Signs Someone Is Falling for You: What the Research on Early Love Actually Shows

What falling for someone looks like from the inside

Start with the uncomfortable fact that shapes everything below: falling for someone is something that happens inside a person, and every sign on every list is an attempt to see it from outside.

Researchers have looked at the inside part directly. Arthur Aron, Meg Paris and Elaine Aron recruited first- and second-year university students at the start of an autumn term and tested them five times over ten weeks, each time asking whether they had fallen in love since the last session. In the first study, 329 students wrote open-ended descriptions of themselves; after falling in love those descriptions changed more, and spread across a wider range of domains, than the descriptions of students who had not. In the second study, 529 students completed measures of self-efficacy and self-esteem, and both rose after falling in love. The authors checked that the change was not simply a better mood. Their reading is that falling in love expands the self: the other person's perspectives, interests and resources start to be counted as partly one's own.

The measure that grew out of that idea is almost comically simple. Aron, Aron and Smollan's Inclusion of Other in the Self scale is a single item — a set of pairs of circles overlapping by increasing amounts, and you pick the pair that describes you and the other person. In their validation work it tracked longer closeness measures, was barely affected by the urge to give flattering answers, and predicted whether romantic relationships were still together three months later. A later synthesis of 137 studies and 37,761 people, by Benjamin Le and colleagues, put inclusion of the other in the self alongside commitment, love and dependence as the strongest predictors of whether a dating relationship lasted.

And the researchers who built the standard measure of passionate love, Elaine Hatfield and Susan Sprecher, wrote intrusive thinking into its definition. Items on their scale include the feeling that someone is always on your mind and that you cannot control where your thoughts go.

Now notice what all of that has in common. Self-expansion, felt overlap, preoccupation: every one is a state you can only report about yourself. None of these studies measured an outward sign, and nobody has published a validated list of behaviours that reveal that another person is falling for you. What follows are the behaviours that the research on closeness, commitment and responsiveness says tend to travel with those states. They are inferences from that work, and each section says how far the inference stretches.

They come back to what you told them

The best-supported idea in the whole literature on intimacy is not a gesture. It is a feeling on the receiving end: that the other person understood what you said, valued it, and cared about you because of it. Harry Reis and Phillip Shaver built their model of intimacy around it in 1988, and Reis, Margaret Clark and John Holmes later argued that perceived partner responsiveness organises most of what closeness research has found.

It has been tested outside the model. Jean-Philippe Laurenceau, Lisa Feldman Barrett and Paula Pietromonaco had people fill in a record immediately after their social interactions, for one week in one study and two in the other. Intimacy in an interaction rose with what the person disclosed and with what the other person disclosed back, and that effect ran partly through how responsive the other person seemed. In the second study, disclosing feelings mattered more than disclosing facts. Natalya Maisel and Shelly Gable followed 67 cohabiting couples day by day and found something sharper: support from a partner reduced sadness and anxiety and lifted relationship quality only when the support was responsive. Help that missed what the person actually needed did not do the same work.

What responsiveness looks like from the outside is ordinary. Someone remembers the thing you mentioned once and circles back to it a week later without being prompted. They ask how the difficult conversation with your manager went, because they kept track of when it was happening.

The honest limit is the one that runs through this whole page. Responsiveness is what care looks like in any close relationship, and nothing in these studies separates the romantic version from the version a good friend offers. That is the wall every attempt to tell flirting from friendliness runs into. And the small everyday version of it, the bids and turning toward that Gottman's lab coded, is smaller and more cautious in the published papers than the famous percentages suggest, as what the research on couples who last actually reports sets out. A single return to something you said is a data point. A habit of it, over weeks, is a pattern.

They ask the second question

One behaviour has been connected to romantic outcomes in a real dating context, and it is not a subtle one.

Karen Huang, Michael Yeomans, Alison Wood Brooks, Julia Minson and Francesca Gino ran two experiments in which pairs of strangers chatted and some were told to ask more questions. The people who asked more questions were liked more, and the effect came mostly from follow-up questions — the ones that pick up what the other person just said and ask for more of it. They then turned to speed dating: 110 participants across 1,961 short dates. Daters who asked more questions were more likely to be wanted for a second date. The paper was reanalysed by other researchers, answered by its authors, and corrected in 2025 after an independent audit that left its conclusions standing.

That is evidence about how questions are received, not about what they reveal. But it makes a reasonable inference available. A follow-up question costs attention. It means the person was tracking the answer rather than waiting for their turn. Someone who is becoming invested in you tends to spend that attention, and a follow-up that arrives days later — about something you said in passing — shows the thread was still running in their head after you left.

Two cautions. Some people ask good questions of everyone because that is how they were raised or trained, and some interested people are too nervous to ask many at all. And the same research makes a point that matters on your side of the table: the questions work on the person being asked. We go through that research, and what it means for early dates, in the first-date piece, and through the wider literature on disclosure in how people actually open up.

Disclosure that deepens in both directions

The older version of this page credited the idea that vulnerability is offered first to someone judged safe to a popular book. There is a better source, and it says something more precise.

Nancy Collins and Lynn Carol Miller pooled the research on self-disclosure and liking in 1994 and found three separate effects, all reliable. People who disclose more intimately tend to be liked more. People disclose more to those they already like. And people come to like others more after disclosing to them. The size of each depended on the design of the study, the type of disclosure and the gender of the person disclosing, so none of it is a law. But the second effect is the one that matters here: disclosure follows liking. When someone begins telling you things they do not tell everyone, that is consistent with them already having decided they like you.

The famous demonstration runs the other way. Arthur Aron and colleagues had pairs of strangers work for 45 minutes through a sequence of questions that escalated gradually in how personal they were — the procedure later sold as “the 36 questions to fall in love.” It produced more closeness than matched small talk. Being led to expect that the other person would like you made no significant difference, and neither did being matched on not disagreeing about important attitudes. The outcome measured was closeness between strangers in a laboratory, not love, and the popular framing oversells it. What it does show is that reciprocal, gradually deepening disclosure is how closeness gets built.

So the sign is not a single confession. It is the sequence: small things first, then slightly bigger ones, in both directions, with each step depending on how the last one was received. A sudden flood of intense disclosure in the first week is a different thing, and the last section of this page comes back to it.

One person will show you much less of this than another, whatever they feel. Scott Bradford, Judith Feeney and Lorne Campbell studied 113 dating couples and found that attachment avoidance strongly predicted lower self-disclosure on questionnaire measures. That is not a verdict on anyone's feelings, and a later section explains why it changes how every sign on this page should be read.

The word “we” starts turning up

Christopher Agnew, Paul Van Lange, Caryl Rusbult and Christopher Langston asked people in relationships to describe them, and counted something nobody deliberately controls: pronouns. The more committed people were, the more often they spontaneously used plural pronouns — we, us, our — and the more central the relationship was to how they described their lives. Over time, the relationship ran in both directions. Earlier commitment predicted later “we” thinking, and earlier “we” thinking predicted later commitment. And the links were weak or absent in best friendships, which makes this one of the few markers in the area that looks specific to romance.

It has limits. These were people already in relationships, the data are correlational, and the study measured how people described their relationship to researchers rather than how they talked to their partner. Nobody has shown that one “we” in the third week means anything. But a shift in pronouns over months — “we should try that place” replacing “you should,” plans spoken of as shared by default — is closer to the mental change underneath than most of the gestures lists recommend watching.

Plans that have dates in them

“We should do that sometime” is a placeholder. “That festival you mentioned is in March — are you free that weekend?” is a different act, because it commits the speaker to a version of the future with you still in it.

Scott Stanley and Howard Markman described commitment in 1992 as two things. One is constraint — the costs of leaving. The other is dedication, and among its components are a relationship agenda (wanting the relationship to continue and planning as if it will), couple identity (thinking of yourselves as a unit) and primacy (the relationship ranking high among priorities). Naming a date months ahead is what a relationship agenda looks like in practice. It is not a symptom of commitment standing outside it. In the measurement literature, it is part of what commitment is made of.

Commitment in turn predicts whether relationships last. Benjamin Le and Christopher Agnew's meta-analysis of the investment model, 52 studies and 11,582 participants, found commitment associated with staying rather than leaving at r = .47. That figure rests on only twelve studies, two of them about jobs rather than partners, and the authors say plainly that most of what decides a breakup is not captured by commitment.

The inverse is often clearer. When the future stays permanently vague after months, that is one of the markers of a situationship, and the vagueness is usually the message.

They bring you into their world

There is a deliberate version of this — meeting the friends, then the family — and a slower one, when you discover that people in their life already know your name because you have been part of how they talk about their week.

The research here measures something adjacent: what happens to relationships once the surrounding people have an opinion. Diane Felmlee followed 290 people in dating relationships and modelled who broke up. Relationships were more stable when people felt their own friends approved of the partner, and when they felt the partner's family approved of them. Perceived approval predicted better than what the network actually thought. One result complicates the easy version: approval from a person's own family went the other way in the full model. Susan Sprecher and Felmlee's earlier three-wave study of 101 dating couples found support from parents and friends associated with relationship quality, and the woman's perceived network support predicted which couples stayed together. Le's later synthesis listed network support among the robust predictors of whether dating relationships survived.

Reading that back into a sign takes one step, and the step should be named. Introducing someone to the people who matter to you exposes the relationship to their judgement and makes it harder to walk away from quietly. People tend to do that when they expect it to continue. It is a reasonable inference. It is not a finding, and some people keep partners away from family for reasons that have nothing to do with how they feel — distance, a difficult family, culture, safety.

What should not be read as integration is the opposite movement: a partner who wants you inside their world while pulling you away from yours. That is covered below, because it is not a sign of love.

How they react to your good news

Most advice about testing someone's investment focuses on how they behave when something goes wrong. The research points at the other end.

Shelly Gable, Harry Reis, Emily Impett and Evan Asher found across four studies that telling someone about a good thing that happened to you lifts your wellbeing beyond the good thing itself, and that the benefit is larger when the listener responds actively and enthusiastically — asking about it, sharing the pleasure — rather than passively or dismissively. Couples in which partners typically responded that way reported more intimacy and daily satisfaction. Gable, Gary Gonzaga and Amy Strachman then videotaped 79 dating couples taking turns to talk about recent good and bad events. Two months later, how partners had responded to good news was more closely related to relationship wellbeing and to breakup than how they had responded to bad news.

The sample is small and the design correlational, so it cannot say that enthusiasm causes anything. But as an observation it is useful precisely because it is unglamorous. Someone falling for you will usually be happy for you in a way that has some energy in it. They ask what happened next. They remember it. They do not turn your promotion into a story about themselves or change the subject. This is also the behaviour that has held up best in the wider research on what makes relationships work, which is covered in what the evidence on emotional maturity supports and what it does not.

The phone, the voice, and what the body can tell you

The phone first, because this page used to get it wrong. Where a phone sits on the table — face up, face down, in a bag — is not a readable sign. Jonathan Courtright and Scott Caplan pooled six studies of the mere presence of a phone, 601 participants, and the effect was not statistically distinguishable from zero. What does have evidence is picking the phone up while the other person is talking: across 37 studies and 7,515 participants, using a phone during an interaction was associated with more negative perceptions of the person using it, at around r = −.35, and a meta-analysis of 52 studies of couples put the association between partner phone snubbing and relationship satisfaction at about −.22. So the thing worth noticing is whether the conversation keeps being handed back to a screen, across several meetings rather than one distracted evening. The full evidence is in what phone use on a date actually shows.

The voice is the channel most lists never mention, and it has the most specific finding. Juan David Leongómez and colleagues recorded 110 English and Czech speakers talking to video clips of attractive and less attractive people. Speakers changed their pitch when addressing attractive targets, and listeners who did not understand the language could pick out the speech aimed at an attractive person — and could not do the same when asked to judge friendliness instead. It was a laboratory scenario with students, and the authors ask for replication. We go through it with the rest of the nonverbal evidence in what body language actually predicts.

Then the limit, and it is a large one. The behaviours most associated with liking across studies — eye contact, smiling, laughing, proximity, talking more — are the behaviours of rapport in general, and people are poor at reading them. In one study observers judged the interest of speed daters: they read men's interest well above chance and women's barely above it, and the errors ran towards seeing interest that was not there. We go through those numbers in the piece on reading someone's interest. Whatever gender you are reading, the body gives you evidence of liking. It cannot tell you which kind.

The first item in that list is the one people weight most heavily and the one where the limit is clearest, because how long a look feels comfortable varies by culture and by neurotype before anyone's feelings enter into it. What eye contact actually predicts sets out what has been measured and what has not.

Who says it first

The original title of this page assumed a man who is falling in love and hiding it. The research does not support the assumption that men are the ones who hold back.

Joshua Ackerman, Vladas Griskevicius and Norman Li asked people who they thought usually says “I love you” first. In their first study most said women: 64.4% expected women to confess first, and 84.4% expected women to think about getting serious first. Then they asked people about their own relationships. Among 111 undergraduates, 61.5% recalled that the man had said it first, and men reported having first thought about saying it at around 97 days into the relationship, against around 139 days for women. In a sample of couples, 28 of the 40 who agreed on who went first said it was the man. Michele Harrison and Jennifer Shortall surveyed 172 college students and found the same pattern: both sexes believed women fall in love and say it first, and men reported doing both earlier.

The limits are real. These are small, heterosexual, mostly student samples, and the timing figures rest on people remembering when they first thought something. Ackerman's later studies also found that how people felt about hearing it depended on whether it came before or after sex, which is a more complicated story than who is more romantic. But the direction consistently contradicts the stereotype the genre is built on.

The practical point is gender-neutral. If someone has not said anything, the likeliest explanation is not a secret they are managing. People of any gender fall at different speeds, and some are wary of saying something before they are sure it will still be true. Silence on its own is weak evidence either way.

The same feelings can look very different

Adult attachment research measures two separate dimensions. Kelly Brennan, Catherine Clark and Phillip Shaver built the standard questionnaire around them in 1998: anxiety, meaning worry about whether a partner will stay and be available, and avoidance, meaning discomfort with closeness and depending on others. Both change what this page's signs look like, in opposite directions.

Avoidance changes what the other person shows. In Bradford, Feeney and Campbell's study of 113 dating couples, more avoidant people reported disclosing less, focusing less of what they shared on the relationship itself, and drawing out less disclosure from their partner. Someone high in avoidance can be invested and still offer few of the behaviours above, particularly in the early months.

Anxiety changes what you see. In the same study, anxiety — especially the partner's — was tied to more negative evaluations of everyday interactions. An anxious reader tends to discount real signals and amplify ambiguous ones, which is part of why the same pattern can feel reassuring on Monday and alarming on Thursday. The places where love and anxious attachment feel alike and where they come apart are covered in the difference between love and anxious attachment, and our attachment style test gives you a rough read in about two minutes, with a plain statement of what a result like that cannot tell you.

One claim this page used to make does not have a source: that calm, rather than butterflies, is how you know. There is no study showing calm is the mark of love, and no study showing that nervousness is its absence. What can be said is narrower. Intensity is not evidence of depth, and the next section is about the case where it is evidence of something else.

When intensity is not a sign of love

Some behaviours get read as proof of deep feeling and are warning signs instead. The US National Domestic Violence Hotline's list of warning signs of abuse includes extreme jealousy of your friends or of time spent away from them, preventing or discouraging you from seeing friends and family, and preventing you from making your own decisions, and it adds that even one or two of these is a red flag. The same page makes the point this section exists for: many abusive people appear to be ideal partners in the early stages, and the warning signs tend to emerge and intensify as the relationship grows. Early on, control can arrive wrapped in intense affection and be easy to mistake for devotion.

The confusion is common enough to measure. In one Portuguese study of 148 adolescents, a third agreed that jealousy is a proof of love at the start of the study; after a school programme on the subject, fewer than one in ten did. In a Chilean sample of 1,001 people aged 18 to 25, jealousy was associated with both committing and suffering online abuse in dating. Both are convenience samples, not population rates, and they are cited here for the pattern rather than the percentages. Jealousy is an ordinary emotion, and what our piece on the psychology of jealousy covers is the difference between feeling it and acting on it through surveillance and restriction. The acting is control.

“Love bombing” is the popular name for overwhelming early attention that later turns into control. The term is everywhere; the research is thin. The most cited study is a student-journal survey of 484 college students that linked self-reported love-bombing behaviour with narcissism and insecure attachment, which is not enough to diagnose anyone. What is well documented is coercive control itself: a systematic review of 31 qualitative studies by Arshia Amin Choudhury, Natasha Martland and Olga Luzón describes the loss of agency, fear and unpredictability survivors report, and how hard it becomes to leave.

So the distinction worth keeping is between attention that makes your life larger and attention that makes it smaller. The research on falling in love, from the first section of this page, is about the self expanding. Someone who is falling for you in the sense this page means wants you to keep your friends, your time and your privacy. If what you are reading as signs is shrinking those, trust that over the intensity, and talk to someone — the support lines at the top of this page are free and confidential.

What actually resolves it

The last thing the lists never say is that no amount of observation replaces asking, and that the research has a name for the alternative.

Leslie Baxter and William Wilmot asked people how they found out where they stood in a relationship without raising it directly, and catalogued what they called secret tests: seeing whether the other person initiates if you stop, making them jealous to see the reaction, testing how they respond to a hypothetical, asking friends to find out. In a separate study the same researchers found that the state of the relationship was the most commonly avoided topic between close people. The tests exist because the question is hard to ask. They are poor instruments, though. Silence can mean indifference, a bad week, or a correct reading that you have pulled away, and a manufactured jealousy test measures whether somebody can be provoked.

What this page can offer is a way of looking that is honest about its error rate. Watch for the cluster rather than the item — responsiveness, follow-up, disclosure in both directions, “we,” plans with dates, being brought into their world, enthusiasm when things go well for you. Give it weeks, because a repeated pattern carries information one evening cannot. Allow for the fact that some people show much less of this than they feel, and that your own anxiety can make any pattern look like its opposite. Then, when you have enough to want an answer, ask — once, plainly, in a way that makes it easy to be honest with you. More of this under Signs, and the question of whether what you are seeing is worth investing in is picked up in the signs a new relationship has real potential.