Project ScienceSpace was conceived as an effort to design and evaluate immersive virtual environments for science education from the elementary school through the early years of college (see www.vmasc.odu.edu/vetl/html/ScienceSpace/Sciencepace.html and www.virtual.gmu.edu). PaulingWorld is the part of the ScienceSpace Project that deals with molecular display and manipulation. A feature that differentiates PaulingWorld from other ScienceSpace applications is the ability for two or more distant users to share PaulingWorld and to collaborate on the tasks of understanding a complex molecular structure.
In PaulingWorld, the user is represented with an icon that communicates the user’s identity, location, and status (that is, whether the user is in control of the orientation and representation of the molecule under investigation). Therefore, as the user moves around and/or exercises control over the molecular structure of interest, the icon will react accordingly. Users employ a menu system attached to the user’s virtual hand to execute the options provided by the application. The selection of a menu item is accomplished through a ray attached to the user’s other virtual hand and by pressing a button on the wand—a mouse-like device associated with the user’s virtual hand. The users can also navigate around the virtual environment with the menu system or by using the tracking devices that monitor the user’s head and hand positions during the execution of the application.
In the virtual environment, PaulingWorld allows the user to examine the structure of both small and large molecules from any viewpoint and in a number of single or mixed representations. The menu system is used to change representations. The molecules can be represented in the familiar ball-and-stick form, as vanderWaals’ spheres, as coded sticks, as a backbone, or as icons that replace repetitive structures. Figure 1 shows a molecule in the vanderWaals’ sphere representation and Figure 2 shows a molecule as icons that replace repetitive structures.
To support the rapid examination of various molecules, structural data can be directly read in using the protein database format. The structural data in this format is widely available on the Web. Thus, the user can begin to interact with a new molecular structure in the virtual world within minutes.
When used in its distributed mode, PaulingWorld supports collaborative visualization of molecular structures among multiple users within the same virtual environment. The participants in the collaboration can see each other in the virtual environment through their icon representations. PaulingWorld uses a distributed database method to manage and maintain the models used within the virtual environment. Each site retains a copy of all the models used in the distributed virtual environment (DVE). Therefore, only state change information must be propagated to all the participants in order to synchronize the DVE. Any transformations or translations performed by one user on an object within that user’s virtual environment are propagated to all other users. Since state change information is primarily transmitted to the users, the bandwidth required to maintain the entire DVE is not very demanding.
In addition to state change data, the application also must transmit and process information that enables the transfer of the virtual menu (see Figure 3) among the users of the DVE. This is accomplished with the application process initiating and maintaining a connectionless socket connection between the user application processes of each of the user sites. This connection also requires a minimum amount of bandwidth since only a single character is transmitted between the user application processes to effect the menu transfer.
Experiments have been conducted between Houston, TX, and Norfolk, VA, to determine the utility of the icon representations of the users and of the shared virtual menu. Subjects in Houston interacted with trained graduate students in Norfolk and carried out a series of prescribed operations within PaulingWorld. Pre- and post-questionnaires were used to obtain feedback from the subjects as well as objective measures of their actions, time on task, and errors. The results of these experiments showed that users were comfortable with the their representations, found them adequate for collaboration activities, and were easily able to exchange control of the menu. In addition, we investigated the usability of 3D versus 2D widgets for manipulation of a molecular structure. The post-use survey showed a significant preference of users for the 3D widget.
Much work remains to be done to demonstrate the educational efficacy of shared virtual environments like PaulingWorld. The research reported here provides guidance on the design of the environment and on user interactions with the environment.
Join the Discussion (0)
Become a Member or Sign In to Post a Comment