1 EmpiricalBased, Evolutionary Design of FLEAgents

1
1) Empirical-Based, Evolutionary Design of
FLE/Agents

• Background
• Knowledge building environments
• Method
• Empirical study and design
• Pedagogical agent system

2
DoCTA NSS project

• Design and use Of Collaborative Telelearning
  Artefacts Natural Science Studios
• Goal Study social, cultural and pedagogical
  aspects of shared artefacts in distributed
  collaborative learning and apply the findings to
  the design of new learning environments

3
Knowledge building

• A model for collaborative learning (Scardemalia
  Bereiter, 1994)
• Students learn and interact by talking
  (reasoning aloud) with peers to develop
  explanations of scientific phenomena
• Formulate research questions, answering them
  independently, and finding arguments
• Computer supported knowledge building
  environments
• CSILE, Knowledge Forum
• FLE

4
Method

• Empirical-based design
• Identify need for improved knowledge building
  based environment based on data from using an
  existing system
• Evolutionary design
• Identify ways to integrate software agents with
  existing system

5
Empirical study

• Two secondary school classes in Norway (9th
  grade)
• 3 week pilot 4 week field trial (2001, 2002)
• Collaborative learning in small groups
• Discussing science problems
• Knowledge domain Ethical aspects of
  biotechnology
• Computer supported environment without software
  agents (FLE system)

6
FLE (Future Learning Environment)

• Developed at Media Lab, University of Art and
  Design Helsinki (http//fle3.uiah.fi/)
• Open-source groupware implemented in the Python
  language using SOAP platform
• Designed based on pedagogical and psychological
  research in networked learning and knowledge
  building at University of Toronto and University
  of Helsinki (Hakkarainen, Lipponen Järvelä,
  2002)

7
FLE interface
Viewing mode (threaded list of previous postings)
Writing/reply mode (editor with message
categories)
8
FLE message categories

• Categories of progressive inquiry
• Problem
• My working theory
• Deepening knowledge
• Summary
• Meta-comment

9
Physical setting in one of the schools
10
Method

• Observation
• Video recording
• Data logging
• Interviews
• Interaction analysis

11
Data excerpts Interview

• When asked about the usefulness of the FLE
  categories, a student said
• It was kind of smart! Because you can see what
  it the message is about. Thats reliable
  knowledge and thats a summary pointing to two
  KB notes on the screen. You know immediately
  what it is.
• However, when later asked to demonstrate his
  understanding of the difference between My
  Working Theory note (MWT) and a Summary note
  he says (modifies his initial misunderstanding of
  two categories)
• if we had sent this to them pointing to a
  note he has labeled MWT and you ask what it is
  supposed to mean - is it a working theory or is
  it a summary, right? But you see it first by its
  small category abbreviation oh -it is a
  summary after all, okay!

12
Summary of findings

• Students had difficulties choosing knowledge
  building categories
• Instructors had difficulties following the
  distributed collaboration process and guiding the
  students
• Tentative conclusion Need to find alternative
  ways of facilitating distributed knowledge
  building

13
Design implications

• Software agents can be useful as scaffold in
  semi-structured knowledge domains
• Pedagogical agents
• Agent system features

14
Pedagogical agents

• Pedagogical agents can be autonomous and/or
  interface agents that support human learning in
  the context of an interactive learning
  environment.
• Johnson, et al. (2000)

15
Agent system features

• Agent as an observer
• Collect information
• Participant, activity, timestamp
• Last log on, last contribution (for each
  participant)
• Compute statistics
• Agent as an advisor
• Present statistics
• Encourage non-active students to be more active
• Advice students on the use of knowledge-building
  categories

16
System architecture
FLE3
WebTop
Knowledge Building
Chat
Admin
User Interface Assistant GUI
Advice Generation
Observation Student activities Instructor
activities
Updates Who-is-online
Statistic Generation
DBlog
KBrules
Advice feature analysis Message feature Student
feature Instructor activity Confidence factor
Learning CN2
Assistant
17
Technical detail

• Languages
• Python (Fle3) and Java (agent applet interface)
• Database
• MySQL
• Learning algorithm
• CN2 (Clark Niblett, 1989)
• Knowledge representation
• RuleML

18
Student Assistant Agent
19
Instructor Assistant Agent
20
References

• Jondahl, S. and Mørch, A. (2001). Simulating
  Pedagogical Agents in a Virtual Learning
  Environment, Proceedings IRIS-24, pp. 15-28.
• Chen, W. and Wasson, B. (2002) An Instructional
  Assistant Agent for Distributed Collaborative
  Learning. Proceedings ITS-2002, pp. 609-618
• Dolonen, J., Chen, W. and Mørch, A. (2003).
  Integrating Software Agents with FLE3.
  Proceedings of CSCL 2003, Bergen, Norway. Kluwer
  Academic Publishers, pp. 157-161.
• Mørch A, Dolonen J, Omdahl K. (2003). Integrating
  Agents with an Open Source Learning Environment.
  Proceedings of International Conference on
  Computers in Education 2003 (ICCE 2003), Dec.
  2-5, Hong Kong AACE Press, 393-401.