Bifocal Modeling is a new framework to inquiry-driven science learning. It challenges students to experience a scientific phenomenon through two lenses in parallel – a real experiment and a model. In these activities, students explore a scientific phenomenon such as heat diffusion, properties of gases, or wave propagation by designing an experiment and a model and connecting them in real time. Often the students design and develop their own experiment and collect data using embedded sensors. In parallel, they design and develop a model of the same phenomenon; their goal is to build a model which behavior matches the data they collected. Bifocal Modeling activities can serve as an approach to teach scientific content, modeling skills, use of instrumentation, computational thinking, and refine students’ scientific epistemology.
In recent years virtual experimentation has received increasing emphasis as an alternative to conducting experiments in a physical environment. Much of the research in this area has focused on the question of the advantages of virtual experiments over physical experimentation (Jaakkola & Nurmi, 2008; Klahr et al., 2007; Triona et al., 2003), but more recently researchers have started to examine the effects of combining the virtual and physical modalities (Zacharia et al., 2003; Zacharia et al., 2008; Jaakkola et al., 2011; Liu, 2006).
Additionally, studies report that alternation between these distinct experimental modalities in the course of individual experiments can affect and often improve learning outcomes (Zacharia, 2012; Smith, 2010; Gire et al., 2010). Zacharia and colleagues (2008, 2012) suggested that the best way to develop a framework portraying the optimal combinations of physical and virtual manipulation is to employ the learning objectives of each experiment as the criteria for blending them. Nevertheless, there are two under-researched areas in this literature: The first is that most of the virtual experiments were simulated versions of a real experiment, often mimicking the appearance of the lab equipment, with the goal of trying to make the real and virtual experiments as similar as possible (Blikstein, 2014). The second is that the research has mostly focused on predesigned physical and virtual experimentation. Simulation tools have been among the preferred means for providing environments for virtual experiments, but the rules and models behind these simulations often remains hidden from the students.
Recent advances in inquiry learning research have sought to implement activities in which virtual experimentation is supplemented by opportunities to design computer models (Mulder et al., 2011), and the aim of our present work is to examine the learning outcomes of designing these computer models that are explicitly meant to be different that the real ones, in order to promote students’ critical stance towards their own models and hypothesis.
The creation and critical evaluation of models are important components of scientific practice which have been increasingly recognized as a valued educational goal (Levy & Wilensky, 2008; Blikstein, 2010).
Our goal is to present a new way of combining physical and virtual experimentation.
Bifocal Modeling (Blikstein & Wilensky, 2007; Blikstein et. al., 2012; Blikstein, 2010, 2014) is an approach to inquiry-driven science learning that challenges students to design, and compare in real time physical and virtual models in order to identify their respective limitations.
Figure 1. The bifocal modeling schematic
In these activities, students explore scientific phenomena such as heat diffusion, the properties of gases, and wave propagation by conducting physical experiments, designing virtual models, and connecting the experiments with the models in real time through iterative comparisons with empirical data. During the physical phase of the process, students design and develop their physical experiment, and they run the experiment while collecting data with embedded sensors or a time lapse camera. Concurrently, they design and develop a virtual model for the same phenomenon, and compare the behavior of the virtual model with their observations from their physical experiment. When they identify a discrepancy, students have the opportunity to redesign their models and re-iterate the process.
Bifocal Modeling includes various distinctive sub-activities as described in the figure below (Blikstein et al. 2012).
Different ways of resource allocation among these sub-activities could yield various modes of implementation.
A. Design – Students were encouraged to use external learning resources, such as the web or books, to gather initial information about the phenomenon. Students select questions they would like to answer, generate hypotheses about what they will observe, and design physical experiment and virtual model that can potentially answer those hypotheses. In designing the virtual model, students typically define the possible variables, and conceptualize micro-rules or equations to describe the phenomenon.
B. Construct – Students construct structure of their physical experiment (e.g. a ball and ramp) and virtual model (e.g. a computer model of a ball rolling down a ramp) that will capture the phenomena under study.
C. Interact – Students gather data from their physical models using embedded sensors. Similarly, they gather data from the virtual model by changing parameters, running the model, observing the results, and recording data.
Paulo Blikstein, Tamar Fuhrmann, Daniel Greene, and Shima Salehi. 2012. Bifocal modeling: mixing real and virtual labs for advanced science learning. In Proceedings of the 11th International Conference on Interaction Design and Children (IDC ’12). ACM, New York, NY, USA, 296-299.
Fuhrmann, T., Greene, D., Salehi, S., & Blikstein, P. (2012). Bifocal Biology: the link between real and virtual experiments. Proceedings of the Constructionism 2012 Conference, Athens, Greece.
Salehi, S., Fuhrmann, T., Greene, D., & Blikstein, P. (2013).The Effect of Bifocal Modeling on Students’ Assessment of Credibility. American Educational Research Association. Review.
Tamar Fuhrmann, Shima Salehi, and Paulo Blikstein. 2013. Meta-modeling knowledge: Comparing model construction and model interaction in bifocal modeling. In Proceedings of the 12th International Conference on Interaction Design and Children (IDC ’13). ACM, New York, NY, USA.
Furhmann, T., Salehi, S. & Blikstein, P. (2014). A Tale of Two Worlds: Using Bifocal Modeling to Find and Resolve “Discrepant Events” Between Physical Experiments and Virtual Models in Biology. Proceedings of the International Conference of the Learning Sciences (ICLS 2014), Madison, WI.