Principal Investigators: Robert Goldstone & Linda Smith
Title: Grounded and Transferable Knowledge of Complex Systems Using Computer Simulations
Purpose: One of the most important questions confronting education involves when and how students transfer what they have learned to new situations. Considerable research suggests that students often do not spontaneously transfer what they have learned when there are even superficial differences in learning situations. This is disturbing because teachers often give students content not only for its own sake, but so that students will apply what they have learned to relevant new situations. This research team believes that students can be taught in such a way as to allow them to transfer scientific principles across different areas. In determining how best to support this transfer of knowledge, this research project examines the relation between the specific fact and details through which a phenomenon is presented, and the abstraction of deeper scientific principles underlying the phenomenon. These researchers are developing perceptually based and interactive computer simulations that provide perceptual support to students as they master abstract scientific principles. At the conclusion of this research project, the team will have developed new curricula and supporting materials for teaching complex systems at the college and K-12 levels.
Setting: Both the undergraduates and the middle and high school students are located in a moderately sized Midwestern community.
Population: Participants in the laboratory studies are college undergraduates. The middle and high school participants are students in 7th-10th grade science classes. Both the undergraduates and middle and high school students are diverse in terms of race and socioeconomic status.
Intervention: Fifteen computer simulations teaching abstract concepts have already been developed, and 10 more are being developed. These simulations include supporting laboratory materials, such as user manuals and laboratory assignments.
Research Design and Methods: Research methods include both controlled laboratory experiments and classroom-based studies, both investigating the role of perceptually based simulations in fostering students’ scientific understanding. By observing how active exploration of one simulation benefits understanding of a subsequently presented simulation in a different content area, but based on the same principle, the researchers can assess whether the scientific principle has been successfully abstracted. Experiments explore the roles of graphical concreteness, narrative contextualization, language specificity, and diagrams on students’ implicit and explicit knowledge of scientific principles. Outcomes of this research include prescriptions for how and when concrete and highly contextualized materials should be used, compared to idealized and decontextualized materials. The researchers are also continuing to examine the process of concreteness fading, in which concrete graphical elements within computer simulations are gradually replaced with progressively idealized elements. The classroom-based research occurs both in a freshman seminar on Complex Adaptive Systems and in an 8th grade science class. Student use of simulations is being examined using controlled and counterbalanced studies.
Control Condition: Control conditions vary as a function of the purpose of the experiment, but students are randomly assigned to conditions that allow answering theoretical and practical questions of interest across the nine experiments.
Key Measures: Student learning is measured with quizzes, problem-solving tasks, and homework assignments. In addition, protocol data is gathered during student work within the simulations.
Data Analytic Strategy: Analysis of variance and other appropriate statistical techniques are used to evaluate student outcomes of participating in the different manipulations. Protocol analysis techniques are used on the protocols collected during student problem-solving.