Agent-based study on social broadcasting – ScienceDaily
Over the past year, handshakes have been replaced by punches or nudges in greeting. This shows that age-old social conventions can not only change, but do so suddenly. But how does this happen? Robotics engineers and marketing scientists from the University of Groningen have joined forces to study this phenomenon, combining online experiments and statistical analysis into a mathematical model that shows how an engaged minority can influence the majority to overthrow long-standing practices. The results, published in Nature Communication September 29, can help stimulate sustainable behavior.
How does complex human behavior take shape? This is studied in several ways, mainly relying on a lot of data from observations and experiments. Ming Cao, professor of networks and robotics at the Faculty of Science and Engineering, University of Groningen, studied complex group behavior in robots using, among other methods, agent-based simulations . These agents follow a limited number of simple rules, often inspired by nature, which can lead to realistic complex behaviors. “Swarms of birds or schools of fish are a good example,” explains Cao, “their movements can be reproduced by agents who follow a few simple rules to keep a certain distance and point in the same direction as their neighbors. “.
At the same time, the Marketing Research Group of the Faculty of Economics and Business Sciences, led by Dr Jan Willem Bolderdijk, Dr Hans Risselada and Professor Bob Fennis, carried out various research projects on human behavior, but little using this type of agent-based models. After a discussion with Cao and his colleagues, both groups saw possibilities for such models. Therefore, marketing doctoral student Zan Mlakar and the two post-doctoral researchers in Cao’s group, Mengbin Ye and Lorenzo Zino, worked together to create an online experiment to collect data on the social diffusion of new trends. behavioral.
They have developed an online game in which 12 participants act as board members of a company that is considering launching one of two potential products. Participants must vote on the product to be launched. The catch is that the decision must be taken unanimously. Participants cannot discuss their choice, they vote in 24 consecutive rounds, and they only see the distribution of votes at the end of each round. If unanimity is reached, participants receive a reward.
Unbeknownst to the participants, between two and four participants in the groups studied were computer robots, programmed to stick to their choice. “If the majority voted for product A in the first round, the robots would have to vote for B in an attempt to overthrow the majority,” says Ye, who now works as a senior researcher at Curtin University in Australia. During this time, the votes of the human participants on all the tours studied were recorded. The vast majority of over 20 of these online gaming rounds resulted in a unanimous vote, with humans eventually siding with the bots to vote for Product B. The results of all games were then analyzed for evidence. trends in human voting decisions. participants.
Ye: “In quite a few cases we saw a delay before the votes started to change, but when they did, the group would reach unanimity within a few rounds of voting. The overall voting behavior could be reproduced in an agent-based model with three simple rules: do like the majority, stick to your previous decision and go with the trend. “These rules are recognized in the literature as group coordination, inertia, and trend research,” Ye explains. “They have been studied separately in human behavior, but never combined in a model; this combination was essential to capture social change.
The results of experiments and simulations show that new conventions can suddenly arise when the influence of a committed minority reaches a threshold. A small group of “activists” can therefore change social conventions. Cao: “However, this only happens if the minority is also able to influence other members of its network. And it depends on how much risk-taking is present among other voters. The team is now interested in exploring what might improve or inhibit this risky behavior. “We now have a solid framework and a model, which can be used to examine environmental factors that might make people more inert or more sensitive to trends,” Ye says.
The three basic rules could help guide the behavior of large groups. “Of course, we can’t control people,” Cao points out. “But we can provide guidance, for example on how to get people to change their behavior.” This could be useful in the energy transition, or to get people to reduce their meat consumption. “Governments are already spending money to convince people to adopt more sustainable behavior. Our research can help them spend it more efficiently. ‘
Source of the story:
Material provided by University of Groningen. Note: Content can be changed for style and length.