By Eric Spitznagel

Like so many things on social media, patients invented the term before medicine did. They called it “food noise.”

They weren’t talking about ordinary hunger. They were describing something more intrusive and repetitive, a mental broadcast that kept reminding them what was in the pantry, what could be ordered, what could be eaten now and regretted later. Then some patients started taking GLP-1 medications for diabetes or weight loss, and the broadcast faded.

The observation was easy to dismiss at first as a side effect of feeling full. But the reports had a stranger quality. Patients didn’t always describe food as less pleasurable. They described it as less commanding. The cookie was still there, and the drink was still there, but the impulse no longer seemed to have a hand on the steering wheel.

That shift is the question now moving through GLP-1 research. These medications were intended as treatments for diabetes, then became accidental weight-loss drugs. Now they may be accidental bad-habit drugs — quieting compulsion and opening a much wider conversation about appetite, reward, craving, and what it means when a desire stops feeling like a choice.

Less Impulsive Violent Crime and…Nail-Biting?

Daniel Drucker, MD, endocrinologist at the University of Toronto, Toronto, Ontario, Canada, whose work helped define the biology of GLP-1, is quick to note that the weight-loss effects weren’t entirely a surprise. GLP-1s’ effect on insulin secretion was confirmed in experiments published in 1987, and by 1996, researchers had shown that GLP-1 administered into the brains of rats and mice inhibited food intake.

Human confirmation followed years later. The new surprise lies elsewhere. The drugs appear to be doing something that extends beyond blood sugar and beyond appetite, touching systems that govern reward, motivation, and compulsive consumption. That territory was not on the original map.

That possibility is still early, but it’s difficult to ignore.

In 2025, a small randomized clinical trial found that low-dose semaglutide reduced alcohol consumption in a laboratory self-administration procedure among adults with alcohol use disorder. Over 9 weeks, semaglutide reduced some drinking measures and significantly reduced weekly alcohol craving.

A larger trial published in The Lancet pushed the signal further. In that study, 108 treatment-seeking patients with alcohol use disorder and obesity received either semaglutide or a placebo for 26 weeks. Heavy drinking days decreased more in the semaglutide group. Those patients also showed larger decreases in total monthly alcohol consumption, self-reported craving, harmful alcohol use measures, and blood biomarkers related to liver damage.

A 2026 phase 2a trial of adults who smoked daily was mixed: Semaglutide didn’t significantly reduce cigarettes smoked per day, but it did reduce nicotine craving.

There’s more. While alcohol use and impulsive behavior are linked to violent crime, these associations were weaker in people on GLP-1s, according to a June 2026 study in Criminology. Results suggest that “GLP-1 Ras [receptor agonists] may attenuate widely documented behavioral risk mechanisms like impulsivity linked to aggression,” the researchers wrote.

Meanwhile, search for “GLP-1s and nail-biting,” and you’ll find multiple anecdotal threads of people on Ozempic with past nail-biting habits who stopped. “I have been unable to go a day without biting for 30+ years…but I just did it,” wrote one Reddit user. “All of a sudden, self-control exists for me.”

Wanting, Liking, and Altered Intensity

The emerging evidence doesn’t suggest that GLP-1 drugs switch off compulsive behavior. They may instead alter the intensity of an urge, the salience of a cue, or the amount of work the brain is willing to do for a reward.

Kent Berridge, PhD, neuroscientist at the University of Michigan in Ann Arbor, Michigan, whose work helped define the difference between “wanting” and “liking,” sees that distinction as central. Liking is the hedonic pleasure of a tasty food or other reward. Wanting is the motivation to pursue and consume it. Dopamine, once thought to mediate pleasure, turned out to be more involved in wanting. That distinction helps explain why people can keep pursuing rewards that no longer give much pleasure.

In some people, the mesolimbic dopamine system can become sensitized by addictive drugs, gambling, or other rewarding pursuits. Cues then trigger stronger wants, sometimes far stronger than the person’s original desire.

“The excessively strong ‘wanting’ can arguably rise to the level of a motivational compulsion,” Berridge said.

If GLP-1 medications reduce food noise, Berridge said, they may be reducing the dopamine wanting system’s response to food cues. In rodent studies, GLP-1 agonists injected directly into mesolimbic circuitry reduce food seeking and consumption. Some GLP-1 drugs may reach this circuitry more directly, while others act through hunger and satiety circuits in the brainstem and hypothalamus that indirectly inhibit wanting. Either way, the same basic idea emerges. The medication may quiet the brain’s insistence on a reward, not just the stomach’s demand for food.

For clinicians, that distinction matters. Hunger, pleasure, craving, and compulsion are braided together in daily life, but they aren’t the same thread. A patient who’s less hungry may eat less, but a patient whose food cues are less salient may experience a different kind of relief.

Food Noise, But Not Every Food

Ashley Gearhardt, PhD, psychologist at the University of Michigan who studies compulsive eating, said “food noise” is not a scientific diagnosis, but it captures a consistent patient experience. She also thinks the phrase may be misleading in one important way.

“Most people aren’t describing intrusive thoughts about apples, vegetables, beans, or salmon,” she said. “They’re describing intrusive thoughts about a relatively small subset of highly rewarding ultraprocessed foods.”

Gearhardt defines food addiction as a pattern marked by intense cravings, loss of control, repeated failed attempts to cut back, and continued consumption despite negative consequences. She now describes it more specifically as ultraprocessed food addiction because the foods most consistently linked to addictive patterns are rich in refined carbohydrates and fats, engineered to deliver rewarding ingredients rapidly and efficiently.

In that environment, GLP-1 medications may be doing something more subtle than suppressing hunger. “What strikes me about GLP-1 medications is that people often don’t just report being less hungry,” Gearhardt said. “They report being less preoccupied.”

The brain data are beginning to catch up with patient language. In a 2026 Nature study in mice, Ali Deniz Güler, PhD, assistant professor of biology at the University of Virginia in Charlottesville, Virginia, and his colleagues identified a brain reward circuit inhibited by next-generation GLP-1 drugs. GLP-1 receptor-expressing neurons in the central amygdala, they found, influenced consumption of highly palatable food through downstream dopamine signaling. When the circuit was engaged, the mice consumed less rewarding food.

“What surprised us was how clearly this pointed to a separable reward-related component of GLP-1 drug action,” Güler said.

Güler is careful about the limits. Mice can’t report food noise. But the central amygdala is well positioned to integrate emotionally and motivationally relevant information, and the dopamine pathway it connects to is broadly involved in many forms of motivated behavior. That makes the findings suggestive beyond palatable food while stopping well short of proof.

That separable component is what makes the field both exciting and easy to overstate. Reward is not one thing, and neither is compulsion. Güler identifies specificity as one of the field’s major unanswered questions. Researchers need to know whether GLP-1 receptor agonists reduce pathologic motivational drive or broadly dampen reward processing. A useful therapy, he said, should reduce harmful compulsive behavior without flattening normal motivation or pleasure.

Lose the Compulsion, Keep the Enjoyment?

Some patients have described emotional flattening, reduced motivation, or diminished interest in pleasures beyond food while taking GLP-1 medications — another social media naming opportunity, “Ozempic personality.” Berridge raised the same issue directly. A key question, he said, is whether GLP-1 drugs reduce only problematically strong cravings or also reduce mesolimbic desires more broadly, diminishing enthusiasm for “life pleasures and goals.” The evidence remains mixed, but the question matters.

Lorenzo Leggio, MD, PhD, senior investigator at the National Institute on Drug Abuse (NIDA) and the National Institute on Alcohol Abuse and Alcoholism and NIDA’s clinical director, has studied GLP-1 pathways in alcohol and addiction for years. His team published a 2015 mouse study showing that a GLP-1 receptor agonist reduced alcohol drinking, alongside human findings linking genetic variants of the GLP-1 receptor to risk for alcohol use disorder.

The mechanisms remain incompletely understood. Several may be involved simultaneously: dopamine-related reward processing, satiety signals that extend beyond food, effects on stress and anxiety, and possibly inflammatory pathways. Stress is a key trigger for craving in many patients, and GLP-1 drugs appear to influence that system as well.

The broader data are beginning to suggest that these effects are real, not incidental. A 2026 cohort study of more than 600,000 US veterans with type 2 diabetes found that starting a GLP-1 receptor agonist was associated with lower risk for several incident substance use disorders compared with starting an SGLT2 inhibitor. Among veterans with preexisting substance use disorders, GLP-1 use was also associated with fewer substance-related emergency department visits, hospital admissions, deaths, overdoses, and suicidal ideation or attempts. The study was observational, and the authors couldn’t rule out residual confounding. Still, the signal adds to the sense that the field has entered a new phase.

Where Are We Going?

Leggio is clear-eyed about what still needs to happen. Alcohol use disorder and tobacco use disorder are the areas with the most data, including randomized controlled trials. Data on opioid and stimulant use disorders are beginning to look promising. But the next step for all of these areas is identifying who responds and who doesn’t. Even if GLP-1s prove useful for some forms of compulsion, the drugs won’t work for everyone, just as they don’t work for everyone seeking weight loss.

The clinical challenge is to translate a mechanistic insight without turning every difficult habit into a drug target. Leggio said addiction treatment, if GLP-1s prove useful, will need to be personalized. Combining pharmacotherapy with behavioral treatment may produce better outcomes than either alone.

Gearhardt said something similar about compulsive eating. She’d be excited by studies combining GLP-1 medications with behavioral skills and access to higher-quality food. The medication may quiet the urge, but patients still live in the same food environment that helped train the urge in the first place.

That is the strangest part of this new chapter. GLP-1 medications were developed to help regulate blood sugar. Then they became incidental blockbuster weight-loss drugs. Now they’re forcing researchers to reconsider what it means to want something too much. The answer is unlikely to be one drug, one circuit, or one tidy mechanism.

But the drugs have already done something valuable. They’ve taken experiences patients often describe with shame and placed them inside biology. The pantry calling after midnight, the wine bottle that seems to suggest itself before dinner, the cigarette that feels less like a choice than an instruction. Even daily nail-biting. All of these may involve systems that medicine is only beginning to map.

Compulsion may be quieter, louder, or more treatable depending on the conversation between the gut, the brain, and the reward circuits that decide what deserves attention. The drugs began as tools for glucose control. They may end up helping medicine understand why the brain keeps reaching.

The experts quoted reported having no relevant disclosures. Disclosure information for study authors is available in the original study publications.