The science of maintaining friendships

There is a lot of confident, slightly breathless writing about friendship online. It tends to promise that one habit will transform your social life, or that loneliness is quietly killing you, or that you can reverse-engineer closeness with a formula. The actual research is more careful than that, and in some ways more reassuring. It does not tell you to do something dramatic. It tells you that ordinary, repeated contact does most of the work, and that the instincts that keep you from reaching out are usually wrong.

This piece tries to be honest about what the evidence does and does not show. Some findings are strong and have been replicated many times. Others are suggestive, drawn from a single influential study or from correlations that cannot prove cause. Where that line falls matters, so it is flagged throughout rather than smoothed over.

Close friendships take sustained time, not a moment

One often-cited study, led by the communications researcher Jeffrey Hall, tracked how acquaintances became friends and estimated rough thresholds: somewhere around fifty hours of time together to move from acquaintance to casual friend, more than a hundred to reach friend, and a couple of hundred to count as a close friend. Those specific numbers get repeated as if they were physical constants. They are not. They came from particular samples, they vary widely between people, and the hours matter less than what you do with them. A hundred hours of small talk will not build what a much shorter run of honest, present conversation can.

The useful idea underneath the numbers holds up, though: closeness is built through accumulated, shared time, and especially through repeated meaningful interaction rather than one big bonding event. This is why work friendships and school friendships form so easily. The hours pile up without anyone planning them. It is also why adult friendships are harder to start and to keep. Once life stops handing you shared time for free, you have to find it on purpose, and most of us underestimate how much of it a close bond quietly required in the first place.

Dunbar's circles: a model of finite capacity

The anthropologist Robin Dunbar is best known for the idea that humans can maintain only a limited number of stable relationships, often quoted as roughly 150. Less famous, and more useful, is that he describes those relationships as nested layers rather than a single number. The pattern he and colleagues have described looks roughly like this:

• Around 5 in the innermost layer: the people you turn to in a crisis and stay closest to.
• About 15 good friends, the wider group you would call close.
• About 50 friends you would invite to a large gathering.
• About 150 meaningful contacts, the people you know well enough to have a real relationship with.

It is worth being clear about what this is. The layered numbers are an average pattern observed across several datasets, not a hard limit stamped into every brain. Individuals vary, the figures are approximate, and some researchers dispute how firmly the 150 number holds. So treat Dunbar's circles as a model, not a law. As a model, though, it captures something most people recognise from experience: attention is finite, the inner layers cost the most to maintain, and trying to keep everyone equally close usually means no one gets enough.

Why the layers are practically useful

The honest value of the model is permission. If your capacity for close relationships really is limited, then choosing where your limited attention goes is not cold or calculating. It is the only way the inner layer gets what it needs. People who feel guilty for not keeping up with everyone are often trying to maintain a hundred relationships at an intimacy that only a handful can realistically hold. Naming the layers, even loosely, makes it easier to give your closest few the consistency they actually require.

The link between connection and wellbeing is strong

If one area of this research deserves to be taken seriously, it is this one. Across a large body of work, social connection is repeatedly associated with better mental and physical health, and social isolation with worse outcomes, including shorter life expectancy. A frequently cited meta-analysis by Julianne Holt-Lunstad and colleagues found that stronger social relationships were associated with a meaningfully higher likelihood of survival over the studies' follow-up periods, on a scale comparable to other well-known health factors.

Two cautions keep this honest. First, much of this evidence is correlational. People who are healthier may find it easier to stay social, so the arrow does not point cleanly in one direction, and the size of any single estimate should be held loosely. Second, this is not a miracle claim. Friendship is not a treatment, and no one should read a study as a promise about their own lifespan. What the evidence does support, consistently and across many populations, is a real and substantial link between feeling connected and doing well. That is a strong enough finding to take staying in touch seriously, without dressing it up as medicine.

Consistency beats intensity

A theme runs through much of the friendship research, even where the individual findings are modest: relationships are sustained by frequency and regularity of contact more than by occasional intensity. A friendship topped up with small, regular signals tends to stay warm. One kept alive by rare, dramatic catch-ups tends to cool between them, because the relationship spends most of the year with nothing flowing through it.

This fits how memory and habit work. A short message every few weeks keeps a friend present in your mental life and you in theirs. The bond stays in motion. A single long dinner twice a year, however lovely, leaves long stretches where neither person is part of the other's ordinary world, and ordinary-world presence is a lot of what closeness is. The practical lesson is freeing: you do not need to engineer big moments to stay close. You need to keep a thin, steady thread from breaking. If you want a concrete sense of cadence, how often should you reach out works through it.

You underestimate how welcome your message is

One of the more useful findings for anyone who hesitates before reaching out: people systematically underestimate how much others appreciate hearing from them. In a set of studies, researchers including Peggy Liu found that recipients of an unexpected check-in valued it more than the senders predicted, and that the gap was widest exactly when the gesture was more of a surprise, such as after a long silence.

The mechanism is intuitive once named. When you reach out, you are focused on the awkwardness, the long gap, the possibility of bothering someone. The person receiving the message is not running that calculation. They mostly feel that you thought of them. This is the same trap that lets friendships drift in the first place: two people each assume their outreach is unwelcome or overdue, so neither sends it, and a bond that both would happily keep simply goes quiet. The research is clear enough on this to act on. The message you are nervous to send is almost always more welcome than you think.

What the science means in practice

Translated into ordinary behaviour, the findings point in a remarkably consistent and undramatic direction. None of it asks for a personality transplant or a packed social calendar. It asks for small, repeatable contact aimed at the people who matter most.

1. Protect the inner layer first. If close capacity is finite, spend it deliberately. A few relationships kept genuinely close beat a wide circle kept all at arm's length.
2. Favour frequency over occasion. A short, regular signal does more for a friendship than a rare grand event. Keep the thread thin but unbroken.
3. Lower the bar to reach out. Since contact, not depth, is what keeps a bond warm between deeper talks, a thirty-second message counts. It does not have to be profound to matter.
4. Send the message you are unsure about. The evidence says it lands better than you expect. Hesitation is the bias, not the signal.
5. Build in a reminder. The research explains why friendships fade, but it does not solve the simple fact that busy weeks crowd contact out. A small external nudge does what willpower alone tends not to.

That last point is the gap between knowing all of this and living it. Almost no one drifts from their friends because they disagree with the science. They drift because nothing reminds them, the gap grows, and the next message starts to feel heavier than the last. The findings explain the problem cleanly. Closing it is a matter of building a habit that the science merely justifies.

Holding the evidence honestly

It is tempting to turn research into rules, because rules feel safe. But the friendship literature is most useful when it is held loosely. The strong parts, that connection matters and that contact sustains it, deserve to shape how you live. The precise parts, the exact hours and the exact tiers, are best treated as rough orientation that helps you think, not targets to hit. A friendship is not a quota, a streak, or a study to optimise.

What the science really offers is permission and direction. Permission to accept that your attention is finite and to spend it on the people who matter. Direction toward small, steady, unembarrassed contact rather than rare intensity or anxious silence. If you want the practical version of all this, how to stay close to people who matter and how to keep in touch with friends turn these findings into a routine you can actually keep.