A habit is an automatic behaviour that a context triggers and that you can stop without much trouble. Addiction is continuing despite significant harm to your health, work or relationships, while genuinely wanting to stop and repeatedly failing to. The line falls at that combination of harm and impaired control, and it does not fall at frequency, which is where most people look first.
That matters because the frequency question produces two opposite errors. Someone who drinks every evening decides they must be an alcoholic when nothing is going wrong. Someone who binges once a month decides they are fine because most days are clear. Both are reading the wrong number.
What makes something a habit
A habit is behaviour that has become automatic through repetition, cued by a context rather than chosen fresh each time. The cue does most of the work: arriving home, finishing a meal, opening a laptop. You are not deliberating, which is the point of a habit, and it is why habits are efficient rather than pathological.
Most are neutral or useful. The relevant feature for this comparison is not whether a habit is good for you but that it remains available to a decision. When a habit genuinely costs you something and you decide to stop, stopping is uncomfortable and awkward, and it works.
What makes something an addiction
Addiction is a chronic health condition characterised by compulsive use continued despite serious harm and a genuine desire to stop. It produces lasting changes in the brain circuits governing reward, motivation and self-control, which is why the gap between deciding to stop and stopping is so much wider than it is for a habit. [volkow-2016-brain-disease]
Two things follow from that, and both cut against the intuitive framing:
Wanting to stop is not evidence against addiction. It is part of the definition. People who are addicted overwhelmingly want to stop, have tried, and have not managed it, and that pattern of failed attempts is diagnostic rather than reassuring.
Willpower advice has a poor track record here for structural reasons. If the difficulty sits in the circuits that generate motivation and self-control, then asking for more self-control is asking the damaged system to fix itself.
How the two are connected
They are not separate categories. The most influential neuroscience account describes a progression: voluntary, goal-directed use gives way to habitual use, and habitual use gives way to compulsion, with control shifting from deliberate decision-making toward automatic systems. [everitt-2016-habits]
Two caveats are worth keeping attached to that. The model rests substantially on animal studies plus human neuroimaging rather than on watching individual people move through the stages. And most habits never progress at all, so the model describes a route rather than a destiny.
What it does explain well is why the transition is so hard to notice from inside. Nothing announces itself at the boundary. The behaviour looks the same from outside for a long time, and the change is in how much of it is still a decision.
A schematic illustration of the progression described by Everitt and Robbins, not measured data. The transition is not this tidy in any individual case.
What the diagnostic manual actually says
DSM-5-TR does not use the word addiction as a diagnosis. It diagnoses substance use disorder, on at least two of eleven criteria met within a twelve-month period, with severity graded mild, moderate or severe by how many criteria are met. [apa-2022-dsm5tr] The criteria cluster around four themes: impaired control over use, social problems caused by use, risky use, and the pharmacological markers of tolerance and withdrawal.
Two consequences of that structure are worth knowing. It is a spectrum, not a threshold you either cross or do not, and a mild substance use disorder is a real diagnosis rather than a near miss. And tolerance and withdrawal are only two of eleven criteria, so their absence does not rule anything out, which is the single most common misreading.
On behavioural addictions, DSM-5-TR recognises gambling disorder and lists internet gaming disorder as needing further study. Terms like shopping or food addiction are widely used and do not have that status, which is a statement about the evidence rather than about whether the distress is real.
Which side of the line does this sit on?
Think about one specific behaviour over the last twelve months and tick what applies. This is a reflection prompt, not a diagnostic tool, and it produces no diagnosis.
0 of 8 ticked
Several of these map onto the criteria clinicians use, particularly continuing despite harm and unsuccessful attempts to stop. That is not a diagnosis, and only an assessment can give you one, but it is a good reason to book an appointment rather than to keep watching.
Two or more of these features is the point at which a clinician would begin asking more questions. A defined period of stopping, described below, is a reasonable next step, and a GP can assess it properly.
What you have described has the shape of a habit rather than a disorder: automatic, but not carrying harm you are continuing through. If you still want to change it, the ordinary habit-change approaches are the right tools.
No screener on this site measures addiction. None of the seven currently published covers it, and pointing you at one that measures anxiety or low mood instead would be misleading. The tick-list above is a reflection prompt built from the themes the DSM criteria cluster around, and it is not a substitute for an assessment. Our self-assessment hub lists what does exist, and do you have to hit rock bottom covers why waiting for things to get worse is a poor plan.
The test that is actually informative
Stop counting and try stopping. Pick a defined period, a fortnight is usually enough, and pay attention to three things rather than to whether you succeed:
- How much negotiation happens. Finding yourself relitigating the rule, adjusting the definition, or identifying exceptions is more informative than the outcome.
- What happens under stress. A bad day is the real test, and it is the one people usually plan around rather than through.
- What you replace it with. Something appearing in exactly the same slot suggests the behaviour was doing a job, and knowing what that job is matters for changing it.
If the fortnight is straightforward, that is genuinely useful information. If it is not, that is more useful still, and it is the specific thing worth telling a GP, because “I could not stop for two weeks” is a much sharper piece of evidence than “I think I drink too much”.
When to seek help
Speak to a GP if you have continued despite clear harm, if you have tried to stop and could not, or if someone close to you has raised it more than once. You do not need to have hit a crisis, and waiting for one is a worse plan than going early.
Go urgently rather than routinely if you are physically dependent on alcohol or benzodiazepines. Withdrawal from either can be dangerous and occasionally fatal, and stopping abruptly on your own is the wrong move. That is a medical situation, and a GP or emergency service can arrange a safe withdrawal.
If you are having thoughts of harming yourself, contact your GP urgently or use an emergency service or crisis line where you are.
How MyFreud can help
The question this article turns on is what happens when you try to stop, and that is a pattern over weeks rather than a single day you will remember accurately. MyFreud gives you daily mood tracking that shows what the difficult days had in common, so you can bring a GP something more specific than an impression. Seeing that the behaviour clusters around particular states, rather than around particular days of the week, is often the moment the shape of it becomes clear.
Download MyFreud and start today: App Store or Google Play.