In lab treadmill tests, high-intensity running was linked to higher blood endocannabinoids in humans and dogs, but not in ferrets and not during easy walking.
Readers curious about the biology behind the runner's high, comparative exercise physiology across species, and how the endocannabinoid system responds to intensity.
2 of 3 speciesshowed higher endocannabinoids after high-intensity running. Ferrets did not.
The Backstory
For decades, the story was simple: you run hard, your body releases endorphins, you feel euphoric. The "runner's high" was an endorphin phenomenon. Case closed.
There was just one problem. Endorphins are large peptide molecules. They're too big to cross the blood-brain barrier. If running releases endorphins into your bloodstream — which it does — those endorphins can't reach the brain regions where euphoria is generated. The timeline didn't work either: endorphin levels peak after exercise stops, but the runner's high typically arrives during the effort.
Something else was responsible. In 2012, an anthropologist at the University of Arizona figured out what.
The Right Question
David Raichlen wasn't a neuroscientist. He was a biological anthropologist studying human evolution. His research question wasn't "what causes euphoria during exercise?" — it was "why do humans run at all?"
From an evolutionary standpoint, running is expensive. It burns enormous calories, risks injury, and stresses joints. Most mammals avoid sustained running unless they're chasing prey or fleeing predators. But humans voluntarily run — for exercise, for pleasure, for no immediate survival reason. So do dogs. Horses do too. These species are "cursorial" — anatomically adapted for endurance running.
Raichlen's hypothesis: cursorial mammals evolved a neurobiological reward system that makes running feel good, encouraging the sustained locomotion that their bodies were built for. And the reward system wasn't endorphins — it was endocannabinoids.
The Experiment
How They Did It
Three Species, Two Intensities, One Question
Species selection
Three mammal species: humans (cursorial), dogs (cursorial), and ferrets (non-cursorial). If the endocannabinoid reward is an adaptation for endurance running, it should appear in runners but not in non-runners.
Ferrets are agile and athletic but are not built for sustained running — they're ambush predators
Exercise protocol
Each species completed treadmill sessions at two intensities: low-intensity walking and high-intensity running. Blood samples were drawn before and after each session.
If endocannabinoids drive a running reward, they should increase with running intensity
Blood sampling
Circulating endocannabinoid levels (anandamide and 2-AG) were measured from blood drawn before and immediately after exercise.
Blood levels are a proxy — brain endocannabinoid levels can't be measured in living humans
Comparison
Pre-post changes compared across species and intensities. The prediction: high-intensity running should increase endocannabinoids in cursorial species (humans, dogs) but not in non-cursorial species (ferrets).
A clean experimental design that tests an evolutionary hypothesis
Raichlen et al. (2012), J Exp Biol 215:1331-1336
The Results
Cursorial Mammals (Humans & Dogs)
- Blood endocannabinoid levels rose significantly after high-intensity running
- No significant increase after low-intensity walking
- The response tracks with exercise intensity, not exercise itself
- Both species showed the same pattern independently
Endocannabinoid reward is linked to the kind of exercise these species evolved to do
Non-Cursorial Mammal (Ferrets)
- No significant endocannabinoid increase at any exercise intensity
- Walking: no change. Running: no change.
- Ferrets are active animals — they just aren't built for endurance running
- The absence of response supports the evolutionary hypothesis
No endocannabinoid reward for exercise in a species that didn't evolve to run
Raichlen et al. (2012), J Exp Biol 215:1331-1336
The pattern was exactly what the evolutionary hypothesis predicted. Humans and dogs — species built for endurance — got an endocannabinoid surge from running. Ferrets — a species built for short, explosive bursts — didn't. And crucially, the reward was intensity-dependent: walking didn't trigger it. You had to run.
Why This Matters for Cannabis Users
The Endorphin Myth
Myth vs. Reality
The runner's high is caused by endorphins.
Endorphins are released during exercise, but they're too large to cross the blood-brain barrier — they can't reach the brain circuits where euphoria is generated. Endocannabinoids (especially anandamide) are small lipid molecules that cross the blood-brain barrier easily and activate reward-related CB1 receptors in the brain. The runner's high is primarily an endocannabinoid phenomenon, not an endorphin one.
The Evidence
Raichlen et al. (2012) showed exercise-induced endocannabinoid increases in cursorial mammals. A 2015 mouse study (Fuss et al., PNAS) directly demonstrated that blocking cannabinoid receptors eliminated the runner's high in mice, while blocking opioid receptors did not. The endocannabinoid hypothesis has now largely replaced the endorphin explanation in the scientific literature.
Raichlen et al. (2012), J Exp Biol; Fuss et al. (2015), PNAS 112:13105-13108
The endorphin story persisted for decades because it was elegant and intuitive — "natural opioids make you feel good." But the pharmacology never fully supported it. The 2015 Fuss et al. mouse study provided the definitive test: mice with blocked cannabinoid receptors lost the runner's high; mice with blocked opioid receptors kept it. The endocannabinoid system, not the opioid system, is the primary driver.
An Evolutionary Perspective
Raichlen's genius was in framing the question evolutionarily. He wasn't asking "what neurotransmitter does exercise release?" — a pharmacology question. He was asking "why do certain species choose to run?" — an evolutionary question. The endocannabinoid reward provides an answer: cursorial mammals evolved a neurobiological incentive to engage in the locomotor behavior their bodies were designed for.
Humans are running animals. Our anatomy — long legs, short toes, nuchal ligament, sweat cooling, Achilles tendons — is built for endurance running. But anatomy alone doesn't guarantee behavior. You need motivation. The endocannabinoid reward system provides it: run hard enough, and your brain produces its own cannabis-like euphoria.
From this perspective, the modern epidemic of sedentary behavior represents an evolutionary mismatch. We have the hardware for running and the neurochemical reward system to encourage it — but we've built environments where running is optional. The endocannabinoid system that evolved to reward locomotion now goes understimulated.
Limitations
This study was small, measured only blood biomarkers (not brain activity), and didn't collect mood or euphoria ratings. It showed that endocannabinoids increase with running intensity in cursorial mammals — it didn't prove that those endocannabinoids cause the subjective experience of the runner's high. The 2015 Fuss et al. mouse study later provided that causal evidence.
The blood-brain inference — that rising blood endocannabinoids indicate rising brain endocannabinoids — is reasonable but not proven for exercise specifically. Anandamide does cross the blood-brain barrier, but blood levels and brain levels don't always track perfectly.
Related Research
The Endocannabinoid System in Motion
This study connected exercise physiology to endocannabinoid science — bridging two fields that had been studied separately.
Isolation and structure of a brain constituent that binds to the cannabinoid receptor
Devane, Hanus & Mechoulam (1992)
The discovery of anandamide — the endocannabinoid that rises during exercise and creates the runner's high.
The endocannabinoid system and the brain
Mechoulam & Parker (2013)
The masterwork review that explains how endocannabinoids regulate reward, anxiety, and mood — the functions involved in the runner's high.
Evolutionary origins of the endocannabinoid system
McPartland et al. (2006)
The ECS is 500+ million years old — it was regulating locomotor behavior in organisms that preceded vertebrates.
Clinical endocannabinoid deficiency (CECD)
Russo (2004)
If some people have chronically low endocannabinoid tone, exercise may be a natural way to restore it — a non-pharmacological intervention for CECD.
Is the runner's high caused by endorphins or endocannabinoids?
Primarily endocannabinoids. While exercise does release endorphins, these molecules are too large to cross the blood-brain barrier and reach the brain circuits where euphoria is generated. Endocannabinoids (especially anandamide) are small enough to cross the barrier and activate CB1 receptors in reward-related brain regions. A 2015 mouse study confirmed this: blocking cannabinoid receptors eliminated the runner's high, while blocking opioid receptors did not.
Can exercise help with cannabis withdrawal?
Yes. Exercise directly stimulates the same endocannabinoid system that cannabis activates. When chronic THC use has downregulated CB1 receptors, exercise provides a natural endocannabinoid signal that can partially compensate — reducing anxiety, improving mood, and restoring sleep. Moderate-to-high intensity aerobic exercise (running, cycling, swimming) appears most effective.
Why don't ferrets get a runner's high?
Ferrets are non-cursorial mammals — they're built for short bursts of speed, not endurance running. The study found no significant endocannabinoid increase in ferrets at any exercise intensity, supporting the hypothesis that the endocannabinoid reward evolved specifically in species adapted for sustained running. The reward motivates the behavior their bodies were designed for.
How much exercise triggers the endocannabinoid response?
The study found the response required high-intensity running, not low-intensity walking. Most evidence suggests moderate-to-vigorous aerobic exercise sustained for 20-30+ minutes is needed. The threshold likely varies by individual fitness level — the key is reaching an intensity that challenges your cardiovascular system.
What the researchers found
Circulating endocannabinoid levels rose after high-intensity endurance running in humans and in dogs. The same individuals did not show a significant rise during low-intensity walking.
Ferrets, a non-cursorial species, showed no significant endocannabinoid change at either exercise intensity. The pattern supports the idea that an endocannabinoid response tracks with endurance-style locomotion in species adapted for running. The study did not measure mood or euphoria, so it cannot show that endocannabinoids caused a subjective runner's high.
Why it matters
For decades, the runner's high was attributed mainly to endorphins. By 2012, evidence had begun pointing to the endocannabinoid system. Showing that high-intensity running in humans coincides with a blood endocannabinoid surge, and that a similar surge appears in a cursorial mammal but not a non-cursorial one, adds a comparative biology angle. It suggests a physiological signature that may help explain why endurance exercise persists in some species, while avoiding claims about cause.
The numbers in context
- Species included: 3 total. Humans, dogs (cursorial), ferrets (non-cursorial)
- High-intensity running: increased circulating endocannabinoids in 2 of 3 species (humans and dogs)
- Low-intensity walking: no significant increase in any of the 3 species
- Non-cursorial species result: 0 of 1 (ferrets) showed an endocannabinoid increase at any intensity
How the study worked
Researchers measured blood endocannabinoids before and after treadmill exercise in three species: humans, dogs (cursorial), and ferrets (non-cursorial). Each species completed low-intensity walking and higher-intensity running sessions, and pre-post blood levels were compared within species. The abstract does not report sample sizes or which endocannabinoids were assayed. This was a short, acute, within-session comparison without randomization or blinding, and no subjective mood or pain ratings were collected.
Who was studied
Humans, dogs (cursorial mammals), and ferrets (non-cursorial mammals); Country not specified.
What this study cannot tell us
Sample sizes, participant characteristics, and exact analytes are not reported in the abstract. Only circulating endocannabinoids were measured, not brain levels or receptor activity. No mood, pain, or reward ratings were collected, so any link to the runner's high is inferred. Acute, single-session design without randomization or blinding. Treadmill running and blood draws can induce stress that may alter endocannabinoid levels. Species differed in familiarity with treadmills and in natural locomotor patterns, which could confound intensity matching across groups.
How to read the evidence
Rated preliminary: small, cross-species, acute pre-post measurements with no reported sample sizes, blood biomarkers only, and no behavioral outcomes.
When this study was published
Published in 2012, early in the modern wave of research linking exercise to endocannabinoid signaling. Subsequent studies have expanded measurements and added mood and pain outcomes, but details here remain limited to blood biomarkers.
The bigger picture
The endocannabinoid system is engaged by more than external cannabinoids. Exercise can shift these signaling molecules in circulation, which may relate to sensations that many runners describe. This study links that shift to exercise intensity and to a species' locomotor ecology. It does not demonstrate changes inside the brain or establish that endocannabinoids drive mood or reward in this context. Blood levels are only a proxy and can be influenced by stress, temperature, or handling. Even in humans, the study measured biochemistry, not behavior.
Replication
Not stated in abstract.
Funding
Not reported in abstract.
Conflicts of interest
Not reported in abstract.
Questions still open
- Are endocannabinoid increases driven more by intensity, duration, or both in humans when measured alongside mood or pain metrics?
- Do specific endocannabinoids, such as anandamide versus 2-AG, show different patterns across exercise types like running, cycling, or swimming?
- Would other cursorial mammals, such as horses or antelope, show a similar biochemical response under controlled conditions?
- How closely do blood endocannabinoid changes track with central nervous system signaling and actual subjective experience?
- Does training status or fitness level shift the intensity threshold needed to trigger an endocannabinoid response?
Common questions
Did the study prove that endocannabinoids cause the runner's high?
Do dogs get a runner's high?
Why include ferrets?
Read the original research
Wired to run: exercise-induced endocannabinoid signaling in humans and cursorial mammals with implications for the 'runner's high'
Journal of Experimental Biology, 215(8), 1331-1336
Journal of Experimental Biology is a reputable journal that publishes research on the form and function of living organisms.
Citation
Raichlen, David A.; Foster, Adam D.; Gerdeman, Gregory L.; Seillier, Alexandre; Giuffrida, Andrea. (2012). Wired to run: exercise-induced endocannabinoid signaling in humans and cursorial mammals with implications for the 'runner's high'. Journal of Experimental Biology, 215(8), 1331-1336. https://doi.org/10.1242/jeb.063677
Explore the wider topic
- Athletes and Cannabis: Performance, Recovery, and Quitting
- Healthcare Workers and Cannabis: Nurses, Doctors, and Hidden Use
- Quitting Weed for Your Career: How Cannabis Affects Work and What Changes When You Stop
- Cannabis and College: Academic Impact and Quitting Weed as a Student
- THC and Running: The Runner's High Connection
- THC and Golf: The Silicon Valley Open Secret