14 DAY TRIAL // According to a new series of investigations, it would appear that changes in behavior take place faster through tightly-knit social networks, than they do through groups with many distant ties.
This spreading patterns is a lot different from the one employed by news or infectious diseases, which prefer the latter variant. In the case of behavioral changes, redundancy is the key.
In other words, people are more likely to change the way they act if they see larger and larger numbers of individuals doing the same thing. This provides a larger incentive for them to change.
“There has been a lot of theory about the difference between information and behavior spreading,” explains Massachusetts Institute of Technology (MIT) economic sociologist Damon Centola.
“We’ve assumed that they are the same, but you can imagine that behavior is not really like that, that you need to be convinced,” adds the expert, who was also the author of a new scientific paper detailing the findings.
The work appears in the September 3 issue of the esteemed journal Science, Wired reports. The finding has important implications for those seeking to found online communities for changing or maintaining some type of behavior or another.
The team leader created an entire Internet-based health community as part of an intricate series of experiments, and invited people from other online communities to join in.
About 1,500 individuals signed up, and they were all divided into two groups. The first was a network with numerous distant ties, while the other was a more clustered network, featuring numerous and overlapping connections.
Centola then set up various experiments within the site, including some in which he advertised a new health forum. He then watched how many people from both groups signed up for the forum.
No participant was able to interact with another one directly, but they could see each other's activities. As a result, more than 54 percent of people in the first group signed up for the forum.
In the case of individuals in the second group, the participation rate was significantly lower, at only 38 percent.
“O feel that the greatest contribution of this study has to do with the very unusual social experiment that it relies on,” Stanford University economist Tomas Barrios says.
“Usually experimental data for social experiments comes from hard-to-swallow lab settings, or if not, very low tension, low risk social situations that can be ethically intervened by experimenters,” adds the expert, who was not involved in the new investigation.