Your circles of people

The famous number 150 is supposed to tell you how many relationships the human brain can handle. Count your own circles of people and see how solid that number really is.

Your own estimateYou count your circles from memory, and the app doesn't grade them. Dunbar's numbers are averages, not a standard. What you enter isn't stored or sent anywhere unless you save your result yourself.

Who could you call at 3 a.m. if something happened?

In the early 1990s the anthropologist Robin Dunbar compared the size of the neocortex in primates with the size of the groups they live in. For humans he estimated about 148 stable relationships, with a 95% interval of roughly 100 to 230. The figure got rounded to 150 and became known as Dunbar's number. Later, Dunbar and colleagues described layers, each about three times bigger than the one inside it (Zhou et al., 2005). They're usually given in round numbers: 5 closest people, 15 good friends, 50 friends and 150 people you have a real relationship with. Dunbar puts acquaintances in further layers of roughly 500 and 1,500 people.

Another view

The number 150 stands on shaky ground. Patrik Lindenfors, Andreas Wartel and Johan Lind (2021) re-ran the analysis with newer primate data and newer statistical methods. Depending on the method, the average human group size came out at 69 to 109 or 16 to 42 people. But the 95% intervals were huge: 4 to 520 and 2 to 336. In the authors' view, data like these can't pin down a single number. They also point out that the method estimates an average, not a ceiling, and that the link between brain size and group size is itself disputed.

Dunbar disagrees. He says they used the wrong kind of regression, that his method gives 158 on the same data, and that other data, such as phone-call networks, confirm the layers. The authors reply that they reached the same conclusion with other methods, and that Dunbar's method gives them something closer to 290. The dispute goes on.

Friendship takes hours, not years. Jeffrey Hall (2018) asked people who had recently moved, and first-year students, how much time they had spent with new acquaintances and how close they felt.

People became casual friends after about 50 hours together, friends after about 90 and close friends after more than 200.

At two hours a week, that's 25, 45 and 100 weeks. Time spent together off work counted for more than time at work.

What to take from it. Don't chase the number 150. It's more useful to look at your two closest circles: how many people are in them, and who you want to give more time to. That's something you can plan.

How it works

Your circles are your estimate, not a measurement. Each outer circle includes the people from the inner ones, so it can't be smaller. The area of each circle matches the number of people (the radius grows with the square root), scaled to the larger of 150 and your outer circle.

Dunbar's layers of 5 / 15 / 50 / 150 are rounded averages from his work, not a norm. That's why the app doesn't judge anything; it only compares your closest circle neutrally with Dunbar's average of 5.

Bring people closer: weeks = the hours needed according to Hall (50, 90 and 200) divided by hours a week, rounded up. Hall studied people who had moved (355 adults) and first-year students (112), so the numbers are rough.

Nothing is saved or sent. The share card is created right in your browser and shows only the circles and four numbers.