Computer Simulation Modeling Using STELLA
to Enhance Investigative Learning in a Biology Curriculum

Steven K. Rice, Grant E. Brown and R. Paul Willing
Department of Biological Sciences
Union College
Schenectady, NY 12308

HOME	PROJECT SUMMARY	INTEGRATING MODELS AND EXPERIMENTS	RESEARCH AND OUTREACH
WHAT IS STELLA?	EXAMPLES FROM UNION	WHAT WE RECOMMEND	CONTACT US

INTEGRATING MODELS AND EXPERIMENTS

Undergraduate science educators have praised the merits of laboratory curricula where students initiate and perform experiments of their own design (hereafter referred to as "experimental" laboratories). These exercises aim to foster process skills (e.g., hypothesis formation, and experimental design, analysis and interpretation), and encourage creative and critical thinking. In a typical experimental laboratory, students are provided with or develop foundational content, practice experimental procedures, and are then presented with a biological problem to solve that requires the experimental techniques. In their best expression, these experiences culminate when students communicate their results orally or in writing. As a first approximation, this process adequately simulates real experimentation and can be an effective way for students to learn investigative skills.

Unfortunately, these experiences simulate the experimental, but not the investigative process. Although experimentation is arguably what distinguishes scientific pursuits from non-scientific ones, the process of experimentation is cradled within an investigative framework. This context is essential for the formation of hypotheses and for the application of experimental results. In experimental laboratory exercises, the lack of context can limit the benefit of student experiments. For example, students often have difficulty forming their own hypotheses that properly relate to biological content and rely on material in the lab text, ideas that are simple and reflect a lack of understanding of the system of study, or from notions applied from distant fields of biology that have little relevance to the system of study. In general, experimental labs result in experiences that weakly connect experiments and their design with biological content. Consequently students in experimental laboratories develop procedural skills, but often do not build strong links between their experiments and foundational knowledge.

In practice, biologists often employ models to link biological content with experiments. Modeling exercises can be used in a similar fashion to enhance investigative experiences in teaching labs. In the teaching laboratory, modeling efforts prior to experimental design can connect student designed experiments to specific biological content, help students design more relevant and sophisticated experiments, and generate quantitative predictions that logically follow from hypotheses. Following experiments, summary models can be used as experimental concept maps that force students to clarify outcomes and apply their results to new situations.

We are developing and implementing modeling exercises to bridge the rift between biological content and student experiments. Many of the exercises will adapt a computer simulation software package that lets students construct dynamic simulation models for their particular experiments. The STELLA software allows students to develop and parameterize pool and flux models to explore model dynamics, and to make quantitative predictions of experimental results. The modeling exercises will not only help students create more specific hypotheses, but they will also provide a context to evaluate their experimental results.