How Attraction Works — What Science Actually Shows
Last reviewed by the Men Women Psychology editorial team.
The evidence
What the research actually shows
Two of the most robust findings are almost embarrassingly simple. Proximity and familiarity breed liking: Robert Zajonc's mere-exposure research (1968) showed that repeated exposure to a person or thing, on its own, tends to increase how much we like it. Similarity is the second: Montoya and Horton's meta-analysis (2013) confirmed the well-documented similarity-attraction effect — we are drawn to people who share our attitudes and values, partly because agreement feels validating. Layered on top is reciprocal liking, the tendency to be attracted to people who appear attracted to us.
Physical attractiveness matters most, and most equally, at first contact. In Walster, Aronson, Abrahams and Rottmann's classic 'computer dance' study (1966), students were randomly paired for a dance, and physical attractiveness was by far the strongest predictor of how much each person liked their partner and wanted to see them again — for men and women alike. The takeaway is not that looks are everything, but that they dominate early, when little other information is available.
What complicates the popular story is the gap between stated and actual preferences. Eastwick and Finkel (2008) found that although men and women report somewhat different ideals — men emphasizing attractiveness, women emphasizing earning prospects — those stated preferences did not predict who they were actually drawn to during live, face-to-face interaction. In the moment, the supposed gender differences largely vanished, a finding that should make us cautious about treating checklist preferences as how attraction really operates.
The mechanism
Why this happens
The basics likely persist because they were useful. Familiarity reduces uncertainty — a known person is a safer bet than a stranger — so mere exposure nudges us toward the people already in our orbit. Similarity smooths cooperation and feels affirming, which is why shared values and attitudes reliably draw people together. Reciprocal liking is partly self-protective: it is easier to risk interest in someone who seems to want us, and being liked is itself rewarding.
The early dominance of physical attractiveness reflects information scarcity. At first meeting, appearance is one of the only signals available, so it carries disproportionate weight before personality, values, and warmth come into view. As people get to know each other, those other qualities increasingly shape attraction, which is part of why first impressions are not destiny.
Attraction is not static, either. Hatfield and Sprecher's work distinguishes passionate love — the intense, arousal-driven early phase — from companionate love, the steadier affection that can grow over time. And Aron and colleagues (2000) found that couples who shared novel and arousing activities reported higher relationship quality, suggesting that excitement and self-expansion can rekindle attraction well beyond the first spark.
In practice
What this looks like in real life
People often 'fall for' a coworker, classmate, or neighbor and assume it was fate, when proximity and repeated exposure did much of the quiet work. The person you see often and find easy to talk to has a real statistical head start, which is why so many couples meet through ordinary shared contexts rather than dramatic chance.
Someone insists they have a strict 'type' and then becomes attracted to a person who fits none of it. This mirrors Eastwick and Finkel's finding that stated preferences poorly predict real-time attraction. The checklist describes an idea of a partner; the actual pull tends to emerge from interaction, similarity, and reciprocal warmth.
A long-term couple feeling flat tries something genuinely new together — a trip, a class, a shared challenge — and notices a spark return. Aron's self-expansion research helps explain this: novel, slightly arousing experiences can be experienced as renewed attraction to the partner, not just enjoyment of the activity.
Myth vs. evidence
What most people get wrong about this
The biggest misconception is that attraction can be engineered with techniques, games, or manipulation tactics. The evidence points instead to slow, mutual processes — proximity, similarity, reciprocal liking, and shared experience — that are about genuine connection, not tricks. Attempts to 'hack' attraction tend to be unreliable at best and corrosive to trust at worst.
People also overestimate how rigidly gendered attraction is. While men and women report somewhat different ideals on paper, those stated preferences predict real attraction poorly, and the famous Walster study found physical appeal mattered strongly for both sexes early on. Treating gender checklists as laws of attraction misreads what the research actually shows.
Why it matters
What this means for relationships
Understanding these forces is less about exploiting them and more about removing obstacles to genuine connection. Shared environments and repeated contact give attraction room to grow; expressed warmth and clear interest, rather than calculated aloofness, tend to invite reciprocal liking. The aim is to be known and to show genuine interest, not to perform a strategy.
For established couples, the practical lesson is that attraction can be nurtured rather than left to fade. Building on similarity and shared values, staying responsive, and deliberately introducing novelty and shared adventure — in the spirit of Aron's findings — can keep attraction alive long after the initial passionate phase has settled into companionate love.
Where it varies
The nuance
Where gender differences in attraction appear, they are averages with large overlap, and they often shrink or disappear in real interaction. Janet Hyde's gender similarities hypothesis (2005) fits the pattern: men and women are drawn to similar core things — kindness, similarity, mutual interest, and yes, physical appeal — far more than the 'men want looks, women want status' shorthand implies. Eastwick and Finkel's work reinforces how unreliable those stated differences are in practice.
Individual variation dominates. Personal history, culture, attachment style, and context shape who any given person finds attractive more than their sex does. The forces described here are reliable tendencies across groups, not a formula that predicts who a particular individual will fall for, and 'chemistry' that defies the averages is common.
Questions people ask about this
What actually makes two people attracted to each other?
Research points to a few reliable forces: proximity and familiarity, similarity in values and attitudes, sensing that the other person likes you, and early on, physical appeal. These tend to work together gradually. 'Chemistry' usually describes the felt result of these ordinary processes rather than a mysterious separate force.
Do looks matter most in attraction?
Physical attractiveness matters most early, when little else is known — the classic Walster computer-dance study found it strongly predicted liking for both men and women at first meeting. As people get to know each other, similarity, warmth, and personality increasingly shape attraction, so looks dominate the start more than the long run.
Do men and women want fundamentally different things?
On paper they report somewhat different ideals, but Eastwick and Finkel found those stated preferences poorly predict real-time attraction, where the differences largely fade. On average, both are drawn to similarity, kindness, mutual interest, and physical appeal. The 'looks versus status' shorthand overstates a difference that is small and inconsistent in practice.
Can you make someone attracted to you with techniques?
The evidence does not support reliable 'techniques' or games. Attraction grows through genuine processes — repeated contact, similarity, mutual warmth, shared experience — rather than manipulation. Showing sincere interest and being present tend to invite reciprocal liking far more dependably than strategies designed to engineer feelings.
Why do I keep falling for people who aren't my 'type'?
Because a stated type is an idea, and research shows such checklists predict real attraction poorly. In actual interaction, similarity, reciprocal liking, and rapport often pull you toward people who do not match your imagined preferences. This is common and well documented, not a sign that something is wrong.
Can attraction be rekindled in a long relationship?
Often, yes. Aron and colleagues found that couples who shared novel and arousing activities reported higher relationship quality. Introducing genuine novelty and shared adventure, alongside continued warmth and responsiveness, can revive attraction. Results vary by couple, but attraction is something that can be nurtured rather than simply fading on its own.
Research sources
These references point to the published research and established frameworks behind this page. They are provided for further reading; see our research methodology for how sources are selected.
- Zajonc, R. B. (1968). Attitudinal effects of mere exposure. Journal of Personality and Social Psychology, 9(2, Pt.2), 1–27.
- Montoya, R. M., & Horton, R. S. (2013). A meta-analytic investigation of the processes underlying the similarity-attraction effect. Journal of Social and Personal Relationships, 30(1), 64–94.
- Walster, E., Aronson, V., Abrahams, D., & Rottmann, L. (1966). Importance of physical attractiveness in dating behavior. Journal of Personality and Social Psychology, 4(5), 508–516.
- Eastwick, P. W., & Finkel, E. J. (2008). Sex differences in mate preferences revisited. Journal of Personality and Social Psychology, 94(2), 245–264.
- Hatfield, E., & Sprecher, S. (1986). Measuring passionate love in intimate relationships. Journal of Adolescence, 9(4), 383–410.
- Aron, A., Norman, C. C., Aron, E. N., McKenna, C., & Heyman, R. E. (2000). Couples' shared participation in novel and arousing activities and experienced relationship quality. Journal of Personality and Social Psychology, 78(2), 273–284.