Contagion is usually bad for one's health, and something to be avoided at all costs — however, if the MIT team's findings are correct, this may be one virus to catch. In all, the group analyzed the fitness tracker data of 1.1 million people, who ran a collective 350 million kilometers in the span of five years. Subjects were chosen on the basis their fitness tracker stats were automatically posted on social media after every run, a research model that by definition circumvents the issues inherent in self-reporting.
exercise can be contagious too https://t.co/Jiz8Fsf3P3 @sinanaral
— NatureCommunications (@NatureComms) April 18, 2017
In that time, these runners also formed about 3.4 million new social network connections — from this group, the team isolated the 2.1 million ties for which they could pinpoint specific information such as geographic location and weather conditions for both users.
The findings were stark — an additional kilometer run by a user's friends influenced them to run an additional third of a kilometer, and an additional kilometer per minute run by friends pushed them to run an additional 0.3 kilometers per minute faster than usual. Moreover, if those friends ran an extra 10 minutes, an individual is likely to run about three minutes longer than they would have otherwise. If those friends burn an extra 10 calories, that person will end up burning 3.5 more calories than usual.
Exercise is Contagious! Our latest paper w @CNicolaides on influence in networks out now in @Nature Comms! #passiton https://t.co/ESMTY1hib0 pic.twitter.com/qu01YQMANg
— Sinan Aral (@sinanaral) April 18, 2017
However, the effect was found to be transitory — the motivational qualities of fitness tracker stats were potent on the day they were posted, but quickly waned. Moreover, not all runners influenced their connections equally, with individuals more likely to up their game in response to increased performance by less active peers than more active ones. Inconsistent runners influenced consistent runners far more than vice versa.
There was a gender divide evident in the findings too, with male runners found to be influenced by the activity of both men and women, and women influenced only by other women.
As for what accounts for this disparity, the team believe social comparison theory, the notion individuals self-evaluate by comparing themselves to others — may provide an explanation.
"If behavioral contagions exist, understanding how, when and to what extent they manifest in different behaviours will enable us to transition from independent intervention strategies to more effective interdependent interventions that incorporate individuals' social contexts into their treatments," the authors wrote.
"Running on a Sunny Saturday in NYC." Visualization by @CNicolaides. From our recent exercise paper in @NatureComms: https://t.co/9eEVtOrSuV pic.twitter.com/37YBmX8MsH
— Sinan Aral (@sinanaral) April 19, 2017
The study could represent a breakthrough, as prior investigations into the ways in which individual decisions affect the decisions of our peers, and how behavioral changes may spread through social networks, have been hampered by a simple but irresolvable quandary — do humans make "upward" comparisons to peers performing better than them, or "downward" comparisons with those performing worse? The study suggests both considerations likely figure in an individual's competitive calculations, albeit to varying degrees.
"Comparisons to those ahead of us may motivate our own self-improvement, while comparisons to those behind us may create competitive behaviour to protect one's superiority. Our findings are consistent with both arguments, but the effects are much larger for downward comparisons than for upward comparisons," the paper said.
In essence, people who individuals perceive to be their closest fitness "peers" — particularly those they think are slightly lower on the totem pole relative to themselves — are most likely to inspire someone to push themselves to the limit.
The team also think their study could have seismic implications for tackling issues such as obesity and smoking. By monitoring such real-time networks, scientists, doctors and engineers may be able to design applications and policies capable of minimizing the spread of social ills, and maximizing social benefits.
In the least, the study certainly indicates focusing on averages in policy construction is an erroneous approach. Different subsegments of the population evidently react differently to social influence, meaning policies tailored for different types of people in different subpopulations will be more effective than policies constructed with only averages in mind.
