The Unfolding of Learning Theories
Its Application to Effective Design
of Collaborative Learning
Seiji Isotani
[email protected]
Riichiro Mizoguchi
[email protected]
SWEL 2007
ISIR, Osaka University, Japan
Agenda
 Introduction to main topics in collaborative learning research
 CHOCOLATO: Concrete and Helpful Ontology-aware
COllaborative Learning Authoring TOol
 Background of CHOCOLATO
 Collaborative learning ontology
 Interaction Patterns
 Learner’s Growth Model
 Collaborative learning theory ontology
 Application for Collaborative Learning design
 Conclusions
Collaborative Learning (CL)
Effective Groups
CL
Design
Group
Formation
Theories
Theories
Teacher/Designer
Sequence of activities
...
Learners
Data Collection
For Group
Re-formation
Interaction
Analysis
Collaborative Learning Support System
Effective Groups
MARI - Main Adaptive Representation Interface
CL
CL
Design
Design
Group
Formation
Theories
Theories
Sequence of activities
Teacher/Designer
Intelligent
Learners
...
Support System
Teacher/Designer
CHOCOLATO
Data Collection
For Group
Re-formation
Interaction
Analysis
Concrete and Helpful Ontology-aware
COllaborative Learning Authoring TOol
The Problem of Using Learning Theories
 Learning theories
 hard to understand
 too complex & ambiguous
 There is not a common vocabulary to describe them
 Different point of views, levels of aggregation, perspective
and emphasis
 How to select and use a theory to form a group?
 How to “unfold” the theories into a set of activities for a
group?
 How to develop theory-aware programs to support
collaborative learning?
Approach
W(A)-goal
Common goal
W(L)-goal
G
What activity does the
group want to do?
How does the group
change its state?
W(L)-goal
Goal state
Primary focus (P)
Role
How does the lea rner
change his/her state?
Role
G
YI-goal
G
YI-goal
P<=S-goal
k./cog. state (Group)
S
Secondary focus (S)
S<=P-goal
Use ontological engineering
to describe theories for CL
I-goal
Goal state
S
k./cog. state
Y<=I-goal
Y<=I-goal
I-role
I-role
Behavior
Why does the learne r want to
You-role
interact with othe r lea rners?
Behavior
I-goal (I)
I-goal
G
You-role
Behavior
Behavior
I-goal (I)
I-goal
G
Ontological structure
Use ontologies to
support the
development of
ontology-aware systems
learning theories
The systems help users to:
propose group formation;
design group activities;
 estimate benefits, etc..
users
teacher/instructor/designer
Background of CHOCOLATO
 Works of Inaba, Ikeda, Go, Thepchai, Mizoguchi
 Collaborative learning ontology
 Learner’s Growth Model
 Interaction Patterns
 This work improves previous achievements by:
 Making tacit characteristics of learning theories explicit;
 Identifying the relationships among interactions, learning
strategies and learning goals;
 Proposing an ontological structure, a model to describe
learning theories and extending the Collaborative Learning
Ontology [Inaba et al, 2000].
Collaborative Learning Ontology (CLO)
Group Goals
LearningLearning
Strategies
by
Apprenticeship
Learning
Goals
Learning by
Guiding
Y<=I -goal(LA<=LB)
Y<=I-goal (LB<=LA)
W-goal({LA,LB})
I-goal(LA)
I-goal(LB)
LA
LB
HOW?
I-goal(LC)
Cognitive Skill Development
(cognitive, associative, …)
LC
Interaction
W-goal({L
A,LPatterns
B,L
C})
Knowledge
Acquisition:
(accretion, tuning, …)
Interaction Patterns
Interaction Patterns
for Learning Theories
proposed by Inaba et al. 2003
Cognitive
Apprenticeship
Peer Tutoring
Anchored
Instruction
LPP
Distributed Cognition
Influential I_L Events
Cognitive
Constructivism
Sociocultural Theory
Observational
Learning
Cognitive Flexibility
Theory
Pattern
Interaction
Interaction
Interaction 2
Interaction 4
Interaction 3
Instructor Event
Learner Event
Role
Action
Role
LA
Action
LB
I-goal(LA)
I-goal(LB)
Y<=I -goal(LB<=LA)
Y<=I -goal(LA<=LB)
Learning Theory
Necessary Interaction
Desired Interaction
CL Theory ontology
Instructor Event
Learning Event
role
Action
role
CL Theory
*
Learning Strategy IT<=LR
LA
Y<=I-goal
I-role
Learner
LB
Action
I-goal(LA)
I-goal(LB)
Y<=I -goal(LB<=LA)
Y<=I -goal(LA<=LB)
You-role
Instructor
Influential I_L event
I-goal
I event
I-goal
G
*
Learning Strategy LR<=IT
Y<=I-goal
Instructor event
Instructor is a learner
Same learners, but
Instructor
playing the role of
different viewpoints
Learner
instructor
I-role
Instructional action
Instructor
Action
You-role
Benefits for the Instructor
Learner
L event
I-goal
G
*
I-goal
Teaching-Learning Process
Interaction Pattern
Necessary Interaction Activity
*
*
I-goal
Influential I_L event
Desired Interaction Activity
Influential I_L event
Learner event
Learner
Learner
Learning action
Action
Benefits for the Learner
I-goal
Applications of CL Theory Ontology
Application for CL Design
To identify the roles, strategies and benefits for
learners;
To clarify the relationship between learners’
development and interaction patterns;
To facilitate the visualization of requirements (preconditions) and expected benefits of CL theories;
Learner’s Growth Model - LGM
Learner’s Growth Model (LGM) [Inaba et al, 2003]
S(0,0)
[Stages of Skill development]
S(2,0)
S(0,1)
nothing (0)
rough cognitive stage (1)
S(1,0)
S(0,2)
S(3,0)
S(1,1)
explanatory cognitive stage (2)
associative stage (3)
autonomous stage (4)
S(3,1)
S(2,1)
S(4,0)
[Anderson, J. 1982]
[Stages of Knowledge acquisition]
S(2,2)
S(1,2)
S(3,2)
nothing (0)
S(4,1)
accretion (1)
S(2,3)
tuning (2)
restructuring (3)
[Rumelhart and Norman, 1978]
S(1,3)
S(3,3)
S(4,3)
S(4,2)
LGM is a graph that represents all possible transitions
during the learner’s development
A learning theory shows one or more possible
transitions in the LGM graph
Facilitating Visualization with LGM
S(0,0)
S(0,1)
S(0,0)
S(0,1)
S(2,0)
S(2,0)
S(1,0)
S(0,2)
S(1,0)
S(1,1)
S(1,1)
S(3,0)
S(3,0)
S(2,1)
S(3,1)
S(2,1)
S(0,3)
S(1,2)
nothing (0)
S(3,1)
S(4,0) S(0,2)
S(4,0)
S(0,3)
S(2,2)
S(3,2)
S(2,2)
S(1,2)
rough cognitive stage (1)
explanatory cognitive stage (2)
S(3,2)
S(4,1)
S(2,3)
[Stages of Skill development]
S(4,1)
associative stage (3)
S(2,3)
autonomous stage (4)
S(1,3)
S(1,3)
S(3,3)
S(4,2)
S(4,3)
S(3,3)
learning by apprenticeship
in Cognitive Apprenticeship
S(0,0)
S(4,3)
learning by guiding
in Cognitive Apprenticeship
S(2,0)
S(0,1)
[Stages of Knowledge acquisition]
nothing (0)
accretion (1)
tuning (2)
S(1,0)
S(0,2)
S(4,2)
S(3,0)
S(1,1)
restructuring (3)
S(3,1)
S(2,1)
S(0,3)
S(4,0)
S(2,2)
S(1,2)
S(3,2)
S(4,1)
S(2,3)
Learning by Discussion
in Legitimate Peripheral Participant (LPP)
S(1,3)
S(3,3)
S(4,3)
S(4,2)
Facilitating Visualization with LGM
S(0,0)
S(0,1)
S(2,0)
S(1,0)
S(0,2)
S(1,1)
S(3,0)
S(3,1)
S(2,1)
S(0,3)
S(1,2)
S(4,0)
S(2,2)
S(3,2)
S(4,1)
S(2,3)
S(1,3)
S(3,3)
S(4,3)
S(4,2)
learning by apprenticeship
in Cognitive Apprenticeship
Learner plays an apprentice` role
following the learner events
[Stages of Skill development]
nothing (0)
rough cognitive stage (1)
explanatory cognitive stage (2)
associative stage (3)
autonomous stage (4)
GMIP: Growth model improved by Interaction Patterns
Cognitive Apprenticeship
Learning by Apprenticeship
[Stages of Skill development]
S(0,0)
nothing (0)
S(2,0)
S(0,1)
1
2
S(1,0)
3
4
2
1
3
5
6
9
7
4
5
2
4
3
6
5
9
S(2,1)
8 2
6
7
4
S(4,0)
6
5
3
S(3,0)
S(3,1)
1
2
8
2
3
2
S(1,1)
S(0,2)
[Interactions]
rough cognitive stage (1)
S(1,2)
4 6
2
9
4
5
3
5
1
2
4
3
5
6
6
8
S(3,2)
2
9
4
7
5
6
S(4,1)
The dashed ellipses means
that the interaction on the
top/left must be followed by
another interaction
bottom/right.
The ellipses means that the
interaction on the top/left will be
followed by another interaction
bottom/right and vice-versa
(cycle)
8
2
S(1,3)
S(3,3)
tuning (2)
y Necessary Interaction
7
S(2,3)
3
2
[Stages
of Knowledge acquisition]
6.Instigating
thinking
7.Requesting problem’s details
nothing (0)
8.Showing a solution
9.Affirmative accretion
reaction(1)
restructuring (3)
x Complementary Interaction
S(2,2)
S(0,3)
1.Setting up the
learning context
explanatory cognitive stage (2)
2.Demonstrating how to solve a problem
associative stage (3)
3.Clarify the problem
4.Monitoring autonomous stage (4)
5.Notifying how the learner is
S(4,3)
S(4,2)
MARI – Main Adaptive Representation Interface
Anchored Instruction
Learning by Being Taught
Peer Tutoring
LPP
Learning by Teaching
Learning by Practice
Anchored Instruction
Learning by Diagnosing
Interaction
Patterns
Cognitive Flexibility
Learning by Self-expression
Search
Results
Conclusions
In our research we have been using ontologies to establish a common
understanding of what a learning theory is by representing it in terms of its
explicitness, formalism, concepts and vocabulary.
 This makes theories understandable and sharable, both by computers and
humans.
 This research presented the main concepts to develop an prototype of an
ontology-aware system to support CL design that:
 Explicitly identify the relationships among interaction patterns, learning
strategies and learning goals;
 Allow graphical visualizations of learning theories. Thus, users can
quickly interpret the theories, their benefits and can propose
sequence of activities in compliance with them.
 For computers, we provides ontologies which allows systems to reasoning
about the theories and the features (actions, roles, etc …) prescribed by
them.
The Unfolding of Learning Theories
Its Application to Effective Design
of Collaborative Learning
Seiji Isotani
[email protected]
Riichiro Mizoguchi
[email protected]
Thank You
ISIR, Osaka University, Japan