In the mid-1990s, a researcher named Carl Cotman at UC Irvine made rats run on wheels and then looked at their brains.
He found something unexpected: the running rats had significantly elevated levels of a protein called BDNF — brain-derived neurotrophic factor — in their hippocampus. And their neurons were healthier, more connected, and more capable of change.
That discovery opened a door that has since led to one of the more remarkable reframings in neuroscience: exercise isn’t just good for your body. It is one of the most powerful interventions we know of for brain health.
What BDNF Is and Why It Matters
BDNF is often called “fertilizer for the brain.” That’s not quite right — it’s more precise than that — but the metaphor captures the essence.
BDNF is a protein in the neurotrophin family. Its primary job is to support the survival and growth of neurons, promote the formation of new synaptic connections, and regulate synaptic plasticity — the brain’s ability to strengthen or weaken connections based on activity.
The hippocampus has the highest density of BDNF receptors in the brain. This is relevant because the hippocampus is where learning and memory consolidation happen, and where new neurons are born throughout adult life (a process called adult neurogenesis).
Low BDNF levels are associated with depression, anxiety, age-related cognitive decline, and neurodegenerative diseases including Alzheimer’s. High BDNF is associated with sharper memory, better mood regulation, and greater cognitive resilience under stress.
And exercise is one of the most reliable ways to raise BDNF.
The Hippocampus Grows Back
The most striking evidence came from a 2011 study by Kirk Erickson and colleagues at the University of Pittsburgh.
They took 120 sedentary adults aged 55-80 and randomized them to either a year of aerobic exercise (walking, three times per week, gradually increasing to 40-minute sessions) or a stretching control group.
The aerobic exercise group showed a 2% increase in hippocampal volume over the year. The stretching group showed a 1.4% decrease — the normal age-related shrinkage.
A 2% increase doesn’t sound dramatic. But consider: typical hippocampal volume loss from aging runs at about 1-2% per year. The exercise group didn’t just slow the decline — they reversed it. Participants who exercised also performed significantly better on spatial memory tasks.
What’s happening mechanically: aerobic exercise increases cerebral blood flow, raises BDNF, and accelerates hippocampal neurogenesis. The new neurons integrate into hippocampal circuits, particularly those involved in memory encoding and spatial navigation.
John Ratey, a Harvard psychiatrist, named BDNF “Miracle-Gro for the brain” in his book Spark. The research behind that description is solid.
Exercise vs. Antidepressants
In 1999, James Blumenthal at Duke University published a study that got a lot of attention — and deserved it.
He randomly assigned 156 depressed adults to three groups: antidepressant medication (sertraline/Zoloft), aerobic exercise, or both. After 16 weeks, all three groups showed equivalent reductions in depression symptoms.
A follow-up 10 months later found that participants who had been in the exercise-only group had the lowest relapse rates. Those on medication alone had the highest.
The study has since been replicated and extended. A 2007 Blumenthal paper found similar results with a larger sample. A meta-analysis by Josefin Rethorst examined 58 randomized trials and found that exercise had a significant antidepressant effect across diverse populations.
The mechanisms that explain this include BDNF, but also:
- Increased serotonin turnover from aerobic exercise
- Reduced hypothalamic-pituitary-adrenal (HPA) axis reactivity, meaning lower cortisol response to stress
- Increased endocannabinoid signaling (the “runner’s high” is more endocannabinoid than endorphin — recent research has shifted the attribution)
- Structural changes to the prefrontal cortex, improving top-down emotion regulation
Exercise doesn’t just treat symptoms. It changes the structure of the brain that generates them.
What Kind, How Much
The research is fairly consistent on the parameters:
Aerobic exercise (moderate intensity, meaning you can talk but not sing) shows the strongest cognitive effects. Resistance training also has positive effects, particularly for executive function, but the BDNF surge is most pronounced with cardio.
Dose: 30-45 minutes per session. Three sessions per week appears to be a threshold for meaningful hippocampal effects. Daily is better for mood but the structural gains don’t scale linearly — three solid sessions may produce most of the benefit of seven.
Intensity: Moderate to vigorous. High-intensity interval training (HIIT) produces significant BDNF spikes, though the research on sustained structural change is still less robust than the aerobic exercise literature.
Consistency: The hippocampal volume effects take months. This is not a two-week experiment. Erickson’s reversal of hippocampal shrinkage took a full year.
The 10,000-Step Misframing
There’s a common framing in popular wellness culture that “10,000 steps per day” is the magic number for brain health. That target actually comes from a 1960s Japanese pedometer marketing campaign — not from neuroscience.
What matters more than step count is elevated heart rate. A 20-minute jog at 65-75% of your maximum heart rate does more for BDNF than 10,000 casual steps.
This doesn’t mean casual walking is useless. It has real cardiovascular and mood benefits. But if you’re specifically trying to grow your hippocampus and improve cognitive function, you want intensity sufficient to meaningfully elevate heart rate.
What This Means for a Normal Life
The implication is straightforward but hard to fully absorb: exercise is not a leisure activity that you do when you have time. It is cognitive infrastructure.
A person who exercises regularly is, on average, operating with a larger hippocampus, higher BDNF levels, lower baseline cortisol, and a more responsive prefrontal cortex. Those are meaningful cognitive advantages.
A person under chronic stress who doesn’t exercise is allowing the stress-related cortisol to gradually suppress hippocampal neurogenesis — the opposite of what exercise does.
The two states compound in opposite directions.
You can think of aerobic exercise as a biological override on the stress spiral. The hippocampus starts growing. BDNF rises. The brain becomes slightly harder to destabilize.
Not magic. Just the molecular machinery, running in your favor.
New research reports every Sunday. Full audio report on Substack.
What state is your brain in right now?