Every few years, a piece of behavioral science escapes the lab, gets loose on the internet, and comes back to us wearing a costume. The “36 questions to fall in love” is one of my favorite examples. People tell me about it at dinner parties as though it were a magic trick — ask a stranger these questions and, poof, love. The truth is both less magical and, I think, considerably more interesting.
Let me first tell you what the researchers actually did.
In 1997, the psychologist Arthur Aron and his colleagues — Edward Melinat, Elaine Aron, Robert Vallone, and Renee Bator — published a paper with the unromantic title “The Experimental Generation of Interpersonal Closeness: A Procedure and Some Preliminary Findings”.
Notice that the word “love” does not appear in the title. Neither does “romance.” Aron was not trying to build a matchmaking device. He had a much more mundane problem, the kind researchers complain about over bad coffee: closeness between two people normally takes months or years to develop, and you cannot ask a grant committee to fund a study that runs that long. He needed a way to manufacture closeness on a schedule — in about the time it takes to watch a sitcom — so that he could then study it.
So he built a ladder. Thirty-six questions, arranged in three sets of twelve, each set slightly more revealing than the one before. Pairs of strangers were sat down together in the lab and given forty-five minutes to work through them, taking turns.
Set I is gentle and almost silly: Given the choice of anyone in the world, whom would you want as a dinner guest? By Set II you are on shakier ground: What is your most treasured memory? And by Set III the floor has quietly disappeared: When did you last cry in front of another person? By yourself? And: Share a personal problem and ask your partner’s advice on how they might handle it.
The control condition is the part almost nobody mentions, and it is the part that makes the study a study. A second group of strangers spent the same forty-five minutes together answering a matched set of small-talk questions — pleasant, harmless, the conversational equivalent of an elevator ride. Same room, same time, same amount of talking.
The pairs who answered the designed questions ended up reporting substantially more closeness to each other than the pairs who made small talk. Forty-five minutes with a stranger, and some of them reported a sense of connection they described as comparable to their closest relationships.
Also, one detail that has been retold so often it has become folklore: a pair from Aron’s lab reportedly went on to marry, and invited the research team to the wedding. Aron himself tells the story. But it is an anecdote.
For eighteen years the procedure lived where such things usually live — in citations, syllabi, and the occasional conference talk, under the affectionate nickname the “fast friends” paradigm.
Then, in January 2015, the writer Mandy Len Catron published an essay in the New York Times Modern Love column called “To Fall in Love With Anyone, Do This.” She had tried the questions in a bar with a university acquaintance. She wrote about it beautifully. The essay went off like a firework, and the Times ran the full list alongside it as “The 36 Questions That Lead to Love.”
On the staring. Catron and her partner finished by looking into each other’s eyes for four minutes, and this has since fused onto the 36 Questions exercise as though it were step 37 of the protocol. It was not part of Aron’s 1997 study at all. The idea comes from a separate line of research — Kellerman, Lewis, and Laird’s 1989 work on mutual gaze, which had strangers hold unbroken eye contact for two minutes and found increases in reported feelings of passionate love. It is a real finding from a real study. It is simply a different study. Catron combined them, which was a lovely piece of writing and a slightly confusing piece of citation.
A few years ago Kristen Berman, my longtime collaborator and the co-founder of Irrational Labs, hosted a dinner for twenty-seven people with exactly one rule: no small talk. Not less of it. None. Guests all had to arrive at the same moment, so that nobody could ease in with the weather while the host took their coat. Where the place cards would normally go, there were index cards with questions on them. And the enforcement was handed to the guests — if a conversation drifted toward traffic, or real estate, or what somebody did for a living, anyone at the table could call it and change course.
People were still talking three hours later. Some of them were still thinking about those conversations a week later.
The line from Kristen’s account of that evening that I have not been able to stop repeating is this: the questions are not the technology, the permission is the technology. Asking someone at a dinner party what they would spend the hour on in a therapy session is prying — you made a unilateral move and you own the consequences. When a card asks it, nobody made a move. The card is the one being nosy, and if the answer gets uncomfortable you can both blame a piece of printed cardboard.
The basic idea is that we are not choosing small talk because we like it. We are simply following the social norm of not asking anything that could be interpreted in a bad light.
That dinner party eventually became a deck. Kristen and the team at Irrational Labs turned the index cards into No Small Talk, tested it in the field, and put up a free online version:
Recently I took that version of the No Small Talk cards and built it into an app for two people, sequenced and tuned for romantic purposes rather than for dinner parties, and it is the version I recommend for couples when people ask me about the 36 questions:
decisionaids.danariely.com/romantic-journey
Three design choices in it are worth explaining, because each one is there to solve a specific failure of the original.
You choose the deck together, before you start. Aron’s list has a fixed depth, decided in 1997 by people who had never met you. Here you decide how far you are both willing to go, and you decide it jointly. That negotiation is not a preamble to the game — it is the first move of the game, and it is important as well.
You alternate. Every card gets two answers. This is the symmetry constraint from the lab, enforced by the software instead of by your goodwill.
You can always skip — and the skip is visible. This one I am fondest of. Forcing an answer turns a game into an interrogation, so passing has to stay available. But a pass that vanishes without a trace is just an escape hatch, and you will use it forty times. So the app keeps count where you can both see it. Nothing happens as a result. Nothing needs to. Knowing that the other person can see that you have skipped four times is quite enough to make you answer the fifth — and if it is not, then the pattern of what you are skipping has become its own conversation.
I recommend this exercise for people who are just starting dating and for people who’ve been together for a long time. Try it out, even if just a few cards to get the conversation started. I’m promising you that it’s going to be more meaningful than you expect it to be.
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.