Insight | Exercise protects the aging brain better than any pill
Huberman, Wendy Suzuki, Lisa Mosconi and Peter Attia describe exercise increasing BDNF, supporting the hippocampus and lowering dementia risk. It's the strongest lever people have against cognitive decline.

The Short Version
Across these conversations, exercise keeps coming up as the most dependable thing a person can do for an aging brain. Andrew Huberman describes a cascade of signals from working muscles, loaded bones and lactate that raise BDNF, feed and protect the hippocampus and strengthen the blood-brain barrier, and he says exercise can significantly slow the natural shrinking of the hippocampus with age [1]. Wendy Suzuki, whose father developed Alzheimer's, says regular aerobic exercise can grow a bigger hippocampus, sharpens attention and memory within a single session, and was linked in one Swedish study to years of extra good cognition [2]. Peter Attia names exercise as one of only four levers that are unequivocal for brain health, next to sleep, lower ApoB and insulin sensitivity, and he calls supplements a rounding error beside them [3]. Neurologist Mark D'Esposito adds the "pill" half of the comparison: the only approved Alzheimer's drugs help symptoms modestly without slowing the disease, while aerobic exercise has matched cognitive therapy for improving executive function [5]. Lisa Mosconi's conversation adds the other half of daily brain care, food, hydration and hormones, which works alongside exercise rather than replacing it [4].
Table of contents
What They Said
Andrew Huberman: the body keeps talking to the brain
In his solo episode on exercise and the brain, Huberman takes on a literature of tens of thousands of studies and finds that nearly every kind of exercise studied improves brain performance: six-second bike sprints, an hour of steady cardio, even a seated single-leg extension [1]. He splits the effects into acute ones, measured right after a session, and chronic ones, measured over weeks or months of training two to four times a week, and says almost all studies find positive effects [1].
The short-term effect is mostly arousal. Huberman argues that roughly 60 to 70% of the acute benefit comes from increased autonomic arousal: higher heart rate and blood pressure, wider eyes, more alertness [1]. Six maximal six-second sprints with a minute of rest between them improve performance on memory tasks and on the Stroop test of cognitive flexibility, and so do 20 to 30 minutes of steady cardio [1]. For something that brief, he says, "I'd be willing to stake let's say six of my 10 fingers on the idea that it's all due to enhanced autonomic arousal" [1]. The pathway runs from the adrenal glands, which release adrenaline; adrenaline doesn't cross the blood-brain barrier, so it acts on the vagus nerve, which signals the nucleus of the solitary tract in the brainstem, which activates the locus coeruleus, which sprays norepinephrine across the brain, raising activity in the prefrontal cortex, hippocampus and other regions [1]. Dopamine release usually comes with it [1]. "It's also true that exercise gives us energy, and this is how it gives us energy," he says [1]. He cites Peter Strick's lab at the University of Pittsburgh, whose circuit tracing found that the motor areas most strongly linked to the adrenals control the core and multi-joint compound movements [1].
The long-term effects go deeper. Here Huberman lists several signals that matter for aging:
- Osteocalcin from bones. When bones are mechanically loaded, they release osteocalcin, which, according to animal and human studies largely from Columbia, crosses into the brain and supports the growth of neurons and their connections in the hippocampus [1]. He says "many of the effects of BDNF appear to be mediated by osteocalcin" [1]. He notes limited evidence that it may even increase neuron numbers in the dentate gyrus [1].
- BDNF. Exercise raises brain-derived neurotrophic factor and nerve growth factor, which stabilize and strengthen neural connections [1]. His rarely discussed point is that BDNF is activity-dependent: it is released by electrically active neurons and works best on neurons that are already active, which may help explain why people who exercise regularly keep healthier brains into later life [1].
- Lactate. During hard exercise, lactate becomes neurons' preferred fuel, sparing glucose for later mental work [1]. In the brain, astrocytes supply lactate to active neurons, which further raises BDNF [1].
- VEGF and the blood-brain barrier. Lactate triggers VEGF, a growth factor that strengthens the blood-brain barrier [1]. That matters for aging because "one of the major features of age-related cognitive decline and one that's greatly exacerbated in Alzheimer's is a breakdown of the blood brain barrier" [1].
His central claim about aging is about the hippocampus: "our hippocampus decreases in volume over time as we age, just naturally even in somebody that doesn't have Alzheimer's dementia, and exercise can adjust the slope of that decline significantly, provided there's enough exercise in the appropriate exercise" [1].
Super-agers. Huberman also discusses the anterior midcingulate cortex, a hub connected to the brain's dopamine, arousal, learning and stress systems [1]. Super-agers, people in their 60s to 90s whose thinking and memory rival much younger people, maintain or even grow this region while it shrinks in their peers [1]. A review by Lisa Feldman Barrett links its activity to grit, persistence and willingness to exert effort [1]. The region responds to doing things you don't want to do, as long as they are safe, and physical exercise is among the examples that have increased its size [1]. He is careful about the evidence: "It's not necessarily causal, but it's strongly correlated with this super-aging phenomenon" [1].
What happens when you stop. In studies of trained people who stopped exercising, brain oxygenation and other markers of brain health declined significantly after about 10 days without any cardio or resistance training [1]. A week off for illness or injury is fine, he says, and you'll likely come back stronger [1].
Wendy Suzuki: from memory scientist to exercise scientist
Suzuki spent much of her career at NYU studying how the hippocampus turns experiences into long-term memories [2]. For her, the structure is personal: "The hippocampus and what it does really defines our own personal histories" [2]. Newer research shows it also supports imagination, so people without it can't imagine situations they've never experienced [2].
Her turn to exercise began with her own life. Working toward tenure, she ate takeout, gained 25 pounds and was very stressed; after being the weakest person on a rafting trip in Peru, she started going to the gym, beginning with a hip-hop dance class [2]. A year and a half later, 25 pounds lighter and writing an NIH grant, she had a thought for the first time: "grant writing went well today" [2]. She could focus longer and remember details more easily [2]. What made it a mission was a call from her mother: her father had gotten lost driving home from a 7-Eleven seven blocks from their house, and he was later diagnosed with Alzheimer's dementia [2]. Her father, an engineer, had not been very active, while her mother played team tennis into her 80s [2].
The mechanism she describes. Every time you move your body, Suzuki says, you release neurochemicals such as dopamine, serotonin and noradrenaline, and aerobic exercise also releases BDNF, which helps new brain cells grow in the hippocampus [2]. She calls a workout "this wonderful bubble bath of neurochemicals," and says regular exercise means "we all have the capacity to grow a bigger, fatter, fluffier hippocampus" [2]. She describes two studied pathways from body to brain: working muscles release a myokine that crosses the blood-brain barrier and stimulates BDNF (shown in running rats), and the liver releases the ketone beta-hydroxybutyrate during exercise, which also crosses into the brain and boosts BDNF [2]. Huberman clarifies that this doesn't mean you need a ketogenic diet [2]. On new neurons, she says newer studies using better techniques suggest they are born in the adult human hippocampus into the ninth decade of life, although it remains debated [2].
She is careful about what exercise can and cannot do. She doesn't claim exercise would cure her father's disease, but she hopes entering her 70s with a bigger hippocampus would delay how much such a disease could affect her memory [2].
The evidence she cites. The minimum dose for mood is about 10 minutes of walking; for bigger effects on the hippocampus and prefrontal cortex you need cardio that raises your heart rate, and a power walk counts [2]. A single 30- to 45-minute aerobic session reliably improves mood, prefrontal attention on the Stroop test and reaction time, and in one of her published studies these effects lasted at least two hours, the longest she measured [2]. In unpublished pandemic data with people in their 20s to 90s, 30 minutes of age-appropriate exercise lowered anxiety, depression and hostility scores and raised energy, and older participants improved even more on attention tests [2]. That last finding is directly relevant to aging: the older people gained more, not less.
Her study of low-fit adults aged 30 to mid-50s compared three months of spin classes two to three times a week with competitive video Scrabble [2]. The exercisers improved on the Stroop test, on a memory task that required telling similar items apart, and on a spatial memory task in a virtual city [2]. A second, unpublished study let moderately fit people add spin classes up to seven a week and found that "every drop of sweat counted": the more they added, the better their mood and hippocampal memory [2]. She also cites a 2018 Swedish study in which women who were highly fit in their 40s "gained nine more years of good cognition later in life" compared with less-fit women, while stressing it is correlational [2].
Her own routine includes a 30-minute cardio and weights workout seven days a week, built up from four or five days [2]. Among the three top tools for attention, she names exercise, meditation and sleep [2].
Peter Attia: one of four proven levers
Attia comes at the question as a physician focused on how people actually die. Neurodegenerative disease is one of his four horsemen, alongside atherosclerotic disease, cancer and accidents [3]. "Alzheimer's disease is both the most prevalent form of dementia and the most prevalent neurodegenerative disease," he says, affecting about six million Americans [3].
His case for exercise is partly a case against the alternatives. He thinks the field may have over-emphasized amyloid: many cognitively healthy people die with brains full of amyloid, a falsified paper on one amyloid variant set the field back, and anti-amyloid drugs have largely disappointed [3]. In the conversation he puts it this way: "there has been no shortage of drugs that target amyloid that have seemed unsuccessful" [3]. Unlike heart disease, there is no reliable chain from biomarker to outcome [3].
So what is certain? Attia lists four things that are unequivocal for brain health: adequate sleep, lower LDL and ApoB, insulin sensitivity (not having type 2 diabetes), and exercise [3]. Avoiding head injury is a likely fifth [3]. If someone had only three hours a week, his team's reading suggests one hour each of zone 2, strength and interval training, though more is better [3]. By comparison, supplements such as Theracurmin, magnesium L-threonate, methylated B vitamins and EPA/DHA are a rounding error [3]. He is also bearish on most hyperbaric oxygen claims, calling many studies terrible [3].
Several other threads in Attia's conversation connect to exercise and the aging brain. Zone 2 cardio is one of his first tools for lowering blood pressure (he recommends three to four hours a week, more than the common 150 to 180 minutes) [3], and exercise, especially combined resistance and cardio training, can help lipids [3], both of which feed into his brain-health levers. ApoE4 raises Alzheimer's risk but is not deterministic, unlike the rare PSEN1, PSEN2 and APP mutations that cause early-onset disease [3]. His general principle for any risk: "If a risk is causal and it is modifiable, it should be modified" [3]. And he points out that people with Alzheimer's usually die from complications like aspiration pneumonia, infections or failure to thrive, which is part of why prevention matters [3].
Exercise also protects the aging brain indirectly by preventing falls. After 65, breaking a hip or femur carries a 15 to 30% chance of death within 12 months [3]. Attia's four pillars of exercise are strength, stability, aerobic efficiency (zone 2) and aerobic peak (VO2 max), with eccentric strength, the body's brakes, key for stepping down and walking downhill [3]. Huberman draws the same link in his own episode, citing Attia on controlled landings protecting against the falls that start a spiral of inactivity and earlier death [1].
Mark D'Esposito: what the pills can and can't do
D'Esposito, a neurologist at UC Berkeley who still sees patients, provides the clearest look at the pharmaceutical side [5]. The only approved Alzheimer's drugs, acetylcholinesterase inhibitors like donepezil, have existed for over 20 years and help symptoms modestly without slowing the disease [5]. He describes families saying a patient is remembering more on them, but adds, "it doesn't really slow the course of the disease That's the problem that disease just carries on even though we're symptomatically improving the symptoms" [5]. He says he doesn't treat his patients much differently than he did 35 years ago, except for better social support, and that a Penn study found showing patients family movies or albums worked better than any drug for behavior [5].
He is frustrated with the drug industry more broadly. On dopamine and working memory, he says "one of the most disappointing things to me in my career has been that pharmaceutical companies have not picked up on this" idea of drugs for specific cognitive processes [5]. And even where drugs do work, they are not simple boosters: in his bromocriptine studies, people with low working memory improved but those who started high got worse, because "more is just not better" [5].
Against that, his summary of the exercise evidence is striking. Groups developing cognitive therapies often use aerobic exercise as a separate treatment, and Art Kramer's group at the University of Illinois found it about as effective as cognitive therapy for improving executive function, including in studies of seven-year-olds on exercise bikes [5]. Neurologists now talk about exercise, nutrition and sleep in a way they never did 30 years ago [5]. D'Esposito also notes that the frontal executive system declines more than other systems with healthy aging because it is the most complex and biologically costly, and he wants the field to move from preventing disease to optimizing the brain; otherwise, "we're always playing defense instead of playing playing offense" [5].
The GIST of every new summary in your inbox, daily or weekly. Free, and you can unsubscribe any time.
Lisa Mosconi: the other daily levers
Mosconi's conversation, as summarized, focuses on women's brains, hormones and food rather than exercise [4]. She highlights that nearly two-thirds of Alzheimer's patients are women, and that this is not solely due to longevity: women's brains go through a sharp drop in estrogen at menopause, and estrogen is vital for energy metabolism, immune response, neuroplasticity and glucose processing in the brain [4]. "Estrogen isn't just a reproductive hormone. It's a superpower for your brain" [4]. She describes menopause as a neurological transition, not just a reproductive one [4].
Her levers are nutritional: berries, dark leafy greens, omega-3-rich fats, phytoestrogens and water with electrolytes [4]. She warns that Western diets lead to brain shrinkage and higher plaque accumulation even in midlife, while Mediterranean-style diets show larger, fuller brain volumes and delayed cognitive decline [4]. "Your brain literally shrinks on a processed food diet—and plaques can form decades before symptoms appear" [4]. Like Suzuki on exercise, she treats brain care as an everyday lifestyle rather than a reactive intervention, and she prefers whole foods to isolated supplements [4]. Her summary doesn't discuss exercise directly, so her contribution to this question is as a complement: the same brain that exercise protects is also shaped by food, hydration and hormones [4]. D'Esposito's lab work fits here too: Emily Jacobs found the frontal lobes may have more estrogen receptors than any other brain area, and that frontal function and working memory rose and fell with estrogen across the menstrual cycle [5].
Where They Agree, and Where They Don't
Broad agreement that exercise is central. Huberman [1], Suzuki [2], Attia [3] and D'Esposito [5] all put exercise at or near the top of what people can do for brain health. Attia lists it among four unequivocal levers [3]; Suzuki names it one of her three top tools for attention [2]; D'Esposito notes neurologists now talk about exercise in a way they didn't 30 years ago [5]; Huberman builds a weekly plan around it [1].
Agreement on BDNF and the hippocampus. Huberman and Suzuki both describe exercise raising BDNF and supporting the hippocampus [1][2]. They emphasize different routes: Huberman points to osteocalcin from loaded bones, lactate and VEGF, and activity-dependent BDNF [1]; Suzuki points to a muscle myokine and the liver ketone beta-hydroxybutyrate [2]. These are complementary rather than competing explanations.
Agreement that drugs have not delivered. Attia on disappointing anti-amyloid drugs [3] and D'Esposito on donepezil-type drugs that don't slow the disease [5] make the same point from different angles. Attia also puts common brain supplements far behind the core levers [3], and Mosconi prefers food to isolated supplements [4].
Where they differ: which exercise. Suzuki's research has focused almost entirely on cardiovascular exercise, and Huberman notes during their conversation that studies of weight training and high-intensity intervals are still needed [2]. In his own episode, Huberman argues for a mix: zone 2, high-intensity intervals, time under tension and jumping with controlled landings [1]. Attia's suggested minimum is also a mix of zone 2, strength and intervals [3]. So the sources agree cardio is well supported, while the case for resistance and impact work rests more on mechanisms like osteocalcin and on fall prevention [1][3].
Where they differ: how much. Suzuki's unpublished data suggest "every drop of sweat counted," with more classes bringing more benefit [2]. Huberman warns that you can overdo it: two hard HIIT sessions in one day reduced cognitive performance, linked to reduced cerebral blood flow, and "you need to be cautious about not overdoing the high intensity interval training sessions" [1]. The two views fit if Suzuki's finding is about weekly volume of moderate cardio and Huberman's is about stacking very intense sessions right before mental work.
Where the evidence is softer. Several sources flag their own uncertainty. New neurons in the adult human hippocampus remain debated [2]. Huberman calls the evidence for osteocalcin raising neuron numbers limited, and says there has been no systematic study of which exercises best release osteocalcin in humans [1]. The Swedish fitness study is correlational [2], and so is the super-ager link to the anterior midcingulate cortex [1]. Suzuki's dose-response and older-adult results are unpublished [2]. None of the sources presents a head-to-head trial of exercise versus a dementia drug; the better than any pill case rests on exercise being among the few proven levers while the drugs described either don't slow disease or haven't been developed [3][5].
Why It Works
The sources describe several layers of mechanism.
Immediate arousal. Movement drives adrenaline, which acts through the vagus nerve and brainstem to release norepinephrine and dopamine across the brain, making the prefrontal cortex and hippocampus easier to engage [1]. Compound and core movements most strongly activate the adrenals, according to Strick's circuit tracing [1]. Suzuki describes the same neurochemical surge in plainer terms, as dopamine, serotonin and noradrenaline released every time you move [2].
Growth factors. Aerobic exercise raises BDNF, which helps new brain cells grow in the hippocampus [2], and BDNF works best on neurons that are already active [1]. Signals from muscle (a myokine) and liver (beta-hydroxybutyrate) cross the blood-brain barrier to boost it [2], and osteocalcin from loaded bones appears to mediate many of its effects [1].
Fuel and plumbing. Lactate fuels neurons during hard exercise and triggers VEGF, strengthening a blood-brain barrier that breaks down in age-related decline and especially in Alzheimer's [1]. Long zone 2 sessions keep the heart and blood vessels healthy enough to deliver blood and fuel to the brain [1]. Attia's lever list overlaps here: exercise helps with blood pressure, lipids and insulin sensitivity [3], and lower ApoB and insulin sensitivity are themselves on his short list of brain-protective factors [3].
Structure over time. The hippocampus shrinks with age even without dementia, and exercise can change the slope of that decline [1]. Suzuki hopes a bigger hippocampus would delay how much a disease like Alzheimer's could affect memory [2]. Super-agers keep the anterior midcingulate cortex that effortful challenges, including exercise, appear to build [1].
Sleep. Exercise, especially high-intensity work early in the day with morning light, improves deep slow-wave and REM sleep, which carry many of exercise's brain benefits [1]. Sleep is also on Attia's list of unequivocal levers [3] and on Suzuki's list of top tools [2].
Where It Applies, and Where It Doesn't
For healthy aging and prevention, the case is strongest. The hippocampus shrinks naturally with age [1], the frontal executive system declines the most with healthy aging [5], and exercise addresses both, through the hippocampus [1][2] and through executive function [5]. Suzuki's pandemic data suggest older adults gain even more on attention than younger ones [2].
For people with genetic risk, Attia says ApoE4 raises risk but isn't destiny, while the rare PSEN1, PSEN2 and APP mutations (about 1% of cases) cause early-onset disease in the 50s [3]. The sources don't claim exercise overrides those rare mutations.
For people who already have Alzheimer's, the sources are cautious. Suzuki explicitly does not claim exercise would cure her father's disease [2]. D'Esposito's clinical tools for established disease are mostly symptomatic drugs, social support and personal materials like family albums [5]. The strongest claims here are about prevention and delay, not reversal.
For women, Mosconi's work adds that the menopause transition is a distinct window of brain risk, shaped by estrogen and diet [4]. Exercise is not a substitute for addressing that, and her summary points to food and, for some women, hormone therapy [4].
When exercise can backfire. Too much intense exercise right before mental work lowers cerebral blood flow and focus [1]. Injuries matter because detraining harms the brain within about 10 days [1]. And Attia's warning on falls cuts both ways: training strength, stability and eccentric control prevents falls [3], but jumping and new eccentric work should be progressed slowly, since they cause soreness and injuries [1]. If you have a heart condition, joint problems, a history of falls or a neurological diagnosis, get a doctor's or physical therapist's advice before starting harder training.
Where the sources leave gaps. Weight training and HIIT are less studied for brain outcomes than cardio [2]. Human osteocalcin research is limited [1]. Several numbers cited are correlational or unpublished [1][2].
What To Do With It
- Get your heart rate up most days. Suzuki's baseline is cardio that raises your heart rate, even a power walk, with 30 to 45 minutes enough to sharpen mood and attention for hours [2]. She has built up to seven days a week [2].
- Include at least one long zone 2 session a week. Huberman suggests 45 to 75 minutes of jogging, rowing, cycling, swimming or weighted hiking at a pace where you can still talk [1]; Attia recommends three to four hours of zone 2 a week for blood pressure [3].
- Add one high-intensity interval session a week, such as one minute all out and 30 seconds rest for seven or eight rounds on an air bike, but don't stack two hard sessions in one day before mental work [1].
- Lift, and load your bones. Include resistance training, and add jumps with controlled landings (box jumps, rope work), starting low and progressing slowly [1]. Attia's minimum mix is an hour each of zone 2, strength and intervals per week [3].
- Train your brakes and balance. Practice eccentric strength and stability, such as slow single-leg step-ups (two seconds up, four seconds down), to prevent the falls that are so dangerous after 65 [3].
- Don't go more than about 10 days without training. A week off is fine; ramp back up gently after illness or injury [1].
- Time exercise before demanding mental work. Suzuki suggests exercising right before you most need your brain, often in the morning [2]; Huberman says before, during or after learning all work if they are close in time [1].
- Do one safe thing you don't want to do, such as a cold shower or a new skill, to challenge the anterior midcingulate cortex that super-agers maintain [1].
- Pair exercise with the other proven levers: sleep well, get ApoB and blood pressure checked and managed, and stay insulin sensitive [3]. Protect your head from injury [3].
- Eat for your brain too, with berries, leafy greens, omega-3 fats and plenty of water, and limit processed foods [4]. Spend your effort on these before brain supplements, which Attia calls a rounding error [3].
Sources
- How to Use Exercise to Improve Your Brain’s Health, Longevity & Performance · Huberman Lab
- Dr Wendy Suzuki Boost Attention Memory With Science Based Tools · Huberman Lab
- Dr Peter Attia Improve Vitality Emotional Physical Health Lifespan · Huberman Lab
- Optimize Your Brain Health & How to Prevent Alzheimer's with Dr. Lisa Mosconi · The School of Greatness
- Dr Mark Desposito How To Optimize Cognitive Function Brain Health · Huberman Lab
Products and resources from this insight
Some of these are affiliate links: if you buy through them, Perfect Summary is paid a commission, at no extra cost to you. As an Amazon Associate, Perfect Summary earns from qualifying purchases.
-
Protocols: An Operating Manual for the Human Body · Andrew Huberman(our summary)Huberman's forthcoming book of health protocols, including exercise.
-
Air bike (Airdyne or Assault bike)Huberman uses it for short, time-efficient high-intensity interval sessions.
-
Outlive: The Science and Art of Longevity · Peter AttiaAttia's book on healthspan, lifespan and emotional health discussed throughout.
-
Flaxseed oilHer plant-based omega-3 source; about a tablespoon a day.





