Originally published in February 2014 and comprehensively updated/edited in July 2026.
Large 2014 reviews strengthened evidence that aerobic exercise protects memory, mood, and brain health, from BDNF to hippocampal growth in aging adults.
In the early 2010s, exercise was still mostly sold to the public as a matter of vanity and cardiovascular arithmetic: burn calories, protect your heart, fit into last year’s jeans. Gyms filled with the first wave of people wearing pedometers, and soon fitness trackers, counting steps the way earlier generations had counted calories. Meanwhile, a quieter trend was emerging: the “brain training” boom, in which apps promised to sharpen memory and attention through puzzles and drills, riding a wave of public anxiety about cognitive decline and an increasingly distracted, screen-saturated world. Few people thought to connect these two trends. The treadmill and the brain-training app seemed to belong to entirely different projects: one for the body, one for the mind.
By 2014, that separation was starting to look like a mistake. A run of large reviews and meta-analyses — pulling together evidence from neuroscience, psychology, and gerontology — had converged on a conclusion that seemed almost too tidy to be true: aerobic exercise, the kind that raises your heart rate for a sustained period, does more for the brain than any commercial brain-training product had been able to demonstrate. It appeared to protect memory. It appeared to buffer mood. And in the brains of older adults, it appeared to do something that sounded almost like science fiction — reverse, in a modest but measurable way, the shrinkage of a brain structure that is central to memory formation.
This article traces that body of evidence: where it came from, what it actually shows, and where the boundaries of scientific certainty sit. The story of exercise and the brain is a good case study in how science moves — not through a single dramatic discovery, but through the slow accumulation of studies from different labs, different species, and different methods, all pointing in a compatible direction.
The Body of Evidence Converges
Interest in exercise and cognition did not begin in 2014. Researchers had been probing the connection for decades, largely in animal models. But by the early 2010s, enough human studies had accumulated that scientists could start doing something new: synthesizing them. Large reviews and meta-analyses are a different kind of scientific product than an individual study. Rather than testing a single hypothesis in a single sample, they pool findings across many studies, looking for patterns that hold up across different populations, methods, and research groups. When a pattern survives that kind of scrutiny, it carries more weight than any one experiment could on its own.
One influential synthesis from this period, led by researchers including Louis Bherer, Kirk Erickson, and Teresa Liu-Ambrose, pulled together the accumulated evidence on physical activity and cognitive and brain function specifically in older adults. Their conclusion was not that exercise makes people smarter in some general sense, but something more specific and, in a way, more useful: physical activity appears to help preserve cognitive abilities that would otherwise decline with age, particularly the mental skills involved in planning, focus, and mental flexibility — collectively known as executive function.
From Correlation to Controlled Trials
A recurring challenge in this field is the difference between showing that two things travel together and showing that one actually causes the other. Early research on exercise and cognition was mostly correlational: people who exercised more also tended to score better on memory and attention tests, and to show larger volumes in certain brain regions on MRI scans. That kind of evidence is suggestive, but it cannot rule out other explanations. Maybe people with healthier brains to begin with are simply more inclined to exercise. Maybe some third factor, like socioeconomic status or overall health, drives both variables at once.
What made the reviews synthesized around 2014 more persuasive was the growing presence of randomized controlled trials — studies in which researchers assign participants to an exercise program or a comparison activity and then track outcomes over time. This design allows scientists to speak more confidently about cause and effect, because the groups start out equivalent on average, and any later difference is more plausibly attributable to the exercise itself. A synthesis published in the Annual Review of Psychology by Ruchika Prakash, Michelle Voss, Kirk Erickson, and Arthur Kramer, first circulated in the fall of 2014, is a good illustration of this shift toward stronger designs. The authors described modest but consistent support for the idea that physical activity helps preserve — and, in some cases, enhance — cognitive functioning in older adults, with the clearest benefits showing up in tasks that depend on the prefrontal cortex and the hippocampus, two brain regions with outsized importance for planning, decision-making, and memory. They were careful to note, though, that the evidence for children and younger adults was thinner, and that the field still needed better-powered trials in those groups.
What Happens Inside a Moving Brain
Knowing that exercise correlates with, and even appears to cause, better cognitive outcomes is only half the story. Scientists also wanted to know why. What is actually happening, at a biological level, when someone goes for a brisk walk or a run?
BDNF: Fertilizer for Neurons
Much of the explanation centers on a protein called brain-derived neurotrophic factor, or BDNF. Researchers sometimes describe it, a little informally, as a kind of fertilizer for the brain — it supports the growth, survival, and connection of neurons. Animal studies going back to the 1990s had already shown that when rodents run on a wheel, BDNF levels rise sharply in the hippocampus, the brain’s primary hub for forming new memories. Foundational work by researchers such as Carl Cotman and Nicole Berchtold helped establish that this molecular response was not a curiosity of rodent physiology; it appeared to be a general biological principle connecting physical activity to brain plasticity, the brain’s capacity to reorganize and strengthen its own connections in response to experience.
Human studies added another layer. Circulating BDNF in the bloodstream tends to rise after aerobic exercise, and higher levels have been associated with better performance on tests of executive function, including in adults well into their seventies. The exact relationship between blood-level BDNF and what is happening inside the skull is still an active area of investigation — blood is an indirect proxy for brain activity, not a direct window into it — but the animal and human findings point in a consistent direction. Aerobic activity appears to nudge the brain toward a more plastic, adaptable state, at both the molecular and structural level. It’s worth noting that researchers did not view BDNF as the sole explanation. Even then, many scientists regarded it as one important piece of a broader picture that likely included improved cerebral blood flow, other growth factors, and additional biological pathways that were still being mapped.
The Hippocampus Study That Changed the Conversation
If one study captured the public imagination during this period, it was a 2011 paper by Kirk Erickson and colleagues, published in the Proceedings of the National Academy of Sciences, which continued to shape the field’s thinking well into 2014 and beyond. The hippocampus normally shrinks somewhat with age, a process linked to memory decline and, in its more severe forms, to dementia risk. Erickson’s team recruited 120 older adults who were not regularly active and randomly assigned them to one of two groups: one walked briskly for about 40 minutes, three days a week, for a year; the other did stretching and toning exercises of similar duration and frequency, without an aerobic component.
A year later, the walking group’s hippocampal volume had grown by roughly two percent — enough, the researchers estimated, to reverse one to two years’ worth of typical age-related shrinkage. The stretching group, by contrast, continued to show the expected gradual decline. The walking group also performed better on tests of spatial memory, and the degree of hippocampal growth correlated with higher levels of BDNF in the blood, tying the structural finding back to the molecular story described above. In interviews at the time, the study’s lead author pushed back on the assumption that hippocampal shrinkage is simply an inevitable feature of aging, pointing out that a year of moderate exercise was enough to grow the structure — evidence, in his view, that the older brain retains more capacity for change than is often assumed.
It’s worth pausing on what this study did and did not show. It was a genuine randomized controlled trial, which strengthens the case for causation. But it involved a specific population — sedentary older adults without dementia — a specific exercise protocol, and a specific outcome measure. Other researchers, including a critique published later that same year, raised methodological questions about whether the memory improvements could be attributed with full confidence to the hippocampal changes specifically, illustrating the normal give-and-take of scientific debate rather than any fatal flaw. The finding held up reasonably well under continued scrutiny and became one of the most cited pieces of evidence in the field, but it was always one strong data point within a much larger, still-developing picture — not a single, final word.
Memory, Mood, and the Aging Brain
Taken together, the reviews from this period suggested that exercise’s brain benefits cluster around a few overlapping domains: memory, executive function, and mood.
Protecting Executive Function
Executive function is something of an umbrella term, covering the mental processes involved in holding information in mind, switching between tasks, ignoring distractions, and planning ahead. In plainer terms, these are the mental skills that let us plan, resist distractions, juggle information, and adapt to changing situations. This cluster of abilities tends to be one of the first cognitive domains to show signs of strain with age, and it is also, conveniently, one of the domains most consistently linked to physical activity in the research literature. A widely cited earlier meta-analysis by Stanley Colcombe and Arthur Kramer, published in 2003, had already found that fitness training produced the largest cognitive benefits specifically for tasks requiring this kind of executive control, rather than for simpler tasks like straightforward reaction time. The reviews synthesized around 2014 largely reinforced that pattern, adding newer neuroimaging evidence to the older behavioral findings.
Exercise as an Ally Against Depression
The mood side of this story runs on a related but distinct mechanism. Depression has long been associated with reduced BDNF activity and, in some studies, with smaller hippocampal volume, echoing the same biological territory implicated in memory. Exercise’s antidepressant-like effects, observed across both animal models and human clinical studies, appear to draw on some of the same neurotrophic machinery. This doesn’t mean exercise is a replacement for established treatments like therapy or medication — the researchers in this field have generally been careful not to overstate that claim — but it does suggest a plausible, mechanistically grounded reason why physical activity shows up so reliably as a helpful adjunct for mood regulation, alongside its cognitive benefits.
How Much Exercise, and What Kind?
A natural question, once the basic effect is established, is what dose of exercise actually matters. Here the evidence from this period was less precise than the headline findings might suggest. The Erickson hippocampus study used moderate-intensity walking, three days a week, for a year — a program well within reach of most sedentary older adults, which was part of its appeal as a public health message. Other studies in the broader literature used different intensities, durations, and modalities, from cycling to resistance training, and the reviews generally concluded that aerobic exercise specifically, more than resistance training alone, was the modality with the strongest link to neurotrophic and cognitive benefits — though resistance training showed benefits of its own, particularly for certain aspects of executive function.
What the field did not have, as of 2014, was a well-established dose-response curve — a clear answer to how much exercise, at what intensity, produces the maximum cognitive benefit, or whether there are diminishing returns or even a point past which additional exercise stops helping. The Prakash and colleagues review explicitly flagged this as a priority for future research, along with a more sophisticated understanding of how genetic differences between individuals might shape how much benefit any given person gets from a given exercise program. That kind of precision — the sort that could eventually turn “exercise more” into something closer to a personalized prescription — was still on the horizon rather than in hand.
The Limits of the Evidence
It’s tempting, when a body of research points so consistently in one direction, to treat the conclusion as settled. The more accurate description, and the one the researchers themselves tended to favor, is that the evidence was strengthening rather than complete. Several caveats are worth holding onto.
Much of the foundational biology — the BDNF pathway, the neurogenesis observed in the hippocampus — was first established in rodent studies. Rodent and human brains share deep evolutionary similarities, and the translational research bridging the two species had grown considerably more convincing by 2014, but biological mechanisms don’t always transfer cleanly across species, and the human evidence for some of the more granular molecular claims remained less direct than the animal evidence. Blood-based measures of BDNF, in particular, are a proxy for what’s happening in the brain rather than a direct measurement of it.
The human trials themselves, while improving, still tended to involve relatively small and short-duration samples by the standards of, say, pharmaceutical research. The Erickson hippocampus study’s 120 participants was considered a large sample for this kind of neuroimaging trial, but it is modest compared to the sample sizes typically required for high confidence in medical research more broadly. And as the Prakash review noted, the evidence outside of older adults — particularly in children and younger adults — was comparatively underdeveloped, resting more on correlational and cross-sectional data than on the kind of randomized designs that had begun to anchor the older-adult literature.
None of this undermines the core finding. It simply means that “exercise changes the brain” is better understood, as of this period, as a well-supported and mechanistically plausible hypothesis with real experimental backing, rather than a fully resolved fact with every detail nailed down. It’s also worth being honest about scale: the cognitive benefits observed in these studies were generally modest rather than transformative — about a two percent increase in hippocampal volume and a meaningful but not dramatic improvement on memory tests — not a cure for age-related decline. But modest effects can matter a great deal when they’re sustained across years of aging, which is part of what made this research compelling in the first place. That distinction matters, because it shapes how confidently the finding can be translated into public health advice, and it leaves room for the more precise, more individualized understanding that later research would go on to pursue.
Conclusion
The picture that emerged from the exercise-and-brain research converging around 2014 was, in its own understated way, one of the more hopeful stories in neuroscience at the time. It suggested that an intervention already known to be good for the heart, the lungs, and the metabolism was also, through pathways involving proteins like BDNF and structures like the hippocampus, good for memory and mood — and that this benefit was not limited to the young or the already fit. Sedentary older adults, the population perhaps most anxious about cognitive decline, appeared to have the most to gain, and the exercise required to see a measurable effect was moderate rather than extreme.
The broader significance of this research extends beyond any single finding. It marked a shift in how scientists and the public thought about brain health — away from a purely fixed, use-it-or-lose-it model of an aging brain, and toward a more dynamic picture in which everyday behavior, not just genetics or luck, plays a meaningful role in shaping cognitive trajectories. That reframing would go on to influence a great deal of subsequent research, from studies on sleep’s role in clearing metabolic waste from the brain to the growing science of how physical activity interacts with diet, stress, and social connection to shape long-term brain health. The treadmill and the brain, it turned out, were never as separate as the marketing of the early 2010s made them seem.
Scientific References
Erickson, K. I., Voss, M. W., Prakash, R. S., Basak, C., Szabo, A., Chaddock, L., Kim, J. S., Heo, S., Alves, H., White, S. M., Wojcicki, T. R., Mailey, E., Vieira, V. J., Martin, S. A., Pence, B. D., Woods, J. A., McAuley, E., & Kramer, A. F. (2011). Exercise training increases size of hippocampus and improves memory. Proceedings of the National Academy of Sciences, 108(7), 3017–3022.
Prakash, R. S., Voss, M. W., Erickson, K. I., & Kramer, A. F. (2014/2015). Physical activity and cognitive vitality. Annual Review of Psychology, 66, 769–797.
Bherer, L., Erickson, K. I., & Liu-Ambrose, T. (2013). A review of the effects of physical activity and exercise on cognitive and brain functions in older adults. Journal of Aging Research, 2013, 657508.
Colcombe, S., & Kramer, A. F. (2003). Fitness effects on the cognitive function of older adults: A meta-analytic study. Psychological Science, 14(2), 125–130.
Cotman, C. W., & Berchtold, N. C. (2002). Exercise: A behavioral intervention to enhance brain health and plasticity. Trends in Neurosciences, 25(6), 295–301.
van Praag, H., Shubert, T., Zhao, C., & Gage, F. H. (2005). Exercise enhances learning and hippocampal neurogenesis in aged mice. Journal of Neuroscience, 25(38), 8680–8685.
Coen, R. F., Lawlor, B. A., & Kenny, R. (2011). Failure to demonstrate that memory improvement is due either to aerobic exercise or increased hippocampal volume. Proceedings of the National Academy of Sciences, 108(18), E89.
