The Liking Gap: Why I Assume People Didn't Enjoy Talking to Me

The liking gap: research finds people underestimate how much others enjoyed talking with them, and self-critical thoughts explain most of it.

Gizella Nagyne Palinkas

11/17/20267 min read

Psychologists call it the liking gap — a measured, repeatable tendency to underestimate how much the person you just spoke to liked you and enjoyed the conversation. In the study that named it, people's partners rated them more warmly than they expected to be rated, and the thing that best explained the gap was not anything the partner did. It was how critical the speaker's own thoughts were about their own performance.

The signals were there. People were too busy reviewing themselves to notice them.

What is actually happening

Walk away from a conversation and two things happen. The other person forms a rough, generous impression and moves on. You start a review.

The review has access to material they never saw: the sentence you meant to say, the joke that landed flatly in your own ears, the pause you thought was too long. You are grading a performance against your own intention. They are not grading anything.

The authors describe this directly — people, they write, "are too focused on their own self-critical thoughts to notice" how much their partner liked them. Their paper calls it a neglected-signal account: the warmth was broadcast, and went unread.

They offer three further reasons in their discussion:

  • Self-criticism may be useful. Picking apart your own performance helps you do it better next time. The listener has no reason to run the same analysis, so only one of you is being harsh.

  • You hold yourself to a higher standard. You are comparing what you said to the ideal version. They are comparing it to nothing in particular.

  • You think your nerves are visible. People overestimate how obvious their own nervousness is to others.

What the research actually found

The study is Boothby, Cooney, Sandstrom and Clark (2018), "The liking gap in conversations: Do people like us more than we think?", published in Psychological Science, volume 29, issue 11, pages 1742–1756.

It is six studies, not one. In the first, 36 community members had roughly five-minute conversations using ice-breaker questions. Partners' actual liking averaged 5.82, while participants' guesses averaged 5.17 — a reliable difference.


The rest of the studies push on it from different angles:

  • It is not just modesty. Trained coders watched video of 15 of those conversations. What the coders could see predicted how much the partner actually liked the person — but not how much the person thought they were liked. The observer and the partner agreed; the speaker was the odd one out.

  • The self-critical thoughts carry the effect. In a second study of 84 people, the negativity of participants' thoughts about their own performance statistically explained a significant part of the gap.

  • Talking for longer does not fix it. Among 102 people with conversations running from 2 to 45 minutes, the gap did not shrink significantly with length. That study found an enjoyment gap too, by a larger margin than the liking gap itself.

  • It can get worse after talking, not better. Among 100 people at "how to talk to strangers" workshops, participants already expected to be found less interesting than they found their partner before talking — and the gap was wider afterwards.

  • It lasts months among people who live together. First-year college suite mates were surveyed at five points across an academic year. The gap was there from September through February, and was no longer statistically significant at the final point in May. The authors say it may be because people finally knew each other well, or because decisions about who to live with next year prompted more revealing conversations. They do not claim to know which.

Where this research is limited

The authors name most of them. The first study was small and collected opportunistically, because nothing like the effect had been shown before. Several samples were capped by how many people were willing to take part. The design and analysis were not preregistered. And the participants skew heavily toward students and workshop attendees, which is not everybody.

The pattern replicated across six studies with different designs, which is the reason to take it seriously. It is still a pattern measured in particular rooms with particular people, not a law.

An everyday example

You leave a work call having spoken for six minutes. On the walk to the kitchen you remember that you said "yeah, exactly" three times and that you cut across someone at the start.

The other four people on the call remember that you sounded like you knew the project.

Neither account is wrong. You had access to the draft; they only ever saw the final version. The research's finding is that this asymmetry is systematic — not occasional bad luck, and not a sign that this particular conversation went badly.

What this does not necessarily mean

This is the part worth reading twice, because a finding like this is easy to over-apply.

It does not mean everyone likes you. The research measured a gap between actual and perceived liking in short conversations between mostly strangers. It did not find that all conversations go well, and it does not mean a specific person who seemed cool toward you was secretly warm.

It is not about assuming people are annoyed with you. The studies measured liking and enjoyment, not irritation. Those are related but not the same, and I am not going to stretch the data to cover the second one.

It is not a diagnosis of anything. It describes a tendency found in ordinary samples of people, not a condition — a pattern rather than a verdict on you or anybody else.

It is not a reason to stop reflecting. The authors suggest the self-critical review may genuinely help you improve. The cost is not the reviewing. It is treating the review as evidence about what the other person thought.

Practical things to try

None of these come from the study — it measured the gap, it did not test fixes.

  1. Name the asymmetry when it starts. "I am reviewing my own draft; they only saw the final version." Putting the mechanism into words makes the review feel less like information.

  2. Separate two questions. Did I say what I wanted to? is worth answering. Did they like me? is the one your evidence is poor on. Answer the first and leave the second alone.

  3. Treat missing warmth signals as unread, not absent. The coder study is the useful one here: outsiders could see the liking that the speaker could not.

  4. Give the review a container. A fixed ten minutes, written down, then closed — rather than an open loop that reopens at 11pm.

  5. Notice if the loop is the real subject. If the replaying is what costs you evenings, the rumination has become the topic. That is a different problem with different tools.

When professional help may be appropriate

Underestimating how you came across is common and, on its own, is not a reason to seek treatment. It is worth talking to a doctor or a qualified therapist if:

  • the reviewing is persistent enough to disturb your sleep or take over your evenings;

  • you are avoiding work, study or social situations you want to be part of;

  • it comes with low mood, or anxiety that lasts beyond the situation that triggered it;

  • or it has changed noticeably from how you usually are.

Mind Decoded explains patterns in everyday behaviour. It does not diagnose anybody, and nothing here substitutes for a clinician who can ask you questions.

Frequently Asked Questions

Q: What is the liking gap in simple terms? A: It is the difference between how much someone actually liked talking with you and how much you think they did — with your estimate being the lower of the two. It was named by Boothby and colleagues in a 2018 paper in Psychological Science. In their first study, partners' actual liking averaged 5.82 while participants guessed 5.17. The effect showed up across six studies using different conversation lengths and settings.

Q: Does the liking gap mean everyone secretly likes me? A: No, and that is the most common over-reading of it. The research found a systematic gap between actual and perceived liking in short conversations, mostly between strangers. It did not find that every conversation goes well, or that a particular person liked you more than they appeared to. The honest takeaway is narrower: your estimate of how you landed is probably biased low, so it is weak evidence rather than a readout.

Q: Why does my own view differ from what an observer sees? A: Because you have information the observer does not. You know what you intended to say, so you grade the gap between intention and delivery. In the study, trained coders watching video could predict how much the partner liked the person, but could not predict what the person thought — the speaker's self-assessment was the outlier, not the partner's judgment.

Q: Does the gap go away if you know the person? A: Eventually, maybe. Among first-year college suite mates surveyed across a year, the gap was present from September to February and was no longer statistically significant by May. The authors explicitly offer two explanations — people had finally got to know each other, or housing decisions for the next year prompted more revealing conversations — and say both could contribute. So it faded in that one setting, over months, for reasons that were not pinned down.

Q: Is this the same thing as social anxiety? A: No. The liking gap is an ordinary estimation bias measured in general samples of people, not a clinical condition, and this article is not in a position to diagnose anything. The 2018 study did note that the gap was larger for shyer participants in its first study — but "larger when shy" is a long way from a disorder. If worry about how you come across is disrupting your sleep, work or relationships, that is the point to speak to a professional.

Take the two-minute check

Is this an ordinary estimation gap for you, or has the reviewing become the main event?

If the replaying is the part that costs you

The Overthinking Worksheets — a seven-page printable: a thought record, a worry-time sheet, an "is this solvable?" sorter, a good-enough decision sheet and a night mind dump. A4 and US Letter, instant PDF. €7.00.

These are self-reflection pages, not therapy and not medical advice.

Sources

Both sources were fetched on 16 November 2026.

  1. Boothby, E. J., Cooney, G., Sandstrom, G. M., & Clark, M. S. (2018). The liking gap in conversations: Do people like us more than we think? Psychological Science, 29(11), 1742–1756. https://doi.org/10.1177/0956797618783714 — read from the author's copy at https://faculty.wharton.upenn.edu/wp-content/uploads/2020/05/The-Liking-Gap-in-Conversations.pdf

  2. Young, E. (2018, 24 September). "Others Like Us More Than We Think, According To New Research On The 'Liking Gap'." BPS Research Digest, British Psychological Society. https://www.bps.org.uk/research-digest/liking-gap

Read more on the blog

Connect

Questions or thoughts? Reach out anytime.

Email

subscribe to our newsletter

hello@our-minddecoded.com

© 2026. All rights reserved.