The munchies are one of cannabis's most recognizable effects. A meal that seemed satisfying an hour ago suddenly isn't enough. Ordinary snacks become appealing, and sweet foods can seem especially tempting. For decades, researchers have investigated why cannabis changes the experience of eating. The answer is not one mechanism but several working together. THC activates cannabinoid receptors involved in hunger, smell and reward, while the body's own cannabinoids can influence sweet-taste responses. Yet one of the most carefully controlled human studies found that marijuana did not actually change measured taste intensity or pleasantness. The science raises a surprising possibility: cannabis may make food more desirable without necessarily changing what the tongue tastes.
Scientists Gave People Marijuana and Watched What They Ate
Some of the clearest evidence for the munchies comes from controlled human laboratory experiments conducted in the 1980s.
In 1986, researchers led by Richard Foltin studied nine men living in a residential laboratory for periods of up to 25 days.
Participants smoked marijuana containing 1.84% THC or placebo.
A single active cigarette before the private work period did not increase food intake.
But two or three active cigarettes during the social period increased daily calorie consumption.
The increase came from additional between-meal snacks rather than larger regular meals.
A follow-up study published in 1988 examined six men living in a residential laboratory for 13 days.
With marijuana containing 2.3% THC, total daily calorie intake increased by 40%.
Once again, regular meal intake remained unchanged.
Participants instead ate on more occasions between meals.
And when researchers examined the types of snacks consumed, the largest increase came from sweet solid foods, particularly candy bars and cookies.
Sweet drinks and savory snacks such as chips did not account for the same increase.
These experiments established an important distinction: cannabis did not simply make participants consume larger meals.
It changed how often they wanted to eat, especially between meals.
The Brain's Fullness Neurons Started Encouraging More Eating
Why would someone want another snack after already eating enough?
A 2015 study published in Nature uncovered an unexpected mechanism in mice.
Researchers led by Marco Koch, working in the laboratory of Tamas Horvath at Yale, investigated how cannabinoids affect neurons in the hypothalamus, a brain region involved in regulating hunger and fullness.
Their focus was a group of cells called pro-opiomelanocortin neurons, or POMC neurons.
Under ordinary conditions, these neurons help signal satiety.
They release a molecule called alpha-melanocyte-stimulating hormone, or α-MSH, which acts as a signal to stop eating.
Scientists might therefore expect cannabinoids to make these neurons less active when stimulating appetite.
Instead, the opposite happened.
When researchers activated cannabinoid type 1 receptors, known as CB1 receptors, the POMC neurons became more active.
But their output changed.
CB1 activation selectively increased the release of beta-endorphin, an opioid peptide involved in promoting feeding, without increasing release of the satiety-related peptide α-MSH.
The same cells normally associated with telling the body to stop eating were now helping encourage more food consumption.
The researchers traced the switch to changes involving mitochondria and a protein called UCP2.
When they administered naloxone, a drug that blocks opioid receptors, the cannabinoid-induced feeding response was blocked.
The mice ate despite already being sated.
This was a striking demonstration that cannabinoids could alter how the brain's existing satiety circuitry functions.
There is an important limitation: the experiment involved mice and a cannabinoid CB1 agonist, not people smoking cannabis.
The findings reveal a biological mechanism that may help explain cannabinoid-driven feeding, but they do not establish that precisely the same neuronal switch occurs in humans consuming THC.
THC Made Mice More Interested in Smells, Then They Ate More
Hunger is only part of the experience.
Food can also become more appealing because of how it smells.
In 2014, researchers led by Edgar Soria-Gómez published a study in Nature Neuroscience examining how the endocannabinoid system influences food intake through olfactory processing.
The experiments focused on CB1 receptors located on cortical feedback neurons projecting into the main olfactory bulb, the brain structure involved in processing odors.
When cannabinoids activated these receptors, they reduced excitatory input onto inhibitory granule cells within the olfactory bulb.
That change improved odor detection.
In experiments involving fasted mice, THC increased odor investigation and food intake through a CB1-sensitive olfactory circuit.
The researchers also found that mice lacking CB1 receptors in this particular circuit no longer showed the same effect.
That helped connect the behavioral changes to a specific cannabinoid-sensitive neural pathway.
The significance becomes clearer when considering how humans experience food.
Much of what people describe as flavor actually comes from smell.
The tongue detects basic tastes, but the aroma of food contributes substantially to the overall eating experience.
A sharper sense of smell could therefore make food seem more flavorful without changing the tongue's response to sweet, salty, sour or bitter stimuli.
However, the direct experimental evidence here comes from mice.
The study does not establish that THC produces an equivalent increase in human odor sensitivity.
The Tongue Has Its Own Cannabinoid Switch for Sweetness
Another experiment suggests that cannabinoids can influence taste at an even earlier stage, directly on the tongue.
In 2010, researchers led by Ryusuke Yoshida published a study in Proceedings of the National Academy of Sciences examining how endocannabinoids affect taste responses in mice.
The researchers investigated two cannabinoids naturally produced by the body: anandamide and 2-arachidonoylglycerol, commonly called 2-AG.
Both increased gustatory nerve responses to sweet substances.
The effect was dose-dependent, meaning stronger cannabinoid stimulation produced greater enhancement of the sweet response.
But the change was remarkably selective.
The researchers found no corresponding enhancement of salty, sour, bitter or umami responses.
The mechanism involved CB1 receptors located on type II taste cells that also express T1r3, a component of the sweet-taste receptor.
In mice genetically lacking CB1 receptors, the sweet-enhancing effect disappeared at the cellular, nerve and behavioral levels.
The researchers also identified an opposing signal.
Leptin, a hormone involved in regulating energy balance, suppressed sweet responses in those same cells.
The findings suggest that the tongue contains a system capable of adjusting sweet sensitivity in response to signals related to energy regulation.
Endocannabinoids increased the sweet response, while leptin reduced it.
This creates an intriguing parallel with the human laboratory studies from the 1980s, in which marijuana increased snacking particularly on sweet solid foods.
But the two findings should not be treated as a proven causal chain.
The taste-cell experiments used the body's own endocannabinoids in mice.
They did not demonstrate that THC enhances sweet-taste receptor activity in humans.
No human study in the supplied research establishes that THC directly makes human taste buds more sensitive to sweetness.
Yoshida et al., Proceedings of the National Academy of Sciences, 2010 — the study's abstract, via PubMed (PMID 20080779).
The Human Taste Test That Found No Difference
If cannabinoids can enhance sweet-taste responses in mice, does marijuana actually make food taste stronger or more pleasant to humans?
A carefully controlled human experiment produced a surprisingly different result.
In 1994, Richard Mattes, Leslie Shaw and Kenneth Engelman published a series of controlled experiments in Chemical Senses.
The principal double-blind, placebo-controlled trial involved 57 adults described as light marijuana users.
Additional experiments examined different routes of cannabinoid administration and repeated dosing.
Across the tested conditions, researchers found no drug effect on measured taste intensity or pleasantness.
Participants did not show a measurable increase in how intensely they perceived the tested tastes.
They also did not rate those tastes as more pleasant because of marijuana.
One physiological effect did emerge clearly.
Salivary flow decreased as blood cannabinoid levels increased.
In other words, the dry-mouth effect associated with marijuana was measurable, even though the researchers did not detect changes in taste intensity or pleasantness.
The authors suggested that reports of food tasting better during marijuana intoxication might reflect changes in memory and cognition rather than altered gustatory function.
This finding is an important counterweight to the animal research.
It does not invalidate the mouse experiments, but it demonstrates why results involving taste cells and cannabinoid signaling cannot automatically be extended to the human experience.
The feeling that food tastes better may emerge from several processes operating together, including hunger, smell, reward and how the eating experience is remembered.
People Still Recognized Fullness, but Cannabis Made Them Want to Eat
A companion study published by Mattes and colleagues in 1994 examined cannabinoid-related appetite stimulation.
The researchers found that appetite stimulation was real, but highly variable between individuals.
The researchers found comparable total daily energy intake under fed and fasted conditions, suggesting that the tested cannabinoid regimens had not impaired satiety mechanisms.
Appetite effects nevertheless varied substantially between participants and dosing conditions.
That distinction helps explain why the munchies are more complicated than ordinary hunger.
The cannabinoid system influences both the motivation to eat and the reward value associated with food.
Someone may recognize that they have already eaten enough while still finding another snack appealing.
The research therefore points toward overlapping processes rather than a single hunger switch.
Cannabis can affect the drive to seek food without necessarily eliminating the body's ability to recognize satiety.
Scientists Blocked CB1 and Turned the Munchies Into a Weight-Loss Strategy
The importance of CB1 receptors became especially clear when pharmaceutical researchers tried the opposite approach.
Instead of activating CB1 signaling, they developed a drug designed to block it.
The medication was rimonabant, marketed as Acomplia.
Rimonabant acted as a CB1 receptor blocker and inverse agonist.
The reasoning was straightforward: if cannabinoid signaling can stimulate appetite, reducing that signaling might help people lose weight.
Large clinical trials supported its weight-loss effects.
The Rimonabant in Obesity program, known as RIO, involved approximately 6,600 patients.
Major studies included RIO-Europe, published in The Lancet in 2005, and RIO-North America, published in JAMA in 2006.
The trials demonstrated significant weight loss.
In June 2006, European regulators authorized rimonabant for weight management.
But the drug's history soon revealed the risks of interfering with a receptor system involved in more than appetite.
The FDA never approved it in the United States, and an FDA advisory committee recommended against approval in June 2007 because of neurological and psychiatric safety concerns.
Serious psychiatric safety concerns emerged, including depression, anxiety and suicidal ideation.
In October 2008, the European Medicines Agency recommended suspending the drug.
Its marketing was suspended across participating European Union countries in November, and the European Commission formally withdrew authorization in January 2009.
Its history demonstrated two important facts.
First, CB1 signaling plays a powerful role in regulating appetite and body weight.
Second, that receptor system is also deeply connected to mood and psychological functioning.
Blocking the receptor could reduce weight, but the broader effects made the treatment unacceptable.
The same biological system associated with the munchies was powerful enough to become a target for obesity medication. But rimonabant's psychiatric risks made that particular treatment unacceptable.
The Munchies Became an FDA-Approved Medical Treatment
The appetite-stimulating effects of THC are not limited to recreational cannabis use.
They have also been used in medicine.
Dronabinol, a synthetic form of THC, was first approved by the FDA in 1985 for chemotherapy-related nausea and vomiting.
In December 1992, its approval expanded to anorexia associated with weight loss in patients with AIDS.
Early clinical research explored its effects on appetite and nutritional status.
Plasse and colleagues reported appetite improvement in preliminary studies, while a small 1993 trial by Struwe and colleagues, using 5 mg twice daily, found possible nutritional benefits but had substantial participant dropout and did not establish a statistically significant improvement in appetite scores.
The same general effect that can lead someone to reach for another snack can have medical value when loss of appetite is a serious problem.
The goal in that setting is not to make food more entertaining.
It is to help stimulate appetite in people experiencing clinically significant appetite loss.
Dronabinol's approval demonstrates that cannabinoid-related appetite stimulation has an established therapeutic application, even while the precise sensory experience of the munchies remains scientifically complex.
So Why Does Food Seem Better After Cannabis?
The research points to several mechanisms that may contribute to the munchies.
Human laboratory experiments demonstrate that marijuana can increase food consumption, especially through additional snacking and greater intake of sweet solid foods.
Mouse experiments show that cannabinoid signaling can change the output of hypothalamic satiety neurons, improve odor detection and selectively enhance sweet-taste responses.
CB1 signaling also participates in the reward processes that influence how desirable food feels.
But those findings do not establish that THC directly increases human taste-bud sensitivity.
The 1994 controlled human taste experiment found no measurable change in taste intensity or pleasantness ratings.
That leaves an important distinction between food tasting different and food feeling more desirable.
Cannabis can increase appetite and food intake in humans, while mouse research suggests that cannabinoid signaling can also enhance odor detection. Whether THC produces comparable sensory changes in humans remains uncertain.
The strongest conclusion is not that THC simply makes everything taste better.
It is that cannabinoid signaling influences multiple parts of the eating experience, from the brain's appetite circuitry to sensory processing and food reward.
The munchies are not just about taste.
They are about how the brain and body decide that food is worth eating.
Read next
Sources
- Koch, Marco, et al. "Hypothalamic POMC Neurons Promote Cannabinoid-Induced Feeding." Nature, vol. 519, no. 7541, 2015, pp. 45–50. https://doi.org/10.1038/nature14260 Mechanism of cannabinoid-induced feeding. Mouse experiments demonstrating that CB1 activation increases POMC neuron activity while shifting output toward appetite-promoting beta-endorphin. Identifies involvement of mitochondrial UCP2 and shows that naloxone blocks cannabinoid-induced feeding.
- Soria-Gómez, Edgar, et al. "The Endocannabinoid System Controls Food Intake via Olfactory Processes." Nature Neuroscience, vol. 17, no. 3, 2014, pp. 407–415. https://doi.org/10.1038/nn.3647 Smell and food intake. Mouse research identifying CB1-sensitive feedback circuits in the olfactory bulb. Shows that cannabinoid signaling enhances odor detection and feeding, including increased odor investigation and food intake following THC exposure.
- Yoshida, Ryusuke, et al. "Endocannabinoids Selectively Enhance Sweet Taste." Proceedings of the National Academy of Sciences, vol. 107, no. 2, 2010, pp. 935–939. https://doi.org/10.1073/pnas.0912048107 Cannabinoid signaling on the tongue. Demonstrates in mice that anandamide and 2-AG selectively enhance sweet-taste responses through CB1 receptors on sweet-sensitive taste cells. Does not establish the same effect in humans.
- Foltin, Richard W., Joseph V. Brady, and Marian W. Fischman. "Behavioral Analysis of Marijuana Effects on Food Intake in Humans." Pharmacology Biochemistry and Behavior, vol. 25, no. 3, 1986, pp. 577–582. PMID: 3774823. Human evidence of increased snacking. Residential laboratory study of nine men: a single active cigarette before the private work period had no effect, but two or three during the social period increased daily calorie intake through additional between-meal snacking rather than larger meals.
- Foltin, Richard W., Marian W. Fischman, and Mary F. Byrne. "Effects of Smoked Marijuana on Food Intake and Body Weight of Humans Living in a Residential Laboratory." Appetite, vol. 11, 1988, pp. 1–14. Sweet-snack preference in humans. Residential laboratory study of six men showing a 40% increase in daily caloric intake, with the principal increase occurring through more snacking occasions and greater consumption of sweet solid foods.
- Mattes, Richard D., Leslie M. Shaw, and Kenneth Engelman. "Effects of Cannabinoids (Marijuana) on Taste Intensity and Hedonic Ratings and Salivary Flow of Adults." Chemical Senses, vol. 19, no. 2, 1994, pp. 125–140. https://doi.org/10.1093/chemse/19.2.125 Controlled human taste experiment. Double-blind, placebo-controlled study of 57 adults finding no cannabinoid effect on taste intensity or pleasantness ratings for sweet, sour, salty or bitter stimuli. Documents reduced salivary flow.
- Mattes, Richard D., et al. "Cannabinoids and Appetite Stimulation." Pharmacology Biochemistry and Behavior, vol. 49, no. 1, 1994, pp. 187–195. PMID: 7816872. Appetite versus satiety. Companion human study showing variable appetite stimulation while finding that satiety mechanisms were not impaired. Supports the distinction between recognizing fullness and continuing to eat.
- Van Gaal, Luc F., et al. "Effects of the Cannabinoid-1 Receptor Blocker Rimonabant on Weight Reduction and Cardiovascular Risk Factors in Overweight Patients: 1-Year Experience from the RIO-Europe Study." The Lancet, vol. 365, 2005, pp. 1389–1397. CB1 blockade and weight loss. Clinical trial demonstrating significant weight reduction with rimonabant, supporting the role of CB1 signaling in appetite and weight regulation.
- Pi-Sunyer, F. Xavier, et al. "Effect of Rimonabant, a Cannabinoid-1 Receptor Blocker, on Weight and Cardiometabolic Risk Factors in Overweight or Obese Patients: RIO-North America: A Randomized Controlled Trial." JAMA, vol. 295, 2006, pp. 761–775. Clinical evidence for CB1 inhibition. Randomized trial documenting rimonabant's weight-loss effects. Provides context for the drug's subsequent regulatory history and psychiatric safety concerns.
- Plasse, Terry F., et al. "Recent Clinical Experience with Dronabinol." Pharmacology Biochemistry and Behavior, vol. 40, no. 3, 1991, pp. 695–700. https://doi.org/10.1016/0091-3057(91)90385-F Medical appetite stimulation. Open pilot studies reported improved appetite, with weight gain in seven of ten patients with symptomatic HIV infection.
- Struwe, Melissa, et al. "Effect of Dronabinol on Nutritional Status in HIV Infection." Annals of Pharmacotherapy, vol. 27, nos. 7–8, 1993, pp. 827–831. https://doi.org/10.1177/106002809302700701 Therapeutic application of THC. Double-blind crossover trial of dronabinol 5 mg twice daily in 12 patients with HIV; only five completed the trial and the appetite-score improvement was a non-significant trend (p = 0.14), while percent body fat and symptom distress improved.
- U.S. Food and Drug Administration. Marinol (dronabinol) approval history. Regulatory history. Dronabinol was first approved in 1985 for chemotherapy-induced nausea and vomiting; in December 1992 the approval expanded to anorexia associated with weight loss in patients with AIDS.
0 comments