Almost every big life change, from starting a new school to changing jobs or moving towns, elicits the same advice: get involved and make friends. While there's some truth to that guidance, it takes more than simply participating in activities to grow a social network, according to research by an international team.
The researchers, led by Makoto Chikaraishi , professor at the Graduate School of Advanced Science and Engineering , Hiroshima University, assessed the current scientific literature on social networks — the concentric and overlapping circles of connections among individuals — and developed a novel framework to account for the dynamic nature of how these networks evolve. They published their work, which included simulated scenarios on how various policies — such as those that restricted social interactions during COVID-19— have on social networks, on Aug. 27 in Transportation.
Scattered seeds of social networks only grow when nourished
"Social networks are dynamic, not static," Chikaraishi said. "We set out to capture social networks as adaptive networks that evolve with daily activities. We found that activities scatter countless 'seeds,' but few grow into lasting ties — and policies shape the soil in which they take root."
To better understand the evolution of social networks, the researchers surveyed newly enrolled international students at the University of Tokyo in spring of 2019, asking for details of each person's social network — specifically, with whom they spend leisure time. Then, the students completed a three-week joint activity diary. Each participant reported the details of joint activities conducted, including names and other details of those involved. The activities were divided into four types: going to an event, going out to eat, attending a party, and some other leisure activity like hiking. At the end of the three weeks, the researchers again surveyed respondents on their social networks.
While the sample group was small, with only 16 respondents, Chikaraishi said it offered a basis for developing a framework. They found that the 16 individuals made 1,344 activity engagement decisions over 21 days, resulting in 435 joint activities, or activities where more than one of the individuals participated. Of the 435 joint activities, 550 new encounters were made and 183 of those new connections were described in the second social network surveys.
When averaged across participants, international students logged about 86 person-encounter events over three weeks — not unique individuals, as repeat meetings and prior acquaintances were included — while only 13 unique friendships endured.
Using this information, the team developed a framework to simulate this data and found that these networks were dynamic and could shift. To further explore the framework, the team developed a baseline simulation comprising 200 international students from 10 countries.
"Our empirical results show that the proposed framework can successfully trace the social network growth of international students," Chikaraishi said. "We also show that the model can be used to get policy inferences: we confirmed that restrictions on out-of-home activities would considerably limit their social network expansion — the simulation shows that banning social activities can cut network growth by almost one-third."
Next, the researchers said they plan to scale up their adaptive network framework and integrate it with transportation simulations and large datasets.
"Recognizing social networks as adaptive systems opens new ways to evaluate urban and transport policies through their social consequences," Chikaraishi said. "Our ultimate goal with this work is to provide policymakers with a tool that links mobility, urban design and social resilience — ensuring cities are planned not only for movement, but for connection and community."
