Fowler's friendship study has its roots in physics.
The field underwent a revolution in the late 1990s, Fowler said, when researchers began to notice that just about everything that forms connections—from atoms to neurons to communications systems—follows simple rules of organization. The connection points, or nodes, of such networks are interchangeable.
Using the example of a computer network, Fowler said, "any router can be replaced with any other router, and you'll end up with exactly the same network structure."
But then researchers began to encounter exceptions to the rule.
Ginestra Bianconi, a postdoctoral researcher and networks expert at the the Abdus Salam International Center for Theoretical Physics in Italy, co-authored a 2001 paper that found mathematical parallels between the behavior of ultra-cold gases and human networks.
Her paper was the first to propose that the nodes of a complex network—in this case, Web pages on the Internet—differ in their degrees of fitness, and thus in their ability to compete for links.
Such research led the way to Fowler's. Humans, as it turns out, also differ within their networks—we're "more like snowflakes than interchangeable cogs," he said.
But the differences in humans can't be chalked up to fitness in the same way as Web pages, because a greater number of connections between people isn't always the best evolutionary strategy.
Past researchers have proposed that natural selection could be influencing social styles, but Fowler reasons that if one strategy were better than the others, it would have become dominant by now.
He thinks evolution has allowed for the maintenance of a variety of social networking styles, presumably because each is adaptive in some way.
Members of close-knit networks, for example, potentially benefit from increased support but are also exposed to a greater risk of conflict.
Bianconi called the new paper "very interesting."
"This paper is new because it shows surprising fact that the [variability] of social behavior might have a genetic component," she said.
But Bianconi pointed out that the study doesn't account for how human social systems evolve with time.
Because the study population comprised adolescents, a follow-up would be worthwhile to see if genes hold as much sway as people mature, she said.
Fowler agrees that more work needs to be done. He and his colleagues are looking to extend their experiments to social networking sites such as Facebook, where larger sample sizes abound. And they've continued to search out new links in the genes-social networks-human behavior loop.
His next paper, co-authored with Christopher Dawes, also a political scientist at UCSD, identifies a gene potentially linked with the tendency to join political groups.
Fowler strongly believes in choice, however. If half of our social-networking preferences are genetic, the other half emerges from experience—which is partly influenced by the friends we choose.
"It's going to be a combination of your susceptibilities and your lifetime experiences," he said. "You have control over your life experiences."
SOURCES AND RELATED WEB SITES