"I Only Do It When I Drink": When One Habit Hides Behind Another

The thought draws a neat line around the problem. You do not smoke, not really. You smoke when you drink. You do not overspend. You overspend after a few glasses of wine. You do not text your ex. You text your ex at two in the morning after the pub.

The qualifier “when I drink” does two things. It explains the behavior, which feels like understanding, and it limits it, which feels like control. The problem is smaller than it looks, you tell yourself, because it only exists inside this one context.

What “only when I drink” really means

Alcohol myopia theory, proposed by Steele and Josephs, describes how intoxication narrows the field of attention to the most salient, immediate cues while reducing the influence of more distant, inhibitory ones [1]. Sober, you weigh the consequences: the smell on your clothes, the health cost, the person you will bother at two in the morning. Drunk, those considerations dim and the immediate impulse brightens.

The behavior is not appearing from nowhere. It is being released. Alcohol does not create the desire; it removes the barrier. And “I only do it when I drink” is less an explanation than a description of the conditions under which your inhibition is low enough for the behavior to surface.

The habit chain

Research on habit formation describes behaviors as context-cued automatic responses [2]. The context can include time, place, mood, social company, and, critically, internal state. Alcohol is an internal state change that functions as a powerful cue. If a behavior has been performed repeatedly in the context of drinking, the drinking itself becomes the trigger [5].

This creates a chain: the decision to drink is the first link, and the linked behavior is the second. Addressing the second link while leaving the first intact is difficult, because alcohol impairs exactly the executive function you would need to break the chain in real time [4].

A fictional example

The following is an illustrative example, not a real person’s story.

Chris has not bought cigarettes in two years. He considers himself a non-smoker. But every time he goes out drinking with friends, he cadges cigarettes from whoever has them, and by the end of the night he has had five or six.

• What he thinks: “I’m not really a smoker. I only do it when I drink, and I don’t drink that much.”
• What he feels: Mildly annoyed the next morning, quickly forgotten.
• What he does not account for: He drinks two to three times a week. At five cigarettes per occasion, that is ten to fifteen cigarettes a week, more than many people who call themselves smokers.

The “only when I drink” frame makes it sound like an occasional, incidental thing. The maths tells a different story.

The second question

The obvious question is: “Why does the linked behavior happen when I drink?” The subtler question is: “Does the linked behavior give me a reason to drink?”

This is uncomfortable territory. For some people, the pairing works in both directions. The drinking enables the behavior, and the prospect of the behavior makes the drinking more appealing. A night out is not just about the drinks; it is about the permission the drinks grant.

Marlatt’s relapse-prevention model describes “seemingly irrelevant decisions,” small choices that move a person toward a high-risk situation while appearing harmless [6]. Deciding to go to the pub is not deciding to smoke. But it is deciding to enter the context in which smoking reliably happens. If you keep choosing the context, you are, in effect, choosing the behavior.

Disrupting the pairing

Research on habit change suggests that disrupting the context is one of the most effective interventions [5]. For a behavior that is cued by drinking, the context disruptions include:

1. Change the drinking environment. Different venues, different company, different times. If the behavior is linked to a specific context within drinking, changing that context may be enough.
2. Change the amount. The more intoxicated you are, the weaker the inhibition. Drinking less in the same context may preserve enough executive function to break the chain.
3. Change the drinking itself. If the linked behavior is causing genuine harm, and reducing or changing the drinking context does not stop it, the question of whether the drinking needs to change becomes harder to avoid.
4. Pre-commit sober. Make a specific, concrete plan before the first drink about what you will and will not do, and tell someone.

None of these is easy under the influence, which is exactly the point. The best intervention happens before the drinking starts, when your full set of values and inhibitions is still available to you.

For the broader question of whether you can control a habit within limits, see “I’ll just have one”. For the thought that coping is a sufficient justification, see “I need it to cope”.

When to get more support. If a behavior linked to drinking is causing repeated problems, if the amount you drink to enable the behavior has increased, or if you have tried to break the pairing and cannot, talk to your GP or a specialist service. In the US, SAMHSA’s National Helpline (1-800-662-4357) offers free, confidential referral. In the UK, Drinkline is on 0300 123 1110. If you are in distress, call or text 988 in the US, or contact Samaritans on 116 123 in the UK and Ireland.