MAKE CONTACT NOTES
Talk to Strangers: Why We're Wrong About How It'll Go
What a Chicago train experiment, the liking gap, and a decade of research reveal about the conversations we avoid
We expect talking to strangers to be awkward and unwelcome. Research from Chicago commuter trains says we're wrong, and it's costing us connection.

Note: names in the story are changed.
Our parents told us not to talk to strangers. Most of us took it a little too seriously.
Think about the last time you sat next to someone on a train, a plane, or in a waiting room. You probably put your headphones in, looked at your phone, and made sure neither of you had to say a word. I do it too. It feels polite. It feels safe. It feels like what everyone wants.
Last week I caught myself doing exactly that, and decided to test it.
I was at my local cigar lounge, a place I go to regularly. The main room was closed for a charity poker tournament, raising money to help people coming out of incarceration get back on their feet. So I ended up outside on the lawn with a few other people. Every one of us was on our phone.
And I noticed something in my own head. I'm an extrovert. I genuinely like meeting people. This is a place where you're supposed to meet people. And still, there was this quiet resistance: they probably don't want to be bothered. This might be awkward. Just stay on your phone.
So I picked up a little courage and turned to the guy next to me. I'd seen him around; he was the one running the tournament. I asked him what it was all about.
It turned into one of the best conversations I'd had in weeks. Mark turned out to be a hardware engineer at a chip company, and we went deep on tech and AI. An older gentleman sitting next to us, Walt, a retired marketing exec, joined in, and the three of us talked for a long time. Walt didn't always connect with our tech tangents, and I kept trying to pull him back in, which was its own small lesson in how much work a good conversation takes.
Mark and I swapped numbers, and we've even talked about hosting a poker night together in San Francisco. Now every time I walk in, there's a friend there. The friendship is still being built. But it exists, and it only exists because I ignored the voice in my head that said nobody wanted to talk.
That moment stuck with me, because I'd predicted it completely wrong. And it turns out that's not just me. It's one of the most consistent findings in the research on human connection.
The Chicago train experiment
In 2014, University of Chicago researchers Nicholas Epley and Juliana Schroeder ran a simple experiment on commuters heading into the city by train and bus. They split people into three groups. One group was asked to strike up a conversation with a stranger. Another was asked to sit in silence. The third just rode the way they normally would.
The people who talked to a stranger had the most positive commutes. The people who sat in solitude had the least enjoyable ones.
Here's the part that matters: when the researchers asked a separate group of commuters to predict how each option would feel, they predicted the opposite. They expected silence to be the better ride.
Why? Mostly because they assumed other people didn't want to talk. Commuters guessed that more than half of strangers would brush them off. In the experiment, none of the people who tried were rebuffed.
That was me on the lawn. Everyone on their phones, each of us assuming the others wanted to be left alone.
We're bad at forecasting how connection will feel
Psychologists have a name for predicting how future experiences will make us feel: affective forecasting. We do it constantly, and we're often wrong. When it comes to talking to strangers, our forecasts are wrong in a very particular direction. They're too pessimistic.
And the train study isn't the only place this shows up. The research points to three predictions we get wrong, one after another.
Mistake #1: We think the conversation will be unpleasant. That's the train study. We expect awkwardness and rejection. We usually get a better commute.
Mistake #2: We think they won't like us. In 2018, Erica Boothby and colleagues had strangers talk and then asked two questions: how much did you like your partner, and how much do you think they liked you? People consistently underestimated how much they were liked. The researchers called it the "liking gap." It showed up in short conversations and long ones, and among college dorm mates it lasted for months. Shy people had the biggest gap of all.
The reason is simple. After a conversation, you replay your own awkward moments, the joke that didn't land, the pause that went too long. You don't have access to the other person's thoughts, only to their behavior, which was probably warm and perfectly fine.
Mistake #3: We think going deeper will be awkward. This is the one that matters most to me. In 2022, Michael Kardas, Amit Kumar, and Nicholas Epley asked pairs of strangers to talk about either shallow topics or deep, personal ones. Both kinds of conversations went better than people expected. But people overestimated the awkwardness of deep conversations the most, and they felt happier and more connected after the deep ones. The reason? We assume strangers will care less about what we share than they actually do.
Why this matters more than a better commute
On its own, a nicer train ride isn't life-changing. But put these three mistakes together and you get something bigger.
Every friendship you've ever had started as a conversation with a stranger. Every one. Mark was a stranger until I asked him one question. If we systematically predict that those conversations will go badly, that people won't like us, and that going deeper will be awkward, then we're quietly closing the door on relationships before they begin. And when we do talk, we stay on the surface, because we've convinced ourselves that's what the other person wants.
Sitting down and actually talking with the people around you is becoming a lost art. And it's still how friendships start.
That's the connection to everything I've been writing about. In my last note, I wrote about having 12,000 LinkedIn connections and barely knowing most of them. Surface connections are easy to collect. Deep ones require a moment of courage at the start: saying hello, and then saying something real. The research says that moment is far less risky than it feels.
This matters most for anyone starting over: travelers, digital nomads, people who've just moved to a new city. When you're new, strangers are all you have. Every connection you'll make in that place starts with talking to someone you don't know.
A few honest caveats
I want to be fair to the research. These studies were mostly short conversations, mostly in the U.S. Norms about talking to strangers vary a lot between cultures and settings. And in the deep-conversation experiments, participants had agreed to discuss the questions. So the lesson isn't that every stranger wants to chat, or that you should skip reading the room.
The lesson is narrower and more useful: the voice in your head that says "they won't want to talk" or "this will be awkward" is skewed pessimistic. It's usually wrong.
Try it this week
- Say one more sentence than usual. With the barista, the person next to you in line, the neighbor in the elevator. Then notice how it actually felt.
- Ask one real question. Not "what do you do?" but "what's been the best part of your week?" or "what brought you here?"
- Pull someone in. If there's a quiet person on the edge of a conversation, ask them something. It's a small kindness that changes the whole room.
- Assume you were liked more than you think. After a conversation, when you start replaying the awkward parts, remember the liking gap.
- Predict, then check. Before a conversation you're dreading, rate how you think it'll go from 1 to 10. Afterward, rate how it actually went. Keep track. Your forecasts will start to recalibrate.
Why we're doing this at Coffee Club
This research is the whole idea behind Coffee Club, our small, low-pressure meetups for people who want real connection. It's a room full of strangers, which sounds intimidating, until you're actually there.
At our next Coffee Club, we're running the experiment ourselves. We'll ask everyone to predict how the morning will go, then rate how it actually went. I'll share what we find in a future note.
The stranger next to you probably wants to talk more than you think. And they'll probably like you more than you'll ever know.