The psychology of attraction: what draws people together

You meet someone at a party. There's something about them. You can't name it exactly. Chemistry, maybe. A vibe. That ineffable quality that makes one person magnetic and another person forgettable. You walk away thinking it's mysterious, that attraction is some unpredictable alchemy that either happens or doesn't.

Psychologists have been studying this "alchemy" for over seventy years, and they've found something slightly deflating for the romantics among us: attraction is far more predictable than it feels. The variables that determine who you're drawn to are, for the most part, measurable, replicable, and remarkably mundane.

This doesn't ruin the magic. Understanding gravity doesn't make sunsets less beautiful. But it does mean that the story we tell ourselves about attraction, that it's spontaneous, inexplicable, based on some deep cosmic compatibility, needs updating.

Does physical proximity actually matter that much?

More than almost anything else, and it's the finding that people resist most.

In 1950, Leon Festinger, Stanley Schachter, and Kurt Back studied friendship formation at Westgate, a housing complex for married MIT students. The buildings were assigned essentially at random, so the researchers had a natural experiment: who became friends with whom, and why?

The answer was embarrassingly simple. People became friends with their neighbors. Not just in the same building, but on the same floor, and especially in adjacent apartments. Residents were far more likely to name the person next door as a close friend than someone two doors down. People who lived near stairwells or mailboxes (high-traffic spots where casual encounters were more frequent) had more friends in the building overall.

This is the proximity effect, and it has been replicated dozens of times in dozens of contexts since 1950. Your social network, your friendships, and your romantic options are shaped primarily by who you physically encounter in your daily routine. The person you marry is, statistically, someone you met through work, school, mutual friends, or your neighborhood: contexts defined by proximity.

The proximity effect works through a mechanism Robert Zajonc identified in 1968: the mere exposure effect.

Why does familiarity increase attraction?

Zajonc's research showed that repeated exposure to a stimulus (a face, a word, a symbol, a melody) increases liking for that stimulus, even when the exposure happens below conscious awareness. In his studies, participants were shown photographs of faces at varying frequencies. They consistently rated the faces they'd seen more often as more attractive, even when they couldn't remember having seen them before.

The effect is strong and has been replicated with faces, geometric shapes, Chinese characters (presented to non-Chinese readers), and even sounds. The threshold is low, as few as one or two exposures increase preference, and it works even for subliminal presentations.

The evolutionary logic is straightforward: a stimulus you've encountered before and survived is probably safe. Novelty carries unknown risk. Familiarity signals safety, and at the most basic biological level, we're attracted to safe. This is why the person you see at the coffee shop every morning starts to seem more attractive over time, and why your coworker who initially registered as average-looking becomes genuinely appealing after months of daily interaction.

The mere exposure effect also explains something about online dating that frustrates people: profiles lack the familiarity that builds attraction in person. You're making a judgment about a complete stranger based on a photograph and a few sentences. The proximity and repeated-exposure channels that generate attraction in offline settings are completely absent.

Do opposites actually attract?

Mostly, no. This is one of the most persistent folk beliefs about attraction, and the research has consistently failed to support it.

Donn Byrne's work through the 1960s and 1970s established what researchers call the similarity-attraction paradigm. Byrne demonstrated across dozens of studies that people are attracted to others who share their attitudes, values, and beliefs. The more similar the attitudes, the greater the attraction, in a linear, dose-dependent relationship.

Byrne's "bogus stranger" paradigm was elegant: participants filled out an attitude questionnaire, then received a questionnaire supposedly filled out by a stranger (actually fabricated by the researchers to reflect varying levels of agreement with the participant's own views). Attraction ratings tracked directly with attitude similarity. The more the "stranger" agreed with the participant, the more the participant liked them.

Since Byrne's work, the similarity finding has been extended to personality traits, communication styles, values, intelligence, and even physical attractiveness. A meta-analysis by Matthew Montoya, Robert Horton, and Jeffrey Kirchner in 2008 confirmed that actual similarity (not just perceived similarity) predicts attraction across a wide range of domains.

Why is the "opposites attract" myth so persistent? Probably because complementarity does exist in specific domains. A person who likes to talk might enjoy being with someone who likes to listen. A person who's decisive might pair well with someone who's easygoing. But these functional complementarities sit on top of a broad base of similarity in values, worldview, humor, and communication style. The couple who seems like "opposites" (she's outgoing, he's quiet) usually shares the same fundamental beliefs about loyalty, fairness, family, and how to treat other people.

How much does physical attractiveness actually matter?

The research here is complicated by what people say versus what they do.

In a classic 1966 study, Elaine Walster (later Hatfield), along with colleagues, organized a "Computer Dance" where University of Minnesota freshmen were randomly paired for an evening. Participants had filled out personality questionnaires beforehand, and the researchers measured physical attractiveness through independent ratings. The question: what would predict whether partners liked each other and wanted to see each other again?

The answer, overriding every other variable: physical attractiveness. Not similarity, not personality, not intelligence, not the whole battery of questionnaires. How good-looking your dance partner was. For both men and women.

This study launched decades of research on what Walster and colleagues called the "matching hypothesis," the idea that people tend to pair with others of similar attractiveness. Subsequent research has largely supported matching: while everyone prefers more attractive partners, actual couples tend to be matched in attractiveness. People calibrate their aspirations to their own level, and relationships between attractiveness-matched partners tend to be more stable.

But there's a significant wrinkle. Physical attractiveness matters most in initial encounters: the first meeting, the first photo, the first impression. As people get to know each other, other factors gain weight. In a speed-dating study by Eli Finkel and Paul Eastwick in 2008, women who said they valued earning potential over attractiveness didn't actually behave that way during the speed dates. But in longer-term contexts, personality, humor, kindness, and intellectual compatibility become increasingly important.

The bottom line: attractiveness gets you through the door. Everything else determines whether you stay.

Why is reciprocity so powerful?

The simplest and arguably most important finding in attraction research: we like people who like us.

Rebecca Curtis and Kim Miller demonstrated this in a 1986 study at Adelphi University. Participants met with a partner and had a conversation. Beforehand, some participants were told (by the researchers) that the partner liked them; others were told the partner didn't. This information was fabricated, but its effects on behavior were real.

Participants who believed the partner liked them were warmer, disclosed more, disagreed less, and behaved in ways that were objectively more likable. And their partners, who hadn't been told anything, actually liked them more in return. Believing you're liked makes you more likable, which makes the other person actually like you. It's a self-fulfilling cycle.

This works in the other direction too. If you believe someone dislikes you, you become more guarded, less warm, and less engaging, which makes them actually like you less.

The implications for dating are straightforward: showing genuine interest is one of the most attractive things you can do. Not manufactured interest, not playing hard-to-get, but actual curiosity about the other person. The research on "playing hard to get" is mixed at best; what consistently works is selective interest, being generally discriminating but clearly interested in this specific person.

What do self-disclosure and vulnerability have to do with attraction?

Arthur Aron's famous 36 Questions study demonstrated that structured, reciprocal self-disclosure between strangers can produce feelings of closeness that normally take months to develop. Pairs who worked through progressively more personal questions for 45 minutes reported feeling significantly closer than control pairs who engaged in small talk.

One pair from the study actually got married.

The mechanism is multilayered. Self-disclosure triggers the reciprocity effect: when someone shares something personal, you feel trusted, which makes you like them more, which makes you share something personal in return. The escalating vulnerability creates a feedback loop of increasing intimacy.

There's also a neurochemical component. Self-disclosure activates the brain's reward circuitry, the same dopamine-producing regions that light up during other pleasurable experiences. Talking about yourself feels good at a neural level, and doing so with someone who is genuinely listening feels even better. The combination of reward (dopamine) and bonding (oxytocin from the social closeness) creates a strong associative pairing: this person makes me feel good.

This is why good questions matter more than good answers on a first date. The person who asks interesting, genuine questions and listens attentively is creating the neurochemical and psychological conditions for attraction. The person who talks about themselves without reciprocating is missing half the equation.

What has online dating research revealed about attraction?

Eli Finkel led a comprehensive review of online dating research published in 2012 in Psychological Science in the Public Interest. The paper's conclusions challenged several assumptions of the dating industry.

Matching algorithms don't work the way companies claim. Dating sites that promise to match you based on compatibility questionnaires are relying on the similarity-attraction finding, which is real but weaker than the claims suggest. The traits that predict relationship satisfaction (emotional stability, agreeableness, secure attachment) predict satisfaction with any partner, not specifically with a matched one. Finkel's review found no evidence that algorithm-based matching outperformed random pairing.

The browsing format changes psychology. Online dating platforms present potential partners as profiles to be evaluated, a fundamentally different context than meeting someone in person. The evaluation mindset promotes comparison, increases pickiness, and reduces the "mere exposure" and "reciprocity" effects that drive attraction in offline settings. You're rejecting people based on a photo who might have become genuinely attractive to you after weeks of daily interaction.

Texting before meeting creates false intimacy. Extended pre-date messaging can build expectations that real-life chemistry fails to meet. Finkel recommends meeting relatively quickly, not to rush, but to test actual in-person attraction before investing in a text-based fantasy.

Proximity still wins, just differently. Dating apps expanded the pool of available partners but didn't change the underlying psychology. People still prefer partners who are geographically close, who share social contexts, and who feel familiar. The technology changed the discovery mechanism but not the attraction mechanisms.

What does all this mean for how you approach dating and relationships?

The research points to some practical conclusions.

Put yourself in proximity. If you want to meet someone, the single most effective strategy is to be present in contexts where you encounter the same people repeatedly. A regular class, a weekly group, a recurring social event: these create the conditions for familiarity and mere exposure to do their work. The random encounter at a bar is statistically unlikely to produce a lasting connection compared to the person you see at the climbing gym every Tuesday.

Lead with genuine curiosity. Reciprocity and self-disclosure are more powerful attractors than appearance, wit, or status. Ask real questions. Listen to the answers. Share something honest about yourself. This isn't a tactic. It's how human bonding actually works.

Stop waiting for "chemistry." The immediate, lightning-bolt experience of chemistry is mostly the dopamine system responding to novelty and physical attractiveness. It's real, but it's not a reliable predictor of relationship success. Many of the strongest relationships start with mild interest that builds through repeated positive interactions. If you only pursue people who generate an immediate chemical hit, you're screening out a large percentage of potentially excellent partners.

Similarity is your friend. Don't apologize for wanting someone who shares your values, your humor, and your communication style. The research says that's exactly what predicts long-term satisfaction. The "exciting" partner who challenges everything you believe might generate dopamine in the short term but conflict in the long term.

Show your interest. The Curtis and Miller finding is one of the most actionable in all of relationship science. If you like someone, let them know, not aggressively, but clearly. Your warmth makes you more attractive to them, which makes them warmer to you, which compounds. The risk of vulnerability is small compared to the cost of withholding.

Aperi's approach to daily questions for couples is built on the Aron self-disclosure model: structured, reciprocal vulnerability that increases gradually over time. But the same principles apply before the relationship starts. The people who are best at forming connections aren't the funniest, best-looking, or most confident. They're the ones who ask good questions and genuinely care about the answers.

FAQ

Is "love at first sight" real?

What people report as love at first sight appears to be a retroactive reinterpretation of strong initial attraction. A 2017 study by Florian Zsok and colleagues in the Netherlands found that "love at first sight" reports correlated strongly with the physical attractiveness of the partner but not with passion, intimacy, or commitment as measured at the time. In other words, what's happening at first sight is strong attraction, the dopamine system responding to an appealing face. The "love" part gets added to the story later, once the relationship has actually developed. This doesn't make the experience less real for the people involved, but it does mean the instant recognition of a soulmate is more likely instant recognition of an attractive person.

Why am I attracted to people who are bad for me?

This usually involves one of two mechanisms. The first is the familiarity principle working against you: if your early attachment experiences involved chaos, inconsistency, or emotional unavailability, those patterns feel "like home," and you may confuse familiar-unsafe with attracted-to. The second is arousal misattribution, the tendency to interpret physiological arousal (racing heart, butterflies) as attraction when it's actually anxiety. A partner who keeps you on edge generates more arousal than a partner who makes you feel calm, and that arousal can be misread as chemistry. Working through childhood patterns and learning to distinguish anxiety from excitement helps recalibrate what "attraction" means.

Do pheromones play a role in human attraction?

The evidence is surprisingly thin. While other mammals have a well-developed vomeronasal organ for detecting pheromones, the human version is vestigial. The famous "sweaty T-shirt" study by Claus Wedekind in 1995 found that women preferred the scent of men with dissimilar immune system genes (MHC complex), but replications have been inconsistent. Hormonal contraceptives may interfere with the effect. The current scientific consensus is that body odor probably plays some role in human attraction, but it's much smaller than in other species and not well enough understood to draw practical conclusions.

How much do shared interests actually matter?

Less than shared values, more than nothing. Shared activities create proximity and shared experiences, both of which build attraction. But the specific activities matter less than whether you share fundamental attitudes about life: how to treat people, what you find funny, what you prioritize, how you handle conflict. A couple who share no hobbies but share deep values will generally do better than a couple who do everything together but disagree about what matters. That said, having at least some shared activities gives you a context for the kind of repeated, positive interaction that strengthens bonds over time.

---

Whether you're in the early stages of attraction or years into a relationship, the research says the same thing: genuine curiosity and reciprocal self-disclosure build the strongest connections. Aperi sends couples a daily question designed to keep that discovery going, because the psychology of attraction doesn't stop mattering once you've found your person.