Download Model-Centered Learning and Instruction

yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Learning disability wikipedia , lookup

Albert Bandura wikipedia , lookup

Inquiry-based learning wikipedia , lookup

Learning wikipedia , lookup

Educational psychology wikipedia , lookup

Project-based learning wikipedia , lookup

Cooperative learning wikipedia , lookup

Learning styles wikipedia , lookup

Learning theory (education) wikipedia , lookup

Differentiated instruction wikipedia , lookup

Constructivist teaching methods wikipedia , lookup

Concept learning wikipedia , lookup

Learning through play wikipedia , lookup

Tech., Inst., Cognition and Learning, Vol. 1, pp. 59-85
Reprints available directly from the publisher
Photocopying permitted by license only
© 2003 Old City Publishing, Inc.
Published by license under the OCP Science imprint,
a member of the Old City Publishing Group
Model-Centered Learning and Instruction
University of Freiburg, Germany
Department of Educational Sciences, Learning Research and Instructional Design
This article focuses on the paradigm of model-centered learning and
instruction, which is based on the theory of mental models. According to
this paradigm, cognition and learning are the result of mental representations. Individuals form mental representations by organizing symbols
of their experience or thought in such a way that they effect a systematic representation of this experience or thought, as a means of understanding it, or explaining it to others. In this article, the focus is on the
psychological and epistemological aspects of model-centered learning
and instruction. Accordingly, the psychological foundations of modelcentered learning as well as the semantic functions of mental models are
described first. Following this, the focus is on types of instruction that
help learners to construct mental models. A distinction is drawn between
three paradigms of model-centered instruction: (a) Self-organized discovery and exploratory learning, (b) externally guided discovery learning, and (c) receptive learning oriented toward an expert’s behavior or a
teacher’s explanation. Then, the state-of-the-art of two scenarios of
model-centered instruction is summarized: providing students with
model-information and the approach of design-based modeling as
employed in discovery learning. Finally, the results of this research are
discussed with regard to learning and instruction.
Keywords: Mental models, schemata, discovery learning, learning by design,
model-centered instruction.
Model-centered learning and instruction (MCL&I) is both a new and old
paradigm of psychology and education. It concerns a complex phenomenon
which occurs not only in classrooms but also in a variety of informal learn____________________________
*e-mail: [email protected]
ing environments such as museums, zoos, gardens, and activity centres. It
also occurs on many levels (individuals, groups of various size and composition, in diverse cultures). Some decades ago, Bandura (1971) developed
the paradigm of social model-learning, according to which behavioral
changes can occur even when a person is not engaged directly in a learning
process but rather observing the behavior of another person. A learner A who
has a particular repertoire of behavioral traits (a, b, c, ...) at time t1 observes
a model-person B who expresses among others the behavioral traits x, y, z.
In examining the behavioral repertoire of A at a time point t2 one can often
observe that A has adopted the behavioral traits x, y, z of B. Similarly, in several instructional conceptions, such as the Cognitive Apprenticeship
approach of Collins, Brown and Newman (1989), the starting point for cognitive learning is situated in the observation of an expert’s problem-solving
behavior (see also Gibbons, 2002). Evidently, these conceptions have a
strong psychological background. However, an analysis of the literature
indicates that the idea of model-centered learning and instruction also has its
origins in the epistemology of neopragmatics (see, for example, Stachowiak,
1973; Wartofsky, 1979), with the focus on both the model-based construction of theories and the meaningful use of models in different fields of
instruction. In accordance with the model-theory of semantics, Chapanis
(1961) classified models into two broad categories: Reproduction models
(i.e. material-semantic models) such as physical objects, diagrams that serve
the purpose of communication (e.g. in the context of instruction), and symbolic models (i.e. formal-semantic models) that serve the purpose of mental
representation of knowledge. Actually, the central epistemological assumption is that “the person constructs the reality” (Buss, 1979) but not vice
versa. Cognition and learning take place in the use of mental representations
in which individuals organize symbols of their experience or thought in such
a way that they effect a systematic representation of this experience or
thought, as a means of understanding it, or of explaining it to others (cf.
Seel, 1991). Interestingly, we can find a similar argumentation in various
conceptions of model-centered learning which focus on guided discovery
and exploratory learning. Such conceptions have been developed in the
fields of mathematics, physics, and geography education (cf. Hodgson,
1995; Lesh & Doerr, 2000; Penner, 2001).
In all these cases talking about “models” implies asking for the original
to be modelled. Globes are models of the earth. Naturally, a globe is not a
reduced earth but rather it is supposed to give answers to questions about the
locations of different places or distances between places. With regard to the
chemical composition of the earth, a globe is not relevant. This example
illustrates the following: Every model is constructed in accordance with specific intentions in order to simplify its original in several respects. As an idealized reduction to relevant characteristics of its original a model is a concrete, comprehensible, and feasible representation of non-obvious or
abstract objects or phenomena. The representation of the objects’ attributes
and components comes second to the representation of structural relationships, and the functions of a model are defined on the basis of the explicit
intentions of the model-building person. Therefore, in physics and similarly
in other disciplines the term model is always used intentionally to refer to
the following functions:
• Models aid in the simplification of an investigation to particular and relevant phenomena in a closed domain.
• Models support the envisioning (visualization) of complex phenomena.
• Models aid in the construction of analogies in order to identify relationships within an unknown domain (e.g. quantum mechanisms) with
the help of the relationships within a known domain (e.g. Rutherford’s
atomic model). Such models are called “analogy models” and are
heuristic hypotheses about structural similarities of different domains.
• Finally, models aid in the simulation of a system’s processes. This
occurs when an individual interacts with the objects involved in a situation in order to manipulate them mentally in such a way that the cognitive operations simulate specific transformations of these objects that
may occur in real-life situations. Such simulation models operate as
thought experiments, producing qualitative inferences with respect to
the situation to be mastered.
Moreover, an analysis of the literature shows that there are numerous
fields of interest in which the theoretical term “model” plays a significant
role: In the context of scientific theory construction, in the theory of measurement, in cognitive psychology with its emphasis on mental models and
constructing artifacts, and in different subject matter domains, such as mathematics and physics where models serve to explain phenomena of the physical world. The various field of model-centered research may be summarized
as in Figure 1:
FIGURE 1 Fields of model-centered research
In the following sections I will focus on the psychological and epistemological aspects of model-centered learning and instruction. Here, theoretical
concepts such as schemata, mental models and external representations play
a central role for the understanding of cognition and learning.
Evidently, model-centered as well as schema-based learning are theoretical concepts of cognitive psychology, a discipline emerged in the second
half of the 20th century. This shift from the normative paradigm of behaviorism to the interpretative paradigm of symbolic interactionism (called the
“cognitive revolution” for short) occurred in the 1960s and “focused upon
the symbolic activities that human beings employed in constructing and in
making sense not only of the world, but of themselves.” (Bruner, 1990, p. 2).
Thus, cognitive psychology is concerned with the construction of symbolic
models of information processing understood generally as the capacity of
the human mind to construct knowledge, to interpret the world and to reason in deductive or inductive manner (cf. Seel, 2000). Cognitive learning
occurs when people actively construct meaningful representations, such as
coherent mental models, that represent and communicate subjective experiences, ideas, thoughts, and feelings (cf. Mayer et al., 1999). By means of
mental representations an individual is capable of simulating real actions in
the imagination. According to Piaget the fundamental basis for the development of mental representations is based on the development and gradual
refinement of assimilative schemata.
Schemata and mental models
The term schema is one of the central concepts of modern cognitive psychology and is common to the work of Kant, Bartlett, Piaget, Abelson,
Schank, Norman, Rumelhart and others. For some cognitive theorists
(Neisser, 1979; Rumelhart, 1980), schemata are the basic building blocks of
the psychological understanding of cognition. Mandl, Friedrich, and Hron
(1988) have defined them as cognitive structures in which general knowledge is represented in memory. More specifically, in the terms of Piaget
(1943), a schema of an action (i.e. “enactive schema”) is defined as the
structured totality of the generalizable characteristcs of this action, i.e. of
those characteristics which allow one to repeat an action and apply it to new
contents. Therefore, in modern cognitive psychology schemata (or frames
and scripts) are understood as generic data structures which play a central
role in the interpretation of perceptions, the regulation of behavior, and the
storage of knowledge in memory. However, according to Rumelhart et al.
(1986) there is no representational object which is a schema. Rather,
schemata emerge at the moment they are needed from the interaction
between many simpler elements working in concert with one another in
order to interpret the given environment as well as assimilate new information. Actually, in Piaget’s theory of equilibration a schema mainly serves to
assimilate new information into existing knowledge structures. The fundamental mechanism in this process is pattern matching, which enables individuals to quickly “settle” on an interpretation of an input pattern.
Beyond this, schemata also form the basis of the construction of models,
which can be understood in accordance with Piaget as “tools” of accommodation (cf. Seel, 1991). More specifically, Rumelhart et al. divide the cognitive system into two modules (or sets of units). One part—called an inter-
pretation network—is concerned with the activation of schemata, the other
one is concerned with constructing a “model of the world.” This model takes
as input some specification of the actions we intend to carry out and produces an interpretation of “what would happen if we did that.” Part of this
process could be a specification of what the new stimulus conditions would
be like (appearance modeling). The interpretation network (i.e. a schema)
takes input from the world (to be explained) and produces relevant reactions
whereas the second module, i.e. the constructed “model,” predicts how the
input would change in response. In the literature it is common to talk about
a mental model that one would expect to be operating in any case, in-asmuch as it is generating expectations about the state of the world and thereby “predicting” internally the outcomes of possible actions. However, it is
not necessary for world events to have happened. In the case that they have
not, the cognitive system replaces the stimulus from the world with input
from the mental model. That means that a “mental simulation runs” envisioning the events in imagination that would take place in the world if a particular action were to be performed. Thus, mental models one allow to perform actions entirely internally and to judge the consequences of actions,
interpret them, and draw conclusions based on them. In other words, the
cognitive system can build an internal control system based on the interaction between the two modules of the cognitive system.
This theoretical conception of Rumelhart et al. (1986) corresponds to a
great extent with the theory of mental models that emerged in the 1980s to
capture deductive (and inductive) reasoning by “settling” computations of
the human mind in a solution rather than to applying logical operations (cf.
Johnson-Laird, 1983, 1994; Hinnersmann, 1989). Moreover, it has been
argued that comprehension and reasoning in specific situations (e.g., in
schools and real-life situations) necessarily involve the use of mental models of different qualities (cf. Greeno, 1989). Accordingly, the theory of mental models has been applied successfully in several domains such as HumanComputer-Interaction (Ackerman & Tauber, 1990), text and discourse processing (Rickheit & Habel, 1999), operating with complex (technical) systems (Kluwe & Haider, 1990), and so on.
Mental models play a central and unifying role in representing
objects, states of affairs, sequences of events, the way the world is,
and the social and psychological actions of daily life. They enable
individuals to make inferences and predictions, to understand phenomena, to decide what action to take and to control its execution,
and, above all, to experience events by proxy. (Johnson-Laird,
1983, p. 397)
Historically seen, Craik (1943) introduced the concept of “internal models” into psychology with the notion of a working model. Craik argued that,
in contrast to the assumptions of behaviorism, people experience reality only
mediated by mental constructions such as internal models. As in the theory
of mental models, this idea is based on the assumption that an individual
who intends to give a rational explanation for something must develop practicable methods in order to generate appropriate explanations on the basis of
principally restricted domain-specific knowledge and a limited information
processing capacity. Accordingly, the individual constructs a model that both
integrates the relevant domain-specific knowledge and meets the requirements of the phenomenon to be explained—then the model works. Thus,
mental models are intentional constructions of the mind which generate
access to the observable or imagined world and make it plausible. Mental
models reflect the structure of the world because they are generated to structure it, and not because they reproduce or copy a given external structure.
They represent and organize the subject’s knowledge in such a way that even
complex phenomena become plausible.
Models and external representations
As Rumelhart et al. (1986) have pointed out, people are not only good at
pattern matching (by means of schema activiation) but also at modelling
their world and manipulating their environment. The ability of modelling to
create expectations by “internalizing” experiences is crucial to learning as
well as the skill of manipulating the environment allow us to reduce very
complex problems to simpler ones. Interestingly from an instructional point
of view, the suggestion has been made (cf. Norman, 1983) that mental models are constructed from the significant properties of external situations,
such as designed learning environments and the subject’s interactions with
them. According to this view, the external environment becomes a key
extension of our mind and its processing is real symbol processing and the
primary symbol processing that we are able to do. Accordingly, imagination
is highly dependent on our experience with external representations which,
therefore, play a crucial role in thought and its communication. This idea
that we learn and think on the basis of mental models which result from the
internalization of external representations is especially relevant for instruction insofar as the construction of external representations provide us with
the means to easily influence the learners’ construction of mental models
intentional. I will refer to this point later in this article. Here, it should be
emphasized that modeling basically involves the interactions between three
types of systems:
• internal conceptual systems of a model building person,
• representational systems that function both as externalizations of the
internal conceptual systems and as internalizations of external systems,
• external systems that are experienced in nature, or that are artifacts that
were constructed by humans (cf. Lesh, 1998; Rumelhart et al., 1986).
Accordingly, Kluwe and Haider (1990) have distinguished between the
following basic types of semantic models that are of interest in this article:
• Semantic models have their field of reference in a more or less complex
system S of the physical world.
• In order to explain a particular system S individuals develop distinctive
internal representations that are called mental models of S, MM (S), and
that represent the individuals’ distributed knowledge with regard to S.
• On the basis of their mental models, scientists develop conceptual models CM (S) which represent the shared knowledge of a discipline with
regard to the system to be explained.
• Cognitive psychologists, for their part, develop psychological models of
model building activities: PM [MM (S)]. Actually, this chapter is concerned to a large extent with the description of such psychological models of model building activities.
• Finally, so-called design and instructional models DIM [CM (S)] are of
special interest from an instructional point of view. They are instruction-oriented descriptions of the conceptual models of the system S that
are used for the construction of interfaces (e.g., learning tasks, manuals
and trainings) in order to guide the learners’ construction of mental
The interplay between these different kinds of semantic models and the
world to be modelled may be illustrated as in Figure 2.
Kluwe and Haider (1990) have pointed out some important relationships
between these kinds of models. For example, there is a strong relationship
between mental models and conceptual models insofar as the shared knowl-
FIGURE 2 Varieties of semantic models
edge of a discipline—expressed through conceptual models—presupposes
the distributed knowledge of the individuals who constitute a scientific community of practice. I argue that conceptual models evolve from the learningdependent progression of the subjective mental models which scientists construct (cf. Norman, 1983). A second important relationship exists between
MM (S) and the field of cognitive psychology in which explanatory models
are constructed with regard to model building activities of learners. A third
kind of relationship can be identified between the design and instructional
models. We argue that these models are related to all other types of models.
However, in the case of subject matter oriented learning there is a clear preference for teaching conceptual models in order to improve this kind of learning. Finally, there is a mutual relationship between DIM [CM (S)] and PM
[MM (S)].
Kluwe and Haider (1990) have pointed out that in the literature the distinction between these different types of models is often not clear insofar as
several psychological approaches (e.g. di Sessa, 1986; Williams et al., 1983;
Young, 1983) are actually concerned with the notion of design and instruc-
FIGURE 3 Relationships between different types of semantic models.
tional models. This leads to the next section of this article, concerning the
instructional implications of model-building activities of learners.
As I have mentioned in the introduction of this article, the question how
we can influence model-building activities of learners has been at the core
of various educational approaches for a long time. In addition, in the field of
research on mental models we can find a strong pedagogical impetus from
the very beginning (cf. Anzai & Yokoyama, 1984; Mayer, 1989; Norman,
1983; Seel, 1986). Johnson-Laird (1989) distinguished between three different sources for constructing mental models:
• The learner’s ability to construct models in an inductive manner, either
from a set of basic components of world knowledge or from analogous
models that the learner already possesses;
• everyday observations of the outside world, and
• other people’s explanations.
This differentiation corresponds, to a large extent, with several paradigms
of model-centered instruction:
(a) Self-organized discovery and exploratory learning,
(b) externally guided discovery learning (for example in the form of
“learning by design,” cf. Kolodner et al.,, and
(c) receptive learning oriented toward an expert’s behavior or a teacher’s
According to Carlson (1991) instruction can be designed to involve the
learner in an inquiry process in which facts are gathered from data sources,
similarities and differences among facts noted, and concepts developed. In
this process, the instructional program serves as a facilitator of learning for
students who are working to develop their own answers to questions. On the
other hand, instructional programs can present concepts with clear definition
followed by clear examples. A designed conceptual model may be presented ahead of the learning tasks in order to direct the learner’s comprehension
of the learning material. Clearly, there might exist environments that can initiate a form of learning based on free exploration by invention, but in
instructional contexts (especially in schools) we regularly operate with wellprepared and designed learning environments that constrain the student’s
learning processes to various extents. Johnson-Laird (1989) has expressed
the central point of instructional guidance for model-centered learning as
Although mental models may differ markedly in their content,
there is no evidence to suggest that they differ in representational
format or in the processes that construct and manipulate them.
What is at issue is how such models develop as an individual progresses from novice to expert, and whether there is any pedagogical advantage in providing people with models of tasks they are trying to learn. (Johnson-Laird, 1989, p.489)
This paradigm of model-centered instruction concerns the learningdependent progression of mental models, defined (in accordance with Snow,
1990) as a specific kind of transition between preconceptions (i.e. the initial
states of the learning process) and causal explanations (i.e. the desired end
states of learning). I have illustrated this paradigm of instructionally guided
model-centered learning as follows (cf. Seel, Al-Diban & Blumschein,
FIGURE 4 The mental model as created through an instructionally designed conceptual model.
In contrast with this paradigm, several authors (such as Johnson-Laird,
1989; Penner, Lehrer & Schauble, 1996; Stewart et al., 1992) have emphasized the role of self-organized discovery learning for the construction of
effective mental models. In accordance with this paradigm the learner has to
search continuously for information in the given learning environment in
order to complete or stabilize an initial mental model which corresponds to
an “a priori understanding” of the material to be learned. All current existing instructionally guided exploratory models for discovery learning were
designed with a specific endpoint in mind. The goal is to create microworlds
in which objects follow specific sets of rules. One example is a microworld
in which balls fall in accordance with Newton’s laws of motion (cf. White,
1993). Students explore this model by developing hypotheses and then varying input parameters to investigate how well their conjectures align with the
model. In math education the defining characteristic of this kind of
exploratory learning is that students explore conventional mathematical
symbolizations in experientially real settings. The instructional intent is usually well defined and typically involves the students’ development of the
mathematical understanding necessary for the mature use of symbolizations
(cf. the MathCars of Kaput, 1994). This author suggested that one might
facilitate students’ learning by supporting their efforts to explore the physi-
cal linkages between the mathematical symbolizations and their own experiences, and to generate new hypotheses. Moreover, he argued that “as the
model develops […] the mental model based in the mathematical representations comes to relate more directly to conceptualizations of the setting”
(Kaput, 1994, p. 390). However, with regard to settings of self-guided discovery learning, Doerr (1996) stated students themselves develop expressive
models to explain phenomena using a variety of (software) tools. Here, students invent models on their own that express their developing interpretations of the phenomena in question. Doerr determined that this model-building approach begins with students’ informal understandings and progressively builds on it. Self-guided learning occurs as a multi-step process of
model-building and revision (Penner, 2001). In terms of model theory, this
process can be conceived as a procedure called “fleshing out” (JohnsonLaird, 1983, p 452) that continuously examines whether a model can be
replaced with an alternative model or not. I have defined this as a Reductio
ad absurdum that is at the core of mental modeling in general (Seel, 1991).
However, self-guided discovery learning is very ambitious insofar as the
learners must have previously achieved adequate problem-solving and
metacognitive skills to guide their learning process. Therefore, for novice
students it can be argued that self-organized discovery learning is closely
associated with learning by trial-and-error but not by insight. Incidentally,
Briggs (1990) demonstrated in a case-study that an instructional strategy
aiming at discovery learning may dramatically increase the probability that
faulty initial mental models may be stabilized. Consequently, a substantial
conceptual change does not take place, and relatively stable intermediate
states of causal understanding often precede the instructionally intended
conceptual mastery (Galili, Bendall & Goldberg, 1993). In sum, as a matter
of fact self-organized learning aiming at the creation of mental models can
be rather pretentious. It is a process which even an expert might sweat over
Regardless of the choice of one of the aforementioned paradigms of
model-centered instruction, I agree with Gibbons (2002) on the fact that
instructional designers have to create effective learning environments and
materials as well as a variety of time and space structures to influence
model-centered learning. However, “the events of instruction, which are the
structures we design, serve human learning processes under the ultimate
control of the individual. Instruction, therefore, does not cause learning but
supports learning intentions to which the learner commits. [...] Some of
these processes (such as the initial processing of visual or auditory informa-
tion) are involuntary, but many of them (focusing attention, finding and
selecting associations, etc.) are completely voluntary.” (Gibbons, 2002, p.3)
If we intend to initiate model-centered learning we have to take into consideration the preconceptions with which learners enter into a given learning
environment and their motivation to engage in learning. Furthermore, we also
have to take into consideration the quality of the learning material and the
information which must be understandable, coherent, plausible etc. in order
to persuade the students to engage in the learning environment. Actually, the
learners’ preconceptions and motivational states as well as the quality of the
given learning materials strongly influence the patterns of participation and
persuasion (cf. Dole & Sinatra, 1998), understood here as central factors of
effective model-centered instruction. By combining these factors with the
theoretical framework of model-based learning and instruction developed by
Buckley and Boulter (1999), one may describe the interplay between modelcentered learning and instruction can be described as in Figure 5.
The top of the diagram is concerned with model-based teaching. It focuses on the patterns of participation, persuasion and model-building which
individuals follow in the classroom to construct their understanding of some
phenomenon. This is accomplished mainly through discourse with and about
external representations, provided and guided by the teacher who facilitates
negotiation among the discourse participants—including those not present
such as the scientists who developed the conceptual models in the domain
and the instructional designers who developed the materials and activities
which are to facilitate of the learners’ understanding of the phenomenon.
The bottom half of the diagram is concerned with model-based learning.
It focuses on individuals’ construction of mental models of the phenomena
under study. During the course of this process learners form an initial model
of the phenomenon, either intententionally to meet some learning goal or
spontaneously in response to some task. When the model is used successfully, it is reinforced and may eventually become a precompiled, stable
model. If it turns out that the model is unsatisfactory, it may be revised or
rejected in a progression of mental models.
In the following section I will describe the state-of-the-art of related research
on model-centered learning in the context of this theoretical framework.
In accordance with the foregoing argumentation I distinguish between
FIGURE 5 The interplay between model-based learning and instruction
two basic settings or scenarios to influence the students’ construction of
mental models through instructional intervention programs:
• Scenario 1: Providing students with relevant information (e.g. presentation of a conceptual model) aiming at receptive model-centered learn-
ing and
• Scenario 2: Design-based modeling as employed in discovery learning.
Research on scenario 1: Providing students with model-information
When we consider the history of mental model research in the field of
instruction we can observe a clear predominance of studies aiming at modeloriented learning explicitly directed by a teacher or an instructional program. Mayer (1989, p. 47) has expressed the pedagogical idea of this
approach, commenting that “students given model-instruction may be more
likely to build mental models of the systems they are studying and to use
these models to generate creative solutions to transfer problems.”
Accordingly, the learners are provided with a conceptual model (i.e. an
external representation) constructed in accordance with instructional principles to illustrate the main components and relationships of a complex system with the help of graphical diagrams or object-based replica. Actually,
many researchers are convinced of the effects of graphical diagrams or even
“helpful video” on the construction of mental models, insofar as they consider these instructional media to be effective with regard to the creation of
dynamic images that constitute the frame of reference for the construction of
mental models (see, for example, Hegarty & Just, 1993; Schnotz & Kulhavy,
1994; Sharp et al., 1995). However, this presupposes that the learner is sensitive to characteristics of the learning environment (Anzai & Yokoyama,
1984), such as the availability of certain information at a given time, the way
the information is structured and mediated, and the ease with which it can
be found in the environment (cf. Seel, 2000).
It has been argued that it is often easier for a beginning learner to assimilate an explanation (provided through a conceptual model) rather than to
induce one individually. In this case, the conceptual model provided will be
functionally incorporated in the thinking process, and related information
can be progressively integrated in a more or less consistent manner to
achieve substantial conceptual changes. Meanwhile, there are numerous
studies which have demonstrated that the presentation of a conceptual model
actually affects the construction of a task-related mental model depending
on the stage in the learning process at which the conceptual model is presented. From my point of view, the seminal work of Mayer (1989) represents
a milestone in this research. Mayer indicates that the presentation of modelrelevant information at the beginning of the learning process seems to
increase both the quality of comprehension during the learning process and
the quality of causal explanations at the end of the learning process.
According to these findings, we presupposed in our own research that a conceptual model provided at the beginning of a lesson facilitates the construction of an adequate mental model for cognitively mastering the demands of
the learning situation (cf. Seel, 1995; Seel & Dinter, 1995).
However, the research on mental models in the 1980s has been criticized
by several authors (such as Royer, Cisero & Carlo, 1993; Snow 1990)
because it has been typically performed piece-meal, in small scale, specialized contexts. In order to overcome these shortcomings of experimental
research, my research group looked for a more comprehensive instructional
approach as a fundamental basis for initiating and directing model-oriented
learning. Accordingly, our research in the past six years has centered around
three main topics:
(1) The investigation of the learning-dependent progression of mental
models (more specifically of analogy models of dynamic systems of
economics) in the course of a comprehensive instructional program;
(2) how this progression can be guided by way of a particular instructional intervention, designed as a multimedia environment in accordance with principles of the Cognitive Apprenticeship approach, and
(3) how to assess mental models and their learning-dependent progression as influenced by model-centered instruction.
Actually, when we started in 1994 with the development of a multimedia
environment aiming at an externally guided goal-oriented, and systematic
influence upon the learners’ progression of mental models the Cognitive
Apprenticeship approach of Collins et al. (1989) proved to be the only promising instructional strategy which corresponded with the idea of providing
the students with model-instruction in the aforementioned sense. This immediately becomes clear when we consider the instructional methods of
Cognitive Apprenticeship: Modeling, coaching, scaffolding, articulation,
reflection, and exploration. Evidently, the instructional intervention of
apprenticeship starts with the presentation of an expert’s conceptual model
of the tasks to be accomplished, and then the students are coached and scaffolded to adapt this model for their own solutions (exploration) of the
designed learning tasks.
In the course of the 4M-project1 more than 400 students (secondary
school, 17 years on average) accomplished the multimedia program
From 1994 until 2001 we gratefully acknowledge financial support for this research from a grant
provided by the German Research Assocation (Deutsche Forschungsgemeinschaft) with Grant-No.
Se 399/4.
Dynamic Systems of Economics, which was designed in accordance with the
principles of Cognitive Apprenticeship. This project and its results have
been described in several articles (e.g. Al-Diban & Seel, 1999; Seel et al.,
2000). In five replication studies we investigated:
• the effects of providing a conceptual model at the beginning of the
learning process on the construction of a corresponding mental model;
• the long-term effectiveness of the multimedia learning program on both
the acquired domain-specific knowledge and the stability of the initially constructed mental models.
Over the project’s life it was possible to realize and evaluate the methods
of the Cognitive Apprenticeship approach with the help of a step by step
formative evaluation strategy. Parallel with this, we investigated the learning-dependent progression of mental models with the help of a specific diagnosis, causal diagrams, that confirmed central assumptions of the theory of
mental models. In fact, causal models (for more details consult, Seel, 1999;
2001) proved to be suitable instruments for the assessment of a learner’s
subjective assumptions about the causality of a dynamic system as well as
for appropriate representations of the learner’s mental model of a dynamic
On the whole, the results of our evaluation studies allow the conclusion
that the Cognitive Apprenticeship approach can be considered an appropriate framework for the instructional design of environments aiming at modelcentered learning. So far the results correspond with observations and
empirical results of other studies, such as those of Casey (1996), Chee
(1995), and Lajoie and Lesgold (1989). This suggests that Cognitive
Apprenticeship principles are suitable for the instructional design of learning environments in spite of the following reservations:
• It proved to be very difficult to realize the methods of articulation and
reflection in multimedia environments, and
• the methods of scaffolding and exploration, which focus on the
improvement of self-regulated learning, proved to be as less effective
than expected.
Numerous studies in the research field of conceptual change (cf. Dole &
Sinatra, 1998) have indicated that students dynamically modify and restruc-
ture their knowledge bases when externally provided information is judged
to be more plausible and convincing. Accordingly, we especially investigated the permanence and stability of mental models generated after a student’s
early exposure to a designed conceptual model. Thereby, we focused on two
research questions:
• Do the learners adapt the conceptual model designed by an expert they
are provided with in the multimedia program and do they retain it in the
course of the learning process? Therefore, are the learners’ causal diagrams similar to the conceptual models they are provided with?
• Are there significant similarities between the causal diagrams constructed by the learners at several measurement points in the course of
the multimedia program? In other words: Do the learners retain the
causal diagrams they produced initially (defined here in terms of external representations of mental models) or do they vary depending on the
changing demands of the learning tasks?
Surprisingly, our investigations indicated that the subjects exhibited only
a minor tendency to adopt the conceptual models they were provided with
in the multimedia-program. Incidentally, they typically drew causal diagrams which indicated varying degrees of similarity to the provided conceptual models. Therefore, a conclusion from our research with far-reaching
consequences for instruction is that mental models are not fixed structures
of the mind, but are rather constructed when needed to master a learning situation with its specific demands. This contradicts the assumption that learners tend to adopt provided conceptual models when no relevant preconceptions are available in order to construct idiosyncratic mental models of the
tasks they are trying to learn. Consequently, we cannot suppose a factual
persistence of a mental model over time. Rather, it is more plausible to
assume a situation-dependent reconstruction of previously generated mental
As a consequence of the results of our investigations it could be concluded
(1) that the learners constructed situation-bound problem solutions independently on according to the specific instructional strategies they
were provided with (such as subsumption of learning tasks under a
common schema vs. induction of a schema from analogous task
examples); and
(2) that an effective design of successful learning environments presupposes the provision of cognitive tools which facilitate and support an
individual model-building and revision for problem solving.
Research on scenario 2: Design-based modeling for
exploratory learning
For this scenario, the research on model-centered learning is not strongly embedded in the theory of mental models but rather integrated in the field
of instructional investigations with an emphasis on the curriculum of different subjects, such as mathematics and science. Stewart et al. (1992) have
delineated the central idea of these instructional approaches, commenting
that “a science education should do more than instruct students with respect
to the conclusions reached by scientists; it should also encourage students to
develop insights about science as an intellectual activity.” (p. 318; see also
Hestenes, 1987)
Accordingly, advocates of this approach argue that “given that we wish
to involve students in the practices of scientists, we focus primarily on
model building.” (Penner et al., 1998, p. 430) Indeed in science, some of the
most important goals of instruction are to help students develop powerful
models so that they can make sense of their experiences involving light,
gravity, electricity, and magnetism; it has also become evident that young
students invent models of their own and that changing students’ ways of
thinking must involve challenging and testing these models. The modelbuilding approach provides a significant challenge in understanding how to
nurture, accommodate, and respond to the partial and incomplete models
that students are likely to build with regard to phenomena of physics (see,
for example, Clement, 1989; Vosniadou & Brewer, 1992, 1994).
Interestingly, we can find a similar argumentation with regard to the
learning of mathematics insofar as several authors, such as Doerr (1996),
Gravemeijer (1999), Hodgson (1995), Lesh and Doerr (2000) and others,
argue that mathematizing (e.g. quantifying, visualizing, or coordinating) is a
particular form of modeling. It generally involves the use of specialized languages, symbols, graphs, pictures, concrete materials, and other notation
systems for the development of mathematical descriptions and explanations.
This obviously constitutes a heavy demand on learners’ representational
capabilities. “The primary role of algebra at the school level is to develop
confidence and facility in using variables and functions to model numerical
patterns and quantitative relationships.” (National Council of Teachers of
Mathematics, 1994) Accordingly, Lesh and Doerr (2000) as well as other
authors discuss models that students should develop in attempts to produce
mathematical descriptions or explanations of systems of the physical world.
These authors argue that helping students to develop powerful models
should be among the most important goals of science and mathematics
In science and mathematics, where there is an emphasis on symbolic representations, it is often assumed that students do not (and cannot) develop
appropriate symbol systems to make sense of systems in the external world
that involve mathematical entities such as directed quantities (negatives),
multivalued quantities (vectors), ratios of quantities, changing or accumulating quantities, or locations in space (coordinates). However, when confronted with the need to create meaningful models of experientially real situations, students can invent significant mathematical solutions on the basis
of model constructions. Such “modeling is seldom a one-shot process;
rather, model testing and evaluation are most often followed by model revision.” (Penner et al., 1998, p. 433). An analysis of the literature shows very
good examples of this approach but unfortunately there is a substantial lack
of empirical research including quantitative data which could be interpreted
in accordance with the theoretical assumptions in the literature. This also
holds true with regard to the currently prospering field of the Learning-byDesign approach, which focuses on the sampling of qualitative data (cf.
Kafai & Ching, 2001).
In view of this research situation we have recently started a long-term
project which focuses on model-centered discovery learning (according to
the theory of mental models) as guided through properly designed multimedia learning environments involving a particular tool to teach students analogical reasoning and the production of analogy models.2 At the moment, the
research group is engaged in the construction and design of the learning
environment for which we have defined several design features, such as the
use of representational formats of multimedia technologies, context boundedness of the learning tasks, explications of the semantic depth structure,
scaffolding, and networking by way of analogies between different domains.
These design features are linked with several learning activities, such as the
active construction of structural knowledge (as in Scandura’s Structural
Learning Theory), the step by step development and refinement of a gener____________________________
2 Again we are subsidized by a grant fromf the German Research Association (Deutsche
Forschungsge meinschaft) with Grant-No. Se 399/8-1. The research group consists of Bettina
Couné, Pablo Dummer, Katharina Schenk, and Susanne Steiner.
al problem solving schema, cognitive experimenting (thought experiments),
and conceptual networking.
The first experimental study (in accordance with the idea of “enlarged
design experiments,” Dummer, 2002) is planned for December 2002. As in
the case of our former experiments in the aforementioned project, we hope
to find quantitative data that confirm our theoretical assumptions about
model building in the course of externally guided discovery learning.
Thereby, we follow Scandura’s verdict that any theory of teaching and learning must include some way of finding out what students know at any phase
of learning. And, “in addition, a fully adequate theory of teaching and learning must allow for the growth of knowledge over time as learners interact
dynamically with a changing [...] environment.” (Scandura, 1988, p. 53)
Model-based learning and instruction is a complex phenomenon which
occurs not only in classrooms, but also in a variety of informal learning environments. It also occurs on many levels (individuals, groups of various size
and composition, in diverse cultures). Nevertheless, the classroom is
undoubtedly a prominent place for the improvement of model-based learning in various subject matter domains. Those who advocate model building
and revision in science and mathematics consider modeling to be a specific
kind of problem solving, and thus, they argue that modeling in the classroom
would help to develop students’ problem-solving skills (cf. Hodgson, 1995).
Moreover, Tanner and Jones (1994) believe that model building activities in
the classroom improve students’ metacognitive competence. This is not the
place for a detailed commentary on these beliefs but it is noteworthy to mention that some enthusiasts have a tendency to misplace the emphasis of discovery learning in science and mathematics by not paying sufficient attention to what it is that is being discovered. Therefore, Lesh and Doerr (2000)
have pointed out that model building and revision in the classroom must correspond to the conceptual models as well as the main constructs of the scientific disciplines that underlie the curriculum in mathematics and science.
Nevertheless, the “big idea” of those who advocate model building and revision in the math and science classroom is to provide students “with the skills
they will need to accomplish this in the real world. This is the objective of
...modeling.” (Hodgson, 1995, p. 353) However, to do a good job we need a
comprehensive and empirically valid theory of instructional design of
model-centered learning in various instructional settings. To date, there have
only been few attempts to develop a theory of instructional design for
model-centered learning environments. One of them was developed by
Gravemeijer et al. (2000) in close connection with Freudenthal’s (1973)
comprehension of mathematics as an activity that involves solving problems, looking for problems, and organizing subject matter resulting from
prior mathematizations or from reality. An important heuristic method for
instructional design focuses on the role that emergent models play in individual students’ learning and in the collective mathematical development of
the classroom community. It is possible for the designer to lay out a proposed developmental route for the classroom community in which students
first model situations in an informal way (this is called a model of the situation) and then mathematize their informal modeling activity (this produces
a model for reasoning). The transitions from a model of to a model for are
generally consistent with Sfard’s (1991) historical analysis of the process of
reification. Sfard argues that the history of mathematics is characterized by
repeated processes of reification in which understandings that initially existed only in action were objectified, thus creating mathematical entities that
were experienced as independent of activity.
However, Lesh and Doerr (2000) have pointed out that models are interacting systems based in moe complex conceptual systems, and as such models cannot simply be handed to students in a meaningful form. This is the
basis for the claim that models (constructs) must be constructed. But it is
also the case that
• construction can lead to a great many things (such as complex systems
of low level facts and skills) that are not models for making sense of
experiences) and
• construction is not the only process that contributes to the development
of models. For example, models evolve by being sorted out, refined, or
reorganized at least as often as they evolve by being assembled (or constructed).
Thus, some enthusiasts of “discovery learning” tend to overemphasize
the process of discovery without paying much attention to what it is that is
being discovered. Whereas some authors, such as Kafai and Ching (2001),
focus on model-building activities of children acting as designers whereby
they are not confronted with the shared conceptual models of experts I suppose that if instruction and learning focus on models and modeling, then it
is absolutely critical to focus on models that correspond to the “big ideas” or
main constructs and conceptual systems that underlie the curriculum in
mathematics and science.
Ackermann, D. and Tauber, M.J. (Eds.) (1990). Mental Models andHuman-Computer
Interaction 1. Amsterdam: Elsevier.
Al-Diban, S. and Seel, N.M. (1999). Evaluation als Forschungsaufgabe von
Instruktionsdesign – Dargestellt am Beispiel eines multimedialen Lehrprogramms.
Unterrichtswissenschaft, 27 (1), 29-60.
Anzai, Y. and Yokoyama, T. (1984). Internal Models in Physics Problem Solving. Cognition
and Instruction, 1, 397-450.
Bandura, A. (1971). Social Learning Theory. New York: General Learning Press.
Briggs, P. (1990). The Role of the User Model in Learning as an Internally and Externally
Directed Activity. In D. Ackermann, and M.J. Tauber (Eds.), Mental Models and
Human-Computer Interaction 1 (pp. 195-208). Amsterdam: Elsevier Publ.
Bruner, J. (1990). Acts of Meaning. Cambridge, MA: Harvard University Press.
Buckley, B.C. and Boulter, C.J. (1999). Analysis of Representation in Model-based Teaching
and Learning in Science. In R. Paton & I. Neilsen (Eds.), Visual Representations and
Interpretations. (pp. 289-294). London: Springer.
Buss, A.R. (1979). A Dialectical Psychology. New York: Irvington.
Carlson, H.L. (1991). Learning Style and Program Design in Interactive Multimedia.
Educational Technology Research and Development, 39 (3), 41-48.
Casey, C. (1996). Incorporating Cognitive Apprenticeship in Multi-media. Educational
Technology Research and Development, 44 (1), 71-84.
Chapanis, A. (1961). Men, Machines, and Models. American Psychologist, 16, 113-131.
Chee, Y.S. (1995). Cognitive Spprenticeship and its Spplication to the Teaching of Smalltalk
in a Multimedia Interactive Learning Environment. Instructional Science, 23, 133-161.
Clement, J. (1989). Learning via Model Construction and Criticism. Protocol Evidence on
Sources of Creativity in Science. In J. Glover, R. Ronning and C. Reynolds (Eds.),
Handbook of Creativity: Assessment, Theory, and Research (pp. 341-381). New York:
Collins, A., Brown, J.S. and Newman, S.E. (1989). Cognitive Apprenticeship: Teaching the
Crafts of Reading, Writing, and Mathematics. In L.B. Resnick (Ed.) (1989), Knowing,
Learning, and Instruction (pp. 453-494). Hillsdale, NJ: Erlbaum.
Craik, K.J.W. (1943). The Nature of Explanation. Cambridge: Cambridge University Press.
diSessa, A. (1986). Models of Computation. In D.A. Norman & S.W. Draper (Eds.), User
Centered System Design (pp. 201-218). Hillsdale, NJ: Erlbaum.
Doerr, H.M. (1996). Integrating the Study of Trigonometry, Vectors, and Force through
Modeling. School Science and Mathematics, 96, 407-418.
Dole, J.A. and Sinatra, G.M. (1998). Reconceptualizing Change in the Cognitive Construction
of Knowledge. Educational Psychologist, 33 (2/3), 109-128.
Dummer, P.N. (2002). Probleme and Perspektiven bei der Evaluation von
Modellsimulationen. Freiburg i.Br.: Albert-Ludwigs-Universität, Institut für
Erziehungswissenschaft (unpubl. Master thesis)
Freudenthal, H. (1973). Mathematics as an Educational Task. Dordrecht: Reidel.
Galili, I., Bendall, S., & Goldberg, F. (1993). The Effects of Prior Knowledge and Instruction
on Understanding Image Formation. Journal of Research in Science Teaching, 30 (3),
Gibbons, A.S. (1998, 2002). Model-centered Instruction. Paper presented at the Annual
Meeting of the American Education Research Association, San Diego, CA, April 1998.
Gravemeijer, K. (1999). How Emergent Models may Foster the Constitution of Formal
Mathematics. Mathematical Thinking and Learning, 1, 155-177.
Gravemeijer, K., Cobb, P., Bowers, J. and Whitenack, J. (2000). Symbolizing, Modeling, and
Instructional design. In P. Cobb, E. Yackel & K. McClain (Eds.), Symbolizing and
Communicating in Mathematics Classrooms. Perspectives on Discourse, Tools, and
Instructional Design (pp. 225-273). Mahwah, NJ: Erlbaum.
Greeno, J.G. (1989). Situations, Mental Models, and Generative Knowledge. In D. Klahr &
K. Kotovsky (Eds.), Complex Information Processing (pp. 285-318). Hillsdale, NJ:
Hegarty, M., & Just, M.A. (1993). Constructing Mental Models of Machines from Text and
Diagrams. Journal of Memory and Language, 32, 717-742.
Hestenes, D. (1987). Toward a Modeling Theory of Physics Instruction. American Journal of
Physics, 55 (5), 440-454.
Hinnersmann, H. (1989). Theorie and Training deduktiven Urteilens. Frankfurt/M: Lang.
Hodgson, T. (1995). Secondary Mathematics Modeling: Issues and Challenges. School
Science and Mathematics, 95 (7), 351-358.
Johnson-Laird, P.N. (1983). Mental Models. Towards a Cognitive Science of Language,
Inference, and Consciousness. Cambridge: Cambridge University Press.
Johnson-Laird, P.N. (1989). Mental Models. In M.I. Posner (Ed.), Foundations of Cognitive
Science (pp. 469-499). Cambridge, MA: The MIT Press.
Johnson-Laird, P.N. (1994). Mental Models and Probabilistic Thinking. Cognition, 50, 189209.
Kafai, Y.B., & Ching, C.C. (2001). Affordances of Collaborative Software Design Planning
for Elementary Students’ Science Talk. The Journal of the Learning Sciences, 10 (3),
Kaput, J.J. (1994). The Representational Roles of Technology in Connecting Mathematics
with Authentic Experience. In R. Biehler, R.W. Scholz, R. Sträber and B. Winkelmann
(Eds.), Didactics of Mathematics as a Scientific Discipline (pp. 379-397). Dordrecht,
The Netherlands: Kluwer.
Kluwe, R.H. and Haider, H. (1990). Modelle zur internen Repräsentation komplexer technischer Systeme. Sprache & Kognition, 9 (4), 173-192.
Kolodner, J.L., Camp, P.J., Crismond, D., Fasse, B., Gray, J., Holbrook, J. and Ryan, M.
(2003). Promoting Deep Science Learning through Case-based Reasoning: Rituals and
Practices in Learning by Design Classrooms. In N.M. Seel and S. Dijkstra, S. (Eds.).
Curriculum, Plans and Processes of Instructional Design: International Perspectives.
Mahwah, NJ: Lawrence Erlbaum.
Lajoie, S.P. and Lesgold, A. (1989). Apprenticeship Training in the Workplace: Computercoached Practice Environment as a New Form of Apprenticeship. Machine-Mediated
Learning, 3, 7-28.
Lesh, R. (1998). The Development of Representational Abilities in Middle Schools
Mathematics: The Development of Student’s Representations during Model Eliciting
Activities. In I. Sigel (Ed.), Development of Mental Representation (pp. 323-349).
Mahwah, NJ: Erlbaum.
Lesh, R. and Doerr, H.M. (2000). Symbolizing, Communicating, and Mathematizing: Key
Components of Models and modeling. In P. Cobb, E. Yackel and K. McClain (Eds.),
Symbolizing and Communicating in Mathematics Classrooms. Perspectives on
Discourse, Tools, and Instructional Design (pp. 361-383). Mahwah, NJ: Erlbaum.
Mayer, R.E. (1989). Models for Understanding. Review of Educational Research, 59 (1), 43-64.
Mayer, R.E., Moreno, R., Boire, M. and Vagge, S. (1999). Maximizing Constructivist
Learning from Multimedia Communication by Minimizing Cognitive Load. Journal of
Educational Psychology, 91 (4), 638-643.
Neisser, U. (1979). Kognition and Wirklichkeit. Prinzipien and Implikationen der kognitiven
Psychologie. Stuttgart: Klett-Cotta.
Norman, D.A. (1983). Some Observations on Mental Models. In D. Gentner & A.L. Stevens
(Eds.), Mental Models (pp. 7-14). Hillsdale, NJ: Erlbaum.
Penner, D.E. (2001). Cognition, Computers, and Synthetic Science: Building Knowledge and
Meaning through Modeling. Review of Research in Education, 25, 1-35.
Penner, D.E. and Klahr, D. (1996). The Interaction of Domain-specific Knowledge and
Domain-general Discovery Strategies: A Study with Sinking Objects. Child
Development, 67, 2709-2727.
Penner, D.E., Lehrer, R. and Schauble, L. (1998). From Physical Models to Biomechanics: A
Design-based Modeling Approach. The Journal of the Learning Sciences, 7 (3 and 4),
Piaget, J. (1943). Le développement mental de l’enfant. Zürich: Rascher.
Rickheit, G. and Habel, C. (Eds.) (1999). Mental Models in Discourse Processing and
Reasoning. Amsterdam: Elsevier.
Royer, J.M., Cisero, C.A., & Carlo, M.S. (1993). Techniques and Procedures for Assessing
Cognitive Skills. Review of Educational Research, 63 (2), 201-243.
Rumelhart, D.E. (1980). Schemata: The Building Blocks of Cognition. In R. Spiro, B. Bruce
and W. Brewer (Eds.), Theoretical Issues in Reading Comprehension (pp. 33-58).
Hillsdale, NJ: Erlbaum.
Rumelhart, D.E., Smolensky, P., McClelland, J.L. and Hinton, G.E. (1986). Schemata and
Sequential Thought Processes in PDP models. In J.L. McClelland, D.E. Rumelhart, &
The PDP Research Group (Eds.), Parallel Distributed Processing. Explorations in the
Microstructure of Cognition. Volume 2: Psychological and Biological Models (pp. 757). Cambridge, MA: MIT Press.
Scandura, J.M. (1988). Role of Relativistic Knowledge in Intelligent Tutoring. Computers in
Human Behavior, 4, 53-64.
Schnotz, W. and Kulhavy, R.W. (Eds.) (1994). Comprehension of Graphics. Amsterdam:
Seel, N.M. (1986). Wissenserwerb durch Medien and “mentale Modelle.”
Unterrichtswissenschaft, 14 (4), 384-401.
Seel, N.M. (1991). Weltwissen and mentale Modelle. Göttingen: Hogrefe.
Seel, N.M. (1995). Mental Models, Knowledge Transfer, and Teaching Strategies. Journal of
Structural Learning and Intelligent Systems, 12 (3), 197-213.
Seel, N.M. (1999). Educational Diagnosis of Mental Models. Assessment Problems and
Technology-based Solutions. Journal of Structural Learning and Intelligent Systems, 14
(2), 153-185.
Seel, N.M. and Dinter, F.R. (1995). Instruction and Mental Model Progression: Learnerdependent Effects of Teaching Strategies on Knowledge Acquisition and Analogical
Transfer. Educational Research and Evaluation, 1 (1), 4-35.
Seel, N.M. (2000). Psychologie des Lernens. München: Reinhardt.
Seel, N.M. (2001). Epistemology, Situated Cognition, and Mental Models: „Like a Bridge
over Troubled Water“ Instructional Science, 29 (4-5), 403-427.
Seel, N.M., Al-Diban, S. and Blumschein, P. (2000). Mental Models and Instructional
Planning. In M. Spector and T.M. Anderson (Eds.), Integrated and Holistic Perspectives
on Learning, Instruction and Technology: Understanding Complexity (pp. 129-158).
Dordrecht, NL: Kluwer Academic Publishers.
Sfard, A.(1991). On the Dual Nature of Mathematical Conceptions: Reflections on Processes
and Objects as Different Sides of the Same Coin. Educational Studies in Mathematics,
22, 1-36.
Sharp, D.L.M., Bransford, J.D., Goldman, S.R., Risko, V.J., Kinzer, C.K. and Vye, N.J.
(1995). Dynamic Visual Support for Story Comprehension and Mental Model Building
by Young, At-risk Children. Educational Technology Research and Development, 43 (4),
Snow, R.E. (1990). New Approaches to Cognitive and Conative Assessment in Education.
International Journal of Educational Research, 14 (5), 455-473.
Stachowiak, H. (1973). Allgemeine Modelltheorie. Wien: Springer.
Stewart, J., Hafner, R., Johnson, S. and Finkel, E. (1992). Science as Model Building:
Computers and High-school Genetics. Educational Psychologist, 27(3), 317-336.
Tanner, H. and Jones, S. (1994). Using Peer and Self-assessment to Develop Modeling Skills
with Students Aged 11 to 16: A Socio-constructive Perspective. Educational Studies in
Mathematics, 27, 413-431.
Vosniadou, S., & Brewer, W.F. (1992). Mental Models of the Earth: A Study of Conceptual
Change in Childhood. Cognitive Psychology, 24, 535-585.
Vosniadou, S. and Brewer, W.F. (1994). Mental Models of the Day/night Circle. Cognitive
Science, 18, 123-183.
Wartofsky, M.W. (1979). Models. Representation and the Scientific Understanding.
Dordrecht: Reidel Publ.
White, B. (1993). ThinkerTools: Causal Models, Conceptual Change, and Science Education.
Cognition and Instruction, 10 (1), 1-100.
Williams, M., Hollan, J. and Stevens, A.L. (1983). Human Reasoning about a Simple Physical
System. In D. Gentner & A.L. Stevens (Eds.), Mental Models (pp. 131-154). Hillsdale,
NJ: Lawrence Erlbaum.
Young, R. (1983). Surrogates and Mappings: Two Kinds of Conceptual Models for Interactive
Devices. In D. Gentner and A.L. Stevens (Eds.), Mental Models (pp. 35-52). Hillsdale,
NJ: Erlbaum.
Norbert M. Seel is Professor of Learning Research and Instructional Design and the Head
of the Department of Educational Science of the Albert-Ludwigs-Universität of Freiburg,
Germany. His areas of interest include basics and applied research in model-based learning
and instruction, evaluation and psychometrics, instructional design and ISD.