The 36 questions on a video call: what the study found

The 36 questions are famous for something the study behind them never claimed. The 1997 paper didn't set out to make anyone fall in love. It set out to build a way of making two strangers feel close for an afternoon, so psychologists could study closeness in a lab. That it works at all, in 45 minutes, is the interesting part. That it works for couples who already know each other well is a reasonable guess the research hasn't fully tested.

What the 1997 study actually did

Arthur Aron, Edward Melinat, Elaine Aron, Robert Vallone and Renee Bator published the procedure in Personality and Social Psychology Bulletin in 1997. Pairs of strangers spent 45 minutes working through tasks that escalated gradually, mostly questions in three sets, each more personal than the last. The two took turns answering.

• Set I is warm-up territory: who you'd invite to dinner from anyone in the world, what a perfect day looks like, what you're most grateful for.
• Set II moves to values and memory: what you value in a friendship, your most treasured memory, your relationship with your mother.
• Set III turns towards the person in front of you and the hard stuff: statements about "we", what you already like about each other, embarrassing moments, crying, death. It ends with each person sharing a personal problem and asking the other for advice.

In the first study, pairs who did this felt closer afterwards than pairs given comparable small-talk tasks. Two follow-up studies found that some things you'd expect to matter didn't change the result: matching pairs on important attitudes, telling them they'd probably like each other, or making getting close an explicit goal.

The authors were careful about what it doesn't do. As the Association for Psychological Science noted in 2015, they wrote it was "unlikely that the procedure produces loyalty, dependence, commitment" and the other parts of a relationship that take time.

Where "fall in love" came from

In January 2015, Mandy Len Catron wrote a Modern Love essay in The New York Times, To Fall in Love With Anyone, Do This. She tried the questions with a university acquaintance and finished with four minutes of silent eye contact on a bridge at midnight. She wrote that they did fall in love, though it may have happened anyway, and the essay went viral. The Times printed the full list alongside it.

Two details from that version get blurred with the science. The four minutes of eye contact isn't part of the 45-minute question procedure the 1997 paper describes, but it now travels with the questions almost everywhere. Even the Times's note on the list calls the staring less well documented than the questions. And the story that a pair from Aron's lab later married is an anecdote, not a finding.

What later research says

The procedure has become a standard lab tool, often called "Fast Friends", and later studies have tested parts of it.

Taking turns matters. In an experiment by Susan Sprecher and colleagues, pairs of strangers talking over Skype either went back and forth or had one person disclose for a whole session before swapping. The turn-takers liked each other more, felt closer and enjoyed it more.

Video works as well as sitting together. Sprecher's 2021 study randomly assigned new pairs to the closeness questions, the small-talk questions or an unstructured getting-to-know-you chat, either face to face or over video chat. The closeness questions produced more closeness than the other two, especially small talk, and whether the pair met in person or on video made no significant difference.

Couples have been studied, but with company. Welker and colleagues gave romantic couples a high- or low-disclosure version of the task, either alone or with another couple. The rise in passionate love showed up when couples did the deep version with another couple, and a second study found that how responsive the other couple was explained the effect.

That leaves an honest gap. Most of the evidence is about strangers, and the couples research points to doing it with friends rather than alone. For a couple apart, the 36 questions are a well-built conversation with a good track record, not a proven fix.

How to do the 36 questions with your partner, apart

Set up the call properly

You need about 45 minutes without interruptions, which is harder than it sounds across time zones. Pick a time when neither of you is at the end of a long shift or about to fall asleep. A Friday night in Toronto that's a Saturday morning in Sydney works better than squeezing it into a lunch break.

Before you start

Copy checklist

• Block 45 to 60 minutes with no hard stop afterwards
• Both of you open the same list in a separate tab or print it
• Laptop on a steady surface, camera at eye level, not a phone in your hand
• Headphones on, notifications off, door closed
• Agree that either of you can pass on any question

0 of 5 done

Keep the structure that makes it work

• Alternate who answers first, and both of you answer every question.
• Give each set about 15 minutes and move on when the time's up, even if you haven't finished it. Set III is the part least like an ordinary chat, so don't run out of evening in Set I.
• Read each question aloud. Over video, it keeps you both on the same one and gives the other person a beat to think.

Adapt it for someone you already know

Strangers get novelty from every answer. You won't, unless you change how you play. Before your partner answers, guess their answer out loud. Being wrong is the best part: it shows you where you've stopped paying attention, and being right is its own small compliment.

For questions you've already discussed, ask the follow-up: "Has that changed since we met?" Someone's perfect day at the start of a relationship and three years in can be very different answers.

The Set III questions about each other land differently too. Saying what you like about someone you've been with for years, out loud and on purpose, is rarer than it should be.

The eye contact problem on video

On a video call, you can't look into each other's eyes. If you look at their face on screen, you seem to be looking slightly down. If you look at the camera, you can't see them.

Three ways to handle the four minutes:

1. Move their video right under your camera and shrink the window. The gap shrinks with it.
2. Swap the stare for silence. Four minutes, cameras on, both looking at each other's face on screen, no talking. It still feels strange in the useful way.
3. Save it for your next visit. Do the questions now and the eye contact in person. It gives the visit a small ritual.

After the last question

Don't analyse it. Say one thing you're glad you heard, then do something light: put on an episode, play a game, say good night. The Truth or Dare questions are a good gear change if you're both still awake.

Key takeaways

Copy

• The 1997 study showed that 45 minutes of escalating, turn-taking questions made strangers feel closer than small talk did; it didn't claim to create love or commitment.
• The four minutes of eye contact and the marriage story come from the 2015 essay's retelling, not the 1997 findings.
• A 2021 study found the questions worked just as well over video chat as face to face.
• With your partner, alternate who answers first, keep about 15 minutes per set, and guess each other's answers before you hear them.
• On video, put their window under your camera or swap the stare for four minutes of shared silence.