The goal of this proposal is to support learners (from high school students to undergraduates and beyond) in reasoning about complexity and in exploring the connections and interdependencies between complex systems. In particular, we propose an extension to the NetLogo agent-based modeling language and environment (Wilensky, 1999); an extension that enables an important and intrinsic expansion to the process of agent-based modeling itself. The technologies proposed will enable learners to link two or more independently-created agent-based models together, specifying the ways in which each model will affect the behaviors, conditions, or parameters of the other(s). Linked models will be able to run simultaneously and interact in real time. As a result, learners will be able to create systems of models, raising the practice of agent-based modeling to a new level. The inquiry process of constructing an agent-based model (ABM) of a complex system often gives rise to questions about the nature of the relations between that system and other systems to which it is connected. LevelSpace, the key innovation of this project, will introduce a conceptual as well as a technical infrastructure for this facet of modeling, enabling a core advance in our capacity to imagine and reason about complex systems and the interactive relations among them. Moreover, the addition of LevelSpace will enable the integration of two other crucial expansions to agent-based modeling that have been explored in recent years—participatory simulations (PartSims) and bifocal models—under the unifying framework of linked models. These approaches to systems modeling have been recognized as powerful settings for developing intuitions and revealing essential features of complex systems, but they have hitherto been pursued by largely distinct subgroups of the NetLogo community, using separate technological infrastructures. A unified multi-level linked-modeling (MLL) framework that also embraces bifocal models and PartSims will enable transitions and connections among these modalities. Finally, because it will be released as an extension to NetLogo, LevelSpace will be able to enhance the modeling practice of the large and growing community of learners and researchers who make regular use of NetLogo. Our project will develop, trial, evaluate, and refine LevelSpace through implementations focused on the domains of population biology and social policy. For population biology, we will engage with classrooms in partner high schools, and our work will build on our NSF-funded curricular units that have been used effectively and at scale in prior research. For social policy, we will deploy LevelSpace in the context of undergraduate classrooms studying the impact of policy decisions. These two areas are suitable testing grounds for exemplifying a variety of kinds of model linkage (including nested and peer structures), as well as for making connections to PartSims and to external sources of data. Moreover, population biology provides an environment in which the innovation of LevelSpace can be rigorously tested to reveal what it adds over the state of the art of ABM-based curricula. In contrast, social policy provides us the opportunity to work in a discipline where many professional practitioners frequently use agent-based modeling, but where most educational practice has not yet adopted a modeling approach. By helping to illuminate the multi-level systems of social organizations, this project’s proposed enhancement to NetLogo promises to make it even more relevant to learning in t
|Effective start/end date||9/1/14 → 8/31/19|
- National Science Foundation (IIS-1441552)
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.