Enabling Modeling and Simulation-Based Science in the Classroom:
Integrating agent-based models, real world sensing and collaborative networks

Project Goals

ModelSim is a 4-year NSF funded project focusing on the design and scaled implementation of technology-rich agent-based modeling (ABM) science curricula to address challenging topics in the physical and biological sciences. More information about the curricular units can be found on the ModelSim Curricula page.

Our extensive experience engaging children in the creation and exploration of ABMs in domains such as biology, chemistry, electricity, and probability have shown agent-based modeling to be a powerful way for a wide range of students to engage with and make sense of complex phenomena. The ModelSim project builds on this rich research tradition and extends it in two distinct ways:

  1. It seeks to distinguish different modalities of ABM inquiry in terms of their affordances for both enhancing individual thinking and engaging a broad range of students.
  2. It investigates the many different factors that facilitate or impede unsupervised deployment of ABM curricula at scale.

Modeling Modalities

We have identified three core ABM modalities:

  1. Agent-based Inquiry -- creating and exploring virtual models of scientific phenomena
  2. Bifocal Modeling -- pairing the design of a virtual computational model with the construction of a physical entity
  3. Participatory Simulations -- networked simulations where all students act as individual agents in the same model

Curricular Units

ModelSim includes four science units: Evolution, Population Biology, the Particulate Nature of Matter, and Electricity. Modeling activities utilize the NetLogo agent-based modeling environment and the curriculum was delivered using the WISE inquiry environment. More information about the individual units and frequently asked questions about implementing the curriculum can be found at the ModelSim Project page.

Project Plan

In the first three years of the project we engaged in extensive Design-based Implementation Research. This work involved the iterative development of powerful modeling tools to support student inquiry, technologies for simplifying the delivery of technology-rich materials in schools with varying IT infrastructures, and techniques for training teachers for implementation with minimal researcher support.

In the final year of the project our goal has been to investigated the many different factors that facilitate or impede unsupervised deployment of ABM curricula at scale. The most extensive ABM project conducted to date, we have deployed the ModelSim units in over 100 classrooms across three states and have collected a massive corpus of data that includes pre- and post-exams, video recordings of implementing classrooms, pre- and post-teacher interviews, logging data of model use, student-created models and artifacts from all phases of the curriculum, teacher quality measures, as well as student, teacher and school level demographic data.