Organic chemists rely heavily on physical models of molecules, so it not surprising that models are also commonly employed to help teach organic chemistry. It has been assumed that when undergraduate students work with these models, ability to construct an internal representation of the molecular structures is enhanced. But few studies have examined this directly and empirically. In addition, virtual models (3-D computer visualizations) have replaced concrete models as powerful computers become available, inviting comparison between learning with real and virtual models.
Researchers at the University of California Santa Barbara and the University of Maryland will conduct a series studies to examine how undergraduate organic chemistry students use models to advance their understanding of molecular structures that require visualizations in three dimensions. This study investigates the benefits of students' use of concrete or virtual models, asking: - Does model use improve students? ability to translate between different representations of molecular representations? - What aspects of models are most effective for representation translation? - Does instruction in model use alter the frequency and quality of model use? - Do models support students of low spatial ability more or less effectively than students of high spatial ability?
This study uses experimental methods to systematically explore the uses of both concrete and virtual models in promoting meaningful learning in organic chemistry in a series of studies set in the controlled environment of the psychology lab. An important dependent variable for this study measures students' ability to translate between alternative diagrammatic representations of molecules. This is an essential skill that all students of organic chemistry must master. Measures include accuracy of the representations that students produce in representational translation and students' interactions (including gestures) with the concrete and virtual models during task performance. The study also tests its emergent theory of model use in university classroom settings by comparing the performance of undergraduates who are trained to use models with those who have had no such training.
This study is important because it systematically studies model use in organic chemistry. If students learn to use models more effectively, then they may find more success in undergraduate organic chemistry, a gatekeeper course for advanced STEM professions. The ability to visualize molecules may be especially challenging for students with low spatial abilities, and the deliberate training in model use may allow more students to not only pass organic chemistry, but stimulate their ability to use models in eventual STEM professions. This study should result in a better understanding of the effectiveness and appropriateness of concrete and virtual models for enhancing student learning in chemistry and new instructional activities for use in chemistry classrooms.
Providing students with concrete models of molecules is a common but not ubiquitous practice in chemistry courses. The assumption is that a concrete molecular model will enhance the ability of the learner to construct an internal representation of the molecular structure and that this, in turn, will help students discover the molecule’s functional characteristics, which can be used to predict its reactive nature. More recently, virtual models (3-d computer visualizations) have replaced concrete models as powerful computers become available. Few empirical studies have investigated the benefits of either concrete or virtual models, and the studies that have been conducted often study the use of models when they are integrated with other instructional practices, so that it is difficult to isolate the effects of models independently of other factors. This project employed experimental methods to systematically explore the uses of both concrete and virtual models in promoting meaningful learning in organic chemistry. The main task in our experiments included translating between alternative diagrammatic representations, an essential skill that all students of organic chemistry must master. Our measures included accuracy of the representations that students produce in representational translation tasks, and students’ interactions (including gestures) with the concrete and virtual models, during task performance. We addressed the following research questions: (1) Does model use improve students’ ability to translate between molecular representations? (2) What aspects of models are most effective for representation translation? (3) Does instruction in model use alter the frequency and quality of model use? (4) Do models support students of low spatial ability more effectively than students of high spatial ability? Broadly, the products of this project included a better understanding of the effectiveness and appropriateness of concrete and virtual models for enhancing student learning in chemistry and new instructional activities for use in chemistry classrooms. With regard to each research question we found that (1) model use improved students’ ability to translate between representations by approximately 2sd, but this effect was only observed in cases where students (2) physically manipulated a model. Gestures were also found to produce this beneficial effect. We also found that students require (3) direct and explicit instruction in constructing and spatially transforming models before any increase in frequency or quality of model use is obtained. Importantly, this last finding revealed that simply providing models to students yields no benefit absent direct instruction in how to use the model. Finally, we observed that (4) model use explains more variance in student achievement than spatial ability. These findings have been reported in five research journal publications and sixteen conference presentations. Intellectual Merit: The project will contribute to theories of diagrammatic and model— based reasoning, individual differences in cognition, and their implications for instruction in science. It will provide basic research on meta-representational competence, which has been defined as ability "to select, produce and productively use representations" and "to critique and modify representations" (diSessa & Sherin, 2000, p. 386). Our studies will elucidate the relationship between metarepresentational competence and cognitive abilities, and examine similarities and differences in the cognitive processes involved in using concrete and virtual models. Finally they will provide new information about the extent to which representational competence can be enhanced by direct instruction in an authentic instructional setting, and how this impacts student achievement. Broader Impacts: Besides chemistry, our research is relevant to instruction in all scientific domains (e.g., geology, physics, biology) in which conventional diagrammatic formats must be mastered by students. It will provide new information regarding the extent to which virtual models can replace physical models in science. The study also addresses questions about who can succeed in science. Specifically, gender-specific differences in visuo-spatial ability have postulated a deficit model for women, so that lower spatial ability is indicated as a causal mechanism for the shortage of women in science. The proposed work offers methods of alleviating difficulties that low-spatial individuals face in science learning and leads to the design, implementation and evaluation of new pedagogies to help students master critical representational skills.
