Since my student days, I have consistently researched the theme of "social dilemmas." A social dilemma refers to the phenomenon where the more rational choices people make (i.e., choices that are convenient for themselves), the more undesirable the resulting state of society becomes. Humans inherently tend to make rational choices, which easily creates an incentive to "free-ride" on the efforts of others without bearing the cost of cooperation. For example, a clear example in environmental issues is energy conservation. What is desirable for society is for everyone to slightly raise the temperature setting of their air conditioners in order to reduce electricity consumption in the summer. However, for individuals, this compromises the comfort of their rooms, so the psychology of "free-riding on the cooperation of others without bearing the cost" comes into play. In other words, while some people are willing to cooperate in considering energy consumption and environmental issues, others choose to keep their air conditioner settings low and stay in a comfortable room. If everyone falls into this mindset, no one will conserve energy, leading to a vicious cycle where electricity consumption becomes strained and CO2 emissions increase. This series of negative chains is a classic example of a social dilemma. This phenomenon is deeply connected to various social problems, from a shortage of volunteers to the depletion of natural resources. While investigating and observing such cases, we are studying the mechanisms that encourage cooperative behavior among people from the perspective of "computational social science."
Computational social science is a field of study that combines computational technology with social science to logically explain human behavior and social phenomena through the analysis and modeling of big data. In this field, knowledge from social sciences such as political science and economics is utilized in addition to information technology.
My first project as a student was modeling mutual aid in labor unions. In the field of labor sociology, a successful example of a certain community union* was well known, but there had been little attempt to theoretically explain why mutual assistance among union members worked so well. So, I decided to re-examine it from the perspective of social dilemmas, rather than from the perspective of conventional labor research. Community unions have a high turnover of members and consist of very fluid social relationships. The fact that cooperative behavior continues even within such a system is theoretically very interesting and a phenomenon that is difficult to explain with general theories of cooperation. In reinterpreting these phenomena from the perspectives of sociology and economics, I realized that there are many mysteries and unexplained structures in society.
*A community-based labor union. Anyone can join, from temporary workers to foreign workers.
Labor union survey (Professor Obayashi is on the far right)
Recently, we've been focusing on a service offered by a certain mobile phone company that allows users to share data packets (data capacity). This service pools unused data packets among subscribers and replenishes them when data capacity is insufficient. If each individual were to try to "get the best deal," everyone would subscribe to slightly less data and try to draw from the pooled packets to make up the difference, which should cause the service to fail. However, in reality, users are "sharing" data packets with each other, and the service is running smoothly. In other words, it's a case of "shared resources" in economic terms working effectively. So, why is this a success story? By analyzing the subscribers' behavioral data, we found that some people perceive it as "a system where you can get other people's unused data packets," while others see it as "a place for mutual assistance," "a place to practice altruistic behavior," and "a piggy bank where you can save leftover data capacity." It became clear that the service is built on the intersection of diverse values, and is supported by people with various perspectives. We fed the insights gained back to the company in question, and they used them to improve the service. My research activities also involve logically unraveling people's behavior by utilizing not only sociological insights but also computational techniques and big data.
