Andrew T. Stull - US grants

Many strong claims have been made about the benefit of learning with physical and virtual (computer-based) models, especially in science, technology, engineering, and mathematics (STEM) disciplines. Chemistry is an ideal domain to investigate these issues because it is a spatially-rich domain that relies heavily on a variety of 2D (diagrams) and 3D (models) representations, and mastering these representations (i.e., developing representational competence) is challenging for students, but essential for their growth as chemists or scientists in other domains. A team of investigators from the University of California, Santa Barbara, will investigate (1) whether models might serve as a crutch or a scaffold for learning, (2) the extent to which interacting with models affects representational competence, (3) whether the ?medium? of the model (physical or virtual) is important in model-based learning, (4) which perceptual (i.e., visual and haptic) cues and cognitive factors contribute to meaningful learning, and (5) how gender and differences in spatial ability affect model-based learning. This project will yield research-based (a) practices for the development of curricula that integrate concrete and virtual models as effective learning aids and (b) principles for the design of meaningful model-based instruction.

The proposed project has the potential to inform teachers and curriculum designers about how better to use models and other representations in chemistry and other highly spatial STEM domains (e.g., geology, astronomy, anatomy, and mechanical engineering), by contributing to our basic understanding of what cognitive and perceptual factors affect learning. The knowledge gained from this project will inform the effective design and delivery of new media made possible by the explosive growth in availability of smart-phones and tablets in the classroom. Finally, results of this project will inform our understanding of how gender and differences in spatial ability contributes to learning with models in STEM disciplines.

A team of investigators from the University of California - Santa Barbara, together with researchers from the University of Georgia, will conduct research on how to improve the design of online undergraduate instruction in biology and chemistry that involves video. The researchers will test the hypothesis based on a leading theory of multimedia learning that when instructors face students throughout the lesson (by using a transparent whiteboard), students should learn more than when instructors turn their backs on the camera and face the board while drawing, as is the case with conventional whiteboards. They should be better able to integrate what the instructor is saying with visualizations such as diagrams and graphs in real time. They should also be able to take advantage of social cues and feel more of a connection with the instructor, increasing motivation. Transparent whiteboards may be particularly beneficial for learning in STEM disciplines where there is a heavy demand on connecting verbal explanations with complex spatial and analytical representations. This project will extend existing theories of how students learn from social cues incorporated within multimedia lessons, and provide research-based principles for improving the design of video-based instruction in STEM fields. The project will also support best teaching practices for promoting STEM learning in the burgeoning arena of online classrooms such as through MOOCs (massively open online courses) for students learning outside of traditional courses and flipped classrooms (presenting lectures via online lectures and using class time for learning activities and discussion) for traditional courses. The project is funded by the EHR Core Research program, which supports fundamental research that advances the research literature on STEM learning, and has implications for education in both formal and informal settings.

Four studies will examine how two different whiteboard technologies affect students learning in STEM domains. Students will learn about concepts in organic chemistry or genetics with the aid of either transparent whiteboards (in which the instructor faces the students while drawing and explaining) or conventional whiteboards (in which the instructor does not). Dependent measures include student accuracy of interpreting representations, solving spatial problems, and learning transfer at immediate and delayed testing. In addition, eye-tracking analytics (fixations and saccades), physiological measures (electrodermal activity and heart rate), and individual difference measures (spatial and reasoning ability) will be collected. This project will yield research-based (a) practices for the development and delivery of video-based lessons that are grounded in empirical research and that address limitations of conventional practice, (b) guidelines for the design of instruction in STEM domains that promotes learning transfer that are based on educational theory, and (c) principles for the elimination of unnecessary barriers to learning and the mitigation of difficulties in spatial learning and reasoning required for a productive STEM workforce. The project expands research and theory related to principles of multimedia learning and learning from social cues in multimedia learning environments. Specifically, the efficacy of a potentially useful system for delivering online lectures will be systematically evaluated.