Trends Need More Than Shepherds and Sheep

Influencing tastes across social networks is a tricky business: a love of “Love Actually” spreads differently than a love of “Pulp Fiction.”
  • Transcript


Question: How can we manipulate social networks to spread desirable effects?

Nicholas Christakis: Well there are two distinct issues here. One is how do networks naturally come to evince particular spreading processes and the different one is how might we manipulate these ideas to seed networks affirmatively with desirable properties and the way we think about this in terms of policy interventions is there are two broad classes of things we might do. We might manipulate the pattern of connection between people or we might manipulate the pattern of contagion between people.

So simple examples of connection are ones most people are familiar with. For example Alcoholics Anonymous or Weight Watchers are artificial social networks. You take a group of people and you form a set of ties between them and having done so you achieve or exploit a kind of power of social networks to magnify whatever they are seeded with, so you get this positive reinforcement between these people and just like the bucket brigade example we discussed earlier, these people now assembled into a little artificial miniature network that is focused on alcohol you know cessation or weight loss are now able to do something or at least do something better than the individuals acting alone. And it’s not just that the weight loss like if you had 10 people and you gave them each an intervention and they each lost a pound they would lose a total of 10 pounds, but you put these 10 people together in a room and you have them interact with each other in order to lose weight and they might lose a total of 20 pounds. So the group, now having been assembled and this interaction having been fostered you get more bang for the same efforts, same people, but now with the ties between them you can get more benefit.

So one set of interventions would be focused on manipulating the connections between people, creating connections, cutting connections, particular patterns of connections forming the kinds of networks, the prototypical networks that we described earlier. A completely different set of interventions might be focused on manipulating patterns of contagion within the network. Here, for example, you might map the network of a set of school kids in a classroom and target specific kids to receive a smoking cessation message or a seatbelt use message. With the idea that if you were able to persuade that one individual you would get, sort of, cascade benefits to other people. So it’s not just that one person that benefits, but their change in behavior influences others and so forth, so that you get… you can collect all of that benefit just for the one cost that you bore in changing the behavior of one individual. So here the idea might be that there might be particular individuals who are more influential within the network, but this is actually a very complicated area because it’s not enough to have influential individuals. You also need influencable individuals. And it’s not enough to just have shepherds and sheep, but you also need them to be connected in very particular ways, so I don’t want people to think that "It’s just very simple. You just map the network. You find somehow the most influential person and you just target them and then everything is terrific." First of all, behavior change is difficult regardless, but second it’s not always obvious how best to intervene and manipulate networks.

So I’ll give you one simple example. We mapped a network of college students and we looked at their tastes in things. We used actually Facebook in this particular case to get our data. Most of our data is real life, face-to-face networks or experimental networks that we create. This was an online network example. And we looked at what these students listed as their favorite taste in movies and we found that of the top 10 movies that they listed not all of the movie tastes spread. So for instance, if I listed "Lord of the Rings" as my favorite movie, it didn’t affect my friends to also be interested in "Lord of the Rings." Maybe they were interested in "Lord of the Rings" anyway from advertising or they heard about it or God knows what, but it wasn’t because I chose "Lord of the Rings" that it affected them. But certain movies if I listed a taste in this movie it did affect my friend’s interest in the movie. For instance, we found that if people picked "Love Actually" as their favorite movie it affected whether their friends liked "Love Actually" or if they picked "Pulp Fiction" as their favorite movie that it would influence their friends to like "Pulp Fiction" as well. Okay, fair enough. Well then when we mapped the network what we found was that the seemingly most influential individuals or at least individuals you might superficially think of as influential, those in the center of the network they were all "Pulp Fiction" fans and the "Love Actually" fans were located on the edge of the network. So if you’re someone that is trying to sell "Love Actually" and you think I’ll just map the network and find the central individuals it probably wouldn’t work because those are the "Pulp Fiction" fans. Trying to get them to watch "Love Actually" isn’t going to be so effective. So in this example the point is that a simpleminded kind of sense that "Oh, the central individuals—all I need to know is the structure of the network, I know who is central, I’ll just give them my message," isn’t likely to be effective because you need to know more about the system, more about the pattern of ties, the individuals within them and the processes by which things flow before you can really have an intervention.

Recorded March 31, 2010
Interviewed by Austin Allen