Position Paper for CHI 96 Basic Research Symposium (April 13-14, 1996, Vancouver, BC)

Combining Pedagogical and Technological Paradigms for Educational Software

Dan Suthers
and the
Advanced Cognitive Tools for Learning Project

Learning Research and Development Center
University of Pittsburgh
3939 O'Hara Street, Pittsburgh, PA 15260

suthers+@pitt.edu
412-624-7046 voice
412-624-9149 fax
http://advlearn.lrdc.pitt.edu/

Keywords: educational software, groupware, WWW.

Abstract

In this paper I describe the research of a new project, formed from two prior projects experienced in the construction of groupware for collaborative learning (Suthers & Weiner 1995) and coached apprenticeship in a learning-by-doing context (Katz & Lesgold 1994). I describe how we are combining these approaches to learning for a more effective pedagogy. I also describe how we combine several corresponding genres of educational software, and obtain more widespread access to advanced educational functionality by using WWW technology. Our work is proceeding at a rapid pace, with the first deliverables and formative evaluations beginning March 1996.

Introduction

Although some highly effective and interesting educational software has been developed, each genre seems to require a tradeoff between advantages and limitations. For example, conventional computer-based training uses pedagogically primitive "drill and practice"; "micro-worlds" simulations don't provide students with guidance on how to learn from them; and "intelligent tutoring systems" based on Artificial Intelligence technology are pedagogically more advanced but require expensive computers. Furthermore, all these genres differ in "look and feel," are not designed to work together, and are not designed to extend to group learning situations. Our approach is intended to address these apparent tradeoffs:

Our software is being designed for both collaborative and individual learning.
We can deliver individualized coaching and interactive simulations along with guidance in their use to users on a variety of inexpensive platforms at different locations.
The interfaces will be familiar to anyone comfortable with a web browser such as Netscape.
The databases and materials used by our software will be usable by other software as well.

Combining Pedagogical Approaches

Our work combines three approaches to learning: collaborative learning, coached apprenticeship, and problem-based learning (a form of learning-by-doing). Each approach is supported by a corresponding genre of educational software.

Collaborative learning, in which groups of students work together on projects, is an important component of our approach to learning critical inquiry skills. Working in groups, students experience the "give and take" of debate while learning how to work together. Collaborative learning has been shown to correlate with more time on task and greater learning. Groupwork motivates students' engagement in learning tasks. Peer support enables students to perform better than individuals might, because it makes a greater diversity of ideas and skills available, and because peers can check each others' ideas and work (Braddock & McPartland 1993, Scardamalia & Bereiter 1991, Slavin 1990). The corresponding software genre is "groupware for learning."

In problem-based learning, students gain first-hand experience with the ideas and materials of the learning domain and learn the target skills by doing them. This kind of learning is often supported with simulations, exploratory environments, or "microworlds." Students can experiment with different simulated situations to see how things work, or even change the rules of the simulated worlds to see what happens. They gain a sense of ownership of knowledge gained through their own activities.

However, merely working together and using microworlds is not enough. Social interaction does not guarantee success Brown & Palincsar, 1989). Students are usually not able to model or articulate the knowledge and skills to be learned to each other. Simulations lack an explicit learning agenda as well as support for learning in the form of explanations and coaching. Students need guidance, which coached apprenticeship, or learning from an agent who can model and explain the desired skills and knowledge, can provide. The corresponding genre of software is "intelligent tutoring systems."

Our educational research agenda is to investigate the combination of collaborative learning and coached apprenticeship in the context of learning by doing. Our empirical methodology is to observe small groups of students working together on a problem, being asked to resolve things between themselves as much as possible, but with a human coach available for consultation if needed. We analyze transcripts of the interaction to identify what the peers are able to offer to each other, and what help they needed from the coach.

Combining Technological Approaches

We are also combining the corresponding software genres of groupware, simulations, and intelligent tutoring systems. This work includes the following elements:

Groupware that enables students to collaborate with other students on a project over a period of time, without the students always needing to be at the same place or working at the same time.
A collaborative inquiry notebook that keeps track of their inquiry process, including a statement of the problem, hypotheses that have been proposed and evidence offered for and against them, as well as references to information resources and experimental records.
Diagrammatic interfaces that help students visualize the important ideas in a critical inquiry debate as concrete objects that can be pointed to, linked to other objects, and discussed; and that can display the record of their inquiry process from different points of view to make information relevant for certain tasks more salient.
Extensions to simulations and other online materials, to provide semantics needed for coaching, and to enable these hands-on materials to be referenced in the collaborative notebook.
Coaching strategies that are designed to stimulate productive collaboration towards identifying effective solution strategies, as well as coaching strategies that model effective solution strategies and solutions themselves.

The World Wide Web is rapidly becoming the "common ground" of human-computer interaction, no longer limited to being merely a convenient delivery mechanism for text and graphics. Our strategy for taking advantage of the Web's potential for collaboration and to make our software affordable and widely available has three components:

Use of widespread media and communications standards, so that all media and most functionality are accessible from software (web browsers) rapidly becoming available on all platforms. This provides universal access from a uniform, familiar interface.
Client-server technology to enable users of "lightweight" platforms to have access to "heavyweight" functionality, namely artificial intelligence-based simulations and coaches available on an adequately configured server machine.
Use of Java to handle cases where specialized software must be run on the user's machine, without requiring previous installation of that software. Java is a programming language that can to a client embedded in an HTML document, and interpreted by a platform-specific Java interpreter or (increasingly) Java-aware Web browser.

The following figure illustrates the organization of our current global communication architecture.

Scenario for Use

In a typical usage scenario, a student starts a session by accessing a project "home page" in Netscape. The home page has buttons that can start different programs: one of the buttons is for "Belvedere", the diagrammatic interface to the collaborative inquiry notebook. When selected, the Java source for Belvedere is sent and executed, resulting in the appearance of a Belvedere window containing the project group's previous work.

Back in Netscape, the home page also shows a list of project topics. If the student clicks on one of the project labels, such as "Mass Extinctions," an index of things one can do in the project appears. This includes:

Browsing online "libraries" of relevant background information.
Going to a "conference" where the student can get newly published papers.
Asking any of a number of pictured "specialists" what they think about the issue.
Going to a museum to look at specimens.
Going to a field location to collect relevant data.
Running simulations of relevant physical systems.

The student accomplishes all of these things by using Netscape. When she finds some information that she would like to use, she clicks on a button that says "Use This." This invokes a Java "applet" within which the student is asked for a title used to refer to the information, and optionally to summarize the information. A diagrammatic reference to the new information appears in the NB window. The student can then link the new information to other ideas, or go on to do something else. (Students can also create new ideas of their own, or bring in information from offline sources.) Other students who are on the same project and connected to the collaborative database will see the new information appear in their displays. Thus, the collaborative database displays are like an online "bulletin board" on which students post and organize their ideas. At any time students can ask for advice from a coach by clicking on a button in NB which sends a request to the session server. Coaching functionality may also be available from Netscape via CGI as shown.

Conclusion

Combining collaborative learning, coached apprenticeship, and learning by doing under the guidance of relevant cognitive research (including our own), we can match the strengths and weaknesses of these pedagogical approaches. Utilizing the Web as a universal computing substrate, we can obtain strong cross-platform capabilities with minimal effort. This enables us to deliver functionality that formerly required more powerful computers to run than the average user's PC.

References

J. Braddock & J McPartland (1993).: Education of early adolescents. Review of Research in Education 19, 993, p. 135-170.
A.L. Brown & A.S. Palinscar (1989).: Guided, cooperative learning and individual knowledge acquistion. In L. Resnick (Ed.), Knowing, Learning, and Instruction: Essays in Honor of Robert Glaser. Hillsdale, NJ: Lawrence Erlbaum Associates.
S. Katz & A. Lesgold (1994).: Implementing post-problem reflection within Coached Practice Environments. In P. Brusilovsky, S. Dikareva, J. Greer, and V. Petrushin (Eds.), Proceedings of the East-West International Conference on Computer Technologies in Education (pp. 125-30), Crimea, Ukraine.
M. Paolucci, D. Suthers, and A. Weiner (1995).: Belvedere: Stimulating Students' Critical Discussion. CHI95 Conference Companion, May 7-11 1995, Denver CO, pp. 123-124. Postscript available via FTP here. Also available in HTML here.
R.E. Slavin (1990).: Cooperative Learning: Theory, Research, and Practice. Prentice-Hall, Englewood Cliffs, NJ, 1990.
D. Suthers and A. Weiner. (1995).: Groupware for developing critical discussion skills. CSCL '95, Computer Supported Cooperative Learning, Bloomington, Indiana, October 17-20, 1995. HTML version available here (from CSCL site in Indiana). Binhexed MS Word 6.0 available via FTP here.
D. Suthers, A. Weiner, J. Connelly, and M. Paolucci. (1995).: Belvedere: Engaging students in critical discussion of science and public policy issues. AI-Ed 95, the 7th World Conference on Artificial Intelligence in Education, August 16-19, 1995, Washington DC. Postscript available via FTP here.