The research program will explore the abilities of the learner using this tool in a variety of contexts. We have three major goals for this research program. First, is the establishment of a research base on dynamic 3D visualization by learners of a broad age range. This project is the first known by the authors to make dynamic 3D visualization tools publicly available to learners as young as 8 years old and will give us unprecedented opportunities to develop this research base.  Second, we are interested in tool design issues for learners. We would like to develop and test LCD features that are empirically derived, refined, and then generalized within various learning environments. Third, we will explore small group construction of knowledge about visualization and atmospheric phenomena through a series of studies where the DWS is used in a variety of situations adopting various pedagogical strategies.  These strategies will include open exploration, challenge questions, guided discovery, and anchored instruction, the latter being the focus of the next section.

   The Saving Eric Adventure is a prototype macrocontext that Ken Hay, Rick Duschl and Kirsten Ellenbogen at Vanderbilt University, and myself are developing.  Central to leveraging the DWS as an instructional tools is the development of a macrocontext, which will help anchor students’ weather experience within a context that is meaningful to students (Cognition and Technology Group at Vanderbilt—CTGV, 1990). Founded on theories of problem-based learning, complex, video-based macrocontexts are intended to overcome inert knowledge by anchoring learning within the context of meaningful problem-solving activities.  The anchor sets up a problem, which provides the justification and motivation for why students need to learn the information, as well as how it can be used to address real-world problems.  Potential solutions are not distilled into passive minds, rather they are "embedded" in the curriculum to be perceived by the learner.  Conceptual content is then extracted and given meaning by students. 
      An anchor is essential to the CTGV’s notion of anchored instruction, referring to instruction in which the material to be learned is presented in the context of a specific topic that serves to anchor or situate the material, and, further, allows it to be examined from multiple perspectives.  In contrast to the disconnected sets of "application problems" located at the end of textbook chapters, macrocontexts refer to stories that take place in semantically rich, open-ended environments (CTGV, 1993).  In these anchored macrocontexts students begin with a higher-order problem and then use top-down strategies to generate the necessary sub-goals to reach the final state.  This top-down processing helps students learn the lower level skills (i.e., mathematical algorithms and facts) in a manner that also gives them insights into the relationships between the skills being learned and the reciprocal opportunities for using them. 
      Anchors help set up goals that, when adopted, constrain which information a learner attends to and provides a legitimate reason for learning the information (Young & Barab, in press).  In creating a legitimate use for the information being learned, a well-designed anchor has the potential to transform content from facts to be memorized to rich, conceptual tools whose function is meaningful to the students (Bransford et al., 1992).  It is with the goal of setting up a motive for students to enact the atmospheric science content that provides the conceptual core of the DWS that we have designed a surrounding macrocontext,  Specifically, the anchor problem that we have designed, "Saving Eric," involves students working in dyads to create a flight plan from Minneapolis to Indianapolis to save a child with a bad heart.  The initial problem is first presented as a digitized video on the computer, with various surrounding hotspots that allow users to locate other relevant information (other students modeling how they used the DWS, pilots discussing past flights during good and bad weather conditions in a similar plane, meteorologists giving weather reports), and a link to launch the DWS tool.  However, our goal was not to develop an environment where students would simply "acquire" information but actually have opportunity to participate in its making using the DWS to visualize atmospheric phenomena.
      Central to any macrocontext are a series of constraints that make the problem more motivating, and that provide a reason for learning the content that is the conceptual core of the macrocontext. The two main constraints of Saving Eric are time and safety.  The time constraint results from the fact that users need to save a little boy in Indianapolis who needs a heart transplant, which has just been removed from a body in Minneapolis.  The safety constraint is that the only plane at the hospital is a two-seater, which is highly sensitive to weather conditions.  Therefore, in planning out the route the user needs to use the DWS to envision weather patterns from 2,000-5,000 feet, and needs to consider multiple variables (i.e., wind speed, temperature, humidity, air pressure).  In addition to considering the variables in isolation, the dynamics among the variables are also important (e.g., beginning in a rainy area and flying in to a cold front may result in ice on the wings). 
    The decision making process meets the epistemic goals for the macrocontext, including decision making and argumentation (Duschl & Gitomer, 1997;Toulmin, 1969).  Given our goal of facilitating decision making and argumentation skills, our particular macrocontext does not have one "answer," but a series of alternative flight plans from which students have to develop the best plan.  This involves first deciding which data is important to consider, then using the DWS to visualize this data, interpreting the visualization, and making a public display justifying the plan.  In supporting students in determining which data to use, we used an embedded data design so that necessary information was contained in the macrocontext.  For example, there were interviews with air-traffic controllers, pilots, and meteorologists that discussed potentially important atmospheric data, as well as other problem irrelevant information .  The inclusion of relevant and irrelevant data in the macrocontext further addresses and validates the need for students to use decision making and argumentation skills. 
     Once students had used the DWS to visualize data and had drawn up a flight plan, they are expected to make a presentation of their proposed flight plan.  Each presentation will draw on our epistemic goals with students being required to justify which information they used in developing their plan, now it supports their claims, and why this information should be considered accepted as credible evidence (Duschl & Gitomer, 1997).  Central to our research is the notion of situated assessment (Young, Kulikowich, & Barab, 1997), in which we embed the assessment as part of the instructional task.  In this case, our assessment involves evaluating student presentations, what Duschl and Gitomer (1997) referred to as an assessment conversation.  More specifically, each presentation are evaluated in terms of a warrant (explaining how the data, visualization, supports the conclusion), and a backing (explaining why the warrant should accepted as credible evidence) (Toulmin, 1969).