Symphony: Exploring User Interface Representations for Learner-Centered Process Scaffolding

Chris Quintana, Elliot Soloway, Joseph Krajcik, Andrew Carra, Matt Houser, Mike McDonald, Mike Mouradian, Aaron Saarela, Naresh Vyas, & Michele Wisnudel

Center for Highly-Interactive Computing in Education

University of Michigan

1101 Beal Ave.

Ann Arbor, MI 48109

+1 734 763 6988

quintana@umich.edu

 

ABSTRACT

We introduce Symphony, a supportive integrated environment for high-school science students. The process of science inquiry is a complex one that is difficult for novice learners to perform. With Symphony, we are exploring how to provide effective process support for the learner by conceptualizing complex processes in the user interface. We have begun by using flow diagrams and timelines in the interface. As we continue, we will evaluate and characterize these and other representations to see how to best support novice learners in the "doing" of a complex process.

Keywords

Learner-centered tools, process scaffolding, supportive integrated environments, process visualization

INTRODUCTION: PROBLEMS FOR LEARNERS DOING COMPLEX PROCESSES

At the Center for Highly-Interactive Computing in Education, we are designing learner-centered computational tools (e.g., [5], [6]) for high-school science students. With our current effort, Symphony, we are now integrating our tools within a supported environment that students can use to conduct authentic science inquiry.

We follow a "learning by doing" constructivist philosophy [6], that is, we feel that students learn science inquiry by actually "doing" such inquiry. However, the science inquiry process is too complex for the novice learner. Science problems have characteristics of so-called "ill-structured" or "wicked" problems [1]: the problem-solving process is an iterative, non-linear one with no fixed way of doing it. Also, there are many possible activities in the process, although only a subset of

activities may be necessary to solve the problem. Thus novice learners trying to negotiate the complex inquiry process encounter several difficulties:

 

 

 

 

 

 

• Learners do not know what set of process activities are possible.

• Learners do not know how to perform the different activities in the process.

• Learners do not know the rationales behind the different activities.

• Learners need help planning the activities they need to do and tracking their completed activities.

These obstacles make it difficult, if not impossible, for a novice learner to effectively engage in a complex process like that involved in science inquiry. We need to address these difficulties to "allow the doing" of a complex process by a learner.

ADDRESSING THE PROBLEM: PROCESS SCAFFOLDING FOR LEARNERS

Symphony is a supportive integrated environment (SIE), integrating our individual science tools while offering the necessary support that learners need. The integrated nature of a SIE provides advantages for learners, helping them manage tool products and interactions between tools [6]. However, here we focus on process scaffolding, exploring the metaphors and techniques that we can incorporate in the user interface of a SIE to support learners in "doing" the process.

One area we can build on is that of domain-oriented or task-based interfaces (e.g., [3]), which are usable for domain experts because the interfaces reflect the familiar (for the expert) language and activities of the domain. However, many domain-oriented interfaces will not work for learners since they are domain novices. We want an environment whose user interface reflects the language and activities of a domain in a manner that is understood by learners, helping them to "visualize the process" so that they can begin to effectively negotiate that process [2].

Symphony – First Iteration

The first metaphor we used in Symphony was that of a tabbed notebook, a familiar metaphor where sections in the notebook (marked by notebook tabs) represented the different science inquiry process activities (e.g., problem analysis, data collection and visualization, etc.)

The notebook metaphor was problematic because the tabs were not sufficient to show how the different activities in the process were related or connected. Also, activities may have sub-activities (e.g., building graphs is a sub-activity of data visualization) whose processes must be illustrated. However, the notebook representation did not scale well; if we used a tabbed notebook to illustrate an activity, a "sub-notebook" representation for sub-activities was confusing.

Symphony – Second Iteration

Thus as we began to redesign Symphony, we decided to explore a different approach for the interface. Symphony needed better conceptual representations of the inquiry process, representations that show the structure and meaning of the process [4] in a familiar (for the learner) manner. The conceptual representations we use in the second version include flow diagrams to display processes and a timeline grid for planning and logging completed activities (figure 1).

Figure 1: Main Symphony process diagram and plan/log grid

A flow diagram exemplifies a conceptual representation of a classificational diagram, which describes an active process and shows the interconnections between the parts of the process [4]. The flow diagram is easily understood, it illustrates to the learner the space of possible activities in the process, and it scales better, that is, we can use flow diagrams to illustrate the workflow involved in both the main inquiry process and in sub-activities. Finally, activities in flow diagrams can be augmented with help to describe the purpose of each activity. Thus with these augmented flow diagrams, learners can begin to see what activities they can perform, they can see how to perform them, and they can see the rationale behind different activities.

A timeline exemplifies a conceptual representation of a temporal analytical diagram, which relates a part-whole structure as successive stages [4]. The timeline, or "plan/log grid", offers another process visualization, displaying the student’s current investigation (i.e., the whole) in terms of the status of the different inquiry activities (i.e., the parts) in a manner that supports flexible planning and re-planning along with a comprehensive view of the investigation history.

Also we are seeing that a composition of representations is also important. For example, the main process diagram and the plan/log grid (figure 1) are linked in that students drag activities from the process diagram to the plan row of the grid. As activities are completed, they move from the plan row to the log row. This linkage between representations is needed since planning requires knowledge about both the process and process history. Thus, not only are different representations of the process important, but so too is the linking of the representations.

CONCLUDING REMARKS AND NEXT STEPS

With Symphony, we are exploring how to best conceptualize complex processes in the user interface of a SIE. As we continue with the project, we hope to describe and characterize for future designers of SIEs the representations and compositions of representations needed to provide effective process support for novice learners trying to negotiate a complex process.

We will test the current version of Symphony with groups of junior/senior high-school science students who will be using Symphony in week-long trials, followed by a semester’s worth of iterations in which the software will be refined and tested in the classroom.

ACKNOWLEDGMENTS

This research is funded by the NSF and the National Physical Science Consortium.

REFERENCES

1. DeGrace, P., and Stahl, L.H. Wicked Problems, Righteous Solutions: A Catalog of Modern Software Engineering Paradigms. Yourdon Press, Englewood Cliffs, NJ, 1990.

2. Favorin, M. and Kuutti, K. Supporting Learning at Work by Making Work Activities Visible Through Information Technology. Machine-Mediated Learning 5, 2, 109-118.

3. Fischer, G. and Lemke, A.C. Construction Kits and Design Environments: Steps Toward Human Problem-Domain Communication. Human-Computer Interaction 3, 179-222,.

4. Kress, G., and van Leeuwen, T. Reading Images: The Grammar of Visual Design. Routlege, London, 1996.

5. Norman, D.A., and Spohrer, J.C. (eds.). Learner Centered Education. Communications of the ACM 39, 4, 24-49.

6. Soloway, E., et. al. Learning Theory in Practice: Case Studies of Learner-Centered Design. Proceedings of CHI ‘96 (Vancouver, BC, Canada, April 1996), ACM Press, 189-196.

Back to hi-ce Office