Towards a Pattern-based Adaptive Approach for Instructional
Design based on Teacher's Pedagogical Design Scheme
Jean-Pierre Clayer, Claudine Toffolon and Christophe Choquet
LIUM Laboratory, PRESUNAM, Le Mans
Avenue Laënnec 72085 Le Mans Cedex 9, France
Keywords: Instructional Design, Pattern, Pedagogical Design Scheme, Design based Research Approach.
Abstract: Learning is changing in deep, activities of teaching practitioners have to evolve. They need to structure and
formalize their internal designs as models to be implemented but do not have competences in instructional
design. We propose to use patterns, semi-structured description of an expert's method for solving a recurrent
problem to elicit and express theirs needs. We propose an engineering design process framework based on
patterns, and pedagogical design schemes to support instructional design. A tool has been developed to
support this process. We experiment a design session with two trainers of an association dedicated to back-
to-work programs within an iterative co-participant design based research method.
Learning and teaching is changing due, on one hand
to the evolutions of the society expectations and, on
another hand to the widely spreading of new
technologies. Teaching, an unstructured, informal
world of much professional practice need design
principles. Designing learning is a complex task
(Rohse and Anderson, 2006). Teaching practitioners
need to structure and formalize their scenarios
(lessons plans) as models to be implemented but do
not have specific designer competence to easily
achieve it. To design, they require a mean to elicit
and express their needs (Bonnardel, 2009).
The research field has produced a set of design
process, methods and tools offering effective
mechanisms and notations to design different issues
in pedagogical or instructional topics. We get
attention in particular to those approaches which
deal the instrumentation of instructional design,
specifically learning scenarios centered. The well-
known modeling approach by the help of
Educational Modeling Languages (EML), as the
IMS Learning Design specification (Koper and
Olivier, 2004) leads to enable the design of
computational models (in the meaning of
understandable by a computer) which could be
enacted by compliant systems. However these
specifications are not really usable by teachers, need
pedagogical engineering expertise and do not
enforce design processes that support the creation of
pedagogically sound designs (Hernandez-Leo et al.
2010). Some limitations and shortcomings have been
identified to deal with the teaching practitioners new
tasks of learning design. Teachers found difficult to
express their needs or to reuse their design products.
Most of the time, not enough assistance and
guidance are provided. A second approach is to
allow designers (may be assisted by modeling
specialists) to define their own EML by specifying a
domain-specific and to use it for building their
scenarios (El-Kechaï and Choquet 2006). Our
research work is based on this second approach and
states that teachers are able to build their learning
scenarios by the uses of patterns.
We adopt a practitioner-centered point of view to
deal with the learning design activity based on
patterns approaches (Hernández-Leo and al 2006);
(De Moura 2008); (Emin et al., 2010). Patterns are
semi-structured description of an expert's method for
solving a recurrent problem which includes a
description of the problem itself and the context in
which the method is applicable (Mor and Winter,
2007). Patterns are “good solutions” to deal with
complexity characterizing the education field (Rohse
and Anderson 2006). According to Laurillard’s
works (Laurillard 2012), this approach is
immediately relevant to teachers as it presents means
by which a community can participate in design.
Clayer J., Toffolon C. and Choquet C..
Towards a Pattern-based Adaptive Approach for Instructional Design based on Teacher’s Pedagogical Design Scheme.
DOI: 10.5220/0004573805320538
In Proceedings of the 15th International Conference on Enterprise Information Systems (ICEIS-2013), pages 532-538
ISBN: 978-989-8565-60-0
2013 SCITEPRESS (Science and Technology Publications, Lda.)
This formalism offer the opportunity to the teacher
to externalize his knowledge (Goodyear, 2005) and
to express his practice which can become a best-
We propose an engineering design approach
based on pattern notation to benefit its potential to
facilitate the expression of pedagogical concerns for
teaching practitioners. We defined an engineering
iterative design process framework and an
associated tool to support the design activity of
practicing teachers. This tool may support the design
activity of scenario of teacher by proposing
adaptations, according to the design context and
pedagogical design scheme. We experiment a design
session with the trainers of an association dedicated
to back-to work programs within a design research
based approach.
In the next section we present the research
context. The concepts of patterns and pedagogical
design schemes are explained in the third section.
Then we propose our engineering design process
framework. Next we present the three main steps of
the experiment with trainers of the association. Then
we discuss the resulting set of design research
guidelines produced for researchers and teachers.
Patterns have been used in the 70’s in the field of
building architecture, by Christopher Alexander
(Alexander et al., 1977). Several repositories of
patterns exist for various disciplines and offer design
expertise reuse to the corresponding communities as
the object oriented programming community or the
HCI community. A pattern is defined by three main
properties: a problem, a context of this problem and
a solution (Alexander et al., 1977), (E-LEN, 2012).
Each pattern captures the best practice to answer to a
problem in a particular context. The formalism of
patterns differs depending on the type of problem
that the pattern solves. Each formalism adds specific
information for the category of problem to solve. By
their formalism, patterns support the creativity
without constraints. Users are guided rather than
forced in the use of patterns (Rohse and Anderson,
2006). This approach is particularly interesting with
novice teachers. A pattern language describes the
relations between patterns (associate, compose of…)
that captures the whole design process and can guide
the designer through step-by-step design guidelines
(Alexander et al., 1977).
In particular in the e-learning community many
projects as Pedagogical Patterns Project (PPP, 2011)
or Design Patterns for recording and analyzing
Usage of Learning Systems (DPULS, 2005)
proposed a catalog of patterns concerning learning
strategies types of problem. In this way designers of
new or existing LMS, especially inexperienced
designers, through the use of catalog of patterns can
take advantage of previous design expertise (PPC,
2012), (E-LEN, 2012), (Delozanne et al., 2007).
Researchers in education get increasing interest
with pattern-based design approach. COLLAGE, a
collaborative learning flow pattern (CLFP) editor
(Hernandez-leo, 2006) proposes a pattern based
visual design approach implemented in RELOAD.
This approach is based on the IMS-LD specification
which enables the modelling of learning processes.
The MDEduc project proposes a Pedagogical
Patterns Editor for the design of learning scenario
using the formalism and syntax of patterns (De
Moura, 2008). ScenEdit and the model ISIS support
also a pattern based approach to design learning
scenarios (Emin et al., 2008). These approaches help
teachers to better express their pedagogical needs
but they still lack process and tool support for the
design activity of learning scenarios. They are also
based on existing design plan. We deal with these
limitations in the next part of the paper.
According to the design problems highlighted
previously in teaching domain, we explored the TEL
engineering domain best practices and the software
engineering patterns solutions implemented in the
information system domain. A pedagogical situation
is composed of different elements defining a
scenario: Learning strategies, Learning situations,
Objectives, Activities, Human resources, Material
resources. We take into account of these elements
within four categories of pedagogical problem:
Activity design, Learning situations design,
Ressources design and Pedagogy. We identify four
formalisms of patterns to solve the four types of
Pedagogical patterns are describing learning
strategies and objectives. These types of patterns
seem to us well suited to present the information on
the pedagogical method in use (Bergin, 2000).
Analysis Patterns are well-suited to describe
material and human resources because of their
structure which allows defining the use of a material
or a role (Fowler, 1997).
Process Patterns are well-suited to describe
pedagogical activities and tasks realized as a
workflow of the actions or sub-tasks (Amber, 1998).
Design Patterns are well-suited to describe
learning situations. as they are used to resolve a
problem of learning design (Gamma et al., 1995).
Teaching design depends on learning theories. For a
learning theory, many approaches may be defined.
Project-based learning is a kind of pedagogical
approach related to the active learning theory. A
typology (Villiot-Leclercq, 2007) distinguishes the
different approach for each learning theory. A set of
methods and tools are defined for a specific
pedagogical approach and underpinned the creation
of the scenario and its pedagogical elements. The
choice of the pedagogical approaches and the
learning strategies allows identifying the type of
learning scenarios (Paquette, 2004).
The design process of a scenario according to a
specific pedagogical approach defines what we
named pedagogical design schemes (PDS). A
pedagogical design scheme is the approach for
designing a course, in our context, a learning
Early instructional design approach developed
concepts for systematically designing instructional
materials. The IMSL-LD specifications are based on
a learner activity driven approach (Koper and
Oliver, 2004). This type of languages provides best
practices guides driven by the efficiency of the
modeling rather than pedagogical design schemes.
Editors/tools developed for these languages
instrument usually the modeling activity guided by a
given design approach, underlying implicitly the
interfaces design.
Most of pattern-based design approaches are
driven by PDS. In (Emin, 2010), the design process
supported by the ISIS model is driven by a strategies
and intentions pedagogical design scheme, that mean
the design of scenario is driven by intentions and
strategies. The COLLAGE approach with the
collaborative learning flow patterns is driven by the
learning activities PDS (Hernandez-Leo 2006).
The design of pedagogical scenario is
constrained in these approaches by a specific PDS.
When teaching practitioners design a learning
scenario with the same objective and learning
situation, their pedagogical design schemes may not
always be the same. For the same design problem,
many pedagogical design schemes may exist.
A Patterns
According to the DSM approach, a meta-model
(Clayer et al., 2012) has been defined to describe the
language of patterns. This language use four
formalisms of patterns: Pedagogical Patterns,
Analysis Patterns, Process Patterns and Design
Patterns. According to this language, we represent
the pedagogical elements of a scenario with these
The figure 1 illustrates the simplified graphical
form of a pattern. This form presents only the most
important element of a pattern: a name, a problem, a
context and a solution.
Figure 1: Simplified Graphical Form of Process Pattern.
We represents the activity “Realize an individual
evaluation/training” by a process-pattern according
to the meta-model (Clayer et al., 2012). The problem
and context are in a textual form. The solution is
composed with a list of activities and the role of
B Pedagogical Design Schemes (PDS)
To assist the designers, we identify their learning
design activity based on their PDS. PDS is defined
by an oriented graph of the pedagogical elements.
The beginning of the oriented graph and the first
pedagogical element give the orientation of the PDS.
For example, figure 2 illustrates a pedagogical
design scheme collected during the experiment. This
pedagogical design scheme is driven by the
The oriented graph contains only one occurrence
of each pedagogical element. Designers can’t design
each type of pedagogical elements when they are
applying a given PDS. The link between the
pedagogical elements give us the next possible
design activity for the current design activity. The
design activity is identified to know the next and the
Figure 2: Objectives Driven PDS.
previous design activity in order to anticipate the
next action of the designer and propose assistance
(the selection of the next pedagogical element
according to the pedagogical design scheme
identified) or guidance (invite the designer to check
its current design).
C Design Process
Figure 3: Engineering design framework.
We propose an engineering design framework to
support the design process based on patterns. The
design process is compliant with any kind of
pedagogical design scheme and support the
identification of the current activity and design
context to offer guidance to the designer based on
adaptive rules. The design process is iterative and
consists of the following steps (Figure 3):
1. Instantiate the Specifying Requirements Pattern:
as designer, the teacher/trainer has to define the
requirements of the pedagogical situation
(objectives, resources, tools, learning strategy,
actors, roles, etc.);
2. Select a Pedagogical Element: the designer can
choose an existing pattern or creates a new one.
Designer does not manipulate patterns but the
pedagogical element represented by the patterns.
The pattern complexity is hidden to the designer.
According to the learning context, the designer
defines parameters to select the most relevant
pedagogical element. He can choose one element
among the pedagogical elements proposed by the
editor: learning situations, human or material
resources, activities, pedagogical strategies or
3. Merge the Element: The designer has to
associate the selected element to the ones he has
already selected during previous iterations. The
scenario is updated with the selected element and
both user's actions and element properties are
collected to be analyzed.
4. Collect Information: The information collected
are analyzed. According to this analysis, the
design context is updated and the adaptation
rules could be fired.
5. Adapt the Solution: A set of possible adaptations
is proposed to the designer, according to the fired
rules set. Adaptations could lead to provide some
recommendations on the design method, or to
propose pedagogical element related to the user's
action, better suited for the design context.
D Editing Tool for Pedagogical Scenario
We develop an editing tool to support the design
process. The editor is generated into the EMF-GMF
framework (Kelly, 2007), from the meta-model of
Patterns (Clayer et al., 2012). The adaptation part of
the process has not been implemented yet in the
editing tool. It will consist in an adapting system
based on the design context and able to adapt the
domain specific-models and the editing interface
according to the DSM features. The interface of the
tool allows the designer to visualize the patterns
(pedagogical elements) he instantiates as frames
divided in four boxes: one for the pattern name, one
for the context, one for the statement of the problem
and one for the graphical representation of the
solution (Figure 4-A). A toolbox (Figure 4-B)
provides the design primitives (pedagogical
elements: objective, pedagogical strategy, material
resource, human resource, activity, learning
situation). Finally, information concerning the
pattern is also accessible through the tab property
(Figure 4-C).
We have realized an experiment with the
professional trainers of a registered association
PARTAGE dealing with professional integration, in
charge with back-to-work programs. PARTAGE
Figure 4: Editing tool.
provides trainings essentially based on formative
evaluations supported by classical pedagogical
resources and methods (teacher-learner based
strategy). Supports for trainings are not always well-
adapted to the public. The public is composed of
adults sometimes not always comfortable with
trainings and classical paper-based support. The
association also faces many changes within its
trainers staff.
PARTAGE has expressed to us their needs based
on two aspects. Firstly, they would like to produce
computerized pedagogical materials in order to
facilitate reuse and adaptation. Secondly, they were
interested by a tool to support their design sessions
in order to produce scenarios better adapted to their
heterogeneous public. They would like to capitalize
the training practices of the constantly evolving
trainers staff and get assistance when designing the
training scenarios, and share those practices between
We worked with two trainers within an iterative
participatory design based research approach.
Trainers took part in the activities of analysis and
design many times in a collaborative work with the
research team. Trainers arrived with scenario paper
based written scenario and follow the design process
proposed in previous section in an iterative way.
A Design of PARTAGE Patterns
During this first session, we have observed the
practices of the association during working meetings
and trainings. The association has defined a process
for their professional integration programs. The
main outcome of this process is to provide
employment to their members depending on their
individual profile and skills.
One of the steps of this process is composed of
training, most of the time based on evaluation.
During this training evaluation step, the skills of
learners are evaluated, according to a category of
work, by a trainer with the support of different
pedagogical resource/material.
The trainer prepares the evaluation training
according to the learner abilities and the available
resources. The trainer designs a learning scenario on
paper to describe the learning situation and the
pedagogical elements associated.
We have collected information on the association
practices by the observation and interviews of
trainers in a need's analysis report. On the basis of
this report, we have designed some patterns (Figure
A set of patterns which describes the learning
situations, the pedagogical elements and some
scenarios we observed has been proposed to the
trainer in charge with the association.
We have proposed other set of patterns to the
association leader. She approved this representation
of their practices within patterns. She reports that
“the design approach based on patterns forced the
trainers to have more rigor and help them to express
their needs”. Nevertheless, she noticed some
difficulties to understand some part of the patterns
formalism. Some terms used to define a section of
patterns were not familiar to the trainers. They
recommended us to support the design of pattern
with a textual description.
During this first step of the experiment, we have
collected a lot of practices and captured them in
B Prototyping Tool Demonstration
We have developed a first version of the editor
prototype to support our engineering design
framework. To reify the design approach, we
planned a working meeting to present the prototype,
and resulting patterns and scenarios designed in the
first step of the experiment. We asked them to
evaluate the resulting scenarios and the tool
implementing them, an attention was given to the
pedagogical design scheme.
We have demonstrated the design of a scenario
with the tool to the trainer. After the demonstration,
we asked for an evaluation of the tool. The trainer
noticed the lack of user friendliness of this first
version prototype. Despite these limits the trainer
has appreciated the opportunities of formalization of
the pedagogical scenario and the expressiveness of
the artifacts resulting from the editing session.
C Collecting PARTAGE Pedagogical Design
We have improved the prototype to take into
account observations made by the trainer during the
previous session. Use of patterns is not explicit for
designers, they handle elements of scenario of the
learning situation: define activity, actors, resources,
and so on.
In a third session we proposed to the trainer to
implement a scenario of a new session of their
training programs using the editing tool. This time
two trainers were implicated. They aimed to
implement the same learning situation but we
noticed that they had two different way of
formalizing it. We identified different pedagogical
design schemes for the same learning situation.
They designed a complete scenario of the
learning situation but following different types of
pedagogical design schemes (PDS).
We noted that trainers were able to handle the
prototype of the editing tool in only thirty minutes
despite a lack of user friendliness of the tool.
D Discussion
Communities of practice in teaching are usually
associated to a given group of practicing teachers
working with a given set of learners identified by an
academic level, by same learning objectives or by
skills to acquire, in the same pedagogical context,
including teaching strategies: within these
communities, the same pedagogical method is
generally adopted, emerged by a consensus or
recommended by a hierarchy. But the way and the
means used for preparing a course, what we call a
Pedagogical Design Scheme, could vary from a
teacher to another and, even, could depend from
many variables. We assume that practicing teachers
have many pedagogical design processes in mind.
The instantiation of these schemes on a given course
design depend all together from the learning
situation, the context of design and the profile of the
We have underpinned this hypothesis by the help
of experimentation with the association PARTAGE.
It was particularly interesting to work with the
trainers employed by this association because they
share in their community a very well described and
tooled pedagogical approach, well-suited both for
their learners characteristics and for the learning
domain (concrete basic skills and procedures). This
context allows us to presume on the effectiveness of
teaching community of practice.
Nevertheless, the part of the experimentation
related supra has proven that, confronted to the same
deign task of a given learning scenario, two trainers
of the teaching community have instantiated two
different pedagogical design schemes, objectives
driven for one and skills driven for the other.
Thus, when one wants to define methods and
tools to instrument the design process of a learning
scenario in a teaching community, he should support
in fact more than one design process, each of them
relevant from a given pedagogical design scheme.
Our engineering approach is based on patterns as a
way to express teachers’ pedagogical needs. Many
experiments have proven the relevance of patterns in
the learning scenario design activity. Our experiment
allowed us to design patterns and collect
pedagogical design schemes emerged from a
community of trainers. Within this community, the
trainers were being able to express a scenario with
the editing tool we proposed and scenarios resulting
were self-expressive.
We proposed an engineering framework process
and an editing tool to manage this design activity,
able to support different pedagogical design
We experienced the capacity of patterns to
represent pedagogical situation and capitalize
schemes of a practice of design through pedagogical
Further research work will be dedicated to
providing support to the designer with proposal of
adaptation during the design activity. This
adaptation of the pedagogical scenario design
activity should take into account the design context
and different pedagogical design schemes. We are
developing a context-awareness and user-awareness
editing tool adaptable to user’s pedagogical design
schemes and able to let him regulate the tool
Alexander, C. Ishikawa, S. Silverstein, M., 1977 A pattern
language, town, buildings, constructions, Oxford
University Press.
Amber, S. W. July 1998, Process pattern – building large-
scale systems using object technology, Cambridge
University Press, ISBN 0-521-645668-9.
Bergin, J., 2000, Fourteen pedagogical patterns, in
European Conference of Pattern Languages of
Bonnardel N., Activités de conception et créativité : de
l’analyse des facteurs cognitifs à l’assistance aux
activités de conception créatives , Presses
Universitaires de France ,2009.
Clayer, J. P. Toffolon, C. Choquet, C, A pattern-based and
teacher-centered approach for learning design,
Computers and Advanced Technology in
De Moura Filho C. O., Conceiving and Implementing a
language-oriented approach for the design of
automated learning scenarios, thesis, 2007.
De Moura, C. O., MDEduc: conceiving and implementing
a language-oriented approach for the design of
automated learning scenarios, ACM SIGWEB
Newsletter, 2008.
Delozanne E., Le Calvez F., Merceron A. , Labat J-M,
Design Patterns en EIAH : vers un langage de Patterns
pour l'évaluation des apprenants. , STICEF Vol 14,
2007, p. 1-19.
DPULS, 2005, DPULS Design Patterns Browser, Action
du réseau d’Excellence Européen Kaleidoscope.
El Kechaï H, Choquet C . Understanding the Collective
Design Process by Analyzing Intermediary Objects.
The 6th IEEE International Conference on Advanced
Learning Technologies (ICALT'2006), isbn : 0-7695-
2632-2, Kerkrade (The Netherlands), p.1047-1051,
July 5-7, 2006.
E-LEN : European project – Minerva programs:
Emin, V., Pernin, J.-P., and Guéraud, V. Times of
Convergence. Technologies Across Learning
Contexts, Lecture Notes In Computer Science, chapter
ISiS: An Intention-Oriented Model to Help Teachers,
2008, Learning Scenarios Design. vol. 5192, 338-343z
Emin,V., Pernin J. P., Aguirre J. L., ScenEdit: an
intention-oriented authoring environnment to design
learning scenarios, TEL: From Innovation to Learning
and Practice, Volume 6383, 2010, pp 626-631, 2010 .
Fowler, M., 1997, Analysis Patterns, Addison-Wesley,
Reading MA.
Gamma, E, Helm R., Johnson R. and Vlissides J., 1995
Design Patterns: Elements of Reusable Object-
Oriented Software. Addison-Wesley. ISBN 0-201-
Goodyear, P., 2005. Educational design and networked
learning: Patterns, pattern language and design
practice. Australasian Journal of Education
Technology, 21(1), 82–101.
Hernandez-Leo D., Jorrın-Abellan I. M. , Villasclaras-
Fernandez E. D., Asensio-Perez J.I., Dimitriadis Y., A
multicase study for the evaluation of a pattern-based
visual design process for collaborative learning,
Journal of Visual Languages and Computing 21
(2010), p,313-331.
Hernández-Leo, D, Villasclaras-Fernández, E. D.,
Asensio-Pérez, J. I, Dimitriadis, Y., Jorrín-Abellán, I.
M., Ruiz-Requies, I., & Rubia-Avi, B., 2006.
COLLAGE: A collaborative Learning Design editor
based on patterns. Educational Technology & Society,
9 (1), 58-71.
IMS-LD, 2003. IMS Learning Design, learningdesign/index.htm.
Kelly S. Domain Specific Languages vs. Generic
Modelling Languages, interview may 2007 :
Koper, R. ,Olivier B., Representing the Learning Design
of Units of Learning, Educational Technology &
Society, 7 (3), 2004, pp 97-111.
Laurillard D, Teaching as a Design Science, Building
Pedagogical Patterns for Learning and Technology,
Routledge edition, 2012, ISBN 978-0-415-8038.
Mor, Y. 2010 Embedding Design Patterns in a
Methodology for a Design Science of e-Learning, in
Christian Kohls & Joachim Wedekind, ed., 'Problems
Investigations of E-Learning Patterns: Context Factors
Solutions', Information Science Publishing, Hershey,
Mor, Y. and Winters, N., 2007, Design approaches in
technology enhanced learning , Journal of Interactive
Media in Education.
Paquette G., Instructional engineering for learning objects
repositories networks, 2nd International Conference
on Computer Aided Learning in Engineering
Education, 2004, pp 25-36, Grenoble, France.
Pedagogical Patterns Collector:
Pedagogical Patterns Project: retrieved September 7, 2011
Rohse, S. Anderson, T, 2006, Design Patterns for
Complex learning, Journal of Learning Design 2006.
Villiot-Leclercq, E. Modèle de soutien à l'élaboration et à
la réutilisation de scénarios pédagogiques, thesis in
computer science, 2007.