Building theories of scientific phenomena: Comparing and Integrating aggregate pattern-based and agent-based computational approaches

Overview: This is a proposal for a STEM + Computing K12 Education (STEM+C) project. The project investigates how 8th grade students can engage in scientific theory building and how theory building can be integrated into their science curriculum. Theory building is a fundamental practice of science, and there are many ways scientists approach building theory. In some cases, they begin with data and abstract a relationship in order to more precisely describe it in terms of a mathematical model (aggregate pattern approach). In other cases, they posit the mechanism for a phenomenon, beginning with assumptions about a system (e.g., its elements and their interactions) and build a computational model that, when run, produces outcomes that can be explored (individual mechanism approach). Despite the centrality of theory building, it is rarely systematically taught in the science classroom. Furthermore, while there have been some studies that investigate student theory building in the context of the aggregate pattern approach, much fewer have done so with the individual mechanism approach. Importantly, no studies have yet sought to characterize the commonalities and differences of the two approaches. This study seeks to address this gap. We propose to study the affordances of each approach to theory building and to use these results to create curricular units that integrate the two approaches. For each approach, the research aims to understand the character of 1) the theories the students produce 2) the processes by which they produce their theories, and 3) the characteristics of student learning that result. These questions are addressed in the first two years through lab-based studies that introduce students to phenomena by guiding them to either 1) articulate the aggregate-level pattern seen in data or 2) develop agent-based computational models that represent the mechanism behind the pattern. This work will be carried out in the context of two scientific phenomena, 1) particle diffusion and 2) predator-prey dynamics. Findings will inform the design of curricular units that will be tested through design-based research in the third year. These findings will illuminate how the two approaches can be integrated to help students learn about both scientific phenomena and theory-building processes. Intellectual Merit: Science education research and standards documents emphasize the need for engaging students in the "practices of scientists." But we still know very little about how to engage students in one of the most central practices of science, theory building. This study begins to address this gap, while comparing two approaches to theory building. We are particularly interested in whether and how theory building can be made more accessible through the support of computational modeling tools. Findings will be germane to literature on engaging students in theoretical practices of science, and science practices more broadly. Findings will also contribute to conceptual change literature, by providing a fine-grained analysis of students’ shifts in thinking as a result of engagement in theory building. Finally, findings will extend Restructuration theory by providing insight into how the production of different representations affords different learning pathways. Broader Impacts: Curriculum that engages students in theory-building is essential to their development of deep understanding of scientific content, skills for engaging in science practices, and more nuanced views of the scientific enterprise. Products of the study will inform the