Sunday, December 21, 2003
Human social networks (as anyone who has read Malcom Gladwells The Tipping Point knows) tend to be organized in a sort of hub and spoke network (described technically as 'scale free' since there is no characteristic number of times any one person is connected to others). Clearly if you could vaccinate just the hubs you would be able to stop the spread of disease efficiently. The problem, until now, has always been trying to figure out who were the hubs.
Cohen and friends suggest doing so by asking individuals for a list of their friends and then randomly innoculating the individuals on the list. The idea is that the hubs (those highly connected indivudals we wish to innoculate) will have many more friendships and will thus stand a reasonably large chance of getting innoculated.