The Liking Gap

At a Glance

1. Quick Summary

After a first conversation with a stranger, most people walk away thinking it went worse, from the other person's point of view, than it actually did. Ask both partners how much they liked each other, and each rates the other higher than they guess the other rated them. Erica Boothby and her colleagues named this pattern the liking gap in 2018 and found it with strangers in a lab, with first-year students sharing a dorm suite for most of a year, and with adults at a personal-development workshop. The gap appears because people spend the conversation monitoring their own slips and silences, and that self-critical commentary crowds out the smiles, laughs and follow-up questions that show the other person is having a good time. It shows up in children by about age five, lingers in work teams for weeks, and quietly steers people away from asking for help, following up with new acquaintances, or starting conversations at all.

Simply About the Bias

The liking gap means that after you talk to someone new, you usually think they liked you less than they really did. You remember the awkward bit where you said something odd. They remember that you were nice to talk to.

Picture the first day at a new school. You sit next to someone at lunch and talk about a game you both play. On the bus home you think, "I talked too much. I laughed at the wrong moment. They probably thought I was weird." Meanwhile, that same kid is at home telling their parents they met someone cool at lunch. You both liked each other. Only one of you knows it.

Here is a trick. When you catch yourself replaying a conversation and grading yourself, remember that the other person was busy worrying about their own performance too. They were not keeping score of your mistakes. Then do one small thing: say hi to them again tomorrow. Most of the time, they are glad you did.

2. The Science Behind It

2.1. Discovery and History

Psychologists had known for decades that people are poor at reading how specific others see them. In 1993 David Kenny and Bella DePaulo reviewed the evidence on meta-perception and concluded that people mostly guess how others view them from how they view themselves. Through the late 1990s Thomas Gilovich's group at Cornell documented two cousins of the liking gap: the illusion of transparency (we think our nervousness shows more than it does) and the spotlight effect (we think others notice our gaffes far more than they do). Both rest on egocentrism. We are the center of our own experience, so we assume we are near the center of everyone else's.

Whether people know how much a new acquaintance likes them was not tested directly until Erica Boothby, Gus Cooney, Gillian Sandstrom and Margaret Clark published "The liking gap in conversations: Do people like us more than we think?" in Psychological Science in 2018. Across five studies with lab strangers, dorm suitemates and workshop attendees, people liked their partners more than they believed their partners liked them. The gap did not shrink with longer conversations, and while shyer participants showed it more strongly, participants of average shyness showed it too.

Two 2021 papers turned a striking finding into a reasonably solid one. Wouter Wolf, Amanda Nafe and Michael Tomasello, in a different lab, ran a preregistered study with 241 children and found that the gap is absent at four and present from five. Andrew Mastroianni, Cooney, Boothby and Andrew Reece extended the effect to groups and work teams. A fair reading today: the liking gap is a well-replicated effect in American samples, with one independent preregistered replication and no published meta-analysis. The popular claim that "everyone likes you more than you think" overstates it. The honest claim is that, on average and after a first meeting, people's guesses about how they came across run below the truth.

Key milestones:

• 1993: Kenny and DePaulo review meta-perception research and find that people barely know how specific others see them.
• 1998 to 2000: Gilovich, Savitsky and Medvec document the illusion of transparency and the spotlight effect.
• 2014: Epley and Schroeder show that Chicago commuters who talk to strangers enjoy the trip more than those who sit alone, despite predicting the opposite.
• 2018: Boothby, Cooney, Sandstrom and Clark publish the five-study paper that names the liking gap.
• 2018: Bruk, Scholl and Bless describe the "beautiful mess effect," a parallel bias about showing vulnerability.
• 2021: Wolf, Nafe and Tomasello find the gap emerging at about age five in a preregistered study of 241 children.
• 2021: Mastroianni, Cooney, Boothby and Reece show the gap in groups and teams, lasting weeks.
• 2021: Sandstrom and Boothby publish a mini meta-analysis of predicted fears versus actual experiences when talking to strangers.

2.2. Key Researchers

2.3. Landmark Studies

"The liking gap in conversations" (Boothby, Cooney, Sandstrom & Clark, 2018)

Two strangers are given a few ice-breaker questions and left to talk. Afterward each privately rates how much they liked their partner and how much they think their partner liked them. Comparing each person's guess with the partner's actual rating gives the gap, and it was consistent: estimates of being liked fell below actual liking, whether the conversation lasted a couple of minutes or the better part of an hour. When participants listed the thoughts that ran through their heads during the conversation, many were self-critical (sounding boring, saying the wrong thing), and the more self-critical the thoughts, the larger the gap. The authors' interpretation was that partners were sending signals of liking that went unnoticed or were discounted because attention was pointed inward.

The dorm study moved out of the lab. First-year students assigned to share a suite rated one another in September and at several points through the year. The gap was there at the start and lasted most of the year, closing only in May, around the time suitemates had to decide whether to live together again and presumably said what they thought of one another. A final study with attendees at a personal-development workshop found the same pattern.

"The development of the liking gap" (Wolf, Nafe & Tomasello, 2021)

In a preregistered study, 241 American children aged four to eleven each played and chatted with an unfamiliar adult, then answered simple questions about how much they liked the partner and how much they thought the partner liked them. Four-year-olds showed no gap. From age five onward children believed their partner evaluated them less positively than they evaluated the partner, and the gap grew with age. Five is also roughly when children begin to understand that others form opinions about them, which fits the idea that the gap rides on concern about evaluation rather than on general pessimism.

"The liking gap in groups and teams" (Mastroianni, Cooney, Boothby & Reece, 2021)

Three studies asked whether the pattern survives with more than two people in the room. In small groups of strangers, each person underestimated how much the others liked them. In student engineering teams working on a semester project, the gap persisted across weeks of collaboration. Among working adults, people who believed their teammates thought less of them were less likely to ask for help, less willing to communicate openly, and felt less included. The gap was largest between peers and depended partly on how much a person dwelt on the negative parts of the impression they thought they had made.

2.4. Neurological Basis

There are no brain-imaging studies of the liking gap itself, so what follows is inference from better-established findings rather than a direct result.

Thinking about oneself and thinking about what others think overlap heavily in the brain. Both draw on the medial prefrontal cortex and surrounding midline regions of the default mode network, the circuitry active when we reflect, remember and simulate other minds. That overlap is one reason egocentric errors are so common: our model of what a partner thinks of us is built from the same material as our model of ourselves, and a self-critical self-model leaks straight into the guess.

The cognitive model of social anxiety developed by David Clark and Adrian Wells adds the attentional piece. In evaluative situations, anxious people turn attention inward, monitor their own performance, and construct an image of how they appear that is far harsher than what observers report. That is a close description of the liking-gap mechanism in its everyday, non-clinical form. Attention spent on self-monitoring is not available for reading the partner's face, and the brain's well-documented tendency to weigh social threat more heavily than social reward means the pauses get stored while the smiles do not.

3. Evolutionary Origins

For most of human history, being liked by the people around you decided whether you ate, whether anyone helped you when you were ill, and whether you were still in the group next winter. Exclusion was close to a death sentence, and the cost of being wrong about your standing was lopsided. Overestimating how well-liked you were could get you expelled; underestimating it cost you some extra effort at being agreeable. A mind tuned by that asymmetry would set its default guess about social standing a notch below the truth.

This is the standard error-management argument, and it explains why the liking gap runs in one direction rather than scattering randomly. A system that errs toward "they may not be sure about me yet" keeps a person attentive, polite and willing to do a little more for the group. The age at which the gap appears in children, around five, is also roughly when children begin to care about their reputation, which suggests the two capacities travel together.

Self-monitoring has its own logic. In small groups where everyone knew everyone, a first meeting with a stranger was rare and high stakes, and reviewing your own behavior for mistakes was a sensible use of attention. First meetings now happen constantly, and a reviewing habit built for rare, consequential meetings now runs after every coffee. The running total of imagined bad impressions becomes a reason to avoid the next one.

4. How This Bias Manifests

4.1. In Everyday Life

• Leaving a party convinced you were boring, while the hosts tell each other you were good company
• Not texting a new acquaintance because "they probably don't remember me," when they were waiting for your message
• Replaying a conversation afterward and fixing on the one awkward pause rather than the twenty minutes that flowed
• Skipping the small talk in a queue or on a flight because you expect the other person to want silence

4.2. In the Workplace

• New hires who believe the team finds them underqualified and hold back questions during the weeks when questions are cheapest
• Junior employees who do not ask a colleague for help because they think the colleague found them annoying last time
• People who skip the follow-up after a networking conversation that went well, because they read the other person's enthusiasm as courtesy
• Interviewees who walk out sure they bombed and stop preparing for the next round the interviewer has already decided to offer

4.3. In Business and Marketing

• Salespeople who misjudge a warm first meeting as lukewarm and either over-chase or give up on a prospect who was ready to buy
• Founders who assume investors were unimpressed and never send the follow-up deck that was expected
• Freelancers who underprice because they believe a client was only mildly satisfied with a meeting the client later described as a highlight
• Referral programs that underperform because customers assume their friends would not want to hear from them

4.4. In Politics and Media

The gap is a quiet contributor to the loneliness that surveys in many countries have reported over the past decade. People who believe others do not much enjoy their company stop initiating, and initiation is where most new friendships begin. Media feed a distorted picture on top of it. Deri, Davidai and Gilovich found in 2017 that people believe others have richer social lives than they do, partly because the visible evidence comes from the most social moments of the most social people. Add a liking gap, and a person sees everyone else connecting easily while believing their own attempts land flat. Civic life suffers too: people who assume their neighbors are indifferent to them are less likely to knock on doors, join a local group, or speak at a meeting.

4.5. In Healthcare

• Patients who leave an appointment believing the doctor found them tiresome and skip the follow-up question they needed to ask
• Clinicians who misread a patient's reserve as dissatisfaction and become more guarded, which makes the next appointment worse
• Trainees who think senior physicians dislike them and avoid the informal conversations where much clinical judgment is passed on
• Support-group newcomers who assume the group did not warm to them and never return, when the group hoped they would

4.6. In Finance and Investing

• Advisers who misjudge a positive first meeting with a prospective client and follow up hesitantly, or not at all
• Clients who think their adviser finds their questions naive and stop asking, so misunderstandings about risk go uncorrected
• Deal teams who read a counterpart's normal reserve as coolness and make unnecessary concessions to win approval
• Investors who avoid asking a peer for a sanity check on a position because the peer "probably thinks I'm an amateur"

5. Real-World Case Studies

Case Study 1: The Suitemates Who Waited Until May

• Context: In the 2018 Boothby et al. paper, first-year students were followed from the week they moved into shared dorm suites through the end of the academic year, rating how much they liked each suitemate and how much they believed each suitemate liked them.
• What happened: In September, students rated their suitemates higher than they believed their suitemates rated them. The pattern held through winter and spring and closed only in May, when students had to decide housing for the following year and, presumably, finally said out loud what they thought of one another.
• The bias at work: Eight months of shared bathrooms and hallway conversations were not enough to correct the initial guess. Liking was being communicated in everyday behavior, but each student kept reading their own side of the relationship as the weaker one.
• Consequences: Students who believed they were less liked than they were had less reason to suggest plans or raise small problems before they grew. The 2018 paper did not track those behaviors, but the 2021 teams research did, and found exactly that pattern.
• Lessons learned: Time alone does not close the gap. Information does. A direct exchange of impressions, which feels risky, is often the only thing that updates a months-old misreading.

Case Study 2: The Engineering Teams

• Context: Mastroianni, Cooney, Boothby and Reece (2021) studied student engineering teams working on a semester project, along with groups of strangers and working adults in teams.
• What happened: Team members underestimated how much their teammates liked them, and the underestimate was still there weeks into the project. Among the working adults, people with the largest gaps were less likely to ask teammates for help, less willing to communicate openly, and felt less included.
• The bias at work: The gap was widest between peers, the people whose evaluation matters most and about whom we ruminate most. The more a person dwelt on the negative parts of the impression they thought they had made, the larger their gap.
• Consequences: Help that would have been freely given went unrequested. Problems that a quick honest message would have solved stayed private. The team's effective capacity was smaller than its actual capacity, and no one could see why.
• Lessons learned: Team leads should not wait for members to feel liked before expecting them to speak up. The feeling lags the reality by weeks. Saying out loud that a contribution was valued, specifically and early, corrects a misreading that otherwise persists.

Historical Example: The Chicago Commuters (2014)

Nicholas Epley and Juliana Schroeder recruited commuters on Chicago's Metra trains and city buses and asked them, before the trip, to predict how they would feel if they talked with a stranger versus sat alone. Most predicted that solitude would be the more pleasant ride and that a stranger would not want to talk. They were then randomly assigned to one of those conditions, or to a control, and reported their experience afterward.

The people assigned to talk with a stranger reported the most positive commute of the three groups. Almost nobody who tried to start a conversation was refused. Each commuter assumed the stranger did not want to talk while the stranger assumed the same about them.

This is the pre-conversation half of the liking gap. Before talking, people assume the other person will not enjoy it; after talking, they assume the other person did not enjoy it as much as they did. A later program by Sandstrom and colleagues that had people in London talk to a stranger every day for a week found that fear fell and confidence rose with practice, which is what you would expect if the belief is wrong and experience is the correction.

6. The Cost of This Bias

6.1. Personal Costs

• Friendships that never form because neither person makes the second move
• Hours spent replaying conversations and grading them, a known contributor to low mood and social anxiety
• Loneliness that persists despite plenty of pleasant interactions, because the interactions are never counted as evidence of being liked
• A self-image of being socially clumsy that the evidence does not support

6.2. Professional Costs

• Help not asked for and feedback not sought, so that avoidable mistakes get made
• Networking contacts left to go cold after meetings both sides enjoyed
• Lower felt inclusion in teams, which the 2021 study linked directly to the gap
• Weaker mentoring, since the junior person usually has to initiate and is the one most likely to assume the senior person is not interested

6.3. Societal Costs

• Fewer weak-tie interactions, which Sandstrom and Dunn showed in 2014 are an underrated source of daily well-being and belonging
• Thinner community life, since neighborhood groups, volunteer work and local politics all depend on people assuming they are welcome
• Public loneliness figures that are partly a measurement of this bias: people reporting that they lack connection while surrounded by willing connectors
• Underused institutions, from libraries to clinics to support groups, whose newcomers do not come back

6.4. Statistical Impact

• The original 2018 paper reported the gap in all five studies, across conversations from a few minutes to about forty-five minutes, with no shrinkage as conversations got longer.
• Among first-year dorm suitemates, the gap persisted for most of an academic year and closed only in May.
• In the 2021 preregistered study of 241 children, the gap was absent at age four, present from five, and larger among older children.
• In the 2021 teams research, larger gaps went with less help-seeking, less open communication and lower felt inclusion among working adults.
• In Epley and Schroeder's 2014 commuter experiments, participants predicted that solitude would beat conversation; the reverse was true, and refusals of attempted conversations were rare.
• No published meta-analysis of the liking gap exists as of this writing, and the samples are overwhelmingly American.

7. The Hidden Benefits

A bias that makes people think they are a little less liked than they are is not all cost.

• It keeps people trying. Assuming the relationship is not yet secure makes a person attentive, generous and quick to repair small slights. Many of the behaviors that make someone well liked are driven by the fear that they are not.

• It guards against the opposite error. The person who assumes everyone adores them is harder to be around than the person who assumes mild indifference. The liking gap errs in the safer direction.

• It fits modesty norms. In most cultures, acting as though you are a favorite is penalized. A default of slight underestimation produces behavior that looks like humility, which others reward.

• It makes correction pleasant. Learning that someone liked you more than you thought is a good feeling, and people who routinely underestimate get it regularly, as long as they stay in contact long enough to receive it.

The benefits depend on continuing to show up. The cost arrives when the gap becomes a reason to withdraw, at which point the self-protective guess never gets tested and hardens into belief.

8. Self-Assessment: Do You Have This Bias?

8.1. Warning Signs Checklist

• After meeting someone new, you usually assume they found you less interesting than you found them
• You replay conversations afterward and focus on the moments that went badly
• You often decide not to follow up with a new acquaintance because "they probably don't want to hear from me"
• When someone is warm to you, you explain it as politeness rather than liking
• You avoid starting conversations with strangers because you expect them to prefer being left alone
• You are regularly surprised when someone says they enjoyed talking with you
• In group settings you feel less included than the group's behavior toward you would suggest
• You hesitate to ask colleagues for help because you think they already find you a nuisance
• You can list the awkward things you said at your last social event but not the things other people laughed at
• You believe other people make friends more easily than you do

Scoring:

• 0-2 checked: Low susceptibility
• 3-5 checked: Moderate susceptibility
• 6-8 checked: High susceptibility
• 9-10 checked: Very high susceptibility

8.2. Self-Reflection Questions

1. Think of the last three new people you met. How much do you guess each enjoyed the conversation, and what evidence, other than your feelings about your own performance, supports that guess?
2. When you replay a conversation, whose mistakes are you reviewing? Can you recall any awkward thing the other person said?
3. Has anyone ever told you they liked you from the start, or wished you had reached out sooner? What did you believe about their view of you before they said it?
4. How many relationships in your life began with the other person making the second move?
5. If you assumed a new acquaintance liked you about as much as you liked them, what would you do differently this week?

8.3. Quick Diagnostic Scenario

Scenario: You spend twenty minutes talking with another parent at a school event. The conversation moves easily; you both laugh a couple of times and discover you live a few streets apart. Near the end you make a joke that lands flat, and the other parent glances at their phone before saying goodbye. The next day you consider sending a message suggesting coffee.

What do you do?

• A) Decide against it. The flat joke and the phone glance tell you they were being polite and would find the invitation awkward → High susceptibility
• B) Wait to see if they message first, and reply only if they do → Moderate susceptibility
• C) Send the message, on the grounds that you enjoyed the conversation, they showed every ordinary sign of enjoying it too, and the phone glance most likely meant they were checking the time → Low susceptibility

9. Identifying This Bias in Others

9.1. Behavioral Indicators

• Does not follow up after meetings that visibly went well
• Apologizes for taking up your time at the start and end of conversations
• Responds to warmth with mild surprise, or deflects compliments about their company
• Waits to be invited rather than inviting, even with people they clearly like
• Seems to enjoy an interaction in the moment and then describes it afterward as having gone badly

9.2. Conversational Red Flags

Phrases people say when under this bias:

• "I'm sure they were just being nice."
• "They probably don't even remember me."
• "I don't want to bother them."
• "That was so awkward, I can't believe I said that."
• "I'll reach out if they reach out."

Types of reasoning they use:

• Treating one flat moment as the verdict on the whole conversation
• Explaining the other person's warmth by anything except liking

Questions they avoid asking:

• "What did the other person actually do that suggests they didn't enjoy it?"
• "If a friend described this conversation to me, what would I tell them?"

9.3. Situational Triggers

• First meetings of any kind, where the gap is largest
• Evaluation contexts: interviews, first dates, meeting a partner's family, the first weeks of a job
• Conversations with peers, where the 2021 teams research found the gap widest
• Being the newcomer in an established group
• Social anxiety, shyness or a recent rejection, which all amplify self-critical monitoring
• Interactions across a status or cultural difference, where the other person's signals are harder to read

10. Cognitive Debiasing Strategies

10.1. Immediate Techniques

• Point the camera outward. During a conversation, give yourself the job of noticing the other person: what they enjoy talking about, when they laugh, what they ask about. Attention spent there cannot be spent grading yourself.
• Count their side of the ledger first. Afterward, before reviewing your own mistakes, list three things the other person did that suggested interest: a follow-up question, a laugh, a mention of meeting again.
• Apply the symmetry correction. Ask how much you liked them, then assume, as a working rule, that they liked you about that much. This guess will be closer to the truth than your instinct.
• Reinterpret the flat moment. A pause, a phone glance or a short answer has many causes. Dislike is rarely the most likely one.
• Send the message anyway. The cheapest test of whether someone enjoyed your company is to propose seeing them again. A polite no costs little, and you will get a yes more often than you expect.

10.2. Long-Term Strategies

• Keep a prediction log for social encounters. After a first meeting, note how much you think the person liked the conversation on a 1 to 7 scale. When evidence arrives (they followed up, introduced you to someone, mentioned the conversation), record it. Most people find their predictions run low, and seeing the pattern in your own handwriting is more persuasive than reading about it.
• Practice with strangers deliberately. Sandstrom and colleagues found that a week of daily small conversations with strangers lowered fear and raised confidence. Treat it as training rather than as a test of your worth.
• Separate performance from reception. You can judge your own conversational performance harshly and still accept that the other person had a good time. The two are far less connected than they feel.
• Learn what liking looks like. Interest shows in questions, in continuing a topic you raised, in staying longer than necessary, in remembering details later.

10.3. Environmental Design

• Build in explicit feedback. Teams, classes and groups can normalize short rounds in which members say what they valued in each other's contributions. Without an explicit exchange, the dorm study suggests, the gap can last the better part of a year.
• Make second contacts easy. Event organizers can share attendee lists or set up follow-up channels, so the second move does not depend on someone overcoming their own misreading.
• Decide the follow-up in advance. Before an event, commit to messaging one person within twenty-four hours. A rule made beforehand does not have to argue with the post-event mood.
• Reduce self-monitoring load. Choose settings for first meetings where there is something to do together (a walk, a task, a game). Shared attention on an activity leaves less for the inner critic.

10.4. When to Seek External Input

• When you are about to end or avoid a relationship on the basis of how you think the other person feels, without having asked
• When your social world is shrinking and you cannot point to anyone who has actually rejected you
• When replaying conversations takes up hours or affects sleep, which can indicate social anxiety that responds well to cognitive behavioral treatment
• When a trusted friend who was present describes an event very differently from how you remember it

Whom to ask: a friend who was there, who can report what they saw rather than what you felt. Or, where the stakes justify it, the person themselves. "I enjoyed that conversation, did you?" is a question most people are glad to be asked.

11. Practical Exercises

Exercise 1: The Two-Column Review

• Objective: Retrain post-conversation review so it collects evidence rather than grading performance
• Time required: 5 minutes after any first or early conversation
• Materials needed: Notebook or phone notes
• Difficulty level: Beginner
• Instructions:
  1. Draw two columns. Left: "What they did." Right: "What I did."
  2. Fill the left column first, with specifics: questions they asked, times they laughed, topics they extended.
  3. Only then fill the right column, including things that went well.
  4. Estimate, 1 to 7, how much they enjoyed the conversation, based on the left column alone.
  5. Write your gut estimate, 1 to 7. Note the difference.
• Reflection questions:
  • Which column was easier to fill, and why?
  • How far apart were the evidence-based estimate and the gut estimate?
  • What would you have concluded if you had only done the right column, as usual?
• Frequency: After every new-person conversation for three weeks, then whenever you notice yourself ruminating

Exercise 2: The Follow-Up Experiment

• Objective: Test your predictions about how welcome you are against reality
• Time required: 10 minutes per week plus the conversations themselves
• Materials needed: A list of recent acquaintances, a simple log
• Difficulty level: Intermediate
• Instructions:
  1. List five people you have met in the past few months and have not contacted since.
  2. For each, predict: will they respond warmly, neutrally, or not at all?
  3. Send each a short, specific message: a reference to something you talked about plus a low-pressure suggestion.
  4. Record what happens over the following two weeks.
  5. Compare predictions with results.
• Reflection questions:
  • How many responses were warmer than predicted?
  • What did you tell yourself about the people who did not respond, and is that the only explanation?
  • What stopped you sending these messages before?
• Frequency: Once a month until predictions and results line up

Exercise 3: The Stranger Week

• Objective: Reduce the pre-conversation half of the bias through direct experience
• Time required: A few minutes a day for seven days
• Materials needed: None
• Difficulty level: Intermediate to advanced, depending on temperament
• Instructions:
  1. Each day, start one short conversation with a stranger: in a queue, at a shop, on public transport.
  2. Before each one, predict on a 1 to 7 scale how it will go and how willing the other person will be.
  3. Afterward, rate how it went and how the other person seemed.
  4. Keep it small. The goal is practice, not a new friend.
  5. At the end of the week, lay the fourteen numbers side by side.
• Reflection questions:
  • Did any stranger react the way you feared?
  • Did the predictions move over the week?
  • How did the conversations feel compared with the anticipation?
• Frequency: One week per quarter, or whenever avoidance is creeping back

Daily Practice

One signal, named

Once a day, after any conversation, name one specific thing the other person did that showed interest. Say it in a full sentence: "She asked a second question about the trip." Over time this builds the habit of seeing the signals the bias discounts.

• Suggested duration: 1 minute
• Best time of day: Right after the conversation, before the replay starts
• How to track progress: A one-line daily note; look back weekly for patterns

Weekly Challenge

Make the second move

Each week, be the one who follows up with one person you would normally wait on. A message, an invitation, a question. Note your prediction before sending and the outcome after.

• Expected outcomes after 4 weeks: A stack of evidence about how welcome your contact is; at least one relationship that would otherwise have gone cold; a shorter post-conversation replay
• Journaling prompts for reflection:
  • Which follow-up was hardest to send, and what was I predicting?
  • What happened instead?
  • What would four more weeks of this do to my social life?

12. For Specific Audiences

For Leaders and Managers

• Assume your team underestimates your regard for them. Felt inclusion lags actual inclusion. Say the specific thing out loud: what the person did, why it mattered.
• Watch for help not asked for. A team member who never asks for anything may be managing a liking gap rather than coping well. Lower the cost of asking by asking them for something first.
• Onboard with explicit welcome. New hires will not infer from ordinary friendliness that they are doing well. Tell them in the first week, and again at the end of the first month.

For Parents and Educators

• Age-appropriate explanation: "After you talk to someone new, your brain tells you they liked you less than they really did. It does this to almost everyone. So when you think a new kid didn't like you, remember that your brain is probably being unfair, and say hi again tomorrow."
• Know the timeline. The gap appears around age five and grows through childhood, exactly the period when children form their first friendships. A child who says "nobody likes me" after a first day at a new school is reporting the bias as much as the facts.
• Teach the signals. Help children notice what liking looks like: someone saves you a seat, asks about your weekend, wants to play again. Then encourage the second move; invitations are more welcome than children expect.

For Healthcare Professionals

• Your patients think you like them less than you do. Reserve in a patient often reflects their guess about your view of them, and it suppresses the questions you need them to ask. A sentence of explicit warmth at the start of a visit changes what they are willing to say.
• Social anxiety widens the gap. For anxious patients the self-focused attention behind the gap is a direct target of cognitive behavioral therapy, and behavioral experiments that test predictions against outcomes suit it well.
• Group programs lose newcomers to this bias. Build a deliberate second-contact step into support groups and rehabilitation programs.

For Financial Professionals

• Follow up as if the meeting went well, because it probably did. Advisers lose prospects by reading a positive first meeting as neutral and following up hesitantly.
• Make it cheap for clients to ask. Clients who think their questions seem naive stop asking, and stop understanding their own portfolios. Say plainly that questions are welcome, then prove it by answering the first one warmly.
• Read counterparts' reserve correctly. In negotiations, a cool manner is more often a tactic or a temperament than a judgment of you. Concessions made to win approval that was already there are pure loss.

13. Interactions with Other Biases

Biases That Amplify This One

Biases That Counteract This One

Common Bias Chains

Liking Gap → Spotlight Effect → Confirmation Bias → Social withdrawal

After a first meeting, a person underestimates how much the other enjoyed it (Liking Gap). Reviewing the conversation, they fix on the moment they fumbled a name and assume the other person noticed and judged it (Spotlight Effect). When no message arrives, the silence confirms the verdict, and a later friendly greeting is filed as mere politeness (Confirmation Bias). They stop initiating, and a relationship both people would have welcomed never happens.

Interrupting the chain: The cheapest break is at the start, by sending a follow-up before the review hardens. The second-best is at the confirmation step: treat silence as uninformative, since the other person is very likely running the same chain in reverse.

14. Cultural Perspectives

Almost all liking-gap research comes from the United States, and nobody has yet published a careful cross-cultural comparison, so what follows is informed inference rather than settled finding.

The mechanism, self-critical monitoring of one's own performance in front of a new person, has obvious cultural variation in its inputs. Cultures with strong modesty norms, where talking about yourself positively is discouraged, may push people toward harsher self-assessment. Steven Heine and colleagues found that East Asian participants are less inclined than North Americans to inflate their own abilities, and it is plausible that the same pattern produces a larger liking gap. On the other hand, high-context communication styles make interest and warmth easier to read for insiders, which could shrink the gap among people who share the code and widen it across cultural lines. Cultures with a norm of direct expression (saying "I enjoyed this, let's do it again" rather than leaving it implied) give people the explicit feedback that closed the gap among the dorm suitemates in May.

For anyone meeting people across cultural lines, the gap is probably wider than usual, because the ordinary signals of liking are harder to read and the "just politeness" explanation is easier to reach for.

15. Myths and Misconceptions

16. Expert Insights

Erica Boothby, Gus Cooney, Gillian Sandstrom and Margaret Clark found across several studies that after a conversation people underestimate how much their partners liked them and enjoyed their company, a pattern they named the liking gap. — Summary of Boothby, Cooney, Sandstrom & Clark, "The Liking Gap in Conversations: Do People Like Us More Than We Think?" Psychological Science 29(11) (2018), pp. 1742-1756 https://doi.org/10.1177/0956797618783714

Commuters expected that sitting in solitude would be more pleasant than talking with a stranger, and that strangers would not want to talk. Both expectations were wrong; those who struck up conversations had the most positive commute. (paraphrasing Epley and Schroeder, 2014)

People are liked more than they know, and the people most likely to underestimate it are the ones who dwell on the negative parts of the impression they think they made. (paraphrasing Mastroianni, Cooney, Boothby and Reece, 2021)

We are so focused on our own appearance and behavior that we assume everyone else is focused on them too, when in fact others are mostly absorbed in themselves. (paraphrasing Gilovich, Medvec and Savitsky on the spotlight effect, 2000)

17. Key Takeaways

1. Your guess runs low. After a first conversation, the average person underestimates how much the other person liked them and enjoyed it. Adjust upward.

2. The cause is where your attention went. The gap grows with self-critical thoughts about your own performance. While you were grading yourself, the other person was sending signals you did not count.

3. It starts early and lasts. The gap appears in children by about age five, and among new acquaintances it can persist for months without an explicit exchange of impressions.

4. It changes behavior, not just feelings. In teams, larger gaps go with less help-seeking, less open communication and lower felt inclusion. Relationships both sides want die of nobody making the second move.

5. The evidence is good but bounded. Five original studies, a preregistered replication in a different lab, and a teams extension, all in American samples, with no meta-analysis yet. Trust the direction; do not treat it as a law.

6. The fix is outward attention plus testing. Notice the other person's signals, count them before reviewing your own mistakes, and send the follow-up. Practice with strangers works.

7. Silence is not a verdict. The other person is probably running the same calculation about you. Whoever breaks the silence usually finds a welcome on the other side.

18. Further Resources

Academic Papers

• 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
• Wolf, W., Nafe, A., & 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, 32(5), 789-798. https://doi.org/10.1177/0956797620980754
• Mastroianni, A. M., Cooney, G., Boothby, E. J., & Reece, A. G. (2021). The liking gap in groups and teams. Organizational Behavior and Human Decision Processes, 162, 109-122. https://doi.org/10.1016/j.obhdp.2020.10.013
• Epley, N., & Schroeder, J. (2014). Mistakenly seeking solitude. Journal of Experimental Psychology: General, 143(5), 1980-1999. https://doi.org/10.1037/a0037323
• Sandstrom, G. M., & Boothby, E. J. (2021). Why do people avoid talking to strangers? A mini meta-analysis of predicted fears and actual experiences talking to a stranger. Self and Identity, 20(1), 47-71. https://doi.org/10.1080/15298868.2020.1816568
• Gilovich, T., Medvec, V. H., & Savitsky, K. (2000). The spotlight effect in social judgment: An egocentric bias in estimates of the salience of one's own actions and appearance. Journal of Personality and Social Psychology, 78(2), 211-222. https://doi.org/10.1037/0022-3514.78.2.211
• Bruk, A., Scholl, S. G., & Bless, H. (2018). Beautiful mess effect: Self-other differences in evaluation of showing vulnerability. Journal of Personality and Social Psychology, 115(2), 192-205. https://doi.org/10.1037/pspa0000120

Books

• Epley, N. (2014). Mindwise: How We Understand What Others Think, Believe, Feel, and Want. Knopf.
• Keohane, J. (2021). The Power of Strangers: The Benefits of Connecting in a Suspicious World. Random House.
• Robson, D. (2024). The Laws of Connection: The Scientific Secrets of Building a Strong Social Network. Pegasus Books.
• Gilovich, T., & Ross, L. (2015). The Wisest One in the Room: How You Can Benefit from Social Psychology's Most Powerful Insights. Free Press.

Book Chapters

• Clark, D. M., & Wells, A. (1995). A cognitive model of social phobia. In R. G. Heimberg, M. R. Liebowitz, D. A. Hope, & F. R. Schneier (Eds.), Social Phobia: Diagnosis, Assessment, and Treatment (pp. 69-93). Guilford Press.
• Epley, N., & Caruso, E. M. (2009). Perspective taking: Misstepping into others' shoes. In K. D. Markman, W. M. P. Klein, & J. A. Suhr (Eds.), Handbook of Imagination and Mental Simulation (pp. 295-309). Psychology Press.

19. Summary Card

20. Glossary of Terms Used

21. Discussion Questions

For book clubs, classrooms, or self-reflection:

1. Think of a friendship that almost did not happen. Who made the second move, and what would have happened if they had waited?

2. The dorm suitemates kept their liking gap for eight months until housing decisions forced a conversation. What explicit exchanges of impressions do your own groups, teams or families allow, and which do they avoid?

3. The gap appears in children at about age five, the same time they begin to care about their reputation. Is the bias a side effect of a useful capacity? Could we have one without the other?

4. The evidence comes almost entirely from American samples. How do you expect the gap to behave in a culture where modesty is enforced, or where warmth toward strangers is a formal obligation?

5. If you assumed for one month that every new acquaintance liked you about as much as you liked them, what would you do differently, and what is the worst realistic outcome?

Laas opgedateer 4 November 2026