For months, your morning walk barely required a decision. Breakfast finished, shoes went on, and you headed outside.
Then work became unusually demanding and the walks stopped. Six weeks later, you have time again, but the routine no longer feels quite so automatic.
Have you lost the habit?
It is an obvious question with a less obvious scientific answer. Habit researchers have learned a great deal about how repeated behaviours become more automatic. They know considerably less about what happens when an established, beneficial habit is left unused for weeks or months and then resumed.
That missing evidence is important. Without it, there is no defensible expiry date for a healthy habit — but there is also no basis for assuming that an old habit waits indefinitely, unchanged, until you decide to use it again.
What survives an interruption seems to be a more complicated question involving the behaviour, its cues and the circumstances you return to.
A habit is more than something you do often
In everyday conversation, almost any repeated behaviour can become “a habit.” Researchers usually mean something more specific.
With repetition, features of a familiar situation can become associated with a response. Encountering that situation can then make the behaviour easier to initiate without requiring the same degree of deliberate decision-making each time.
Frequency helps this process, but frequency and habit strength are not interchangeable.
You can see the distinction in an ordinary kitchen. At home, making coffee may involve a sequence you barely notice: cupboard, mug, machine, milk. In an unfamiliar holiday apartment, making the same coffee suddenly requires thought because the surroundings no longer guide the sequence in quite the same way.
Nothing mysterious has happened to your ability to make coffee. The context is simply doing less of the work.
That becomes useful when thinking about an interrupted exercise, eating or sleep routine. Counting how often the behaviour occurred tells us one thing. Understanding how strongly it became connected to its usual circumstances tells us something else.
The famous habit-formation research answers a different question
Much of what people have heard about habit timelines can be traced back to research on forming habits.
In a well-known real-world study, 96 volunteers chose an eating, drinking or activity behaviour and repeated it in a consistent context for 12 weeks. Automaticity generally increased with repetition among participants who provided sufficient data, but the trajectories varied enormously. The estimated time to reach 95% of the modelled automaticity plateau ranged from 18 to 254 days.
The researchers also found that missing one opportunity did not materially derail the formation process.
That last result is reassuring if you forget your planned walk one morning while trying to establish it. Its relevance becomes much weaker if the question is what remains after an established routine has been absent for two months. The study did not test that situation.
More recent evidence reinforces the variability of formation itself. A 2024 systematic review covering 20 studies and 2,601 participants found reported median formation times of around 59–66 days in the relevant studies, with mean estimates around 106–154 days and substantial variation between individuals. More than half of the included studies were rated at high risk of bias.
This is why the popular “66-day rule” is a poor translation of the evidence. Habit formation has no universal 66-day timetable, and formation studies cannot supply a countdown for losing a habit either.
Look at what happened around the behaviour
Return to the morning walk.
Perhaps the walk originally followed breakfast every weekday. You ate at roughly the same time, left through the same door and followed the same route. Over months, those details became part of the circumstances in which walking normally happened.
Research on context stability helps explain why that matters.
Marco Stojanovic and colleagues examined people who were actively building habits. One experiment followed 95 university students for six weeks while context stability was manipulated. A second dataset covered 308 habits recorded by 218 users of a habit-building app. Across both datasets, greater context stability predicted greater automaticity and better habit-repetition goal attainment.
These studies concern habit building rather than prolonged interruption. Their useful contribution here is narrower: the circumstances surrounding a behaviour can participate in making it easier to repeat.
Now consider two versions of your six-week break.
In the first, work settles down and your old morning returns. Breakfast happens at the usual time, your shoes are where they always were, and the familiar walking route is still available.
In the second, the six weeks included changing jobs. You now commute earlier, eat breakfast elsewhere and no longer leave home at the time when you used to walk.
The calendar records the same six-week interruption in both cases. Behaviourally, you are returning to two quite different situations.
Habits can weaken, although the best decay evidence is not about this kind of habit
There is recent research that actually follows habits as they weaken.
Edgren, Baretta and Inauen tracked 194 people for 91 days while participants were deliberately trying to reduce sedentary behaviour, unhealthy snacking, alcohol use or smoking. Habit strength generally declined, with considerable differences between individuals. Among these people deliberately trying to weaken unwanted health-risk behaviours, person-specific models estimated decay stabilisation times ranging from 1 to 65 days.
Those numbers are interesting precisely because they show variation rather than a neat decay curve. They should not be repurposed into a prediction for your walking routine.
The participants wanted the behaviours to weaken and were actively trying to change them. Their habits involved four health-risk behaviours, and habit strength was measured through self-report. Someone who values a healthy routine but temporarily stops because life becomes difficult presents a different research question.
For now, this study gives us evidence that habit strength can change over time and that decay can look very different between people. It leaves the ordinary interrupted-healthy-habit question largely open.
Sometimes the world around a habit changes first
Researchers have also used major life transitions to study what happens when familiar behavioural contexts are disturbed.
Verplanken and Roy studied 800 people, half of whom had recently moved home. They tested an intervention aimed at environmentally relevant behaviours and found that it was more effective among recent movers.
This was sustainability research, rather than a study of interrupted exercise, nutrition or sleep habits. Its value for our question lies in the broader phenomenon it illustrates: when familiar circumstances change, existing behavioural patterns may exert less influence and people can become more responsive to deliberate change.
A house move is an obvious example, but everyday life contains smaller versions of the same problem. Work moves from home to an office. Children change schools. A lunchtime meeting appears permanently in the slot where you used to walk. The gym beside your workplace closes.
When a routine becomes harder after such a change, the interruption itself may be only part of the explanation. The old route into the behaviour may have changed as well.
What does “back at zero” actually mean?
This phrase creates more confusion than clarity because it bundles several questions together.
You may still know exactly how to perform the behaviour. A previously regular gym-goer knows how to enter the gym, organise a session and use familiar exercises after time away. Someone who used to prepare lunch every evening still knows the recipe, the containers and the sequence.
Practical familiarity can remain even when a behaviour has stopped occurring.
Automaticity is a separate issue. We do not yet have good longitudinal evidence showing how much automaticity from an established beneficial habit remains after different lengths of ordinary non-performance, or how rapidly it returns once the behaviour resumes.
So instead of trying to assign yourself a position on an imaginary zero-to-100 habit scale, separate the questions.
Do I remember how to do this? Does the old cue still exist? Does the behaviour begin easily when that cue appears? Am I having to make deliberate decisions at points that used to feel routine?
Those observations describe the current behaviour more usefully than deciding whether the old habit is technically “gone.”
A broken streak measures the streak
Habit apps make interruptions extremely visible.
Twenty-seven missed days looks precise, and it is precise — as a record of whether the behaviour occurred. It is not a direct measurement of automaticity, cue-response strength or how easily the behaviour will resume.
Researchers studying habits often use measures designed to capture automaticity or self-reported habit strength because repetition count alone cannot answer those questions.
This is one reason streaks should be interpreted narrowly. They can help you see that a behaviour has become less frequent or stopped. They cannot diagnose the internal state of the habit from the number of blank squares on a calendar.
For someone returning after a break, the next few performances may reveal more than the streak count does.
Returning can be an experiment rather than a verdict
Direct evidence on reacquiring interrupted healthy habits remains limited, so a scientifically responsible article cannot offer a validated restart formula.
It can suggest what is worth observing.
Suppose your old environment is largely intact. Try the behaviour in the circumstances where it used to happen. The experience may show that the old sequence still feels familiar, or that deliberate effort is now required.
If life has changed substantially, examine the old routine before blaming the interruption. What used to bring the behaviour into your day? Is that cue still present? Does the timing still make sense? Has something else taken over the same part of the day?
A morning walk that once followed the school run may need another place in the day when the school run disappears. A strength session tied to the gym beside an old workplace may need a different starting point after changing jobs.
These are practical deductions from what habit research tells us about context, rather than a proven reacquisition protocol. The point is to learn what the behaviour needs now, rather than reconstructing an invisible habit score from the number of days you missed.
The evidence ends before the most interesting question does
We can describe habit formation reasonably well. We have evidence that context stability can support automaticity. We can observe habit weakening when people deliberately try to reduce unwanted behaviours, and life-transition research shows how changes in context can loosen established behavioural patterns.
The evidence becomes thinner when the question is the one many people actually ask:
I had a healthy habit. Life interrupted it for a while. What exactly survived, and will it come back more quickly because I did it before?
At present, research does not give us a precise general answer.
That leaves room for a practical response without pretending the uncertainty has disappeared. Resume the behaviour where appropriate, pay attention to how it functions in your current circumstances, and distinguish what still feels familiar from what now requires rebuilding.
The first few repetitions are information.
The bottom line
A healthy habit does not have a scientifically established expiry date after you stop performing it.
Evidence on habit formation shows wide individual variation, and research on context helps explain why the surroundings in which a behaviour occurs matter. Separate studies show that unwanted habits can weaken when people deliberately work to reduce them. None of these evidence streams yet provides a reliable timetable for the passive loss and reacquisition of an established, beneficial habit.
That makes the number of missed days a poor stand-in for the question you really care about.
When you return, notice what happens. Familiar skills may still be there. Old cues may still help, or your circumstances may have changed enough that the behaviour needs a new place in your day.
Research can explain why those possibilities are plausible. It cannot yet tell you, from the calendar alone, how much of your old habit remains.
Sources & Further Reading
- Lally P, van Jaarsveld CHM, Potts HWW, Wardle J. — “How are habits formed: Modelling habit formation in the real world” — European Journal of Social Psychology, 2010.
- Singh B, Murphy A, Maher C, Smith AE. — “Time to Form a Habit: A Systematic Review and Meta-Analysis of Health Behaviour Habit Formation and Its Determinants” — Healthcare, 2024.
- Stojanovic M, Grund A, Fries S. — “Context Stability in Habit Building Increases Automaticity and Goal Attainment” — Frontiers in Psychology, 2022.
- Edgren M, Baretta D, Inauen J. — “The temporal trajectories of habit decay in daily life: An intensive longitudinal study on four health-risk behaviors” — Applied Psychology: Health and Well-Being, 2025.
- Verplanken B, Roy D. — “Empowering interventions to promote sustainable lifestyles: Testing the habit discontinuity hypothesis in a field experiment” — Journal of Environmental Psychology, 2016.