Is sugar addictive?

Published 15 September 2026 · 6 min read

By Loutra · Sources (13)

The idea comes from a rat, and mostly from a deprivation

It's 4 p.m., you open the cupboard without having decided to, and as you fold the wrapper you say the sentence everybody says: I'm addicted to sugar. It sounds obvious. It actually comes from a very specific place, and that place is a lab at Princeton.

In 2008, Nicole Avena, Pedro Rada and Bartley Hoebel published the review that gave the whole field its vocabulary: Evidence for sugar addiction. The protocol is spelled out in it: rats are "food deprived daily for 12 h, then after a delay of 4 h into their normal circadian-driven active period, they are given 12-h access to a sugar solution and chow." After a month on that schedule, the animals show four signs borrowed from drug research: they binge when access opens, they show withdrawal, they seek harder after a break, and they become sensitive to drugs they have never received.

Two details from that same review never survive the summary. First: the "withdrawal" was not produced by simply removing the sugar. It took an injection of naloxone — an opioid blocker, 3 mg/kg — or the removal of all food, to see the teeth chattering, forepaw tremor and head shakes appear. Second is an admission from the authors themselves: the effects are "smaller in magnitude than those produced by drug of abuse such as cocaine or morphine", and "it is not clear from this animal model if intermittent sugar access can result in neglect of social activities as required by the definition of dependency" in the psychiatric manual.

With free access, the rat never becomes addicted

That is the point Margaret Westwater, Paul Fletcher and Hisham Ziauddeen, at Cambridge, put at the centre of their review. Their sentence is blunt: "It is important to emphasise these addiction-like behaviours are only seen with sugar with intermittent access regimes and not with ad libitum access."

And it wasn't the critics of the model who found this. It was Avena. Rats on free access spread their sugar across the day, adjust their calories and don't gain weight — so much so that the Princeton group concluded themselves that such experimental conditions "cannot elicit sugar dependence."

In other words, what the model demonstrates is not that a molecule hooks you. It's that a schedule does: twelve hours without, then access. What the research built in rats looks far less like a smoker than like a day spent "being good" before emptying the cupboard at night.

In humans, the reference review finds no sugar addiction

The same authors compared what brain imaging shows in people dependent on a drug and in people who eat a lot of sugar. Their conclusion is one line, in the abstract: "We find little evidence to support sugar addiction in humans." Bingeing comes, they write, from "intermittent access to sweet tasting or highly palatable foods, not the neurochemical effects of sugar." And they go further than the usual caution: "Given the lack of evidence supporting it, we argue against a premature incorporation of sugar addiction into the scientific literature and public policy recommendations."

"More addictive than cocaine" comes from a saccharin study

It's the most repeated sentence on the subject. In 2007, in Bordeaux, Magalie Lenoir, Fuschia Serre, Lauriane Cantin and Serge Ahmed offered rats a mutually exclusive choice: sweetened water on one side, an intravenous cocaine infusion on the other. 94% chose the sweet taste, including animals already made dependent on cocaine.

The detail that goes missing everywhere: the sweetener in the headline result is saccharin, "an intense calorie-free sweetener". Not sugar. The team then found the same preference with sucrose, which is precisely their argument — it isn't about calories but about "an inborn hypersensitivity to sweet tastants". This is an experiment about taste, in rats, in a two-lever box. Turning it into "sugar is more addictive than cocaine" removes the species, the apparatus and the molecule.

What has an addictive profile is the food, not the nutrient

Erica Schulte, Nicole Avena and Ashley Gearhardt asked the question from the other end: which foods, exactly? They had 35 foods rated by 120 students and then by 384 adults. At the top: chocolate, ice cream, French fries, pizza, cookies. At the bottom: cucumber, carrots, apple, beans, broccoli. Apples and carrots contain sugar. What predicts the ranking is three things: degree of processing, fat content, and glycemic load — how fast the carbohydrate arrives.

The opposing camp says nothing different about the molecule. Ashley Gearhardt and Alexandra DiFeliceantonio argue that these foods do deserve the label "addictive substance": they take the criteria used in 1988 to classify tobacco and conclude that highly processed foods meet them. In the BMJ they cite a meta-analysis of 281 studies across 36 countries — 14% of adults and 12% of children clear the threshold on the Yale Food Addiction Scale; another, across 272 studies, gives 20% (95% CI: 18-21) and 55% with a binge eating disorder diagnosis. But what they point at is "foods that deliver high levels of refined carbohydrates or added fats" — and the combination of the two, never sugar on its own.

A third group, led by Johannes Hebebrand, proposes changing the word: "food addiction is a misnomer" because of the substance connotation, and should give way to "eating addiction". The shift isn't cosmetic — it changes what you would treat.

What the NHS actually says

Its "Sugar: the facts" page fits in a few sentences. "Eating too much sugar can make you gain weight and can also cause tooth decay." The target is "free sugars": added ones, plus those in honey, syrups and fruit juices — not those in milk, whole fruit and vegetables, of which it writes: "We do not need to cut down on these sugars." The adult limit is 30 g a day, "roughly equivalent to 7 sugar cubes".

The fact about the page is elsewhere. On it, the word "addiction" appears zero times, and "craving" zero times too (checked 15 September 2026). And on the NHS page about addiction — "addiction is defined as not having control over doing, taking or using something to the point where it could be harmful to you" — the list of things "it's possible to be addicted to just about anything, including" covers work, the internet, solvents, shopping and sex. Neither food nor sugar.

France reads the same. Its health insurance service lists the recognised "addictions without a substance" — gambling and video games — then those "currently under study": compulsive internet use, sex, physical exercise, compulsive shopping. Sugar is on neither list; eating disorders are filed separately. What France does have is a quantity marker: its food safety agency recommends "no more than 100 g of sugars per day (excluding lactose and galactose) and no more than one glass of a sugary drink", and notes that 20 to 30% of adults and adolescents exceed it.

The only measured human cessation contains two withdrawals

One study, and only one, made people stop sugar to see what happens. Jennifer Falbe and her team followed 25 adolescents aged 13 to 18 who drank at least three sugary drinks a day, then asked them to drink none for three days: more cravings, headaches, and less motivation, concentration and wellbeing.

Except the authors write the thing that makes the result impossible to attribute: caffeine and sugar are both "the primary ingredients" in those drinks. And caffeine withdrawal is a recognised diagnosis, gives a headache to one person in two, and begins 12 to 24 hours after the last dose — exactly the window of this study, detailed in the article on the caffeine withdrawal headache. Nobody can say which of the two products did what.

Withdrawal is not proof of addiction

That's the position this blog took a week ago about coffee, and it applies here unchanged: a symptom when you stop is not a verdict on dependence. The criterion, for the NHS as for the French health service, is loss of control — not a number of grams, not a craving at 4 p.m.

What that changes in practice: if what hooks you is a schedule and a food rather than a nutrient, then "I'm never touching a gram of sugar again" aims at the wrong thing, and looks a lot like the deprivation used to build the model in the first place. No trial has compared the two approaches in humans — that's a gap, not a proof. But it's worth noting that the NHS, on sugar in tea and coffee, writes: "gradually reduce the amount until you can cut it out altogether." It doesn't propose a detox.

Questions people ask

Loutra is free on the App Store

An otter that’s counting on you, a daily check-in, and a breathing round for the minutes that are hard.