For decades, “runner’s high” was almost synonymous with endorphins. The story is appealing: run long enough, release natural opioids, feel euphoric. There is genuine evidence that the opioid system responds to endurance exercise — but modern human experiments suggest that it is not the whole explanation.
A second system has moved to centre stage: endocannabinoids, signalling molecules made by the body that interact with cannabinoid receptors. The surprising part is that euphoria can still appear even when opioid receptors are experimentally blocked.
Quick answer
Are endorphins responsible for runner’s high?
They may participate in exercise-related reward, but current human evidence suggests the classic runner’s high does not depend on opioid signalling alone. Endocannabinoids are strong candidates, although the full mechanism is still not settled.
Key takeaways
- A classic human PET study found exercise-related changes in opioid receptor binding that correlated with euphoria.
- In a randomized double-blind study, 45 minutes of running increased euphoria and reduced anxiety even when opioid receptors were blocked with naltrexone.
- The same study found increases in the endocannabinoids anandamide and 2-AG.
- A systematic review found that 14 of 17 acute-exercise studies reported increased endocannabinoids.
- Runner’s high is a multi-system phenomenon, not a reliable sign that one neurotransmitter has been “maximized”.
Why the endorphin theory became famous
Endorphins are endogenous opioids involved in pain modulation and reward. In a small 2008 PET study, ten athletes were scanned at rest and after two hours of endurance running. Opioid receptor availability changed in several brain regions, and the changes were associated with reported euphoria.
That was important evidence that the brain’s opioid system responds during prolonged running. But showing that a system changes during an experience is not the same as proving that the system is required for the experience.
The experiment that challenged the simple story
In 2021, researchers used naltrexone, an opioid receptor antagonist, in a double-blind randomized experiment. Sixty-three participants ran for 45 minutes or walked. Running increased euphoria and reduced anxiety — and blocking opioid receptors did not prevent those changes.
Meanwhile, circulating anandamide and 2-AG increased after running. That made the endocannabinoid system a particularly strong candidate for core features of runner’s high in humans.
What the endocannabinoid evidence actually supports
A systematic review of human exercise studies found increased endocannabinoids after acute exercise in 14 of 17 studies. That is consistent evidence that the system responds to exercise, but it still does not prove a single molecule is solely responsible for the subjective high.
Mood after exercise is likely influenced by intensity, duration, expectation, fitness, social context, stress state and multiple signalling pathways. Biology rarely gives us one clean “feel-good chemical”.
The twin twist: identical athletes can still respond differently
A small study of nine pairs of trained male identical twins measured several hormonal responses to a 30-minute treadmill session. Some responses resembled each other within twin pairs, while the increase in absolute beta-endorphin concentration did not show significant within-pair resemblance.
The study was far too small to settle the genetics of endorphins. But it illustrates a broader point: even with nearly identical DNA, acute exercise responses can diverge. That is exactly why individual tracking and twin designs remain so interesting.
TwinPare perspective: measure the experience, not just the molecule
If two people run the same route, one may feel euphoric while the other feels flat or exhausted. The useful question is not “which chemical did I boost?”, but what combination of training load, recovery, sleep, stress and individual biology shaped the response.
TwinPare Health & Fitness is being developed around that broader context: combine signals over time and help users understand patterns rather than promise a single biomarker explanation.
Source notes
The sources have been verified and editorially reviewed for this article. The limitations below show which level of conclusion the sources support.
- [boecker-2008] The runner’s high: opioidergic mechanisms in the human brain Henning Boecker et al.. Cerebral Cortex, 2008. Evidence type: Human PET exercise study Limitation: Only ten athletes were studied; the study supports opioid-system involvement but does not prove opioid signalling is necessary for runner’s high. PubMed
- [siebers-2021] Exercise-induced euphoria and anxiolysis do not depend on endogenous opioids in humans Michael Siebers et al.. Psychoneuroendocrinology, 2021. Evidence type: Double-blind randomized opioid-blockade experiment Limitation: The study supports opioid-independent euphoria and anxiolysis after running in this sample; it does not establish one exclusive mechanism. PubMed
- [siebers-review-2023] Do Endocannabinoids Cause the Runner’s High? Evidence and Open Questions Michael Siebers et al.. The Neuroscientist, 2023. Evidence type: Systematic review Limitation: The review found consistent acute endocannabinoid responses but concluded that definitive causal proof in humans remains challenging. PubMed
- [di-luigi-2003] Heredity and pituitary response to exercise-related stress in trained men Luigi Di Luigi et al.. International Journal of Sports Medicine, 2003. Evidence type: Monozygotic co-twin exercise study Limitation: Only nine monozygotic male twin pairs were included; results are exploratory and do not quantify endorphin heritability. PubMed
Editorial source review
This section shows how the article's key factual claims are linked to the source.
Phrasings that require caution
- Do not claim that endorphins are irrelevant to exercise-related reward; the evidence shows the runner’s high can occur despite opioid blockade.
- Circulating endocannabinoid levels are not a direct readout of every brain mechanism.
- Runner’s high is not guaranteed and may differ substantially between individuals and sessions.
| ID | Claim | Source support | Caution |
|---|