TRAILS

1
TRAILS

• A jointly executed research project within
  Kaleidoscope

2
Workshop speakers

• Judith Schoonenboom, SCO-Kohnstamm Institute,
  University of Amsterdam, judith.schoonenboom_at_uva.n
  l
• Jean-Pierre David and Anne Lejeune, IMAG-CLIPS,
  jean-pierre.david_at_imag.fr, anne.lejeune_at_imag.fr
• Canan Blake, The Open University,
  c.tosunoglu_at_open.ac.uk
• Marta Turcsányi-Szabó and Eszter Bodnár, ELTE
  TeaM lab, turcsanyine_at_ludens.elte.hu,
  bodnar_e_at_ludens.elte.hu

3
OVERVIEW of the workshop

• Aim one small step in developing a comprehensive
  view on trails
• Four presentations
• Introduction
• Trails within a learning design
• Visualizing trails
• Computer conferencing trails
• At the end of the introduction, we will identify
  some questions that are relevant to trails and
  CSCL, including your questions
• Each presentation is followed by a discussion on
  the questions
• At the end overall conclusions

4
WHAT ARE TRAILS

• AND WHY ARE THEY RELEVANT TO CSCL?

5
Introduction

• What are trails?
• What is the TRAILS project?
• Results so far

6
The growing importance of navigational learning

• Growth of amount of information
• Navigation through information
• Growth of flexible learning
• Navigation through study materials and courses
• Development of navigational hardware devices
• Navigation led by hardware devices

7
Trail time ordered sequence of learning objects
8
Navigational learning and trails

• Support for navigational learning
• Support for the creation of trails
• Support for the analysis of trails

9
TRAILS AND CSCL

• Trails are relevant in many types of learning
  Formal learning / informal learning, personalized
  learning, collaborative learning
• Support for creating and analysing trails
  support for specific forms of CSCL, e.g.
• Orchestration of group navigation
• Creation by learners of their own objects, their
  own trails, which can be used by others
• Reflection by learners on their own trails within
  a group

10
THE TRAILS PROJECT

• and its relevance to CSCL

11
THE TRAILS PROJECT

• Kaleidoscope Jointly Executed Research Project
  (JEIRP)
• January December 2004
• Project led by London Knowledge Lab, Mark Levene
  (mark_at_dcs.bbk.ac.uk project leader) and Kevin
  Keenoy (kevin_at_dcs.bbk.ac.uk)
• Partners
• SCO-Kohnstamm Institute, University of Amsterdam
• School of Computer Science, Queen's University
  Belfast
• Cognitive Science Section, University of Graz
• Institute of Educational Technology, The Open
  University, UK
• e-Learning Unit of STREAM Technology Center,
  SchlumbergerSema Madrid
• TeaM Lab, Informatics Methodology Group, Eotvos
  Lorand University, Budapest
• Equipe ARCADE, Laboratoire CLIPS - IMAG, Grenoble

12
AIMS

• To apply new technologies to the challenge of
  trail-support in e-Learning
• To investigate how different types of learning
  objects can be represented within an e-Learning
  system supporting trails
• To study an ontology for trails of learning
  objects with an e-Learning system, which is
  compatible with Semantic Web technologies
• To investigate personalisation and collaboration
  within an e-Learning system supporting trails
• To integrate presently scattered activity in
  Europe which is relevant to this challenge

13
RESULTS OF THE TRAILS PROJECT SO FAR

• WP1 deliverable focuses on modelling users
  navigational behaviour and their changing
  knowledge state as they navigate through existing
  learning objects in a learning environment
• WP3 deliverable describes learning objects in the
  form of code, their re-use and metadata that are
  involved in creating trails
• WP2 deliverable is dedicated to pedagogical
  aspects of trails and describes
• user scenarios for working with trails
• support for working with trails
• a taxonomy of trails

14
A taxonomy of trails

• Trails
• Cognitive
• Temporal
• LO trails
• Authored
• Emergent
• Result trails
• Discussion trails

• Graphs
• Mind Map
• Learning Environment
• Simple
• Compound
• Open

15
Different trail types
16
Result 1 the orchestration of group navigation

• Formalized learning designs, e.g. IMS Learning
  Design afford
• Set up of trails that can or must be followed by
  a group or by individuals
• Specific roles are assigned to specific learning
  activities
• Roles and activities are assigned to specific
  persons and specific learning objects
• Orchestration done by
• Authoring different trails for different roles
• Creating moments in which these trails come
  together in one collaborative trail.

17
Result 2 a visual guide to group navigation

• Visualization can be made of the learning objects
  that have been visited by a group or by an
  individual, and of the learning objects that have
  not yet been visited, but which are of interest
• This can function as a guide to the group in
  choosing their route through a set of learning
  objects.
• Visualisation is based on an analysis of the
  aggregate statistics of learning object use by
  particular individuals or groups
• Analysis leads to understanding, predicting and
  recommending the type of learning objects
  learners are interested in

18
Result 3 the analysis of individual trails
within a group

• Analysis of the trail of an individual within a
  discussion
• reveals all kinds of information related to the
  role of the individual within the group, for
  example
• the type of preferred reaction that the
  individual gives to messages of other group
  members,
• the extent to which the individual contributes to
  the discussion, measured by the number of
  messages provided by the individual.

19
Discussion questions

• Why develop a comprehensive view on trails? What
  are the reasons for (not) doing this?
• What kinds of collaborative learning can be
  supported by trails and how?
• What trail techniques are used in supporting
  collaborative learning?
• The role of personalisation in e-Learning systems
  that support trails.
• What is the role of trails within the whole
  learning design?
• What is the role of metadata, both metadata of
  learning objects and metadata of trails?
• What is the anatomy of trails? How can trails be
  classified?
• And what are your questions?

20
Workshop report, October 8 - 1

• Trails is about routes that (groups of) learners
  follow through an environment (VLE), and the
  trails they leave behind trails are time ordered
  sequences of (learning) objects, either
  pre-existing or created by the learner. Trails
  are relevant to CSCL, e.g. collaborative trails,
  or use individual trails to deduce routes that
  might be of interest to other learners.
• Tracking trails could be used for control, but
  also to empower learners, by providing them with
  an overview of what they have done and/or what
  they could do next.
• By inspecting their own trails, students come to
  tell their own story this is an important part
  of the learning. If they tell their story to
  others, then this is also relevant to CSCL.

21
Workshop report, October 8 - 2

• For inspection, maps and visualizations of trails
  are needed, and these act as a form a feedback
  in the workshop two examples have been presented.
• Students must be able to edit their own trails,
  and to summarize them, e.g. by using mind maps.
• A raw trail only tells what happened
  successively, and then and then and then. A
  trail needs structure to become useful, it needs
  to become a story
• We must look into research on narrative, as a way
  to find your way through vast amounts of
  information

22
Discussion questions on the definition of trails

• Is trails only about navigational links between
  objects, or are other types of links involved as
  well?
• What is the relation between different types of
  trails?
• What is your definition of a learning object, do
  you make a distinction between trails between
  learning object and trails within a learning
  object?
• How does the idea of trails relate to the work of
  John Dewey?

23
Questions on trails and interaction

• Ways in which trails are useful in analyzing
  interaction
• Relations between subsequent messages show the
  development of the interaction
• A visual map of the interaction trails can help
  students analyze the discussion and their own
  behaviour
• Trails show patterns of participation
• In general, participants were sceptical on
  (semi)automatic interaction analysis
• Different roles of interaction participants will
  lead to different trail, as was shown in one
  presentation

24
Challenges

• How do we keep track of the harder elements, e.g.
  affective elements of learning?
• Much of the learning occurs in between the
  learning objects, how do we keep track of what
  happens there?
• How do we support incremental changes, e.g. in
  roles?
• What kind of decisions can be supported by
  trails?
• How can we further develop tools to make sense of
  available trails?
• What opportunities are created by the increased
  use of mobile devices?