Scaling up commons dilemma experiments for research and education

当代社会面临的许多挑战是涉及共享资源（称为“公地”）的集体行动问题。例子包括减少排放以减少气候变化风险，以及戴口罩以减少COVID-19的传播。在集体行动中，社区目标和个人目标之间经常存在紧张或冲突。在2019冠状病毒病危机期间，戴口罩和洗手减少了病毒的传播，使曲线变平，尽管这可能会给个人带来不便。这个项目调查了这一研究领域的三个基本难题。首先，是什么让沟通在促进合作成果方面如此有效？众所周知，即使参与者没有能力履行承诺，沟通也会产生重大的积极影响。该项目的目标是了解沟通内容的哪些方面解释了成功的合作。第二，一个群体的规模如何影响其成员的合作能力？在较小的群体中，更容易确定谁在搭便车。然而，小团体可能没有足够的人力来监督资源的分配。第三，解决集体行动问题的团队如何应对不确定性和意外情况。2020年的各种事件，如大流行，证明了在不确定情况下理解集体行动的重要性。研究人员已经研究了风险和集体行动，已知事件的概率，但在理解群体如何应对未知的未知方面存在差距。为了解决这些难题，有必要扩大控制行为实验的规模。该项目为“公共”研究建立了一个强大的平台，也可以作为一个引人入胜的教育游戏。该项目正在指导许多具有不同背景的研究生和本科生进行研究设计、数据收集、分析和编程。项目成果包括大学水平课程的教育资源，以及一个特殊的K12版本的游戏，附带教师的教育材料。这个跨学科研究团队扩展了一个基于网络的实验，即“火星港”，在这个实验中，一群参与者决定投资于共享的基础设施，并采取有利于自己和对团队产生影响的行动。该平台可以通过游戏内聊天交流和随机事件进行大规模控制实验，这些事件将不确定性和变化引入每个游戏过程。大量的参与者（包括大学生和普通民众）参加了一场名为“火星疯狂”的比赛，在比赛中他们竞争成为火星港的冠军。六个实验设计测试了关于沟通、群体规模和不确定性的假设。在比赛期间收集的数据用于训练机器学习模型，该模型将通信数据分类为功能类别，这有助于更好地理解球员之间的通信与团队表现之间的关系。在各种锦标赛中，团队规模在5到50人之间，通过控制玩家对事件的了解和调整触发事件的阈值，可以评估不确定性对集体行动的影响。此外，研究还表明，参与行为实验可以在一定程度上提高对集体行动问题的理解。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Many of the challenges facing contemporary society are collective action problems involving shared resources (known as “commons”). Examples include emission reductions to reduce risks of climate change, and mask-wearing to reduce the spread of COVID-19. In collective action, there is often tension or conflict between the goals of the community and the goals of the individual. Wearing masks and hand washing during the COVID-19 crisis reduces the spread of the virus and flattens the curve, although it might be an inconvenience for the individual. This project investigates three fundamental puzzles in this research area. First, what makes communication so effective at stimulating cooperative outcomes? It is known that communication, even if participants have no ability to enforce promises, has a major positive effect. The goal of the project is to understand what aspects of communication content explains successful cooperation. Second, how does a group’s size impact the ability of its members to cooperate? With smaller groups, it is easier to identify who is free riding. However, small groups may not have sufficient person power to monitor resource appropriation. Third, how do groups that address collective action problems cope with uncertainty and surprises. Various events in 2020, such as the pandemic, are testament to the importance of understanding collective action under uncertainty. Researchers have investigated risk and collective action, with known probabilities of events, but there is a gap in the understanding of how groups cope with unknown unknowns. To address these puzzles, there is a need to scale up controlled behavioral experiments. This project builds a robust platform for “commons” research that also serves as an engaging educational game. The project is mentoring many graduate and undergraduate students, with diverse backgrounds, in research design, data collection, analyses, and programming. Project outcomes include educational resources for college level courses, as well as a special K12 version of the game with accompanying educational material for teachers.The transdisciplinary research team expands a web-based experiment, the Port of Mars, where a group of players make decisions to invest in shared infrastructure and perform actions that benefit themselves and have consequences for the group. The platform enables the conducting of large-scale controlled experiments with in-game chat communication and random events that introduce uncertainty and variance into each play-through. A large number of participants (both college students and members of the general population) take part in a tournament, coined “Mars Madness”, where they compete to become the champion of Port of Mars. Six experimental designs test hypotheses on communication, group size and uncertainty. Data collected during the tournaments is used to train machine-learning models that classify communication data into functional categories which help to better understand how communication among players relate to group performance. In the various tournaments group sizes vary between five and fifty players and they make possible to evaluate the impacts of uncertainty on collective action by controlling players’ knowledge of events and adjusting the thresholds that trigger events. Furthermore, the studies show the extent to which participating in the behavioral experiments improve the understanding of collective action problems.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.