Integrated Simulation and Modeling Environment

Using Participatory Simulations to Investigate the Complementarity of Agent-Based and Aggregate Reasoning for Making Sense of Complexity

Main Page, Papers & Presentations, Tools & Technology, Classroom Resources

Jump to Overview, Rationale, Sites, Team & Contact


The importance of students understanding complexity has been recognized in various national standards documents, such as the Benchmarks for Science Literacy and AAAS. The ISME project is conducting an in-depth study of the learning of secondary students engaged in making sense of complex systems through participating in classroom simulations.

To those ends, we design, develop, and implement Participatory Simulations that use the HubNet technology. The HubNet technology constitutes an extension of the NetLogo environment, enabling multiple participants to interact simultaneously with NetLogo models. In the Participatory Simulations environment, all the students in a classroom are networked into a single hands-on simulated experiment that all can plan, collaboratively run, view, and discuss. Thus, the project seeks to leverage the complexity and dynamicism of the classroom as a cognitive tool for catalyzing and extending student insight related to the structure of complex systems. We study, in particular, the complementarity of students' agent-based and aggregate forms of reasoning as a resource for making sense of complexity.

ISME's intended contributions are both scholarly and design-based: We aim to contribute both to research on learning (in the domain of complexity and technology-assisted education) and by creating the technological infrastructure, tools, and model-based learning environments for learners to engage in communal inquiry into the domain of complex systems.

ISME is funded by an NSF ROLE Grant no. REC-0126227.


The rationale for ISME research rests on three pillars:

Complex systems: the importance of understanding systems and complexity;

Understanding complex systems: the importance for researchers of learning to understand how individual learners? agent-based and aggregate forms of reasoning interact in supporting their understanding of complex systems;

Learning about complex systems: the importance of real-time first-person participation in simulations for scaffolding students? formal understanding of complexity.

Complex Systems

While the study of complexity may at first seem distant from the core of the curriculum, systems ideas are now finding their way into important standards documents. For example, many states have organized their science standards using the “Common Themes” from the Benchmarks document. The first of the Common Themes discussed in the Benchmarks is Systems (Chapter 11, 1993). Fundamentally, the worlds of natural and social phenomena are complex. All natural systems, from a systems point of view, can be seen as composed of myriad components; whether they are individual animals or plants in an ecosystem, molecules in a chemical reaction, or buyers and sellers in a trade economy. Yet, understanding the ideas of systems and emergence are difficult. As AAAS Benchmarks note, children tend to think of the properties of a system as “belonging to individual parts of it rather than as arising from the interaction of the parts.” Consequently, “a system property that arises from interaction of parts is … a difficult idea.”

Consistent with Benchmarks, we argue that not only is such study crucial to the progress of science, but it is vital to successful and meaningful civic participation in a democracy. We can no longer afford to engage the rich complexity and dynamism of our world with over-simplified, relatively static constructs. The absence of these ideas and tools in school-based learning frames and compels an agenda for curricular design, technological innovation, and research.

We seek to address this fundamental difficulty of systems learning while simultaneously highlighting the ways in which assuming a “systems perspective” can be helpful in improving student understanding of difficult ideas in the current curricula. The claim is that many areas in the traditional curriculum – e.g. the spread of disease, the distribution of speeds of molecules in a gas, and even the idea of function in algebra – can be better understood by students, if these topics are approached from a systems perspective. Thus we argue that at both a fundamental conceptual level and at a pragmatic level, the understanding of systems ideas and complexity is a core literacy needed for all students both to understand science and mathematics and to act as informed citizens.

Understanding Complex Systems

Our previous work has shown that when students engage in participatory simulations they tend to use aspects both of top-level, macro-, or aggregate reasoning and bottom-up, micro-, or agent-based reasoning in making sense of complex phenomena like the spread of a disease through a population. In contrast to the claims of some researchers that these forms of reasoning are incommensurate (e.g., Chi, 2001), our research suggests that students are able to shift easily between these two forms of reasoning and to combine them as reciprocal cognitive resources supporting problem solving.

What we mean by aggregate reasoning is reasoning about properties of and changes in populations. Typically, such reasoning employs constructs such as ‘amount,’ ‘average,’ and ‘rate of change’ to describe the state of the system and its time evolution. This form of reasoning has strong antecedents in the mathematical and scientific literature and in formal curriculum (e.g., in mathematical modeling with differential equations).

In contrast, agent-based reasoning is reasoning about the behavior of individual elements of a system. Typically, such reasoning involves exploring rules of behavior for individual agents and their interactions and reasoning about the emergent results of alternative rule sets. Traditionally, this form of reasoning has been less supported by formal curricula, though the advent of powerful computation has resulted in tools that enable such support. We have come to believe that these two forms of reasoning are fundamentally complementary and that through their separate and coordinated development students can make great progress in understanding the behavior of systems changing over time.

These two forms of reasoning are very powerful ways of making sense of complexity in the world – yet, the communities who practice them and the literature describing them are largely separate and distinct. The aggregate and agent-based modeling tools themselves are deployed by different communities -- each community focused on its tool and attendant form of reasoning. We believe that at both the cognitive level and the tool level the time has come for a synthesis of these two approaches. Accordingly, we explore how the two forms of reasoning complement each other in making sense of complexity and change. We expect the ISME project to be a benchmark platform for fostering such conversation amongst learners and consequently a benchmark research tool for investigating the complementarity of the two styles of modeling and forms of reasoning. We also hope/expect that the field of educational research will be enriched by the conversation between the two approaches, moving beyond claims about the superiority of one approach or another to seeing the significant leverage that can be achieved through recognizing the complementarity of the two.

Learning Complex Systems

The third pillar builds on a significant body of work demonstrating the importance of real-time first-person experience to building an understanding of challenging content. The proposed project takes the next step in a long line of work by moving the unit of analysis up from an individual to the whole class. We ask: what is the cognitive and motivational significance of supporting real-time first-person-in-relation-to-an-evolving-whole experience for making sense of complex dynamic systems? What are the roles of collaborative sense making and strategizing in enhancing both individual and class learning?

To support the investigation of this question we employ the functionality of our HubNet architecture to create what we are calling an Integrated (Participatory) Simulation and Modeling Environment (ISME). We have designed and built on this environment to leverage the complexity and dynamism of the classroom as cognitive tools for catalyzing and extending student insight related to the structure of complex systems. Our design amplifies the learning potential inherent in the classroom social space by having students engage in participatory simulations activities where students assume roles and, through interacting with each other in the networked environment, create an emergent pattern of the system. Whether it is the emergence of a traffic jam in a simulated city or the spread of disease in a simulated population, we have found that by having students themselves interact in real-time to create the emergent behavior of a dynamic system, participatory simulations provide a natural connection for learners to the study of complexity in the world around them.

In our research, we study the cognitive affordances of using this integrated simulation-and-modeling environment for advancing students’ understandings of systems ideas like emergence, feedback, and non-linearity. By working closely with our commercial partner, we intend to put the ISME system and curricular materials on a trajectory to become the next generation of near ubiquitous classroom technologies and thereby support, in very practical ways, the possibility of systems learning becoming part of the core of all students’ school-based learning.

Sites and Collaborators

Stone Scholastic Academy that is in Chicago Public Schools
Nichols Middle School that is in District 65
Brown Elementary, Campbell Elementary, and Zavala Elementary, that are in Austin Independent School District
Texas School for the Deaf
Nancy Ares at the University of Rochester
Corey Brady at Texas Instruments

Team and Contact

Principal Investigators:
Uri Wilensky -
Northwestern University

Walter Stroup -
The University of Texas at Austin

Design and Research Teams
Uri Wilensky
Dor Abrahamson
Sharona T. Levy
Josh Unterman
Matthew Berland
Reuven Lerner
Paulo Blikstein
UT Austin:
Walter Stroup
Thomas Hills
Andy Hurford
Sarah Davis
Carolyn Remmler

Development Team (Northwestern)
Uri Wilensky
Seth Tisue
Andrei Scheinkman
Esther Verreau
Craig Brozefsky
Matthew Berland
Josh Unterman