Linking Complex Systems
Linking Complex Systems will create a new technological genre to address multiple, interacting complex systems. By merging agent-based modeling and system diagramming, this new linked hybrid modeling permits learners to move between detailed individual models and global views of stocks and flows for the first time, enabling new modes of experimentation and fostering levels of learner reasoning about complex systems and systems dynamics that are not currently possible.
The Linking Complex Systems project has a single overarching goal: to understand and facilitate learners’ reasoning among and within multiple, interconnected complex systems from multiple perspectives. We will achieve this goal via specific, targeted objectives:
- Unite two complementary technologies into a new technological genre supporting complex system reasoning, marrying the strengths of both to form a new linked-hybrid modeling environment, aimed at supporting learning and reasoning in interconnected complex systems.
- Develop a set of core learning exemplars for high school students involving interlinked complex systems and making extensive use of the project’s new technology genre.
- Investigate student reasoning and technology interactions as students work with the learning exemplars.
- Analyze, present, and publish findings at major research conferences and in peer-reviewed education journals.
- Generalize and disseminate the technology genre and related tools to the education technology and research communities.
Figure 1. Schematic overview of linked-hierarchical modeling showing linked micro and macro views.
MIT's StarLogo and the Concord Consortium's SageModeler will be integrated to form the basis for a new linked-hybrid modeling (LHM) environment. The resulting genre will provide three main functions to aid learners.
First, the new environment will make use of SageModeler's macro-level systems diagram view to aid learners in easily understanding the entities and relationships in a system. By dynamically generating a macro-level SageModeler diagram for any existing or new StarLogo system and auto-updating it as the StarLogo model is modified, the LHM environment will permit learners to easily keep track of a system’s entities and connections as it grows.
Second, this new genre will provide learners a much closer link between the macro and micro levels of the system (see Figure 1). The nodes in macro-level LHM diagrams will typically represent classes of agents within the system, and connections between agents will represent interactions or influences between these agents. When viewing a macro-level LHM diagram, a learner will be able to click on any individual node or connection to bring up a pop-up view of the relevant programming block structure underlying that corresponding agent or interaction. This hybrid view will make the connections between macro-level relationships and micro-level rules and interactions explicit for the first time, fostering connections that are essential for understanding all systems, especially interlinked complex systems.
Finally, the new genre will introduce a new type of quasi-programming interface for these models, accessed through the macro-level model. For a certain subset of interactions, learners will be able to create or remove interactions among the system’s agents by drawing or deleting connections between nodes in the macro-level system diagram.
With these capabilities, the new genre will permit learners to easily view dynamic, high-level representations of systems in a way that is currently impossible.