Aristotle warned that a long silence seems to make friends forget their friendship, but research suggests the friend on the other end would be happier to hear from you than you think

I was cooking curry at home, a dish a group of us used to make together at the university co-op, and somewhere between the stirring and the smell a friend from those years was suddenly in the kitchen with me.

I picked up my phone and typed: Hey, I was just thinking about…

Then I stopped. It felt too out of the blue, too sudden, so I deleted it and went back to the pot. I never sent it, and Aristotle would not have been surprised.

Psychologists have tested the kind of guess I had just made, and on average it runs low: in their studies, the person on the other end rated a check-in more highly than the sender predicted, at least when it came unannounced.

Out of sight, out of mind

In Book VIII of the Nicomachean Ethics, in W. D. Ross's translation, Aristotle separates having a friendship from doing it. Friends who are asleep or living apart, he says, "are not performing, but are disposed to perform, the activities of friendship." Distance "does not break off the friendship absolutely, but only the activity of it."

Then comes the warning. "If the absence is lasting, it seems actually to make men forget their friendship; hence the saying 'out of sight, out of mind'." In H. Rackham's translation the saying becomes a poet's couplet: "Full many a man finds friendship end / For lack of converse with his friend."

The danger Aristotle names is quieter than a quarrel: nobody decides anything, and the friendship simply goes unused until it is hard to find.

He pushes further in the same chapter. People who approve of each other "but do not live together seem to be well-disposed rather than actual friends," because "there is nothing so characteristic of friends as living together." By that standard, the co-op was friendship in its active form, a shared kitchen and a shared pot. What I felt over the stove years later sits closer to being well-disposed, in his phrase.

That leaves a question Aristotle had no way to test. If the warmth is still there on my side, is it still there on theirs? And how would I know without asking?

Notes to old classmates

Peggy Liu of the University of Pittsburgh and three colleagues set out to measure something like it. Their paper, The Surprise of Reaching Out, published in the Journal of Personality and Social Psychology, reports 13 studies with nearly 6,000 participants. Most of the studies compared how much people expected a simple gesture of contact to be valued with how much recipients valued it, or said they would.

Every study was registered before it ran, and the data are public, though it is still one team's work.

One of the simplest studies took place on a US college campus. Students named a classmate they usually had pleasant encounters with but hadn't been in touch with for a while, then wrote a short note to "check in and say hello." The researchers delivered the notes and asked the recipients how much they appreciated them.

On a scale of 1 to 7, the writers predicted on average that their classmates would rate the note at 5.57. The 54 classmates included in the analysis gave it 6.17 on average.

Other studies used small gifts, like a bag of cookies, a box of tea, coffee or a mini plant, along with remembered experiences and imagined scenarios. Whenever the contact came unannounced, the shortfall kept appearing.

According to a footnote, nearly everyone, senders included, rated appreciation above the midpoint of the scale. People already know a message like this is welcome. What they misjudge is how welcome.

The surprise on the receiving end

The authors' explanation centers on surprise. When they asked both sides how much they were thinking about the recipient's surprise, recipients focused on it more than senders did. In the statistics, that difference accounted for part of the gap, and the authors call it "one underlying explanation."

They then tested the idea. In a scenario study of 1,576 people, the gap disappeared when the contact, a note with a bag of cookies, came through a program in which both people expected it. Recipients still said they would appreciate it, but now senders' predictions matched theirs.

Another scenario study, with 1,602 people, compared close friends with looser acquaintances. Senders underestimated both, but by more than twice as much for acquaintances, and gifts from close friends were still rated higher overall. The authors read this as fitting their surprise account: contact from an acquaintance is more unexpected, and that is where the guess fell furthest short.

I deleted my text because it felt too out of the blue. In these studies, that surprise counted in a message's favor: recipients dwelled on it more than senders did, and the more surprising the contact, the bigger the shortfall.

Philosophers discuss a deeper version of the sender's difficulty as the problem of other minds, the old question of how we can know anything of another person's thoughts and feelings, or even that they have them. Standing there, I could feel my own hesitation in detail. Their reaction I could only imagine, and I imagined it from my side of the phone.

Seneca, writing to his friend Lucilius, describes the other side. "I never receive a letter from you without being in your company forthwith," he says in Letter 40, and a letter "brings us real traces, real evidences, of an absent friend." He was writing as the one who receives, which is exactly the view a sender can't quite reach.

A second run, with 742 people

Psychology has spent years watching findings fail on a second attempt, so this one deserves a check. In 2023, Clearer Thinking's Transparent Replications project reran one of the paper's studies with 742 people, more than three times the original 201.

The result held. Once again, people picturing an acquaintance getting in touch said they would appreciate it more than a separate group of would-be senders predicted. The effect came out a little smaller, and the replicators called it small. They also flagged one wording problem: the paper describes that study as testing a "brief message," but participants were never told what form the contact took.

The fine print

The paper is open about its limits, and they matter for anyone holding a half-written text. Almost every main study involved someone the sender usually had pleasant encounters with. Falling-outs were not studied, and the authors raise the possibility that after one, senders might even overestimate how welcome they would be.

What was measured is appreciation in the moment; whether the friendship revived afterward was not studied. The participants were mostly American, from college students writing to classmates to online panels of adults. And the differences are consistent but modest, under one point in every main study.

Knowing it didn't make them send it

The detail I keep returning to sits in the supplementary studies. The researchers informed would-be senders about their findings, then gave them a brief chance to reach out to someone. Interest in reaching out trended upward without reaching statistical significance, and actual sending during that window did not rise significantly. The authors add their own hedge: this was a brief test, and other factors may have shaped what people did.

Liu has described her own workaround. "I then tell myself that I would appreciate it a lot if an old friend reached out to me and that there is no reason to think they would not similarly appreciate me reaching out to them," she told PsyPost.

That self-talk is a way of talking herself into the act, and the act is the part of friendship that, for Aristotle, distance interrupts. For him, friendship rests on a disposition but shows itself in activity, and being well-disposed, what he elsewhere calls goodwill, is not enough. The study suggests the disposition on the other side is in better shape than senders fear. None of that can send the message.

The text I deleted is still the last thing I almost said to them, written and taken back, a message nobody received.

Somewhere there is a person who once stood beside me at a co-op stove. If the research holds for the two of us, they would probably be gladder to hear from me than I expect. Aristotle put absent friends alongside sleeping ones, still disposed to friendship but not acting on it, and I keep wondering how long something can sleep before it forgets it was ever awake.