How do I know if someone actually likes me?

You can’t read someone’s mind, but you can read what they do, and that’s more reliable than any single text. If your last breakup shook your confidence, assume your gut is running a little low, because research suggests most people underestimate how much others like them even on a good day.

Why does it feel impossible to tell?

Meet Anjali, 23. Eight months ago her boyfriend of two years ended it with “I don’t think I was ever as into this as you were.” Now she’s three dates in with Ezra. Last night he walked her to the train and texted “had a really good time, get home safe.” Anjali has read it maybe nine times. “Really good” sounds warm. “Get home safe” sounds like something you’d say to a coworker. By midnight she’s decided he’s being polite.

Split what happened from what Anjali is doing with it. What happened: three dates, a walk to the train, a text that night. The reading: he’s just being nice. The fear: that she’s misread someone again, like she did for two years. The next step: still open.

Part of what’s going on is a well-documented blind spot. In five studies, Erica Boothby, Gillian Sandstrom and colleagues found that after a conversation, people consistently underestimated how much the other person liked them and enjoyed their company. They call it the liking gap.1 It showed up among strangers in a lab, first-year students meeting their dorm mates, and members of the public at a workshop, and among college roommates it lasted for months.1

So even a confident person tends to undershoot. Anjali, eight months after hearing “I was never that into it,” is probably undershooting by more.

What actually counts as a sign of interest?

Forget the tone of a single message. Look at effort and specifics, the stuff that takes attention to fake.

Questions, especially follow-ups. In a set of studies by Karen Huang, Alison Wood Brooks and colleagues, people who asked more questions, particularly follow-up questions, were liked more by the people they talked to, and they were seen as more responsive: listening, understanding, caring.2 In the speed-dating part of the research, people who asked more follow-ups were more likely to get a yes to a second date.2 That cuts both ways. Someone asking “wait, how did the thing with your sister go?” is paying attention. And you asking questions back is one of the better ways to show your own interest.

Interest in you, not in everyone. In speed-dating research by Paul Eastwick, Eli Finkel and colleagues, 156 undergraduates went on four-minute dates with 9 to 13 people each.3 People who were especially drawn to one particular partner tended to get that interest back, while people who said yes to everyone were wanted less by their dates.3 In plain terms, being singled out matters. Remembering your coworker’s name, asking about your exam, picking a place because you mentioned you love dumplings: that’s attention aimed at you specifically.

Plans with a future tense. “We should do that sometime” is friendly. “Are you free Thursday? There’s a trivia night near you” is a plan.

Repair. If you cancel and they offer another day, that tells you more than how fast they reply.

None of this is proof. Some people are shy askers. Some are slow planners and great company. But a pattern of specific attention over a few weeks is the best evidence you’re going to get without asking.

Why does a breakup make you doubt every good sign?

Because a breakup can rewrite your story about yourself, not just your relationship status.

Stanford researchers Lauren Howe and Carol Dweck studied why some people carry rejection longer than others. People who saw a breakup as revealing some core truth about who they are, the “this proves I’m unlovable” reading, held onto it longer and brought it into the future.4 People who believed they could grow were hurt too, but recovered more readily.4

That’s what’s happening when Anjali reads “had a really good time” and hears “you’re about to be disappointed again.” The text is from Ezra. The doubt is left over from the ex.

You don’t have to fix that before you date. You just have to know which voice is which. When you catch yourself discounting something kind, ask: is this about what they did, or about what I decided about myself last year?

So should you just ask them?

Eventually, yes, though not on date three, and not as an exam. Early on, the most useful question is often the one you ask yourself: “Do I like how I feel around this person?” Your interest is half the equation, and it tends to get lost when you’re busy auditing theirs.

When things have gone on for a while, a light, honest line works better than decoding: “I’ve really liked hanging out. Want to do something next weekend?” Their answer, and how easily it comes, tells you more than a week of rereading.

What to do this week

• Write down what they did. Three dates, two follow-up questions, one rescheduled plan. Facts only, no adjectives.
• Notice the discount. Each time you explain away something kind, jot down the explanation. If it sounds like your ex talking, mark it.
• Ask more questions yourself. It’s good for the date, and it gets you off the scoreboard.
• Make one plan with a day attached. Then see what they do with it.
• Give it time. Interest shows up as a pattern over weeks, not in one message.

If the doubt about whether anyone could like you is constant, keeps you up, or is getting in the way of work or friendships, a licensed mental-health professional is the right person to help with it. That’s not a dating problem you’re meant to solve alone.

What doesn’t the research show?

• The liking gap studies were about conversations between new acquaintances and roommates, not romantic interest specifically. They show a general tendency to undershoot, not how any one date feels about you.1
• The question-asking research measured liking after short conversations. More questions helped on average, but a quiet person can still be interested.2
• The speed-dating study used four-minute dates between undergraduates, a very different setting from a few weeks of seeing someone.3
• The breakup research is about patterns across people. It doesn’t diagnose why any one person is struggling, and it doesn’t mean you’re doing recovery wrong.4
• No list of signs can tell you what someone feels. Behavior is evidence, not a verdict, and the only person who can confirm it is them.