How closeness actually forms

You've met someone and you're stuck at small talk. Or it's three months in and it's going flat and you don't know what you did. And somewhere you've read that there's a hormone, a question list, a puzzle, a trick.

Here's what the research says, with the marketing taken out.

The hormone story. Vasopressin does shape pair-bonding in male prairie voles: block it while they mate and they don't bond, add the receptor to a promiscuous species and they start. Evidence: decent That's where "the male bonding hormone" comes from. In humans, one Swedish study linked a gene variant to how bonded men felt; a 2025 study of 70 newlywed couples found the opposite direction early on and no difference by year three. Evidence: worth trying Nasal sprays of oxytocin, across dozens of small trials, don't reliably do anything to trust or bonding in people. Evidence: decent Vasopressin sprays have had only a handful of small trials, with mixed results. Evidence: worth trying There is no vasopressin hack. There's vole biology and a DNA test someone will sell you.

The tricks. "Do a puzzle together and she'll fall for you": no study. It's stitched together from two real findings, that couples who do new, slightly hard things together feel closer, and that people who move in time together cooperate more, neither of which is about a first date. Evidence: worth trying "Mimic her body language": the speed-dating study everyone quotes was retracted. Evidence: solid "Stare into each other's eyes for two minutes": one small paper from 1989, never repeated. Evidence: worth trying "The 36 questions make people fall in love": they made strangers feel closer after forty-five minutes; the same paper found two other tricks did nothing. Evidence: decent

What holds up is less exciting and more useful. Across eleven thousand couples, what predicted a good relationship wasn't either person's personality or background. It was how each one saw the relationship: whether they felt the other was committed and appreciative, how satisfied they thought the other was, how the sex was, and how much they fought. Evidence: solid The closeness effects that hold up in the lab are two-way: taking turns opening up, trying something new together. Laughing at the same thing goes with them, though that one's a correlation. Done to someone rather than with them, it isn't what was studied, and where one-sided versions were tested they came out weaker. Evidence: decent

So the honest advice isn't how to make her fall for you. It's how to give closeness a chance, and how to tell when it's there.