Why do habits form, and how do you change one?
One-line takeaway
BLUF: Repeating a behaviour in a stable context builds an automatic cue-to-action link in the brain, so the cue triggers the behaviour before you deliberate. Changing one works best by altering the cue or the environment rather than relying on willpower at the moment of temptation.
Habits run roughly a third to a half of daily behaviour, so understanding their mechanism beats trying harder.
Cue, response, reward
A habit has three parts: a cue, a response, and a reward. Early on, the behaviour is goal-driven; you do it because you want the outcome. With repetition in a consistent setting, control shifts from goal-directed circuits in the prefrontal cortex toward the dorsolateral striatum in the basal ganglia, which learns the cue-response pairing directly. Neural recordings show activity becoming chunked, firing at the start and end of a routine rather than throughout, as though the sequence has been packaged into a single unit. The practical consequence is that the behaviour starts running on the cue alone, whether or not you still want the reward. That is what people mean by automatic.
An efficiency trade with a cost
The deeper principle is that habits are an efficiency trade. Deliberation is slow and costly, so the brain offloads reliable sequences to a cheaper system, freeing attention for novel problems. The cost is inflexibility. Devalue the reward — make a food unpleasant, or decide you no longer want the outcome — and a goal-directed behaviour stops quickly while a well-formed habit keeps going. That is the standard laboratory test for habitual control. It also explains why habits are so context-bound. Cues are usually environmental: a place, a time, a preceding action, a person. This is why the same person can smoke reliably in one country and barely think of it in another.
Change the room, not the resolve
The most reliable evidence comes from where the cue lives. Studies of people moving house or changing jobs find that habits weaken precisely when the environment changes, opening a window where new routines take hold more easily. This is why the effective interventions are structural: keep the food out of the house, lay out the running shoes, delete the app, take a different route home. Implementation intentions, phrased as if I encounter cue X, I will do Y, work by pre-attaching a new response to an existing cue. In healthcare and public policy the same logic appears as choice architecture, changing defaults and layouts rather than lecturing people about willpower.
The twenty-one-day problem
Common misconceptions begin with a number. Myth: it takes twenty-one days to form a habit. Reality: that figure comes from a 1960s plastic surgeon's observation about patients adjusting to their appearance, not from habit research; in Lally and colleagues' 2010 study the median time to reach automaticity was about sixty-six days, ranging from around eighteen to over two hundred and fifty depending on the behaviour. Myth: you can delete a habit. Reality: the old association tends to persist and can resurface under stress, so replacing the response to a cue works better than erasing it. Myth: missing a day ruins progress. Reality: occasional lapses had no measurable effect on habit formation in that study. Myth: bad habits mean weak willpower. Reality: they mostly mean a well-practised cue.