8.4 — Breaking One, and the Addiction Pattern

Building is easier than breaking, for a structural reason: a habit is not deleted. The learned association stays. What you can do is stop it launching, and build a stronger competing one over the top.

Which is why old habits return so readily under the original conditions — the same bar, the same stress, the same room after ten years. The sequence was never removed; it stopped being triggered.

Work the loop, not the routine

The reflex is to attack the behaviour directly with resolve. That is the hardest of the four points to attack (8.2), and it is the one that requires you to win every single time.

Attack the other three instead.

Remove or change the cue

The highest-yield move by a wide margin. A habit that is not triggered does not need resisting.

Find the actual cue first, which requires observation rather than assumption. For a week, each time you do the thing, note: where, what time, who was there, what you were doing just before, and what you were feeling.

Most unwanted habits are triggered by a state rather than by a situation — bored, tired, stressed, lonely, avoiding something. Identifying which one is most of the work, and it is nearly always one of those five.

Then remove what you can. Delete the app. Do not keep it in the house. Take a different route. Do not sit in that chair. This looks like avoidance and it is simply removing a trigger, which is legitimate and effective.

Change the environment

From 8.2: habits are bound to context. A change of context is the single best moment to break one, and if one is coming — a move, a new job, a holiday — plan to use it.

Where you cannot change the context, change what you can: rearrange the room, sit somewhere else, put the thing somewhere else.

Substitute the routine, keeping the reward

A habit that is only removed leaves a gap, and the gap gets filled by something — usually by the old habit.

So work out what the reward actually is, which is frequently not obvious. Someone who eats at 4 pm may be getting sugar, or a break, or human contact from the walk to the kitchen. Test it: try satisfying each candidate separately and see which one removes the pull.

Then substitute a different routine that delivers the same reward. Break needed — walk outside. Contact needed — call someone. Something to do with your hands — something to do with your hands.

Increase the friction

Twenty seconds is enough to let the deliberate system arrive (8.2). Phone in another room, app deleted from the phone and used only on a laptop, the thing you buy not bought.

Riding the urge out

Urges feel like they will grow until you act. They do not. They rise, peak and fall, typically within about twenty to thirty minutes and often much less, whether or not you do anything.

The technique — urge surfing — is to watch the urge rather than fight it.

Notice it and name it. There is an urge.

Find it in the body. Where is it? Chest, stomach, hands, jaw, a restlessness in the legs?

Watch what it does over the next few minutes, with curiosity rather than resistance. Does it rise? Does it move? Does it change quality?

Breathe. Do not argue with it and do not act on it. You are not resisting; you are observing.

Notice when it falls.

Why not fight it. Fighting is effortful and it makes the urge the centre of attention, which strengthens it. Suppressing the thought rebounds (4.6).

The delay rule is the simpler version, and it works. I will wait ten minutes. Not never — ten minutes. Most urges do not survive ten minutes, and the ones that do are dealt with more deliberately than they would have been.

And the delay-plus-substitute version is stronger still: ten minutes, during which you do something that occupies attention.

Where addiction differs

Everything above is for ordinary unwanted habits. Addiction is a different clinical situation and it needs saying plainly, because treating it as a willpower problem is both wrong and cruel.

What defines it: compulsive use continuing despite clear harm, loss of control over the amount or the timing, craving, tolerance in many cases, and withdrawal in many cases.

What the mechanism actually is. From 8.1: wanting and liking are separate systems. In addiction the wanting system becomes progressively sensitised while the liking system does not, and frequently fades. This produces the state that is hardest for outsiders to understand — someone pursuing something intensely that they no longer enjoy.

Alongside that, the prefrontal control systems are impaired, both by the substance and by the chronic state, so the part of the brain that would intervene is working worse precisely when it is most needed.

And it is not a moral failing. Nor is it purely a chosen behaviour. Nor, in the strongest version of the "brain disease" framing, is it something over which the person has no agency at all — that framing has its own critics and evidence against it. The accurate middle: it is a condition with strong biological, psychological and social components, in which the capacity to choose is genuinely impaired and not abolished.

Some withdrawal is dangerous

This matters medically and it is not widely known.

Alcohol withdrawal can be life-threatening. In someone physically dependent, stopping suddenly can cause seizures and a condition called delirium tremens, with a real mortality rate. A heavy daily drinker should not stop abruptly without medical advice.

Benzodiazepine withdrawal is also potentially dangerous, and requires a supervised taper.

Opioid withdrawal is extremely unpleasant and is generally not life-threatening on its own, though the risk of overdose after a period of abstinence is very high because tolerance falls.

Nicotine, cannabis and stimulant withdrawal are unpleasant and not dangerous.

If in doubt, ask a doctor before stopping. That sentence is the most important one on this page.

What works for addiction

Treatment beats willpower, and the effective options are real.

Medication exists and is under-used. For opioids, opioid substitution treatment substantially reduces mortality — this is one of the best-evidenced interventions in medicine. For alcohol, several medications reduce drinking. For nicotine, nicotine replacement and specific medications roughly double or better the chances of quitting.

Structured psychological treatment — cognitive behavioural approaches, motivational interviewing, contingency management — has good evidence.

Peer support helps many people, and it helps some people more than anything else.

And combinations beat single approaches.

Relapse is expected, not exceptional. Rates are comparable to those for other chronic conditions requiring behaviour change. Treating a relapse as total failure is itself a predictor of a worse outcome, and the shame that follows it is a large part of the mechanism (4.8).

If you are wondering whether your drinking or use is a problem, that question is itself informative, and asking a doctor costs you nothing.

The specific modern ones

Phone, feed, games, pornography, gambling.

Gambling disorder is formally recognised as an addiction and behaves like one. Gaming disorder is recognised by the World Health Organization. The status of "phone addiction" and "social media addiction" as clinical categories is contested, and the underlying mechanism is not in doubt: variable reward schedules (1.2), engineered cues (2.3), and a wanting system that fires at the cue.

What works for these, in practice:

Friction, aggressively. Delete from the phone, use on a laptop only. Greyscale. No notifications. Charger in another room.

Substitute rather than only remove, because the underlying need — boredom relief, contact, escape from a difficult task — will otherwise reassert itself.

And identify the state that triggers it. For most people it is one of: bored, avoiding something difficult, lonely, tired. Naming which one, in the moment, is a large part of the intervention — because I am avoiding this task has an obvious next step and I feel like checking my phone does not.

What to do with this

Observe for a week before changing anything. Where, when, who, what you were doing, and what you were feeling. You will be wrong about the trigger if you guess.

Then remove the cue, not the behaviour. It is the cheapest intervention and the most effective.

Find what the reward actually is, and substitute something that supplies it. A removed habit leaves a hole.

Add friction, generously.

Learn to ride an urge once. Watch one rise and fall without acting. Having seen it fall on its own changes your relationship with the next one, because you now know from experience rather than from being told.

And if this is a substance, get help rather than relying on this page. The treatments work, the odds with them are much better than without, and asking is not an admission of anything except arithmetic.

Next: 8.5 covers the specific failure that looks like a habit problem and is not — procrastination, which is not about time management at all.