Contents:
The Project
Instructional Issues
Research Issues
The Future
Conclusion
The Project The US WEST Virtual Reality Roving Vehicle program (VRRV, pronounced "verve") enables students in grades 4 through 12 to experience and use the technology of Virtual Reality (VR). Funds from the US WEST Foundation have enabled the Human Interface Technology Laboratory (HITL), which is part of the Washington Technology Center at the University of Washington, to establish an outreach program in which vans bring high-end VR workstations to participating schools, none of whom would otherwise have access to VR. The project is staffed by programmers and engineers from HITL and by faculty and students from the University of Washington's College of Education. At the time of writing (February, 1995), a VRRV has visited 28 of the 50 schools participating in Washington State, bringing VR to roughly 1200 children and their teachers. The program will expand out of state next year.
The VRRV project has two goals. The first is to provide information about VR to children and to give them experience in a virtual world. To this end, the van visits a school for one, or at most for two days. VRRV staff give presentations to students and teachers about the nature, potential, limitations and current applications of VR. Throughout the day, students spend time visiting virtual worlds under the guidance of VRRV staff. Most of the time, these are worlds built commercially to show of the latest facets of VR. On occasion, we invite students to visit worlds we have built ourselves or worlds built by other children.
The second goal requires a more extensive engagement with teachers and students. VRRV staff work with teachers and students to build an instructional unit, part of which will require students to design and build a virtual world. World-buidling is a two-step process. First, the objects that will be found in the world must be created using 3-D modelling software running on Macintosh computers. The VRRV project supplies schools with the necessary software and provides training workshops on how to use it. The second step is world assembly. The graphics files that the students create are loaded into a VR workstation (Division Inc.'s Provision 100) and embued with the actions and behaviors the students have specified. At present, VRRV programmers peform this task. The heart of the suite of programs used in world assembly is a scripting language that is not all that straightforward to use. However, VRRV programmers are developing a Macintosh-based tool that will allow teachers and students to complete the world assembly process by specifying objects' actions and behaviors from menus and dialogue boxes.
The world-building activity begins with a one-day workshop for teachers that covers instructional design for VR, what VR is good for, assessment and project management. By this time, teachers have a good idea about the projects their students will work on. VRRV staff work with them in their classrooms on unit development and help them train their students to design worlds and to use the object-building software. They supervise object design and creation and help students decide how objects are to behave and interact, tempering youthful enthusiasm with information about what it is possible to do. VRRV staff will often work directly on curriculum development with teachers and students and will teach content if invited to do so. After the worlds have been assembled, VRRV staff supervise the activities during which the students visit the worlds they have built, assign tasks to perform in the virtual worlds and manage the gathering of data.
So far, we have piloted the world-building phase of the project in one middle school. One hundred and twenty grade seven students, working on a unit on wetlands ecology as part of their biology curriculum, built four worlds that modelled the processes of the water, energy, nitrogen and carbon dioxide cycles. The students' task was to demonstrate the processes operating in each cycle by manipulating the objects they had built and accounting for the results. For example, a student would "grab" some energy from the sun, use it to evaporate water in a pond, pick up the water vapor and lift it into the sky to form clouds and make it rain. Or a student could "feed" plants to herbivores, herbivores to carnivores and observe animal and vegetable material decaying to return nutrients to the soil.
Instructional Issues We cannot stress strongly enough that the VRRV project is an educational program, not just a series of demonstrations of VR. At the heart of the project are two assumptions. First, we believe that many children do not do as well in school as they might because they have difficulty working with the traditional abstract symbol systems typically used in texts, computer programs and even videos. VR has the potential for allowing students to have direct, concrete, first-hand experiences of concepts and procedures that have hitherto been unaccessible except through abstruse symbolic narrative. Second, we believe that VR is the technology most likely to support constructivist pedagogy. When students build worlds, they are literally "constructing" environments in which their own understanding of phenomena is embedded. When students visit worlds, they act in ways that can approximate their natural actions in the real world and, by experimentation and observation, construct their own understanding of the way the world works.
The activities in which the VRRV staff engage with students and teachers are premised upon these assumptions. We give students a great deal of autonomy in deciding how the objects to show in virtual worlds are selected and represented. We allow students to decide how objects interact with each other and with the participant. Object behavior is determined first and foremost by the way in which students want their worlds to behave. One issue that arises from this approach is the extent to which students are guided or coached while they are in the virtual world. Most of the time, we have found it necessary for someone standing beside each student to give guidance about how to act in the virtual world and what to do. The reason for this has been the limited time available for training students to navigate in the virtual world. The VRRV system uses wands rather than gloves for moving and acting in the virtual world. This requires that students learn the symbols system of wand position and button presses at the same time as they are learning content. We expect to study the possibility of using guides, or "avatars", to accompany students in virtual worlds to whom they can turn for help both with navigation and with content. Completely unguided experiences are unnecessarily difficult. Yet too much unsolicited guidance defeats the whole point of constructivist pedagogy. We are anxious to determine where the balance lies.
Research Issues The VRRV project has a carefully thought-out research agenda that consists of three research quesitons. The question, "Can students use VR?" can be answered unequivocally in the positive. Ever since our first world-building experiences with students attending a Summer camp at Seattle's Pacific Science Center, we have known that students in grades 4-12 are perfectly capable of using modelling software, of conceiving of ideas they want to express in three-dimensional space, and of navigating comfortably in virtual space.
The question, "Is VR better (or worse) at helping students learn content?" is still in part unanswered. VRRV's research agenda distinguishes between learning from designing and building worlds and learning from visiting worlds. To date, our emphasis has been on learning from designing and building worlds. The reasons for this are practical as well as pedagogical. HITL simply does not have the resources to convert large pieces of the curriculum to VR. When students build their own worlds, the acts of researching a content area, deciding how what they find can be expressed in the objects and behavior of an entire environment, and making design decisions about how to realize those ideas, are sufficient for them to learn the content. The VRRV team has also built worlds that have particular curricular objectives. The most successful of these is a world that allows students taking grade 11 Chemistry to build atoms and molecules my working directly with virtual protons, neutrons and electrons. We are gathering data that will let us determine the effectiveness of world-building and world-visiting relative to other technologies and pedgagogies. The question. "Why is VR better (or worse) than other approaches?" requires a more experimental, or quasi-experimental, approach. The VRRV team's research hypotheses address the relative effectiveness of VR to teach different subject areas, to teach students with different characteristics and learning styles likely to affect the way they learn, and are aimed at isolating the characteristics of VR that are most likely to affect learning. We have data on the relative effectiveness of different types of feedback, immersion, and the way in which objects and behaviors are represented. It is this research more than any other that will lead to guidelines for the design and creation of effective virtual worlds.
The Future The vans are temporary. In fact, it is fair to say that the vans cause more trouble than they are worth -- concerns about scheduling, insurance, equipment, qualifications and roles for "drivers" and per diems for overnight trips, get in the way of thinking about curriculum development and research into VR's effectiveness. It is HITL's goal to do away with the physical delivery of VR as soon as possible and to provide network access for project schools to link to each other and to HITL's computers for world-building and visiting. HITL's "Greenspace" project is pioneering technology to provide multiple participant distributed VR that will allow many people at geographically seperated sites to share and interact in the same virtual world. An early demonstration of Greenspace placed two participants in Seattle and two in Tokyo in the same virtual world with the task of collaboratively corralling virtual creatures into designated areas in the virtual world. We hope that, in future demonstrations of Greenspace, at least one site will be in one of the VRRV schools.
The VRRV project has already taught us a lot about how VR might contribute to education. There is no doubt that children find experience in a virtual world engaging and motivating. School-age children have proven to be adapt at functioning in a virtual world and of using world-building tools effectively. The worlds they have designed and built are by and large imaginative and effective. Collaboration with other children on world-building project is itself an effective strategy for making discoveries about the natural world. The planned studies that are part of the project will provide a wealth of data about the circumstances under which VR will be most effective for helping students learn.
Return to [VRRV Home Page ] [ HITL Home Page ]