You have a conversation with someone new, and it seems to go well. They laugh, you find things to talk about, and they say it was nice meeting you.
Then you walk away and start reviewing the footage. Why did I say that? Did I talk too much? Were they only being polite?
Four psychologists ran five studies on that exact spiral and found that it’s not just unpleasant, it’s inaccurate. People consistently walk away from conversations thinking the other person liked them less than they did. They called it the “liking gap”.
It showed up in every setting they tested: five-minute chats between strangers, workshops with ordinary adults, college roommates who shared a suite for months.
The gap is real. But it’s not the same size for everyone, and the people who worry about it most are the ones getting the most distorted picture.
The liking gap shows up every time researchers look for it
The setup was simple. Two strangers sat down, talked for a few minutes, then went to separate rooms and answered two sets of questions.
The first was about the other person: did you like them, did you enjoy the conversation, could you see being friends? The second set flipped those same questions around: did they like you?
Almost everyone rated the other person higher than they thought the other person rated them.
Both people walked away thinking they’d been liked less than they’d liked the other person, which can’t be true for both of them at once. Someone has to be wrong, and across study after study, almost everyone was.
Part of what drives the gap is something called the spotlight effect: the tendency to overestimate how much other people are paying attention to your mistakes. You remember the joke that fell flat. They remember that you were funny.
You’re cataloguing your worst moments while they’re forming an overall impression, and those are two very different things happening at the same time.
The researchers even recorded conversations and had outside observers watch the tapes.
Those observers could tell who liked whom. The warmth was visible on camera. The only person who couldn’t see it was the one sitting in the conversation, too busy grading their own performance to notice it coming back at them.
Insecurity makes the gap wider, not the conversation itself
If the gap were just a quirk of first impressions, it would be interesting but not especially important. What makes it matter is that it doesn’t hit everyone equally.
A 2025 study led by Hasagani Tissera at the University of Toronto looked at this across three separate samples totaling 2,753 people.
These ranged from two-minute chats to twenty-five-minute calls and, afterward, everyone answered the same kinds of questions Boothby’s team had used, plus measures of things like self-esteem, social anxiety, and neuroticism.
Tissera’s team found that the size of the liking gap depends on who you are.
If you already have lower self-esteem, your gap is wider. If you tend toward social anxiety or neuroticism, your gap is wider. And this wasn’t a small pattern: it showed up the same way across all three samples, with thousands of people.
What makes it sting: how much the other person liked you didn’t change based on any of those things. The insecure person and the confident person were liked about the same amount. The difference was entirely in what they believed afterward.
One person walked away thinking that went fine. The other walked away thinking I was probably too much. Same conversation, same warmth directed at them, completely different story in their heads.
More Bolde Stories
Past experience may matter more than present anxiety
You might assume that social anxiety is the main thing driving a wider gap. If you’re more nervous in conversations, you misread them worse. That’s the intuitive version.
A small 2025 study led by Isabel Leiva at Temple University tested that by bringing in people who ranged from low to high social anxiety and who also differed in how much peer rejection they’d experienced: some had been excluded or shut out more than others.
Everyone sat down with a stranger and talked.
The liking gap showed up, as expected. But what predicted its size wasn’t how anxious someone felt. It was how much peer rejection they’d experienced. The people who had been treated worst by peers were the ones who most underestimated how much their conversation partner enjoyed talking to them.
That’s a small sample and one study, so it’s thin on its own. But the direction is worth noticing. It suggests the gap may be shaped less by how nervous you feel right now and more by what your social experiences have taught you to expect.
If you’ve spent time being the person who gets excluded or dismissed, your brain builds a model of how conversations go. That model says: they’re being polite now, but they’ll stop calling. And it doesn’t update easily, even when the person across from you is smiling and asking follow-up questions and showing every sign that they mean it.
The evidence you’re ignoring is the evidence that matters
Your post-conversation review feels like evidence. You can name three things you wish you’d said differently. You can picture the moment you talked over them, or the pause where you couldn’t think of anything to say.
But the review is not a neutral replay. It’s a highlight reel edited by the part of your brain that is most focused on what didn’t land. And it leaves out the signals that were there.
They asked you a question back. They laughed at something you said. They stayed in the conversation longer than they had to. They brought up something you’d mentioned earlier, which means they were paying attention.
Those are all signs that it went well, and the research suggests people consistently underweight them because they don’t feel as vivid as the fumble you keep replaying.
The fumble is louder. That’s just how memory works. But loud doesn’t mean important, and the quiet evidence is usually the more accurate kind.
Sometimes the conversation was awkward, and that’s real too
Not every misread is a distortion. Sometimes it was awkward. Sometimes you did talk too much, or the chemistry was off, or neither of you could find a rhythm. Not everyone likes everyone, and that’s normal.
The liking gap doesn’t mean assuming every stranger secretly adored you.
It means your immediate they probably didn’t like me reaction is not especially reliable, because people appear to be systematically pessimistic about the impression they make, and the people who are most pessimistic are the ones who are most wrong.
You’re replaying the conversation on the walk home. You’re stuck on the sentence you wish you could take back. There’s a decent chance the other person isn’t thinking about that sentence at all. They’re thinking they liked talking to you, which, according to the research, is the part people are strangely bad at believing.
