Fogbreaker
Everyone gives the same advice to someone who has just moved: visit the landmarks, find your coffee shop, join a club, say hello to the neighbours. It is not wrong. It is just tourist advice handed to a resident, and it misses the two things about moving that are actually measurable.
The first is that you have less time than you think. The second is that you are not choosing what to see. You are choosing which handful of places will quietly become your entire city.
The window is about three months
Habits are not stored in your head so much as in your surroundings. They run on environmental cues, which is why you can drive a familiar route and remember none of it. Move house and every one of those cues is stripped out at once. That is why the first weeks are exhausting: nothing is automatic, and you are consciously deciding things you have not consciously decided in years.
The useful part is that this cuts both ways. Bas Verplanken at the University of Bath calls it the habit discontinuity hypothesis: when the context breaks, old behaviours lose their grip and people become unusually open to new ones. He and Deborah Roy tested it in a field experiment with 800 adults in Peterborough, published in the Journal of Environmental Psychology in 2016. The intervention worked on people who had recently moved and did not work on people who had not.
The number that matters is how long it lasts. Verplanken's estimate is blunt: the window is limited, probably to a maximum of three months. After that the new cues have settled in, the new routine is automatic, and you are back to running on rails, just different rails.
So the exhaustion and the opportunity are the same thing. When it stops feeling like effort, the window has closed.
You get about 25 slots
Here is the finding that should change what you do with those three months.
In 2018 Laura Alessandretti, Piotr Sapiezynski, Vedran Sekara, Sune Lehmann and Andrea Baronchelli published Evidence for a conserved quantity in human mobility in Nature Human Behaviour. They analysed high-resolution, multi-year location traces from around 40,000 people across four separate datasets, and set out to resolve an apparent contradiction in the literature: some research shows people endlessly revisit a small set of places, other research shows the number of places a person has visited keeps growing.
Both are true, and the resolution is elegant. The number of familiar locations a person visits at any given point in time is conserved, at a typical size of about 25. You do keep exploring. You do keep finding new places. But the size of the active set does not grow. When a new regular place enters, an old one drops out.
You are not accumulating a city. You are running a fixed-size rotation, and new entries cost you old ones.
The authors also found that the size of a person's set of preferred locations correlates with how many social interactions they have, and suggest a link to the Dunbar number, the proposed cognitive ceiling on how many relationships someone can maintain. Which implies the limit is not about time or transport. It is about how much place a person can hold.
Put the two findings together and the situation is uncomfortably clear. About three months to fill roughly 25 slots, after which they mostly stop turning over. Whatever you happen to be doing in week twelve is not a phase. It is your city now.
Which is why sightseeing is the wrong move
A tourist has three days and should absolutely go to the best places. That is optimisation under a hard deadline, and the landmark list is the correct tool. We wrote the version of that argument for people exploring a city they are visiting.
You are in the opposite situation and the same advice actively hurts. You have years. What you do not have is a wide enough pool to choose from. You cannot pick a favourite bakery out of the three you have walked past, and whichever one you settle on in month two is likely to still be the one in year four, because by then the slot is taken.
So invert it. Coverage first, favourites second. The job of the first three months is not to find the best places. It is to make the set you are eventually choosing from as large as possible before the window shuts.
What to actually do
Fill in your radius before you go further afield. A comfortable walking pace is around 5 km/h, so a fifteen-minute walk from your door is roughly 1.2 km, and the circle that draws is about 4.5 km². That circle is where almost all of your unplanned life will happen for as long as you live there. Most people cover maybe a fifth of it, in three or four spokes, and never touch the rest. Finish it before you start taking the train somewhere scenic.
Vary the commute in the first fortnight, not the sixth month. Your commute is the single route most likely to consume slots, it is the one you walk most often, and it hardens fastest because it is the most repeated cue you own. Whatever variations you build into it early will stay. Add them late and they never take, because by then the efficient version is automatic.
Do the boring direction. New arrivals radiate toward the centre, because that is where the reasons to go are. The direction that points away from the centre stays dark for years. It is also, usually, where you actually live.
Spend your errands. You are already going to the post office, the hardware shop, the doctor. Those trips are the cheapest exploration budget you will ever have because the walk is happening anyway. Take a different way there and a different way back and you have covered two new streets at no cost in time or willpower.
Come home wrong on purpose. The single highest-yield habit is refusing to take the same way back that you took out. It requires no planning, costs a few minutes, and doubles the ground covered on every trip you were making regardless.
If you want the structured version of all this once you have found your feet, it turns into walking every street in your city, and the no-destination version is flâneuring.
What the research does not say
Both studies are worth taking seriously and neither is a law.
The Peterborough experiment was about environmental behaviours, shorter showers, recycling, transport choices, and not about exploring a neighbourhood. The habit-discontinuity mechanism is general and well supported, but nobody has run this specific experiment on how people learn a new city. Verplanken's three months is one study's estimate, and he hedges it himself as a probable maximum rather than a deadline.
The ~25 figure is a population typical, not a personal quota. It varied between people, it correlated with sociability, and it describes regularly visited locations rather than every place you will ever set foot. Treat it as a shape, not a number to count against.
What survives the caveats is the direction of both findings: the window is real and shorter than it feels, and your active set is smaller and more static than it feels.
The part you cannot see from inside
The genuine difficulty here is not motivation. It is that a routine feels complete from the inside. Ask anyone who has lived somewhere five years to sketch the parts of their neighbourhood they have never walked and they cannot do it, because the unwalked parts leave no trace in memory. There is nothing to remember. Your mental map of where you have been is made entirely of the places you go.
That is the specific problem Fogbreaker was built for. The map starts under fog and walking outdoors is the only thing that clears it, so what you get is not a recommendation, it is an audit. The shape of the cleared area is the shape of your actual life in the city, and the dark parts are the answer to a question you cannot otherwise ask. For someone three weeks into a new place, nearly all of it is dark, and the first months are when that is most useful.
Two honest limits. It will not tell you which bakery is good, only which streets you have not been down, so it complements local recommendations rather than replacing them. And it clears from GPS movement outdoors, so the fog does not care how many steps you took indoors. As your map fills in, it gets less useful, which is the correct behaviour for a tool whose job is to run out of dark.
Three months, twenty-five slots. You are picking them right now whether you mean to or not. Go home a different way tonight.
