The liking gap: after talking with someone new, people tend to underestimate how much the other person liked them. What five studies found, and the limits.
After talking with someone new, you probably came across better than you think. In a set of five studies published in Psychological Science in 2018, people who had just had a conversation consistently underestimated how much their conversation partner liked them. The researchers call it the liking gap.
The gap is real but not huge, and it has limits. Here is what the studies did, what they found and what they don't show.
What the researchers did
Erica Boothby, Gus Cooney, Gillian Sandstrom and Margaret Clark paired up people who had not met. Afterward, each person rated on a 1 to 7 scale how much they liked their partner, and how much they thought their partner liked them. Because every pair rated each other, the researchers could compare what people guessed with what their partner actually said.
They ran the test in five settings:
- Study 1a: 36 community members near Yale, 5-minute chats in a lab. People liked their partner 5.82 on average but guessed their partner liked them only 5.17.
- Study 2: 84 Yale students and recent grads, 5-minute free chats. Gap of 0.57 points.
- Study 3: 102 people in a Harvard lab, chats from 2 to 45 minutes. Gap of 0.38 points. People also underestimated how much the other person enjoyed the chat.
- Study 4: 100 people at UK "How to Talk to Strangers" workshops, 5-minute introductions. People thought their partner found them less interesting than they found their partner (on a 5-point scale).
- Study 5: 102 first-year dorm suite mates, surveyed 5 times from September to May. The gap showed at the first four check-ins and was gone by May.
A related study (1b) had two coders watch videos of the Study 1a chats. The coders could tell how much people liked each other, which suggests the signals were there and people simply missed them.
Why it happens
In Study 2, people wrote down their most salient thoughts after the chat. Their thoughts about how they came across were more negative than their thoughts about their partner, and that negativity partly explained the gap. In plain terms: we are harder on our own performance than the other person is.
How big is it?
Modest. Across the studies the gap was roughly 0.4 to 0.7 points on a 7-point scale. It showed up again and again, which is why researchers take it seriously, but it does not mean people think you are wonderful while you think they hate you.
What this doesn't show
- It's about new acquaintances. For dorm mates, the gap had disappeared by May, after months of living together.
- Not everyone has it. In Study 1a, people low in shyness showed no significant gap. People high in shyness showed the biggest one.
- The samples were mostly students and young adults in the US, plus one UK community sample.
- The studies were not preregistered, though the data are public on the Open Science Framework (osf.io/dw5fm).
- Later work points the same way. A 2021 study in Psychological Science found the gap in children from about age 5, and a 2026 analysis in the Journal of Social Psychology of a compiled sample of stranger pairs found it was "robust across the individual studies and different conditions within the studies."
The takeaway
If you are replaying a conversation on the way home, the evidence says your partner was probably less critical than you are being. That is a good reason to follow up, say yes to the second coffee or send the message.
Sources
- Boothby, E. J., Cooney, G., Sandstrom, G. M., and Clark, M. S. (2018). The Liking Gap in Conversations: Do People Like Us More Than We Think? Psychological Science, 29(11), 1742 to 1756. DOI 10.1177/0956797618783714. Full text read from the author-hosted PDF, October 23, 2026.
- Wolf, W., Nafe, A., and Tomasello, M. (2021). The Development of the Liking Gap: Children Older Than 5 Years Think That Partners Evaluate Them Less Positively Than They Evaluate Their Partners. Psychological Science. DOI 10.1177/0956797620980754 (abstract).
- Sprecher, S. (2026). Liking gap(s) in getting-acquainted interactions. Journal of Social Psychology. DOI 10.1080/00224545.2025.2576765 (abstract).
