Food addiction: is it addiction at all?

“I am to chocolate what my father was to drink.”

I have heard that sentence more than once and every time it was meant seriously. The person saying it is not being dramatic. They are describing an experience they know from the inside: that something happens outside their decision, that the thought is intrusive, that promises dissolve, and that after all of it comes a wave of shame very like the one people with addiction describe.

And yet the question of whether this really is addiction remains open among researchers. Not open in the sense that some of them know nothing about it, but open in the sense that serious people with good arguments claim the opposite. This piece is an attempt to show you both sides honestly — partly because the word you choose has consequences for how you go about getting help.

What food addiction is supposed to be

The idea is a few decades old, but it spread scientifically after 2009, when a questionnaire appeared that transfers the criteria for substance addiction onto food. The questions were reworked: instead of asking about a substance, they ask about food. Did the person consume more than intended. Have they unsuccessfully tried to cut down. Do they continue despite consequences. Has tolerance appeared, and does discomfort follow when they stop.

People who score high on that questionnaire more often have weight-related difficulties, more often binge eating, more depression and more impulsivity. None of that is disputed. Something else is: what the score means.

The questionnaire cannot separate two possibilities that look identical on paper. The first is that food produced a state resembling addiction. The second is that a state resembling addiction arose from something else — from distress, from years of restriction, from how a person soothes themselves — and that food was simply the means at hand. Because the questionnaire measures how you behave and not why, it cannot decide between them.

What the brain part of the story says

Brain imaging does show that palatable food activates the same regions of the reward system as many substances. Dopamine is released, the anticipation of reward strengthens, and with frequent repetition the response to the taste itself falls while the response to the predictive cue — a smell, a wrapper, a time of day — grows.

Here caution is needed. Those regions also light up for music, sex, money and winning a game. If overlap with the reward system is proof of addiction, then almost everything that pleases us is an addiction. The overlap on the scans tells us that food acts on a system built for survival — not that food is a drug.

Besides, that system was doing its job long before processed food existed. A pull towards sweet and fatty things is not a design fault but a very old mechanism that was useful in an environment where food was scarce. That it turns against us in an environment of abundance is not a sign of individual illness. This is worth saying, because conversations about food slide too quickly into conversations about character. What I have written elsewhere about addiction in general holds here too: this is not a question of weak will.

The key objection: there is no substance

With alcohol, nicotine or opioids we know which molecule produces the addiction. With food, no such molecule has been found. Sugar is the most accused candidate, but human experiments have not shown that it acts as a drug in its own right.

So some researchers have proposed a different name: not food addiction but eating addiction. The difference is not cosmetic. The first says the culprit is in the substance, the second says the pattern is in the behaviour — closer to what we know from gambling. That distinction also changes where you go looking for a solution.

Rats, sugar and a careful conclusion

The frequently cited rat experiments do show behaviour resembling addiction: seeking, agitation on withdrawal, reinstatement on a cue. But these behaviours appear mainly when the animals’ access to sweet things is interrupted — periods without, then available again.

If food is continuously available, that picture largely does not develop. What the experiments demonstrated, then, is less the power of sugar than the power of interruption. That is a very important detail, because interruption is not a natural state in humans — it is something we do to ourselves when we decide to restrict strictly.

Restriction often produces the very thing you fear

From research on dietary restraint we know that strict holding back increases responsiveness to food, increases the intrusiveness of thoughts about it, and increases the likelihood of losing control — often at the end of the day. When that happens, shame follows, and shame leads to an even stricter decision for tomorrow. And round the loop goes.

This is not a theory that denies your experience. On the contrary: it explains it, and in a way that offers a way out. If the loss of control follows from the rules, then loosening the rules is not indulgence but treatment. For many people this is the hardest part, because it sounds like the opposite of everything they have been told.

Overlap with binge eating disorder

Most people who score high on the food addiction questionnaire also meet the criteria for binge eating disorder. That means for the great majority of cases we already have a diagnosis that is better researched and has clearly supported treatment.

Some people fall outside that explanation, of course — those who eat continuously in small amounts and never in a form that would meet the criteria for bingeing. For them the addiction description often fits better, and I do not want to dismiss that experience with an objection about diagnosis. But the same question applies to them: what is food doing in the day that nothing and nobody else is doing.

Why does this matter? Because the name determines the address. If we say addiction, the person looks for help in addiction programmes and starts thinking about abstinence. If we say eating disorder, they go where restriction, body image and emotional regulation are the work — and where we have evidence that it helps. What a diagnosis does and does not tell you I have set out in a piece arguing that a diagnosis is a map, not a name.

Where the case for addiction is strongest

In fairness, the other side has its strongest point, and it is not a weak one.

It concerns ultra-processed food. Combinations that do not exist in nature are engineered so that the reward system responds faster and harder than it would to basic foods. Speed of delivery is a known factor in substance addiction — the same substance delivered faster is rated by the brain as more valuable. If that holds for substances, it is not unreasonable to suppose something similar operates here.

Seen this way the question is not whether food is a drug, but whether we have processed part of our food to the point where it starts behaving more like a substance and less like a meal. That is a serious question and genuinely open in research, not a conspiracy or an exaggeration.

When the label helps and when it harms

The label helps when it removes blame. Believing that a mechanism rather than a character flaw is at work reduces shame and increases willingness to seek help. That is no small thing, because shame reliably keeps people away from the door.

The label harms when it imports the logic of abstinence. You can step away from alcohol; from food you cannot. When someone using that logic declares a whole category of food forbidden, they return to the loop in the second picture, only now it carries a professional-sounding name. In practice I see that the people who have renounced most strictly are the most vulnerable to a sense of total collapse at the first deviation.

The label carries one more quiet cost. Someone who understands themselves as addicted to food often stops trusting their own signals of hunger and fullness, because they read them as the voice of the illness. That loses them the very tool that is most useful in repairing a relationship with food. Instead of learning to listen to the body, they learn to resist it — and in many cases that was where the trouble started.

What this means for getting help

If you recognise yourself in this, the name of the condition matters less than what is done about it. Established treatment works on regularity of eating, on loosening prohibitions, on recognising emotional triggers, and on a response to oneself that is not punitive. Every one of those parts runs against the logic of one more, stricter diet.

With adolescents the picture is different, and there the recommended first choice is family-based treatment rather than individual counselling. Where there is medical risk — fainting, a missed period, complications — the first address is a doctor, not a therapist. This is not a formality; some consequences can develop quietly.

And one more thing that is close to me: behaviour that looks senseless is often the best solution a person has found in the circumstances. I have written about what a symptom is doing and why it is unwise simply to remove it in a piece on the symptom as a message.

Where my doubt may be wrong

So far I have mostly argued for caution about the concept. Now against myself.

First: it is possible that we are where the field once was with gambling. There too it was held for decades that without a substance there is no addiction, and today behavioural addiction is a recognised diagnosis. If the same turns out to be true of food, my doubt will look like a delay.

Second: that the questionnaire measures behaviour rather than substance is not necessarily a fault. Addiction criteria are in any case a description of a behavioural pattern, not a chemical analysis. If the pattern fits, refusing the name may be more a matter of habit than of evidence.

Third and most importantly: my doubt does nothing whatsoever for the people living this. If someone feels addicted to food, telling them the term is not professionally settled removes no distress. The word is theirs; my job is to help with what fills the word — and to make sure self-compassion does not stay a nice idea but becomes how you treat yourself after a bad evening.

What to remember

The experience is real even where the name is contested. Nobody is inventing this.

No substance has been found to blame. That is why part of the field speaks of addiction to eating rather than to food.

The animal experiments showed mainly the power of interrupted access, not the power of sugar.

Strict restriction increases the very thing it is meant to prevent. Loosening the rules is not indulgence.

Most of these cases fit binge eating disorder, for which better-researched treatment exists.

The label helps when it removes blame and harms when it imports abstinence — because food cannot be given up.

With an adolescent the first choice is family-based treatment; where there is medical risk, the first address is a doctor.