Keeping in touch with distant friends, seeing which of your high-school classmates are married, viewing an endless stream of kitten photos … Facebook is good for a lot of things, but not many of them are actually productive. That could change after the release last week of an article suggesting that the power of the social-network behemoth could be harnessed to help influence our behavior for the better—for example, by helping smokers quit.
So says Thomas Valente, head of the University of Southern California’s public- health program and the author of the article, which was published in the latest issue of Science (paywalled). In it, Valente reviews what researchers know about so-called social-network interventions—attempts to kick-start or staunch certain behaviors by leveraging the power of social networks. In other words, seed the behavior you want at node A, and watch it spread to nodes B through Z (assuming you’ve assessed the network correctly and chosen the right intervention).
In an interview, Valente described to The Daily Beast how it would work in practice: “If we’re able to scan all the photographs that somebody has up on their Facebook account, and we see that there are some friends that smoke and some that don’t, [then] we can figure out how to trigger communications across Facebook that say, ‘Look, you’re smoking and your friends don’t. You may think that a lot of your friends approve of your smoking, but they probably don’t. How can we generate some supports for you [that will] enable you to quit smoking?”
Of course, social psychologists have known for decades that our social networks strongly influence us, that what our peers do can greatly impact our behavior. But it’s only now, with the explosion of online social-networking tools like Facebook and Twitter—which generate terabytes of useful data—and the advent of advanced network-mapping software, that researchers are ready to fully exploit the power of social networks.
“There’s a whole bunch of stuff that we know about human behavior: how it works, how to influence it and change it,” Valente said. “And now we have these new platforms where we want to ... graft these theories and ideas onto this new technology.
“Had I written this paper 10 years ago, it would have not really been much of a story,” Valente said.
Valente acknowledged that he and his colleagues aren’t the first to try to change behavior through online social networks. “Marketers are already doing that, and we’re just trying to catch up,” he said. There’s more money to be made in selling products than in stopping smoking, of course, so it’s no surprise that marketers have been the first wave of tinkerers to try to ride Facebook’s reach.
When it comes to antismoking interventions and the like, there are potential downsides, perhaps the most obvious being the risk to personal privacy.
Valente noted that ethical principles prevent scientists from disclosing private data without obtaining consent first, but he was also quick to point out that “there has always been an illusion of privacy, because very few people live in an entirely private world.” In other words, it would be easy for a researcher to ascertain someone’s position in a social network without doing much gumshoe-level snooping. And online, he said, “most people voluntarily give up that information.
“In some sense, the online-privacy ship has sailed,” Valente said.
‘There’s a whole bunch of stuff that we know about human behavior: how it works, how to influence it and change it. And now we have these new platforms where we want to ... graft these theories and ideas onto this new technology.’
Valente’s paper goes through the various strengths and weaknesses of different network-based interventions, a field he says is getting more nuanced and effective by the day.
One key insight, and the bridge that researchers like Valente hope to cross from theory to practice, is that different situations call for different approaches. For example, in one of the most basic type of intervention, so-called “champions” are used to spread a positive behavior—or a proscription against a negative one—through a social network. A champion can be any well-connected, influential person, so opinion leaders are a classic example, and they are often nominated by the members of the network themselves (if you wanted to identify the most influential members of a high-school class, just ask the students themselves).
But as Valente points out in the paper, targeting these sorts of leaders isn’t always the best approach—maybe they “have a vested interest in the status quo,” he writes, such as when the desired change isn’t a popular one. After all, why risk your high social status championing an unpopular policy? In instances like this, Valente said, a better bet is to seek out so-called “bridging individuals,” those people who serve as links between groups that are loosely connected, or not connected at all.
In other situations, it might make more sense to target a group rather than individuals, and sometimes it’s vital to stimulate peer-to-peer interaction instead. The point is that there’s no one-size-fits-all approach. Saying that we should use social networks to solve, say, a public-health problem is like saying we should use water to put out a fire: the devil is in the details.
Valente’s hope is that social-network interventions may not just stamp out harmful behaviors, but also change how the average person views the very notion of behavior in the first place. In the long run, network interventions could help shift people away from judgemental, stigma-laden responses to negative behaviors, Valente said—if you’re fat or smoke or do drugs it’s because of a personal failing, a choice you made—and toward a better understanding of how influenced we all are by the social worlds in which we operate.
“I think that we have had for a long time in the social sciences an overly individualistic view about what motivates most people’s behavior,” he said. But he already sees a shift taking place, at least when it comes to his field.
“I think young people today understand the concept of health disparities, and they do much less individual blame and much more understanding that the system is partly responsible for people’s behavior.”