For personal use only!

Klabbers, J. 1996. Problem framing through gaming:

Learning to manage complexity, uncertainty and value adjustment.

Journal Simulation & Gaming, Vol. 27, No 1, p. 74-92.

Dr. Jan H.G. Klabbers

KMPC - The Netherlands;
e-mail: [email protected]
Problem framing through gaming: Learning to manage complexity, uncertainty and
value adjustment

Jan H.G. Klabbers

KMPC-NL

Abstract

In social systems, an almost limitless number of social issues exist that cannot be dealt
with in isolation. Traditional professional knowledge is not well suited to coping with
complex and unique situations. Problem solving as encountered in mathematics and
physics brings forward a narrow, technical rationality, emphasizing a rationalist
framework for interpreting knowledge. The related problem solving strategies are too
limited in scope. They disregard competing frameworks based on multiple
perceptions. It is against this background of multiple realities that gaming will be
discussed. Gaming, in the historicist tradition, provides a suitable approach for dealing
with competing frameworks. It has proven to be a powerful combination of methods
that can deal with complex, uncertain and unique issues, and with value adjustments.
Gaming provides a language for combining the social-human with the physical,
technological and economic knowledge domains. To connect these knowledge
domains, three types of learning environments are distinguished.

KEYWORDS: autopoietic systems; actor approach; free-form games; gaming;

learning environments; machine approach; non-trivial machine;
rigid-rule games; simulation; trivial machine.

Within the scientific community two distinct preoccupations or cultures that is, social
(re-)constructions of reality, can be distinguished. They refer to different perceptions and
perspectives in shaping social order (Klabbers et al., 1989). One culture represents the
rationalist, the other the historicist conception of reality.
Both games and simulations are constructs that can help us understand and cope with
the complex world in which we live. They have been applied in a variety of areas such as,
corporate and environmental planning, planning of health care, urban planning, general
management, policy formation and education. Gaming and simulation are used for teaching,
training and research. Especially in many areas of professional training they have found
broad acceptance. Although they have much in common, clear differences exist in the way
they are applied. Also, from the viewpoint of the scientific cultures, mentioned above, their
methodological backgrounds have not been clarified sufficiently. This leaves much room for
confusion when particular games and simulations are compared with one another and
evaluated in terms of their validity, reliability and utility. In order to understand what
gaming and simulation have in common as well as what distinguishes them from one another
it is useful to examine the underlying principles of both scientific cultures.
In the rationalist tradition, knowledge is composed of abstract, context-independent,
formally-interconnected domain-specific concepts. Knowledge is mainly related to the way
in which something operates (i.e., its functionality). In the historicist tradition, knowledge is
context dependent. Knowledge about reality is formed by processing meaning and sharing
the interpretive schemata of the persons involved. It is not accumulative per se, because
historical conditions change. This implies that the facts and their context change. As both

2
scientific cultures have different conceptions about knowledge, they pursue different
objectives in knowledge transfer. Consequently, different learning environments take shape.
In the rationalist tradition problems are generally well-defined, related to a specific
knowledge domain that is, a particular body of disciplinary knowledge such as mathematics,
chemistry or physics. Knowledge transfer takes place in the so-called reproductive learning
environment. Staying within the rationalist tradition, entering the realm of ill-defined
problems, well-articulated task-environments are created, where no single answers to the
problem at hand are available. The respective heuristic learning environment encourages
learners to search for acceptable alternative solutions.
In the historicist tradition, multiple realities are recognized. Learners bring order to
their world while processing meaning. Learning becomes a social process of enculturation
through learning-by-doing in a self-organizing learning environment. In different learning
environments, different problem solving approaches are applied. These are discussed below.

The rationalist tradition

The reproductive learning environment . A characteristic example of the rationalist

conception of reality is the so-called Trivial Machine (TM). Von Foerster (1984) defines a
machine as a conceptual device with well-defined rules of operation. The TM is defined as a
black box F processing input x to output y, see Figure 1.

Figure 1: Representation of the Trivial Machine

The inference scheme, that is, the invariable relationship F between x and y provided by
trivial machines, is widely used in science in a whole variety of contexts. The following list
of terms illustrates this point (von Foerster, 1984):

x F y
____________________________________________________________
input operation output
input system output
independent variable function dependent variable
cause Law of Nature effect
minor premise major premise conclusion
stimulus Central Nervous System response
goal system action

As soon as the x-y correspondence through F is established, any x will generate a specific y,
independent of time. Consequently TMs are predictable, history independent, synthetically
deterministic (definition of x-y correspondence) and analytically determinable.
Most problem solving in education, especially in mathematics and science education, deals
with learning to use the inference scheme of the TM. Related problem-solving tasks are well
articulated, and respective problems are well-defined. The representation of TMs is
equivalent to the representation of systems.

3
 Learners, through utilizing the inference schemata of TM, progressively reproduce a
body of disciplinary knowledge. Problem solving in this regard is considered mainly a matter
of applying disciplinary knowledge. Problem-solving skills related to TMs can either refer to
goal-directed strategies applied to specific knowledge domains, that is, task-limited specific
strategies, or to general across-domain strategies. TMs are suitable constructs from the
viewpoint of a rationalist conception of reality. They enable a clear distinction to be made
between the observer and the observed. Reality, thus conceived, is controllable and
manageable. For the type of learning environment provided by TMs, I have used the term
reproductive learning environment (Klabbers, 1989).

The heuristic learning environment. We have come to realize that many phenomena cannot
be investigated fruitfully via the construct of TM. For many phenomena it is impossible to
establish a transfer function F. In many cases, for example in econometric, sociometric and
psychometric research, the error component for variance not explained by the x-y
correspondence is generally too high to generate valid and reliable results. A second
argument suggesting that as a construct the TM is too narrow, is based on the reference
scheme of the Non-Trivial Machine (NTM). The Non-Trivial Machine as presented by von
Foerster (1984) is a more powerful construct for understanding complex dynamic systems
(see Figure 2). NTMs provide us with interesting terminology, for example, recursive
processes and autological, self-referential and second-order concepts (i.e., concepts that are
embedded in their own domain).

Figure 2: Representation of the Nontrivial Machine

As can be inferred from Figure 2, the driving function y = F(x,s) is similar to the TM of
Figure 1. The state function s' = S(x,s) is new and is recursive; that is, s' is defined by s at an
earlier stage. Suppose NTM is a discrete system. In that case the state function can be
represented by:
s(t) = S (x(t), s(t-1))
s(t-1) = S (x(t-1), s(t-2)), etc.
S represents the memory or history of any NTM. The internal state generated through S
influences the processing of subsequent inputs to outputs. Von Foerster (1984) shows that
identification of a large class of NTMs is impossible (transcomputational) because the
machine's driving and state functions cannot be inferred from observed sequences of
input/output pairs (x,y).
NTMs are characterized as synthetically deterministic (constituted by deterministic
driving and state functions), history dependent, analytically indeterminable, and analytically
unpredictable.
Suppose we arrange a problem-solving situation based on NTM and, moreover, putting
learners in the position of observers/experimenters. This will teach them that the simple
notion of causality inferred from prior experiences with the TM no longer holds: A response

4
once observed for a given stimulus may not be the same for the same stimulus given later.
Behaviour of an NTM depends on its internal state, which depends recursively on its history.
Problem solving in interaction with NTMs will require applying heuristic techniques:
continuously planning, monitoring, checking and revising of options and alternatives. System
dynamics models, which are appropriate examples of NTMs, provide learners with well-
articulated tasks for ill-defined problems.
Problem-solving skills related to NTMs may stimulate learners applying various
search strategies to develop their reflective awareness (metacognition). By trying to
understand the dynamic behavior of NTMs learners may become aware of their blind spots
about aspects of the system they initially did not have the faintest idea. Here we enter the
realm of second-order problems. Learners will realize that the problem is not “not knowing,”
but “not knowing that one does not know.” Von Foerster points out that second-order
concepts have an unusual structure, that is, "double negation does not yield affirmation."
Learners interacting with NTMs will realize that accessing knowledge is context dependent.
For example, behavior of a system dynamics model depends on its initial conditions, its
internal structure and the mix of the parameters. To search for acceptable policies, various
scenarios might be suitable. For the type of learning environment generated by NTMs, I have
used the term heuristic learning environment (Klabbers, 1989).
TMs and NTMs are conceptual devices with well-defined rules of operation. Such a
mechanistic approach does not seem to be very suitable for learning about social systems,
simply because we do not know their rules of operation.

The historicist tradition

Another meaning of knowledge. In the historicist tradition, the term knowledge has a
connotation which is different from its meaning in the rationalist tradition. The interpretation
of what constitutes a problem differs as well. In the rationalist tradition, problems are
defined as "situations where one has a good idea about what to accomplish, but no clear idea
about how to go about accomplishing it" (Davis, 1973; Duncker, 1945; Prawat, 1989).
Problem solving then implies applying the right scientific knowledge. In the social realm
however, the key question is: ”What is the right knowledge?” Many situations, according to
the rationalist definition, hardly qualify as problems (Prawat, 1989) because in our pluralisitic
society we tend to disagree on what to accomplish and what resources to allocate. This
applies to health care, crime, education, environmental degradation and numerous other
issues. Apparently, in such a context the word problem connotes a meaning which is
different from its meaning in the rationalist tradition.
Rein (1976) has stated that knowledge presupposes a framework to interpret it, but in a
pluralistic system there are competing frameworks. They are conveyed and defended by
interest groups or stakeholders; in general they are institutional actors. Actors are persons or
groups of persons with roles and interests, carrying out one or more activities in a social
system. Actors:
- engage in meaning processing and frame problematic situations through multiple
circular, recursive interactions;
- construct knowledge;
- produce a collective structure (social system), which can easily go beyond their
comprehension;
- find themselves in a complex situation with its ambiguous and uncertain outcomes.
Instrumental problem solving, in use in the rationalist tradition, is not well suited for dealing
with such problematic situations generated by multiple actors and multiple realities. It

5
presupposes one common definition of reality. Thus, for dealing with multiple realities the
actor approach is better suited than the machine approach.
Progressing from the frame of reference stipulated previously by the machine
approach, the representation of the NTM is broadened. The state function s=S(s,x) is replaced
by Actors and Rules, representing the internal state of the actor-system. This shift
characterizes the switch from the machine to the actor approach. In the actor system, via
communication and coordination, actors develop rules and procedures both for
communicating and for intervening in their internal environment F. Through this recursive
processing of information, they continuously enact their collective structure. The actors
engage in a process of social construction of reality, and their collective awareness (the
system of meaning) is an indicator of the internal state of the system. They engage in a
recursive, time-variant process of change. They reproduce their social system in an
operationally closed setting (Klabbers, 1986; Luhmann, 1986). They shape a system of
meaning and as a consequence ‘make history’.

Figure 3: Illustration of an Actor System

The actor system, as illustrated in Figure 3, is re-arranged into the frame of a game (see Fig.
4).
In a game, actors may use different codes, signs and symbols and apply different
rules and procedures while shaping their natural and/or artificial environment represented by
the reference system. With this general representation (see Fig. 4), the underlying structure
of a whole variety of games such as MONOPOLY, CHESS, GO, computer games, and so on,
can be made transparent. The different boards represent different types of reference systems.
Through their respective rules, the actors intervene in their symbolic environment by moving
pieces. They change the situation on the boards (screen) in recursive processes of monitoring,
planning, revising, moving, monitoring, and so on.

Figure 4: Basic components of a Game

6
The reference system may also be represented via the signs and symbols that are in use in the
formal language of mathematical models; it is the language of computer simulation models. In
Fig. 5, gaming and simulation are integrated in computer-supported, multi-actor systems.
The actors may take the roles of management or (public) policy makers. Accordingly,
complex multi-level, multi-actor games like PERFORM (Klabbers, 1986) and DENTIST
(Klabbers et al. 1980) have been developed. They form a hybrid system, interconnecting the
machine, with the actor approach.

Figure 5: Illustration of a Computer-Supported Multiactor System

In the field of gaming two different types of games are distinguished: rigid-rule games, such
as classical management games or business simulations, and free-form games. These require
different contexts of use and problem orientations, that is, different learning environments.

The heuristic learning environment: rigid-rule games. In rigid-rule games it is assumed that
rules are supposed to have some validity, and that they apply equally to all actors (player
groups). Sometimes those rules are prescriptive; the game leader forces the players to take
steps according to the rules. Sometimes they are descriptive, as they model behavioural
equations (Klabbers & van der Waals, 1989). Rigid-rule games are closed games. Players
receive specific role instructions, and their actions are goal-oriented. The game operator tells
them: "This is the problem: how will you solve it?"
Management games developed during the fifties and sixties have had a great impact on
the notion of gaming. Since the sixties, hordes of students of business administration have
participated in management gaming. Many training programs of large corporations in the
USA, in Europe and elsewhere nowadays utilize games for training and assessing managerial
skills. According to their structure, rigid-rule games resemble very much the setting of a
heuristic learning environment. Designers have the option of choosing either a one-actor or a
multi-actor approach (Klabbers, 1975, 1986, Klabbers et al., 1980). Multi-actor games
stimulate players/learners to articulate their thoughts and engage in policy dialogues.

The self-organizing learning environment: free-form games. In free-form games the actor
approach is most clearly demonstrated. Using free-form games, only a scenario and a simple
accounting system are provided. Players are put in an open situation and are asked: "What
will you do?" Designers of those games presuppose that conception of reality is weakly
known and take into account that competing conceptions or pre-occupations about reality
exist. Playing free-form games enable actors to understand better one another’s concerns as
they emerge in the processing of meaning. Players engage in a process of creating order

7
through multiple lines of action and communication, producing and re-producing their social
system. They construct reality. Learning is related to shaping order, creating knowledge, and
improving mutual-awareness among the participants. These so-called self-organizing
learning environments may not only show that knowledge is context dependent, because of
the changing meaning of knowledge they may also highlight the impact of shifting contexts.
Shubik (1983) points out that one should be able to see immediately that emphasis on the
participation and quality of the individuals must be higher when using free-form gaming as
compared with rigid-rules gaming. "The value of a free-form game may be highly related to
the expertise and sophistication of both the players and the referees" (Shubik, 1983).
Self-organizing learning environments, like free-form games, call into question the
coherence of the assumptions on which context dependent knowledge is based. They help by
articulating the subject matter in a continuous process of re-framing the problematic situation.
They engage in a process of problem framing.

Gaming in the context of the historicist tradition.

Through the successive representations (see Figures 1 - 4) the rationalist tradition of

simulation , illustrated by the TM and NTM, is linked with the historicist tradition of gaming.
In Table 1 the route I have chosen is summarized. Simulation is related to the machine
approach and gaming to the actor approach.

Table 1: Classification of Simulations and Games

It is important to notice that changing Figure 2 into 3 implies a shift from

functionalism to constructivism. This implies that in Figure 3 the meaning of the input
changes. As the NTM reacts to changes in its environment, the Actor-System (AS) is both
reactive and proactive. It is reactive as far as input to F is concerned. It is proactive because
actors, according to Weick (1979), enact or construct their environment and subsequently
select relevant changes for closer attention, internalising the environmental challenge as an
element of internal competence (Weick, 1979). From this point of view the environment is a
construct.
The actor approach allows for shifting between two perspectives.
1. One can take the position of an observer/outsider (functionalist approach)
2. One can take the position of a participant/insider (constructivist approach).

8
Multi-actor systems develop a system of meaning through meaning processing. They produce
self-referential social systems. By definition games too are social systems: collective
structures shaped and maintained through individuals and their interrelationships.
Characterizing multi-actor systems, and thus games, requires that we take the two
perspectives mentioned above into account.
From the spectator/outsider viewpoint multi-actor systems are like NTMs synthetically
indeterministic, history dependent, analytically indeterministic and analytically unpredictable.
From participant's point of view multi-actor systems are synthetically (structurally)
determinable: the actors mutually shape order, dependent on definitions of reality by the
actors (values, beliefs, interests, rules, etc.), and comprehensible on the basis of meaning-
processing (shared language, signs, symbols, images, icons, etc.). It should be realized that
the participant’s perspective presupposes that knowledge is produced from the inside. It is a
social construct.

Teaching problem framing through gaming

In previous sections the scope of knowledge transfer has been widened by moving
from the machine approach (TM and NTM) to the actor approach (Rigid-Rule and Free-Form
Games). The TM is related to the reproductive learning environment, both the NTM and
rigid-rule-games to the heuristic learning environment and free-form games to the self-
organizing learning environment (see Table 1). In professional education, knowledge that is
general, theoretical, and propositional enjoys a privileged position. While teaching
professional knowledge, many methods of didactic education assume a separation between
knowing and doing. Knowledge is treated as an integral, self-sufficient substance,
theoretically independent of the situations in which it is learned and used (Brown, Collins &
Duguid, 1989). Accordingly, professional activity consists of instrumental problem solving
made rigorous by the application of scientific theory and technique (Schön, 1982).
From a practitioner’s point of view there is a growing crisis of confidence in this type
of professional knowledge and consequently in this type of knowledge transfer by our
educational institutions. "Professionally designed solutions to public problems have had
unanticipated consequences, sometimes worse than the problems they were designed to solve.
Newly invented technologies, professionally conceived and evaluated, have turned out to
produce unintended side effects unacceptable to large segments of our society " (Schön, 1982,
1987). It is axiomatic that complexity, uncertainty and value adjustments regarding a
problematic situation are not resolved by transforming an ill-structured problem into a well-
articulated learning task.
In order to be able to cope with complex, uncertain and unique social situations,
Schön has proposed the term problem framing. It is a process in which, “interactively, we
name the elements and attributes to which we will pay attention, and frame the contexts in
which we will pay regard to them” (Schön, 1982). Problem framing is precisely what is
happening in the self-organizing learning environment through free-form gaming.
Brown et al. (1989) pointed out that "knowledge is situated, being in part a product of
the activity, context, and culture in which it is developed and used." Gaming is appropriate
for supporting situated learning; free-form gaming, especially, is a fruitful cognitive
apprenticeship method, as it "may enculturate students into authentic practices through
activity and social interaction in a way similar to that evident and evidently successful in craft
apprenticeship." In free-form gaming, students are guided through a cognitive apprenticeship
in becoming reflective practitioners. Moore and Anderson (1975), in presenting guidelines
for designing learning environments, argue that:

9
“1. Learners should be given the opportunity to operate from various perspectives. The
learner should not just be a recipient of information, but should at times be an agent, a
referee, and a reciprocator.
2. Activities should contain their own goals and sources of motivation, not just represent
means to some end (such as grades). That is, in an effective learning environment,
activities are autotelic.
3. Learners should be freed from a dependence on authority and allowed to reason for
themselves; they are thus made productive in the learning process.
4. The environment should be responsive to the learner's activity. Not only should they
be given feedback, but they should be helped to be reflexive, evaluating their own
progress."

Recent advances in systems theory relevant for gaming and simulation

A debate about the methodological issues raised in the earlier sections will be more
coherent if recent advances in systems theory, dealing with autopoiesis and complexity, are
taken into account. They are relevant in view of (free-form) gaming and the related self-
organizing learning environment. They are also relevant in view of simulation as they assist
better understanding of the dynamic behaviour of non-linear recursive systems and self-
organizing systems. Both advances, although coming from different backgrounds, bring
forward one coherent view on complex systems and consequently on gaming.

Autopoiesis. In 1974, Varela and Maturana introduced the concept of autopoiesis,

focussing on the reproduction of the organization of a living system through the network of
interactions among the components. An autopoietic system is a self-sustaining network that
produces, transforms and destroys its components. The interactions and transformations
among the components regenerate and realize the processes that produced them (Maturana &
Varela, 1980). Although both authors carefully restrict their theoretical notions to living
systems in biology, others like Luhmann (1984) have extended the concept to the social
realm. Luhmann has pointed out that social systems use communication as their particular
mode of reproduction. Klabbers (1986) has mentioned that social systems are constituted
through recursive interactions among the actors, generating a system of meaning that
reproduces itself. The concept of autopoiesis brings forward notions about self-referential and
self-organizing systems. Actor systems, constituting collective networks, are viewed as
autopoietic systems. Examples of such systems at different levels of aggregation are, groups,
organizations, institutions and societies.
Free-form games are considered autopoietic systems as they enable the production of a
self-sustaining collective network of actors that produces the actors. The interactions and
transformations between the actors realize and regenerate the systems of meaning that shape
the actors. It is not the individual that is reproduced in the biological sense but the person in
the social-cultural sense.

Science of complexity. Science is predominantly based on the assumptions of

universality and the context-independent character of natural laws, which is the ideal of
rationality. The traditional scientific approach is prevailingly occupied with investigating
reversible processes, with the notion of trajectories and time that can be reversed. This
conception of science has disclosed a passive nature, a mechanistic world. The systems
concerned are allopoeitic. They have as the outcome of their functioning (output) something
that is different from them. Prigogine and Stengers have pointed out that according to this
conception once the rules or laws governing nature are established, nature will obey them.

10
Evidently nature is trivialized, mastered and controlled (Prigogine & Stengers, 1985).
Technology is the dominant vehicle in accomplishing this goal.
However, recent advances in science, related to the dynamics of dissipative, non-linear
feedback systems, show that the assumption of reversibility of processes is rather weak for
natural, biological, ecological and social systems. These type of systems are understood to be
time-oriented; they do not run backward. With regard to the climate system for instance, since
the sixties, anthropogenic disturbances caused by the emissions of greenhouse gases may
eventually induce a climate change that cannot be reversed by merely decreasing these
emissions to prior levels. These types of systems challenge analysis simply in terms of their
constituent parts. This notion demands that we focus on processes rather than on states, on
becoming rather than being. The science of complexity, dealing with deterministic chaos,
eliminates the myth of deterministic predictability. Consequently, it challenges the purely
rationalist tradition. All features of complexity point at gaming as the basic characteristic of
nature, be it the formation of matter, the organization of matter into living structures, or social
behaviour of man (Eigen & Winkler, 1975). Elements of such games are chances and rules.
Humans participate in this huge evolutionary game with its uncertain results. Management of
complex systems is, as we all know, very wicked, in the sense that the systems resist
conventional analysis and instrumental problem-solving techniques (Kalff, 1989).
Epistemological issues raised by the science of complexity may bring the fields of the natural
and social systems closer together.

Managing complexity, uncertainty and value adjustment

Combining the notions of autopoiesis and complexity, one general conclusion can be
drawn. Complex systems cannot be steered from the outside, as their dynamic behaviour is
based on an internal logic. This internal logic of a social system, expressed in its system of
meaning, identified and explored via problem framing, makes clear the interrelations among
actors, perceptions and positions. Steering depends on these interrelationships as they mould
the system. This implies that unilateral steering-from-the-outside should be replaced by self-
steering through processes of cooperation based on shared purpose. From this point of view
simulation and gaming have very distinct roles.
Simulation based on mathematical modelling is concerned with description and explanation
of general characteristics of the (reference) system involved. The respective system is
considered allopoietic. The nature of a system is reduced to a simple basic structure (e.g.,
causal model), which explains its behaviour. Simulation aims at explaining and controlling
reality. The simulation approach stresses steering from the outside.
Gaming, especially free-form gaming, focuses on making sense of reality in terms of
meaning processing between the participants/actors. Free-form games can be approached
and framed by the actors (participants) in various ways and the outcomes require considerable
judgment, communication and negotiation skills. Games are considered self-referential
systems: understanding is generated from the inside and is context dependent. During a free-
form game, participants have the opportunity to create new social order and consequently
produce situated knowledge. Gaming, if well conceived, can make explicit the internal logic
of the system involved, providing the basis of self-steering. It can make transparent and
manageable the complexity of the pattern of relations.

Learning to manage. Much empirical evidence about learning effects is generated

through educational research. Results, however, are not conclusive because several research
traditions (paradigms) compete with one another about the meaning of the results. Little is
known from empirical research about the learning effects of the heuristic and self-organizing

11
learning environments. Reviewing three decades of business gaming, it is still not clear how
effective they are. Evaluation studies carried out since the early sixties are not conclusive
(Wolfe, 1985). More recently Keys and Wolfe (1990) point out that management games
“have been found to be generally effective and to possess internal validity in the strategic
management type course.” It is generally recognized that, from a methodological point of
view, games are 'unruly' when it comes to evaluating them (Bredemeier & Greenblat, 1981).
Their value in developing managerial skills appears to be limited (Evans & Sculli, 1984). A
basic question to address is the appropriateness of evaluating games (actor systems) with
methods and techniques stemming from the TM approach.
Within the context of the self-organizing learning environment, learning is not merely
concerned with reproducing a body of disciplinary knowledge; rather, it challenges the
perceptions of learners and the way these evolve over time. Issues that arise during a free-
form game stem partially from the content of the material provided. They are mainly
generated through interactions among participants. Participants produce and continue
producing their own social system with its unique identity, largely determined by the actors
and their interactions with one another (Dunbar & Stumpf, 1989). These principles have
been embedded in recently developed management games THE FUNO MANAGERS (1991),
kOMpleKS (1992), ICuMS (1993), ROLE ROTATION (1994) and ISO-2000 (1995) in order
to train senior managers and high-level public officials in the principles of self-organizing
systems underlying strategic management and the management of change. Learning for them
becomes innovative only in so far as they accept that learning is not merely a search for
certainty and security.

Conclusions

In western civilized societies, one important role for education is reproducing the vast
body of knowledge available for sustaining the advanced techno-economic structure.
Reproductive learning environments are aimed at supporting this maintenance learning, based
on the inference scheme of the TM. However, innovative learning is more adequate and
certainly more needed to prepare the younger generation for the emerging social
circumstances. The current educational systems should stimulate more the learner’s self-
organizing capacities for learning, and for learning to learn. The heuristic learning
environment is appropriate as it enables learners to improve their search strategies for finding
acceptable solutions to complex problems on the basis of available knowledge. This type of
learning environment not only stresses domain-specific knowledge transfer, it also
emphasizes the importance of cross-domain or strategic knowledge. Strategic knowledge
facilitates the switching of perspective to shape coherent cross-domain strategies. Interactive
simulation in the form of rigid-rule games aims at improving learners’ mental flexibility.
Most of the problematic situations we encounter nowadays, such as climate change in
relation to sustainable development, drug and crime prevention, health care, starvation and
discrimination, cannot be resolved by narrow technical means based on the purely rationalist
conception of reality (as illustrated in the machine approach). They require new approaches to
societal steering (In ‘t Veld et al., 1991). Evaluation of self-organizing learning environments
should be based on the self-referential capacities of the learners that improve their reflexivity
in evaluating their own progress (or lack of it) in coping with complex issues. Managing
complexity, uncertainty and value adjustments in social systems, is, from the autopoietic point
of view, related to managing self-sustaining collective networks. It is a continuous process of
managing change, while keeping shared values intact. Gaming is an appropriate cognitive
apprenticeship environment for managing complexity, uncertainty and value adjustments.

12
References
Bredemeier, M. E. & Greenblat, C. S. (1981). The educational effectiveness of
simulation games: A synthesis of recent findings. In Greenblat , C.S. & Duke, R.D.
Principles and practices of gaming-simulation. London: Sage
Brown, J. S., Collins, A. & Duguid, P. (1989). Situated Cognition and the Culture of
Learning, Educational Researcher 18(1) , 32-42.
Davis, G. A. (1973). Psychology of problem-solving: Theory and practice. New York:
Basic Books.
DENTIST. In Klabbers, J., van der Hijden, P., Hoefnagels, K. & Truin, G. (1980).
Development of an interactive simulation game: a case study of DENTIST. Simulation
& Games: An International Journal, 11, 59-86.
Dunbar, R.L.M. & Stumpf, S. A. (1989). Trainings that demistify strategic decision-
making processes, Journal of Management Development 8, 36-42.
Duncker, K. (1945). On problem-solving. Psycholical Monographs 270.
Eigen, M. & Winkler, R. (1975). Das Spiel. Naturgesetze steuern den Zufall. München:
Piper & Co.
Evans, W. A. & Sculli, D. (1984). An evaluation of business games with respect to the
development of managerial talent, SAGSET Journal, 14 , 3-13.
FUNO MANAGERS, The. Klabbers, J. (1991). De Baak Management Studiecentrum,
Noordwijk, The Netherlands.
ICuMS. Klabbers, J. & Gust, M. (1993). KMPC, Bemmel, The Netherlands.
In ‘t Veld, R., Schaap, L., Termeer, C. & Van Twist, M. (1991). Autopoiesis and
configuration theory: New approaches to societal steering. Dordrecht: Kluwer.
ISO-2000. Klabbers, J. (1995). Arnhem: PiMedia.
Kalff, D. J. A. (1989). Strategic decision making and simulation in Shell. In Klabbers, J.,
Scheper, W., Takkenberg, C., & Crookall, D. (Eds.) Simulation-gaming: On
the improvement of competence in dealing with complexity, uncertainty and value
conflicts . Oxford: Pergamon.
Keys, B. & Wolfe, J. (1990). The role of management games and simulations in
education and research, Journal of Management. 16 , 307-336.
Klabbers, J. (1975). Interactive simulation: On-line interaction between man and machine
for the study and management of social systems. Informatie , 17 , 535-560.
Klabbers, J. (1986). Improvement of (self-)steering through support systems. In Geyer,
F., & van der Zouwen, J. (Eds.) Sociocybernetic paradoxes . London: Sage.
Klabbers, J. (1989). Methodological aspects for designing learning environments (in
Dutch). In Klep, J. & Kommers, P. (Eds.) Courseware en leerplanontwikkeling
didactische systeemanalyse. Enschede: SLO.
Klabbers, J., van der Hijden, P., Hoefnagels, K. & Truin, G. (1980). Development of an
interactive simulation game: a case study of DENTIST. Simulation & Games: An
International Journal, 11, 59-86.
Klabbers, J., Scheper, W., Takkenberg, C. & Crookall, D. (Eds.) (1989). Simulation-
gaming: On the improvement of competence in dealing with complexity, uncertainty
and value conflicts . Oxford: Pergamon.
Klabbers, J. & van der Waals, B. (1989). From Rigid-rule to free-form games:
Observations on the role of rules. In Klabbers, J., Scheper, W., Takkenberg, C. &
Crookall, D. (Eds.) Simulation-gaming: On the improvement of competence in dealing
with complexity, uncertainty and value conflicts. Oxford: Pergamon.
kOMpleKS. Klabbers, J. (1992). De Baak Management Studiecentrum, Noordwijk, The
Netherlands.

13
 Luhmann, N. (1984). Soziale Systeme: Grundriss einer allgemeine Theorie. Frankfurt am
Main: Suhrkamp.
Luhmann, N. (1986). The autopoiesis of social systems. In Geyer, F. & van der
Zouwen, J. (Eds.) Sociocybernetic paradoxes . London: Sage.
Maturana, H.R. & Varela, F.J. (1980). Autopoiesis and cognition: The realization of the
living. Dordrecht: Reidel.
Moore, O. & Anderson, A. (1975).Some principles for the design of clarifying educational
environments. In Greenblat, C. S. & Duke, R. (Eds.) Principles and practices of
gaming-simulation. Newbury Park: Sage.
PERFORM. In Klabbers, J. (1986). Improvement of (self-)steering through support
systems.
Prawat, R. S. (1989). Promoting access to knowledge, strategy, and disposition in students:
A research synthesis. Educational Research, 59, 1-41.
Prigogine, I. & Stengers, I. (1985). Order out of chaos. New York: Bantam.
Rein, M. (1976). Social science and public policy. New York: Pinguin Education.
ROLE ROTATION (1994). Arnhem: PiMedia.
Schön, D. (1982). The reflective practitioner . New York: Basic Books.
Schön, D, (1987). Educating the reflective practitioner . San Fransisco: Jossey-Bass.
Shubik, M. (1983). Gaming: A state-of-the-art survey. In Ståhl, I. (Ed.) Operational
gaming . Oxford: Pergamon
Von Foerster, H. (1984). Principles of self-organization in a socio-managerial context. In
Ulrich, H. & Probst, G. J. B. (Eds.). Self-organization and management of social
systems . Berlin: Springer-Verlag.
Weick, K. E. (1979). The social psychology of organizing. London: Addison-Wesley.
Wolfe, J. (1985). The teaching effectiveness of games in collegiate business course: an
1973-1983 update. Simulation & Games: An International Journal, 16 , 251-288.

ADDRESS: KMPC
Oostervelden 59 , 6681 WR Bemmel, The Netherlands
[email protected]

14

View publication stats