INTEGRATION OF AN AUGMENTED REALITY SANDBOX INTO A LARGE ENROLLMENT INTRODUCTORY GEOSCIENCE COURSE FOR NON-MAJORS

The augmented reality (AR) sandbox, conceived by Oliver Kreylos at the UC-Davis Institute for Data Analysis and Visualization, attempts to bridge the gap between 2D and 3D visualization by projecting a digital topographic map directly onto a landscape created in a sandbox. As that sandbox landscape is altered, the topographic map dynamically adjusts in real time to match the landscape, giving students the opportunity to discover how to read topographic maps on their own. We seek to test the hypothesis that the AR sandbox is a more effective tool for teaching topographic maps than the traditional, paper-based approach alone. We developed a topographic map-reading assessment which includes questions on identifying landforms, topographic profiles, and campfire exercises. We also collect additional descriptive data, including gender and 3D visualization ability. To generate a baseline for comparison, a large group of students (N = 189) enrolled in an introductory geology class for non-majors took the assessment without exposure to the AR sandbox in spring 2015. Results from the pre-test group show previously identified trends: male students tend to score better on the topographic maps assessment than female students, as do students with higher 3D visualization abilities. In spring of 2016 students enrolled in the same class were randomly assigned to one of two groups. The control group (N=55) completed the traditional, paper-based topographic maps lab. The experimental group (N=66) completed the same lab, but also used the AR sandbox. Students working in groups of 4-6 spent 15-20 minutes working on an instructor guided exercise with the AR sandbox to explore concepts such as contours, landforms, profiles, and gradient. Subsequently all students took the topographic maps assessment. Unfortunately, we identified no statistically significant gains in the experimental group over the control group, even when the results are subdivided by gender and/or spatial visualization ability. These results suggest either (1) larger experiment is necessary to see statistically significant gains or (2) the instructor-guided approach is not the most effective way to engage students with the augmented reality sandbox.