People looking to adopt new health practices are more likely to be influenced by close connections, including people they know well, than by social networks such as Facebook, research suggests.

New research from MIT shows that closely connected communities are more efficient and faster at spreading information and influencing behaviours than larger, more diverse groups. (IStock)Researchers at the Sloan School of Management at the Massachusetts Institute of Technology conducted an elaborate experiment that involved trying to convince people to sign up for an internet-based health community.

Study leader Damon Centola, an assistant professor at MIT, says the result flies in the face of conventional wisdom.

"For about 35 years, wisdom in the social sciences has been that the more long ties there are in a network, the faster a thing will spread," he said.

"It's startling to see that this is not always the case."

Centola placed 1,528 people into one of two distinct social networks. One network featured so-called long ties, a large group of relatively unrelated people, the other a more tightly focused group with people who shared common interests.

Throughout the trial, researchers followed which group was more likely to influence people to join an online health forum website and how quickly it took participants to join.

In the more focused group, 54 per cent joined the forum compared to 38 per cent in the other group. The rate of adoption was also four times faster in the focused group.

Centola says his work shows that groups built around closer ties work best at disseminating information or behaviours than larger, more random groups.

The MIT study is published in Friday's issue of the journal Science.