Over decades, the Kavli HUMAN Project will track 10,000 New Yorkers’ lives.
Anyone who lives in a city is affected and influenced by an evolving network of environmental, economic, behavioral, and biological factors.
It’s an overwhelming concept, but that’s kind of the point for Paul Glimcher, an NYU professor and the director of the Kavli HUMAN Project—a first-of-its-kind study that, once it launches next summer, will track the lives of 10,000 New Yorkers in 4,000 families over the span of decades.
The HUMAN Project is unique in both its scope and its determination to capture everything that feeds into New Yorkers’ existences. Using smartphone technology like surveys, location trackers, and text data, Glimcher’s team will monitor individuals’ diets, medical conditions, neighborhood demographics, social interactions, GPS location, employment status, time spent on the phone, and so much more. The data will be stored in a firewall-protected, continually reencrypted database. Participants will remain anonymous, and only researchers on the project will be able to access the information, which they anticipate will be used to inform everything from urban policy to scientific studies.
To anyone who might say this sounds impossible, Glimcher has a response: it worked for astronomers.
In the 1990s, the Princeton astronomer James Gunn developed the Sloan Digital Sky Survey, a comprehensive data trove of every astronomical body observed at any point in time, doing away with the need for isolated studies.
Glimcher thinks it’s possible to apply this model to people, and create an intricate, informed database that could revolutionize behavioral science. CityLab talked to Glimcher about how his model will take shape, and what potential it has for transforming our understanding of cities and people.
Why have you decided to turn such a comprehensive lens on people and their environments?
As an academic, I work in a field called neuroeconomics, which applies psychology and economics to the study of human behavior. I became interested in the way large-scale forces shape decision-making and outcomes. Around five years ago, I was happily studying these things when I was approached by Miyoung Chun of The Kavli Foundation, which supports public understanding through science. She wanted my help in figuring out a new way to study human behavior; through a series of workshops we held with doctors, psychologists, anthropologists, sociologists, and ethicists, we kept hearing the same thing: that we’ve been going about the study of human behavior in exactly the wrong way.
This is where the Sloan Digital Sky Survey comes in. A scholar at one of the talks Miyoung and I hosted handed me a book on the survey, and said that all of the problems we’re facing now with human behavioral studies sound a lot like issues astronomers faced thirty years ago. If you were astronomer and it was 1985, and you wanted to study, say, the rate at which quasars are moving away from us, you would do so by booking time on a telescope for maybe three days a year, and if you’re lucky, you might get to observe eight or 10 quasars. The idea was that astronomers were, in these individualized, “artisanal” studies, gathering data to answer one specific question—there wasn’t a lot of potential for broad applications.
But after Gunn developed the Sloan survey, scientists could access a telescope that scanned the whole sky, taking spectroscopic images of everything; they poured all that information into a database from which scientists could extract the particular measurements they needed. Just by typing on a keyboard, they could answer questions that otherwise would have taken decades of research to solve. It’s obvious that my view is: Let’s do this for human beings.
How have you gone about translating this astronomical model to the study of people in New York?
The trick of the Sloan Digital Sky Survey is figuring out where to point the telescope. That’s our issue too, but in big data for humans, the telescope is smartphones, it’s web browsing histories, it’s geo-trackers. That telescope is owned by a bunch of large companies, and it’s pointed at a very specific 60 percent of the American population—mostly white people who live above the poverty line, and fall between the ages of 10 and 62. But we wanted to capture a really representative cross-section, so we’ve identified 10,000 New Yorkers who are a perfect mirror of the city, with slight overrepresentation of children, the elderly, and people living in poverty. Apart from medical exams, all of our data collection and surveys will be done through smartphones; we will supply them to people who don’t have one.
Through this smartphone telescope, you’ll be aggregating a huge swath of comprehensive data on these people’s lives, over the course of at least 20 years. How might this information play out in the city at a policy level?
City agencies like the New York City Department of Health and Mental Hygiene already run studies tracking things like how the flu is moving through the city, or how HIV [rates have] changed over time. But the approach previously has been that sort of hyper-focused, “artisanal” model, like astronomers used pre-Sloan survey. We imagine that with our data, we’ll be able to do these studies with more people, greater precision, and faster response time—a lot of the research that’s already going on in the city, we’ll be able to do better and cheaper.
But there are also a bunch of things that we don’t have great traction on, that this data might help to illuminate. One is the effects of economic development. We know a lot about how, say, 10 square blocks of the built environment changes after the city does a project like the Barclays Center or Hudson Yards. We know a lot less about the people who live in that block group, and what happens to them. We know that some of them move out as neighborhoods develop; we don’t have great clarity about where they move to. We don’t know if they turn out better or worse. We don’t know how their social networks change. We know that people’s housing situations are so entangled with health outcomes, but it’s been difficult to apply that information in current policy environments. This is a case where we need a more holistic, longitudinal, and representative view than we have.
Do other people agree with you? This is an unprecedented study to undertake. What has the response been like?
Whether we’re talking to city officials or academic researchers, people have told us that this data would give us the tools to address some pretty difficult concerns: the effects of inequitable wealth distribution; outcomes based on children’s school circumstances; the efficacy of end-of-life care for elderly people; environmental factors that contribute to diseases like Alzheimer’s and cancer. We’re still recruiting individuals to participate in the study, but people we talk to agree that this is a civic project of enormous importance. The thing we hear over and over again is gosh, this could change the world.