Skip To Content

    An Anthropologist Walks Into A Bar

    Helen Fisher, science advisor for Chemistry.com, is one of a growing number of researchers using online dating as a laboratory to study personality.

    At a singles mixer one night in December, anthropologist Helen Fisher was helping a man in his thirties figure out what bracelet he should wear. The event would eventually bring in around 200 would-be daters, but at a few minutes after 6 p.m., the darkened bar held a sprinkling of men nervously sipping beers on their own. The women would arrive later, and more often in groups.

    The icebreaker for the evening was a version of a questionnaire Fisher had designed, dividing people into four personality types — Negotiator, Director, Explorer, and Builder. Each one had a color-coded bracelet. The man was worried his type, Director, didn't really describe him, and Fisher listened with a therapist's manner and a scientist's interest as he explained. "A real Director wouldn't say what you just said," she told him. According to her research, introspective types tend to be Negotiators: "The Negotiator," she said, "spends more time wondering who they really are."

    Fisher identifies herself as a Negotiator, and she's spent much of her life figuring out who people are. "I've always been an anthropologist," she told me. She grew up in a glass house — an actual glass-walled dwelling designed by architect Philip Johnson in Connecticut — as did her neighbors, and at 6 or 7 years old, she would sneak out to watch through their glass walls as they ate dinner. Now, at 67, she's a special advisor to the dating sites Match.com and Chemistry.com, for whom she studies personality — she's less a matchmaker than a taxonomist of people.

    Fisher's personality questionnaire is already an integral part of creating a profile on Chemistry.com, but the December event, part of Match's Stir series, marked the first time a version was used live as an icebreaker. She didn't work the room — she's a self-described introvert — but she did engage a series of people in intense, one-on-one conversation about things like whether they considered themselves tender-hearted. She eventually re-diagnosed the Director man as a Negotiator, and handed him a green bracelet. "Go find yourself some red women," she told him.

    One main tenet of her system is that Negotiators — empathetic, verbal, compassionate people — tend to go for Directors — skeptical, tough-minded folks who are good with numbers (they got the red bracelets at the mixer). Builders, who are conventional, loyal, and value stability, seek each other out; so, too, do Explorers, who, as their name suggests, crave excitement and novelty.

    Fisher said she arrived at these types not through psychology (like the developers of the famous Myers-Briggs personality test) but through biology. The characteristics of each personality type are associated with a different neurotransmitter. She hopes to test her system further by scanning people's brains and determining whether their patterns of activity match their questionnaires.

    Match.com executives tapped Fisher eight years ago, when they were planning to start Chemistry as a new spin-off site: Match.com CEO Mandy Ginsberg says, "We asked her, 'Why do people fall in love with one person rather than another?' And surprisingly, for the first time, I think we actually stumped her."

    Fisher was intrigued, but worried about her reputation as a scientist. Half her friends warned her to stay away. But, she says, working with Chemistry "hasn't ruined my reputation at all. I don't think academia really understands business, and business doesn't understand academics."

    When it comes to dating sites, that may be changing. As Dan Slater notes in his recent history of online dating, Love in the Time of Algorithms, the scientific advisor has become a fixture of many online dating sites (OKCupid is an exception — its cofounder Sam Yagan told me, "We've never allowed psychology to inform any of OKCupid's methods").

    And Slater thinks it may be paying off: "A lot of the users I spoke with felt as if online dating had gotten to a point where there's a fairly high likelihood of them hitting it off, at least on a first date, with the person they're matched with. It seems like the longer online dating is around, the better those algorithms will become, and it will be interesting to see how much the online dating data will influence the science."

    Now that dating sites are moving to mobile and asking users for post-date reviews of their matches, he sees an even greater opportunity for social scientists to learn how humans pair off: "What we've learned is that human beings aren't very good at assessing themselves or what they want. But if we watch how they behave, we'll be able to see what they want." And the new tools will give sites "courtside seats" in the dating process.

    Researchers are already taking advantage of the data that dating sites have mined. Even OKCupid, though it may not use psychology in its own algorithms, has made its data available to a number of professors. And Princeton biologist Lee Silver mathematically analyzed 100,000 questionnaires from Chemistry.com's database and found that "the way a person answered any particular question had an impact on how they answered every other question." This has implications beyond online dating: "It says that personality has a lot more structure than it's been thought to have." And the fact that certain patterns appeared in the questionnaire over and over again, he said, was strong evidence that personality has a genetic component.

    But his findings could be plugged back into the system too. He didn't specifically study which personality types matched up with one another, but he says that with more study, "I'm convinced there are going to be connections that can be made between personality and suitability for each other."

    Sociologist Michael Rosenfeld, coauthor of a longitudinal study called "How Couples Meet and Stay Together," is a bit of a dating-data skeptic. He notes that dating sites' users aren't a nationally representative sample — rather than being selected randomly, they've freely chosen to be on the dating site, so they might not be an accurate representation of society as a whole. And, he says, "as far as the science of relationships, I think that one of the things that's overstated about the dating websites is the 'amazing secret formula' for who they think you're going to like. Because the truth is, only you know who you're going to like."

    Still, he notes that online dating has amplified an existing trend in how Americans couple up: the declining influence of parents over mate choices. "The internet has increased the decline of family but also of friends and coworkers and school," he explains, "because the internet is a little bit more of an efficient marketplace, especially if you are looking for something particular."

    It's possible to imagine a kind of tipping point, where so many people are dating online (Match.com puts the figure at 27.5% of all Americans) that dating sites become nationally representative samples, making broad-based research on them possible, which in turn makes them more efficient than ever.

    Of course, this data-dating utopia (or dystopia, depending on your point of view) is only possible if personality research really can help make good matches. At the mixer, not everyone was convinced. "There's nothing in here that says 'in the middle,'" complained one man, pointing to the categories on the questionnaire. He didn't think he'd be looking at people's bracelets that evening.

    A bubbly woman who arrived with some friends, though, was more appreciative, calling the types a kind of "Cliff's Notes for personality." She was an Explorer, and said, "I would like to date more Explorers. When I don't, I get frustrated."

    With a Builder woman and Explorer man, two types the questionnaire suggests might not always hit it off, Fisher was diplomatic: "She'll be faithful, stable, and honest, and he will be exciting." Both felt their questionnaire's results accurately described them, but the man, at least, seemed undeterred: "I think opposites attract," he said.

    Fisher herself is an Explorer/Negotiator and tends to date men who are Explorer/Directors. "I've just met a new man," she told me. "I hope he's an Explorer. I'm positive he's a Director. He says he speaks two languages: English and math."

    She met him in the real world but says she wishes she met him online. She's a booster of the medium in a way that suggests she might champion it even if she didn't work for Chemistry: "If every single online dating company went out of business, they'd be recreated tomorrow. It's simple, it's easy, it's cheap." She has a profile on Chemistry, which makes clear that she's the author of its questionnaire, and she says she's met several men on the site as well, but none of them turned into a long-term relationship: All "were pretty well retired and I am going a mile a minute."

    She can't help thinking about the four types when she dates, though she tries not to jump to conclusions. "I just can't go out with a Builder," she said. And she's quick to analyze the personalities of people she meets. I'd taken her quiz before I met her, and been diagnosed a Negotiator/Director, two types she told me were opposites. "You might be tough-minded," she said, "but you don't have that aggressive look in your face. That's part of the testosterone system you didn't get."

    For the next version of her questionnaire, she said, she wants to get more specific: "I want to know which parts of the Director you are and which parts of the Negotiator you are. [...] I want to create almost a signature for you." She notes that of 100,000 people who took her personality questionnaire on Chemistry.com, no two answered exactly the same way. And Fisher herself is an identical twin, but her sister isn't just like her — in fact, she's a hot-air balloon pilot. Still, Fisher says they're both creative and risk-taking. "There are styles of thinking," she says, "and every time I do research I see it again."

    Her "signature" might enable daters to look beyond someone's primary type and into all their nuances. It might help Fisher too: "Now if my man has some Builder in him, it's maybe time for me to like that. I need some of that stability and permanence and dedication to the couple."

    Fisher doesn't have data on how effective her questionnaire is at getting people into enduring relationships — she says no dating site really has solid stats on its own success rate. But she did find that of 5,000 Chemistry.com users surveyed after their first date, 81% would go out with the person again.

    We haven't yet hit the data-dating tipping point, and as the crowd got larger and looser, it was easy to see the upside of in-person meeting. One man who had been to two prior Stir events told me he liked mixers because they were "better than looking at pictures." Women were more likely to say they liked the feeling of getting out in the world. At the end of the night, the pile of yellow Explorer bracelets was the most depleted; Stir organizer Heather Truettner said, "the New York woman is an Explorer."

    But even if it doesn't lead to a perfect world of seamless online matching, Fisher's questionnaire has an obvious appeal: People love analyzing their own personalities. Even when daters at the mixer didn't agree with the questionnaire's assessment of their inner selves, they wanted to talk to Fisher about why.

    And she wanted to listen — I got the sense that Chemistry.com is Fisher's glass house now, and its users are the neighbors she loves to observe. She believes her personality questionnaire will one day have applications far beyond dating: "I think and hope it'll get put to use in many ways, perhaps after I'm dead. So bounce me a message off the moon and say, 'Helen, you were right.'"