Two people are talking at a business event or orientation week or at a dinner party where they happen to be sitting next to each other. By most objective measures, it went well. Both were engaged, they found something to talk about, at least one or two moments can be described as warm. As they part ways, one of them thinks: I talked too much, I must have received a bad reception, they were probably just being polite.
The other man walks away thinking: that was a really nice conversation.
According to researchers at Cornell, Harvard, Yale and the University of Essex, the discrepancy is no coincidence. This is a consistent and measurable pattern.
What is the like gap?
a published study Psychological Science He led in 2018 Erica BoothbyGus Cooney, Gillian Sandstrom, and Margaret Clark conducted five studies that tested how well people rated how much their conversation partners liked them. Settings ranged from strangers in a lab paired for a 5-minute icebreaker, to college freshmen dating their dorm roommates over the course of several months, to adults in professional development workshops. The finding was the same in each context.
People consistently underestimated how much their conversation partner liked them and enjoyed their company. Not a bit. In moderation. In every conversation, both people rated their partner as more likable than they believed they were rated—which is logically impossible unless at least one person in every conversation made a systematic error. The data showed that this person was almost always a self-esteem person.
Complicating the explanation of this finding was the fact that the researchers also analyzed video recordings of the conversations. The footage showed real positive signals: smiling, eye contact, engaged body language. Participants do not misread distant or disinterested partners. They were misreading the hot ones.
What does an internal monologue do?
Investigators identified the driver. While one person participates in the conversation, the other person simultaneously makes self-critical comments about how the conversation is going—watching their word choices, mentally replaying what they’ve just said, cataloging moments where they feel they’ve fallen.
as Margaret S. ClarkJohn M. Musser, Professor of Psychology at Yale, noted: “They seem to be caught up in their own concerns about what they should say or say in order to see signals that others like them. The person on the other side of the conversation, on the other hand, does not perform the same audit. They take in the conversation instead of evaluating their own performance within it.
Boothby and Cooney described the result: “When it comes to social interaction and conversation, people are often hesitant, uncertain about the impression they leave on others, and overly critical of their own performance.” In this account, the inner monologue is more than just noise. It creates a really inaccurate picture of what is happening in the room.
Why this is contrary to how we usually see ourselves
The liking gap is different for a reason: it goes against the direction of most self-perception biases. In most domains, people rate themselves above average. Better drivers than most. They are more skilled in their work. Less likely than average to experience serious illness or relationship failure. Self-serving bias is one of the most robust findings in social psychology.
But in conversations, the direction is reversed. People are more systematically pessimistic about a domain where direct, real-time feedback is available than almost anywhere else. As Boothby and Cooney note, “Given people’s widespread optimism in other areas, people’s pessimism about conversation is surprising.”
The researchers’ hypothesis is related to social risk. When the evaluation involves another person’s opinion of you, the stakes of overvaluation feel different. A misreading in a social context is not just a personal mistake, but a potential rejection. Clark put it simply: “We’re self-protective pessimists, and we don’t want to assume that someone else likes us before we find out if it’s actually true.”
Protection is understandable. Research shows that its value is that people regularly invest less in relationships that are already going well.
What does not resolve itself
One of the less obvious findings of the study is that the liking gap is not a first impression effect that resolves as people get to know each other. A follow-up study of university roommates showed that this resulted in months of regular contact. Dating, by itself, doesn’t bridge the gap—at least not to the extent that most people assume.
The distortion also appears regardless of the quality of the conversation. People who had an objectively warm exchange—those videotaped with visible engagement and laughter—were just as likely to underestimate their partner’s feelings about them. The problem is not how the conversation goes. It is in the evaluation after that.
I am not a psychologist and this is not professional advice – these are the results of expert research conducted in specific experimental and naturalistic settings and, as the researchers note, cannot be generalized to every relationship context. But what the research suggests is a challenge to one of the more common post-conversation habits: a quiet rejudgment of how things went, which the data almost always shows turned out to be more negative than the other person’s experience of the same exchange.
The person you just talked to probably liked talking to you more than you thought. This finding was replicated across five studies and across multiple countries, and researchers have yet to find a context in which this reliably changes.
If it’s something that’s seriously showing up in your life, talking to a therapist about social anxiety is really worth considering.
