The ModelSim project approaches its four core curriculum areas (Evolution, Population Biology, Electricity, and the Particulate Nature of Matter) with a distinctive and unified approach: in each of these subject areas, we encourage students to conceive of fundamental real-world phenomena in new ways. And in each case, we provide learning environments for exercising this new way of thinking, which build on students’ intuitions to develop powerful and grounded understandings.
The full lesson plans for those units can be found under the teacher resources and student resources submenu items of the units menu on this page. At this time, only the population biology and evolution units are available on this website. For access to the PNOM (Particular states of matter) unit, please email email@example.com . For access to the Electricity units, please email: firstname.lastname@example.org .
This fundamentally new way of thinking is called “agent-based modeling.” Within this perspective, complex phenomena are viewed as emerging from interactions between systems of simple elements, or “agents.” By focusing on the behavior of these simple agents, and by using computational modeling to simulate interactions between many of these agents, scientists in a wide range of disciplines are introducing powerful new research methods.
We believe that this cutting-edge approach is also powerful for learning science in a wide range of disciplines. This is in part because the agent-based approach taps into student intuitions in subject areas which are traditionally experienced as non-intuitive. Indeed, research has shown that many incorrect ideas that learners have about aggregate phenomena (such as the flow of electric current in a wire, the inheritance of biological traits, or the temperature of a gas) can arise from a “slippage between levels” – that is, attempts to identify agent-level attributes to the system as a whole, or vice versa. Thus, students might conjecture that individual particles of a gas or liquid are “hot” or “wet;” they might talk about current being “used up” as it travels through a circuit; or they might conceive an individual organism’s capacity for speed as resource built up in its lifetime and then “passed on” to its offspring.
These incorrect but vivid ideas are actually powerful conceptual resources. They can be brought to bear in learning environments that show how the aggregate, complex phenomena emerge out of interactions between simple agents in the system. Once learners can identify and computationally interact with entities at this agent level, they can use their intuitions to guide reasoning about the emergent properties of the larger system. For instance, once a student begins to see a gas as a numerous collection of particles that bounce off of each other and the walls of their container, it is possible to envision the effects of changing the speed of these particles. When this is connected with the idea of temperature, the student has the means to reason about not only the meaning of high and low temperature, but also about the relationships between changes in temperature on the one hand, and changes in volume or pressure on the other hand.
To provide students with the means of getting to know systems in biology, physics, and chemistry from agent-based perspectives, we have developed activities based on the award-winning NetLogo modeling environment. NetLogo allows students to model complex systems computationally, first by specifying the behaviors of agents and then by simulating interactions among these agents and visualizing emergent properties at the aggregate level of the system. In addition to on-screen computational simulations, the ModelSim project extends virtual agent-based modeling in two key ways: bifocal modeling and participatory simulations.
In bifocal models, students connect computational behavior in virtual simulations with phenomena detected by physical sensors or produced in the physical world by motors or other output devices. The similarities and contrasts between virtual and physical systems stimulate conceptual reflection and “debugging” processes through which student adjust their physical and virtual models simultaneously. The real-time link enriches both physical and virtual models: data generated in these environments are directly wired into computational representations, and virtual models are grounded in direct and continuous comparison with data. Bifocal modeling activities thus provide a critical “missing link” between laboratory experiences and the construction of explanatory models and theories. Bifocal modeling activities in the ModelSim project are enabled by connectivity between the NetLogo environment and the GoGo and Arduino platforms.
Another means of helping students to make connections with agent-based models is through embodied, role-play activities in which learners take on the perspectives of agents in the system. In these participatory simulations, each student individually controls the behavior of an agent in the simulation. The class as a whole embodies the system and experiences the simulation as a group. Participatory Simulations thus give students the opportunity to gain experience both agent and aggregate perspectives, from vantage points both inside and outside of the systems being studied. Although we all participate as elements in human complex systems, we rarely get to experience these systems at both micro- and macro- levels, as both actors and observers. Through NetLogo’s HubNet architecture, we are able to provide this experience to classroom groups as they reflect on systems in Science. Working in a local network of computers, students use the HubNet client to control the behavior of individual objects or agents and view the aggregated results in the host NetLogo environment, which runs on the teacher’s computer.