation still makes sense to us because the exam matters. Excitement that kills appetite before a big day is the same mechanism with a positive sign. This is why he insists the three-node conversation is dynamic — the bodily state need not match the activity, and can be entirely understandable when it does not.

Mindfulness itself gets the same treatment as consciousness: there is, in his words, no single perfect, universally accepted operational definition, which he translates as specialists disagreeing about what the term should mean for everyone. The one feature he treats as near-consensus is presence — but presence to internal sensation, breath and thought, not necessarily to surroundings. Many practices are built to heighten internal attention while ignoring the external world entirely, which is a distinction that later determines which practice suits which person.

To decompose practice, he picks a generic form and puts a number on its prevalence — roughly 75% of practices, by his estimate, involve stopping, sitting or lying down, and usually closing the eyes. The number is offered as an estimate, not a finding, and he immediately qualifies the eye closure: it is absolutely not required, since many practices are done with eyes open. The generic form is a convenience for analysis, and he says so.