Understanding why loneliness can spread through society like a disease is a key question for social scientists. Credit: E. MARCH/CORBIS

How can we persuade people to look after their health? Why do moods spread like a contagion? How can humanity increase its collective wisdom?

These are some of the most pressing questions that social scientists should tackle, according to a group of leading scholars in the field who hope that their 'top ten' list will help shape the thinking of researchers and funding bodies for decades to come.

In a parallel effort, the US National Science Foundation (NSF) last week unveiled the results of its own agenda-setting exercise, which asked social scientists to identify "grand challenge questions that are both foundational and transformative".

Both groups say that they ran the exercises because they wanted researchers to step back from immediate research priorities and identify the most significant problems in their field. The results demonstrate the growing ambition of the social sciences to tackle difficult issues in a quantitative way, addressing problems from equality and wages to wars and health.

The 'top ten' approach was inspired by a list of 23 major unsolved questions compiled by the mathematician David Hilbert in 1900. The Hilbert problems helped to focus the attention of mathematicians throughout the following century. "He laid out the road map for twentieth-century math," says Nick Nash, a vice-president at General Atlantic, an investment firm based in Greenwich, Connecticut. "What if we had a road map for other disciplines?"

In 2008, Nash was studying for an MBA at Harvard University in Cambridge, Massachusetts, when he proposed the road map to Stephen Kosslyn, then the university's dean of social science. Together, they organized a symposium at Harvard last April that gathered together 'big thinkers' to present unsolved questions and to vote on which were the most important. The results are due to be released this week on Harvard's website (see http://go.nature.com/wa1hwf and 'Top ten social-science questions'). The site will also include a range of questions submitted by members of the public.

At the symposium, Emily Oster, an economist at the University of Chicago, Illinois, focused on a perennial challenge for public-health experts: how to get people to adopt healthier behaviours. For instance, persuading people to eat less and exercise more — to control ballooning obesity rates — might be simple in theory; in practice it is extremely difficult.

Because the rewards of behavioural change are often not apparent for years, Oster thinks the answer lies in programmes that offer an immediate pay-off. Preliminary studies have shown, for example, that cash rewards contingent on hitting weight-loss targets can help1. Even if payments amount to hundreds of dollars a month and, to prevent a relapse, are continued after dieters have shed their excess pounds, the strategy might save society money by reducing future medical expenses.

The approach is not foolproof, though. One recent large-scale experiment, aimed at financially rewarding low-income New York City families for keeping children in school and taking regular medical check-ups, was halted last year after it produced only limited improvements (see http://go.nature.com/eunolm). Oster says that researchers need to experiment with different reward systems, as it is not clear how the best system for a particular problem should be chosen.

Nick Bostrom, a philosopher at the University of Oxford, UK, who was also involved in compiling the Harvard list, wants social science to improve society's "ability to get the important things approximately right". He notes that judgements by specialists are often no better than those made by laypeople2, and suggests that, rather than relying on individuals, society should develop and exploit new methods for aggregating knowledge.

In financial markets, for example, participants buy and sell shares on the basis of expectations about how the market will move. If enough people play the market, the price of the shares reflects traders' collective beliefs about future events. It is also possible to create artificial markets in which traders buy and sell shares related to specific events, such as a politician being elected. These markets also reflect traders' beliefs about outcomes, and can be good forecasting tools3. Bostrom would like to see such 'prediction markets' trialled more widely; they could assist with corporate decisions, such as whether to replace a company chief executive, he suggests.

Harvard social scientist Nicholas Christakis hopes to understand how physiological and psychological attributes, such as obesity and loneliness, can spread through a social network like a contagious disease, a phenomenon he has studied with James Fowler at the University of California, San Diego4. If one of your friends becomes obese, for example, your own chances of putting on weight increase. Christakis notes that there is unlikely to be a single theory that links social and biological factors. In the case of obesity, it may be that having an overweight friend somehow normalizes the idea of gaining weight. And preliminary work in poor neighbourhoods in Chicago suggests that loneliness and fear of crime can alter levels of stress hormones, which in turn can affect people's risk of cancer.

Nash and Kosslyn hope that drawing attention to difficult and important problems will motivate young researchers to work on them, just as young mathematicians were attracted to the Hilbert problems. "Nothing would make us happier than to see future grant applications that mention the 'Harvard problems'," says Nash. The similarity with the NSF's own exercise is not a bad thing, adds Myron Gutmann of the foundation's directorate for social, behavioural and economic sciences in Arlington, Virginia. "I'm delighted by [the Harvard exercise]," he says. "It allows us to look for repeated themes."

The NSF received more than 240 responses to its request for forward-looking ideas, which Gutmann plans to discuss with his advisory committee. "I can imagine that in the next two years we will identify a few ideas that seem especially important and invest in them, in the form of pilot projects and planning grants, with an idea that these investments will position us to make more significant investments 5–10 years from now," he says.

Cary Cooper, chair of the Academy of Social Sciences in London and a psychologist at Lancaster University, UK, is enthusiastic about the Harvard list. If funding were available to support work on the problems, he says, young researchers might feel confident enough to eschew simpler questions. Cooper adds that he will consider asking Britain's Economic and Social Research Council to run a similar exercise.

figure 1