Mountains do something to your head.
How much of it shows in the data?
A first outing on a ridge usually ends with the same sentence: my head feels clear. I do not know a mountain runner who has not said it — nor one who could measure it. This piece is about that gap: what of the psychological benefits of running in nature actually shows up in the research, where popular headlines run further than the data allow, and where the reach of a watch ends.
I am writing this as an introduction for someone still considering a first mountain run. You will not find a list of ten reasons. You will find four things the literature supports, three that are weaker than commonly claimed — and one boundary I state plainly, because my whole product rests on measurement.
What holds up
Attention recovers faster than after an indoor break. A 2025 meta-analysis pooled 273 outcomes from 80 studies. Of eleven cognitive domains, only two showed a reliable benefit: working memory and attentional control. The effect grew up to roughly 30 minutes of exposure and was clearly stronger in people who entered the study already fatigued.
The same effort in green surroundings affects mood differently than indoors. A 2026 network meta-analysis compared 15 randomised trials, 809 participants in total. Outdoor exercise outperformed indoor exercise on a composite of depressive symptoms, anxiety and psychological distress (SMD −0.75; confidence interval −1.32 to −0.18). Notably, indoor exercise alone did not separate significantly from control — the authors read this as the context of activity mattering as much as the activity.
The dose has a threshold, and not an obvious one. A study of 19,806 people found that those spending 1 to 119 minutes a week in nature did not differ in self-reported health or wellbeing from those spending none at all. The difference appeared only from 120 minutes upward, rose to 200–300 minutes, then flattened. It made no difference whether those two hours came in one outing or five.
| What was studied | What came out | How much it weighs |
|---|---|---|
| Attention and working memory | Improvement after nature exposure; threshold about 30 min; stronger in the fatigued | 80 studies, 273 outcomes — but SMALL and highly heterogeneous effects |
| Green vs indoor exercise | Green advantage: SMD −0.75 (−1.32 to −0.18) | 15 randomised trials, 809 people — individual trials small |
| Forest bathing and cortisol | Cortisol and heart rate fall; heart rate −4.00 bpm | Certainty of evidence rated VERY LOW (GRADE) |
| Weekly dose | Threshold 120 min/week; plateau at 200–300 min | 19,806 people, but a CROSS-SECTIONAL study — association, not cause |
Cortisol was already falling before anyone walked in among the trees
This is the most interesting thing I found while preparing this piece — and the most inconvenient. In a meta-analysis of forest bathing studies, cortisol in the forest group was sometimes significantly lower before the intervention. Before participants entered the forest at all. The anticipation itself lowered the very number meant to prove the forest worked.
This does not invalidate the phenomenon. It means that in studies where nobody can be blinded — everyone knows whether they are standing in a forest or in a city centre — expectation is part of the result. That is why the most recent synthesis rates the certainty of this evidence as VERY LOW, even though the direction of effect is consistent and the p-values look strong.
A cautious 2019 review points the same way: after outdoor exercise people report better affect and more enjoyment, but biological markers, perceived exertion and intensity do not differ meaningfully from exercising without nature. The authors rate risk of bias in these trials as high. Two meta-analyses of the same field therefore return different numbers — and that is an honest picture of the state of knowledge, not a flaw in it.
If the number drops before the cause has acted, that cause is not what lowered it.
What the watch will not see
I build a tool that reads heart-rate variability, resting heart rate, sleep and training load every day. So I can be precise about where its reach ends: this data measures the cost of going to the mountains, not the reason for going. Four hours on a ridge will show up as load, slower recovery and the eccentric work of descending. The clear head will not appear in any of those numbers.
I hold to one rule here, the same one as in the engine: what I have not measured, I do not pronounce on. A system that invents a conclusion for a missing value is worse than one that says it does not know — because you will believe the first. That applies to this text too: I will not tell you the mountains cure anything. I will tell you which associations were observed, with what strength and what certainty — and leave the rest to you and your doctor.
What follows if you are only starting out
Four practical conclusions, all straight from the numbers above. First: short outings totalling under two hours a week may not be enough — the threshold in the data sits at 120 minutes, and how you accumulate them does not matter. Two easy one-hour outings do the same as one longer one. Second: the first half hour carries the largest share of the cognitive effect, so a short trip counts too.
Third, and this is the perverse part: the effect was stronger in fatigued people. It works best exactly when you least feel like going. Fourth: mountains load the body differently than flat ground. The same kilometres with 1500 metres of climbing is a different effort and a different recovery — and that part, unlike the rest of this text, is measurable.
If you want to see what that difference looks like on a specific course, I took apart Lemkowyna Ultra Trail 150 km — the profile, the cost of vertical, the places where the route breaks your pace. And if you care what I base my calculations on, I collected it on the scientific basis page.
Test it on your own data — 14 days free
What you feel on a ridge, I cannot measure. I can measure what it cost you and when you will be ready for more — from your own history, not from a template. 14 days free. Card at signup, nothing charged during the trial, cancel any time.
Sources
Bell et al. (2025), meta-analysis of nature exposure and attention — ScienceDirect · Stevenson et al. (2018), systematic review of attention restoration theory — Taylor & Francis · network meta-analysis of green vs indoor exercise (2026) — PMC · Lahart et al. (2019), the cautious review — PMC · Antonelli et al. (2019), cortisol and forest bathing — Springer · synthesis of cardiovascular and mental responses (2025) — PMC · White et al. (2019), 120 minutes a week, n = 19,806 — Scientific Reports.