Did the Miles Shape the Mind? Inside the New Science of Multi-Marathon Cognition

What if the people logging marathon after marathon really do think differently?

That's the question sitting underneath a study published this year in the International Journal of Sport and Exercise Psychology, led by Leo Lundy and Richard B. Reilly's group at Trinity College Dublin. The researchers gave a battery of cognitive tests to 130 multi-marathoners — people who measure their running careers in repeated marathon finish lines rather than a single race — and compared the results against normative data drawn from a large study of aging adults in Ireland. The multi-marathoners, whose average age was 53, made fewer errors on a task that measures sustained attention, responded faster on tests of cognitive and motor reaction time, and scored higher on a standard screening measure of general cognitive function than the age- and gender-matched comparison sample.

It's a tidy, appealing finding. It is also, on its own, close to useless for answering the question everyone wants answered: does running do this to the brain, or do certain brains simply produce people who end up running this much?

Those are not the same claim, and the gap between them is the actual subject of this article. Running, in what follows, is not the point. It's the instrument — a way of asking a much older and more stubborn question about the relationship between behavior and cognition: when we find a group of unusually disciplined, unusually persistent people performing unusually well on tests of attention, what exactly are we allowed to conclude?

What the Trinity Study Actually Measured

It's worth being precise about the study before doing anything else with it, because vague citations are how "runners perform well on an attention test" quietly mutates into "running makes you smarter," which is not what anyone found.

The sample was 130 multi-marathon runners with a mean age of 53.3, 57% of them male. The researchers used three instruments: the Sustained Attention to Response Task (SART), which measures the ability to maintain vigilance over a repetitive task while inhibiting an automatic response; a choice reaction time test; and the Mini-Mental State Examination (MMSE), a brief general screen for cognitive impairment rather than a measure of intelligence in any technical sense. Their comparison group wasn't a matched control sample recruited for the study — it was normative data from The Irish Longitudinal Study on Ageing (TILDA), a large, well-established cohort study of aging in Ireland, matched by gender.

Using principal component analysis and k-means clustering, the authors also identified two distinct cognitive subgroups within the runner sample itself: one characterized by more consistent attentional control and faster responses, the other by greater variability across tasks. Educational attainment differed between these two subgroups; age and gender did not. Across the sample as a whole, cognitive performance didn't vary meaningfully by age.

Crucially, the authors' own conclusions are more careful than any headline built from this study is likely to be. They call explicitly for longitudinal research to determine how these differences actually develop — how much is attributable to running itself, how much to broader training and lifestyle patterns, and how much to the kind of person who ends up pursuing multi-marathoning in the first place. Some details of the full paper — precise effect sizes, the exact operational definition used for "multi-marathoner," recruitment criteria, and any statistical adjustment for covariates — aren't available from the published abstract and aren't reconstructed here.

What This Study Cannot Tell You

The study is cross-sectional: everyone was tested once, at one point in their running careers, and compared to population norms collected separately. Cross-sectional data can establish that two things travel together. It cannot, by itself, establish that one produced the other, and it cannot rule out a third factor producing both.

This isn't a technicality. It's the entire ballgame. There are at least three live explanations for why multi-marathoners might show sharper attention and faster reaction times than age-matched peers, and the Trinity data can't adjudicate between them on its own.

The first is that sustained endurance exercise, accumulated over years, contributes causally to maintaining certain cognitive functions — plausible, and supported by a real body of separate evidence discussed below. The second is that people who are already unusually persistent, attentive, and self-regulating are more likely to become the kind of person who accumulates marathon after marathon in the first place — a selection effect, not a training effect. The third is that both are true simultaneously, reinforcing each other over years in ways a single cross-sectional snapshot can't disentangle.

The Trinity researchers' own finding that educational attainment differed between the two cognitive subgroups is itself a reminder that cognitive-test performance does not emerge in a vacuum. Education is known to influence performance on some cognitive assessments, including general cognitive screening measures such as the MMSE. That does not explain the runners' results, but it gives researchers another variable that future longitudinal work will need to disentangle from training itself.

What Decades of Exercise-Cognition Research Actually Shows

None of this means the causal hypothesis is empty. It means it has to earn its place using different kinds of evidence than a cross-sectional comparison can provide — and a substantial amount of that evidence exists, mostly from research on general aerobic exercise rather than marathon running specifically.

The foundational work here is a 2003 meta-analysis by Stanley Colcombe and Arthur Kramer, which pooled eighteen intervention studies and found that aerobic fitness training produced measurable, selective benefits for cognition in sedentary older adults — with the largest effects concentrated in executive-control processes, the same broad category of function the SART is designed to probe. Later meta-analyses have generally reinforced that finding while keeping the effect modest: a 2018 systematic review by Northey and colleagues in the British Journal of Sports Medicine, covering 39 randomized trials (36 contributing to the pooled analysis) in adults over 50, found an overall cognitive benefit of aerobic exercise with an effect size of 0.29 (95% CI 0.17–0.41) — real, but far from transformative.

The mechanistic story is where things get genuinely uncertain. The most-cited piece of evidence is a 2011 randomized controlled trial by Kirk Erickson and colleagues, published in the Proceedings of the National Academy of Sciences, which reported that a year of moderate aerobic exercise (walking) increased anterior hippocampal volume in older adults by about 2%, while a stretching-and-toning control group continued to lose volume, alongside improvements in spatial memory and higher circulating BDNF. But the cognitive interpretation was contested almost immediately: researchers at Trinity College Dublin pointed out that both the exercise group and the control group improved on the memory task, meaning the study didn't actually establish that the exercise itself caused the memory gain, whatever the headline said. Later, pooled evidence made the structural story more uncertain still. A 2023/2024 meta-analysis of eight randomized trials involving 554 cognitively healthy older adults found no statistically significant overall effect of aerobic exercise on hippocampal volume (SMD = 0.10, 95% CI −0.01 to 0.21, p = 0.073) — even though the same trials showed a real, moderate improvement in cardiorespiratory fitness, and that fitness improvement wasn't associated with any change in hippocampal volume at all. The famous 2011 finding remains an important single result. It should not be read as proof of a settled exercise-to-hippocampus-to-memory pathway.

What almost none of this literature does, at any level of the causal chain, is study marathon running, let alone multi-marathoning, specifically. The jump from "moderate aerobic exercise, a few times a week, in a randomized trial" to "running a hundred marathons produces additional cognitive benefits beyond ordinary fitness" is an extrapolation, not a finding — and it's one this literature doesn't license.

The Kind of Mind That Runs a Hundred Marathons

If the causal story is uncertain, the selection story has real texture — and some of it comes from the very same research group.

A separate 2024 study by Lundy and Reilly, surveying 830 multi-marathoners globally, found a population that is disproportionately older, engaged in the sport as much for social connection and identity as for competition, and self-selected through channels — running clubs, dedicated event ecosystems — that concentrate a particular kind of committed, organized participant. A 2025 follow-up from the same group, using a brief personality inventory with 593 respondents, found multi-marathoners scored higher on conscientiousness than general-population norms — consistent with the discipline needed to sustain years of structured training — but lower on emotional stability, which the authors interpreted as a possible marker of difficulty with stress regulation rather than an asset. That's a more complicated picture than a simple "resilient personality" story, and a reminder that the traits selecting people into this sport aren't uniformly the traits a reader might assume make for a sharper cognitive profile.

Broader research on the demographics of distance running is suggestive here too, though it has to be read carefully. A 2025 paper in Frontiers in Psychology analyzed survey data collected in Poland in early 2008 from 1,539 runners and found that socioeconomic status shaped motivation — higher-SES participants leaned more toward achievement-based goals, lower-SES participants toward health and social connection — though the reported effect sizes were small, and when the study compared recreational runners, marathoners, and ultramarathoners directly, professional or vocational status was the only socioeconomic variable that differed significantly across the three groups. That's a modest, geographically and temporally specific finding, not a portrait of today's global running population, and it shouldn't be stretched into a claim that contemporary marathoners are disproportionately affluent or highly educated. What it does support is the narrower, more defensible point: socioeconomic context measurably shapes who takes up endurance sport and why, which keeps selection effects on the table as a live explanation for cognitive differences observed later.

Is Running Itself a Cognitive Exercise?

There's a less obvious possibility that sport psychologists have been building a case for since the 1970s: that running, especially over the volumes multi-marathoners accumulate, is not a cognitively passive activity but a repeated exercise in self-regulation.

Work by Noel Brick, Tadhg MacIntyre, and Mark Campbell has argued that endurance performance requires ongoing metacognitive work — monitoring bodily sensation, regulating pace and effort, managing the urge to slow down or quit, and adjusting strategy as fatigue accumulates. Their 2016 framework in Frontiers in Physiology treats this as active self-regulation, not automatic locomotion, and their earlier work distinguishes between associative attention (directed at bodily and performance cues) and dissociative attention (directed away from them), with more experienced runners showing more deliberate, flexible attentional control than novices. That's suggestive of running as a form of long-term attentional practice, distinct from its cardiovascular effects.

It's important to be exact about what this literature actually shows, though: it demonstrates that cognition shapes running performance in the moment — how well someone manages attention affects how well they race — not that running durably improves cognition afterward. Those are adjacent claims, not the same one, and conflating them is one of the easier ways an honest hypothesis turns into an overstated one. Separately, acute research on cognitive performance immediately after ultra-endurance events — including a 2022 study in Frontiers in Physiology that tested 24 experienced runners before and after races ranging from 48 to 193 kilometers and found reaction times slowed by 23–31% across several measures, alongside an 8% drop in recognition memory — is a reminder that the short-term effect of a single race on the brain looks nothing like the long-term pattern a cross-sectional study of veteran runners might show. Acute fatigue and chronic adaptation are different phenomena, and evidence for one says little about the other.

What's Left, Honestly

Put the pieces together and a specific, narrower picture emerges — more interesting, if less dramatic, than either "running makes you smarter" or "it's all just who chooses to run."

There is real, randomized evidence that moderate aerobic exercise produces modest, selective cognitive benefits in older adults, concentrated in executive function. The hippocampal-volume pathway sometimes cited as the mechanism behind that benefit is far less settled than a single landmark trial suggests — the best pooled randomized evidence to date finds no significant effect on hippocampal volume at all. There is no direct evidence that multi-marathon-level training produces cognitive benefits beyond what ordinary aerobic fitness already provides. There is credible evidence, from the same research group that produced the cognition study, that people drawn to multi-marathoning differ systematically in personality traits — some in ways that plausibly support sustained training, others that complicate a simple "resilience" narrative — and broader, if dated and geographically limited, evidence that socioeconomic context shapes who becomes a distance runner in the first place. And there is a genuinely distinct, underexplored possibility that the sport itself functions as a long-term exercise in attentional self-regulation, separate from its cardiovascular effects, though the existing evidence for that connects more directly to in-race performance than to durable cognitive change.

None of this settles the question the Trinity researchers themselves left open. It just clarifies what kind of study would actually settle it: something longitudinal, tracking the same people over years as their running histories accumulate, ideally with enough baseline data to separate who they were before the miles from who the miles made them.

Until then, the honest answer to "did the miles shape the mind, or did the mind shape the miles" is that the current evidence can't cleanly separate the two — and that the attempt to separate them is more interesting than either answer would have been on its own. Running may attract a certain kind of mind. Sustained exercise may alter certain brain processes. And years of repeatedly managing effort, pace, and the urge to stop may reinforce habits of attention that matter far beyond the road. All three could be true at once, in proportions nobody has yet measured.

Scientific Sources

• Lundy, L., Reilly, R. B., Fleming, N., Lopez Valdes, A., & O'Keeffe, C. (2026). "Mind over miles: cognitive functioning in multi-marathon runners." International Journal of Sport and Exercise Psychology. https://doi.org/10.1080/1612197X.2026.2716006

• Lundy, L., & Reilly, R. B. (2024). "Demographics, culture and participatory nature of multi-marathoning." PLOS ONE, 19(5), e0302602. https://doi.org/10.1371/journal.pone.0302602

• Lundy, L., Reilly, R. B., Fleming, N., & Wilczyńska, D. (2025). "Unveiling the psychological traits of multi-marathoners: Insights from TIPI personality trait analysis." PLOS ONE, 20(4), e0311647. https://doi.org/10.1371/journal.pone.0311647

• 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. https://doi.org/10.1111/1467-9280.t01-1-01430

• Northey, J. M., Cherbuin, N., Pumpa, K. L., Smee, D. J., & Rattray, B. (2018). "Exercise interventions for cognitive function in adults older than 50: a systematic review with meta-analysis." British Journal of Sports Medicine, 52(3), 154–160. https://doi.org/10.1136/bjsports-2016-096587

• Erickson, K. I., et al. (2011). "Exercise training increases size of hippocampus and improves memory." PNAS, 108(7), 3017–3022. https://doi.org/10.1073/pnas.1015950108

• Balbim, G. M., Boa Sorte Silva, N. C., ten Brinke, L., Falck, R. S., Hortobágyi, T., Granacher, U., Erickson, K. I., Hernández-Gamboa, R., & Liu-Ambrose, T. (2024). "Aerobic exercise training effects on hippocampal volume in healthy older individuals: a meta-analysis of randomized controlled trials." GeroScience, 46(2), 2755–2764. https://doi.org/10.1007/s11357-023-00971-7

• Akbaş, A., Żebrowska, A., Malchrowicz-Mośko, E., Stempień, J., Bezuglov, E., Górka-Chowaniec, A., & Waśkiewicz, Z. (2025). "Socioeconomic status and motivation in endurance sports: insights from long-distance running." Frontiers in Psychology. https://doi.org/10.3389/fpsyg.2025.1527661

• Brick, N. E., MacIntyre, T. E., & Campbell, M. J. (2016). "Thinking and Action: A Cognitive Perspective on Self-Regulation during Endurance Performance." Frontiers in Physiology, 7:159. https://doi.org/10.3389/fphys.2016.00159

• Perrotta, A. S., Jeklin, A. T., Bredin, S. S. D., Shellington, E. M., Kaufman, K. L., de Faye, A., Miles, R. M., & Warburton, D. E. R. (2022). "Effect of an Ultra-Endurance Event on Cardiovascular Function and Cognitive Performance in Marathon Runners." Frontiers in Physiology, 13:838704.

• Cona, G., Cavazzana, A., Paoli, A., Marcolin, G., Grainer, A., & Bisiacchi, P. S. (2015). "It's a Matter of Mind! Cognitive Functioning Predicts the Athletic Performance in Ultra-Marathon Runners." PLOS ONE, 10(7), e0132943.