For decades, the euphoria and pain relief that follow sustained aerobic exercise have been dismissed as a simple endorphin rush. A new integrative framework now challenges that oversimplification, with implications for how clinicians and fitness practitioners think about exercise as a mood-regulating and analgesic intervention — and possibly why some people find endurance training genuinely addictive.
This narrative review synthesizes human and animal evidence to argue that the runner's high is not an either-or product of β-endorphins or endocannabinoids, but rather an orchestrated neuromodulatory response. The endocannabinoid (eCB) system — particularly circulating ligands such as anandamide and 2-arachidonoylglycerol — emerges as the integrative hub. Circulating eCBs consistently rise during moderate-intensity endurance exercise and map onto the four hallmark features of the runner's high: euphoria, anxiolysis, hypoalgesia, and a subjective sense of flow. The review assigns complementary but distinct roles to other players: β-endorphins primarily govern pain modulation and acute stress buffering; brain-derived neurotrophic factor (BDNF) underpins exercise-induced neuroplasticity; serotonin and dopamine regulate hedonic tone and motivational drive; noradrenaline and adrenaline support arousal and attentional focus; and leptin links whole-body metabolic status to movement motivation.
The significance of repositioning eCBs as the coordinator — rather than a secondary actor — is worth placing in broader context. Unlike β-endorphins, which are large peptides that cannot cross the blood-brain barrier efficiently, endocannabinoids are lipid-soluble and are synthesized on-demand in neurons, making them plausible real-time modulators of central affective states. This aligns with rodent studies showing that blocking cannabinoid CB1 receptors, not opioid receptors, attenuates exercise-induced anxiolysis. A key limitation is this review's narrative rather than systematic structure, meaning selection bias in the evidence base cannot be ruled out. Most human eCB data also rely on peripheral blood measures, which are imperfect proxies for central nervous system activity. Still, this integrative model meaningfully advances understanding of why moderate-intensity — not maximal — exercise most reliably produces positive affect, and it sets a clear agenda for mechanistic human trials.
