COMPRAM Method
DeTombe, D.J. (1994) Defining complex interdisciplinary societal problems. A theoretical study for constructing a co-operative problem analyzing method: the method COMPRAM. Amsterdam: Thesis publishers Amsterdam (thesis), 439 pp. ISBN 90 5170 302-3
Dorien J. DeTombe, Ph.D.
Chair Operational Research Euro Working Group
Complex Societal Problems
P.O. Box. 3286, 1001 AB Amsterdam,
The Netherlands, Europe
Tel: +31 20 6927526 E-Mail: DeTombe@lri.jur.uva.nl
http://www.geocities.com/doriendetombe
2 COMPLEX INTERDISCIPLINARY SOCIETAL PROBLEMS-
chapter 2/9
2.0 Introduction
In this chapter, some aspects of complex interdisciplinary societal problems and some aspects of (human) problem handling in general are discussed. The emphasis is given more to the problem and the method of problem handling, rather than the problem solver. The analysis of the problem handling process will be mainly treated from the viewpoint of cognitive psychology. The chapter begins with the question of what constitutes a problem and what is problem handling, followed by a description of different kind of problems. This is followed by a discussion of when a problem is solved, when something is a problem, to whom is it a problem, and about who should handle the problem. Some aspects of problem solving research, which are important for this debate, will be discussed. In a brief historical review, the work of some researchers on the subject of problem solving in the field of cognitive psychology will be discussed. Those researchers are selected whose ideas had a great influence on the work of other researchers in the field. In this review the research of Selz (1922), who used the thinking-aloud method to discover how human beings think, and of De Groot (1969) on problem solving in the chess game, and some of the ideas of Newell & Simon (1972) will be discussed. Newell & Simon's ideas have influenced many researchers. Their ideas have a central place in the theory of problem solving in cognitive psychology and Artificial Intelligence. The discussion closes with research on undefined problems: the research of Crombag (Crombag, de Wijkerslooth & Cohen, 1977; Crombag, 1984) on how judges and physicians handle problems and the recent research of Wierda (1991) about developing interdisciplinary information systems. Based on these discussions the answer to (a part of) the first research expectation can be given. The first expectation is:
the character of the complex interdisciplinary societal problems is itself an important cause of the difficulty of handling these problems
The research questions derived from this expectation are:
1a is handling complex interdisciplinary societal problems so difficult because of the special character of these problems?
1b in what way do the problems that are studied in the field of cognitive psychology differ from complex interdisciplinary societal problems and what are the similarities that are relevant for analyzing and defining complex interdisciplinary societal problems?
2.1 What is a problem and what is problem handling
2.1.1 What is a problem
There are different meanings of the word 'problem':
- The word 'problem' can indicate a discrepancy between the actual situation and the desired situation (Kramer, 1978).
This meaning comes close to the common sense meaning of the word, which is often used to indicate an unwelcome situation. However, not every discrepancy between the actual situation and the desired situation has to be evaluated negatively. One may also wish to change the situation in order to ensure that the situation remains pleasant in the future. If we include this, the definition becomes:
a problem is a discrepancy between the actual situation and the desired (future) situation
- The word 'problem' can also indicate that there is a problem of knowledge.
A problem of knowledge is perceived when there is a discrepancy between the actual state of understanding of a phenomenon or set of phenomena and the desired state of understanding. This knowledge problem can be a negative experience in the sense of not knowing what to do, or a positive experience in the sense of curiosity.
Newell & Simon describe a problem as follows (1972, p. 72):
"A person is confronted with a problem when he wants something and does not know immediately what series of actions he can perform to get it. The desired object may be very tangible (an apple to eat) or abstract (an elegant proof for a theorem). It may be specific (that particular apple over there) or quite general (something to appease hunger). It may be a physical object (an apple) or a set of symbols (the proof of a theorem). The actions involved in obtaining desired objects include physical actions (walking, reaching, writing), perceptual activities (looking, listening), and purely mental activities (judging the similarity of two symbols, remembering a scene and so on)."
Newell & Simon include both meanings of the word 'problem' described above. They consider a problem as a discrepancy between the actual situation and the desired (future) situation and as a problem of knowledge about how to reach the desired state. In the research of Newell & Simon, the goal, in the sense of the desired situation, is clear.
The cognitive psychologists Frijda & Elshout (1974) define a problem as a situation in which the subject is confronted with a task, exercise or difficulty and to which he or she has no immediate answer and to which he or she can find no answer with automated acts
.In the definitions of Newell & Simon and Frijda & Elshout, the problem as such is already defined, the problem is clear. In these definitions, Newell & Simon and Frijda & Elshout connect a problem directly with problem handling. The term problem is often used in connection with the intention to change a situation. A 'problem must be solved' or a 'problem cannot be solved' or 'it is easy or difficult to solve this problem'. Whether a problem must be handled depends on the situation and the persons involved.
We exclude the problem handling process in our definition of the problem because a problem can exist independent of whether it will be handled or not. We not only consider something a problem that causes a discrepancy at this moment, but we also consider something a problem that can cause a discrepancy in the near future between the actual situation and the desired situation.
The definition of a (general) problem we use is:
a problem is a discrepancy between the actual or (near future) situation and the desired future situation and/or a lack of knowledge and/or a lack of know-how
2.1.2 Problem handling
Newell & Simon define problem handling as the actions to be taken to achieve the goal. Newell & Simon and Frijda & Elshout consider something a problem only when the problem handling method or technique is not immediately clear. They consider a problem something on which actions can be taken but also something that needs time to consider. In their definitions of a problem, a problem is defined as a kind of puzzle that must be performed. This definition of a problem is very close to the common sense meaning of the word problem. Duyne (1983) also connects a problem to problem handling. He gives the same kind of definition of a problem. He states that there is no problem when there is a direct answer. He illustrates this statement with the example of a public prosecutor confronted with a case that could be handled immediately by applying rules, because standard procedures were available to handle this
. Like Snoek (1989) and Frijda & Elshout (1974), Duyne does not consider routine activities and standard procedures as problem handling. Duyne states that the knowledge of the goal is the most important aspect in seeing whether one is engaged in problem handling or not. In this he joins the ideas of Newell & Simon.De Groot (1965) considers the problem handling process as a process in which a certain argument supporting a certain decision is made.
Crombag (1984, p. 9) sees problem handling as:"the search for a solution via a sequence of problem transformations: states of the problem on its way to a solution or as an arranged series of operations."
According to these researchers, problem handling may be considered as the human activity aimed at eliminating or reducing a perceived problem. Although we agree with this we make no restrictions concerning any (known) goal in the sense of a desired situation. For complex interdisciplinary societal problems, the desired situation is often not (yet) clear, or not univocal or there may be different, or even contradictory, situations. The goal of the handling process might be defining the problem, or getting more clarity on a certain issue. Nor do we wish to exclude standard procedures and routine activities. We consider all problem handling activities, thinking, applying tools, methods or routine rule application, as problem handling. We define problem handling as:
the process of analyzing a problem in order to gain more insight into the problem, whether or not this leads to influencing the problem in order to reach the desired situation. This process can take place actively or passively, consciously or unconsciously, routinlye or once-only, whether it is by circumventing or by forgetting the problem, by shifting the problem to another party or by (partly) changing the problem, whether through thinking, applying tools and/or methods
2.2 Different kinds of problems
In order to be able to make clear what the distinction is between the kind of problems we focus on in this study and the problems most of the cognitive psychologists use to analyze the problem handling process, we discuss some aspects that can be used to distinguish several kinds of problems. The distinctions should not be considered as dichotomies, but should be seen as two ends of a continuum. Many problems will have aspects that are somewhere along the line of the continuum. Making distinctions between problems based on these aspects is not always exclusive, since there is often an overlap. Distinctions between problems can be made based on:
the presence of the problem:
- present and future problems
At any present time there can be new and old problems. The new problems present themselves for the first time, whereas the old problems have already been there for some time. Future problems are problems that are not causing trouble at present, but will do so in the (near) future if the situation does not change. It is sometimes very important to recognize future problems in time, when there is still a possibility of preventing the problem or of preventing the problem becoming worse, with relatively small effort and means. Where the problem is inevitable, there is more time to think the problem over.
- latent or manifest problems
Latent problems are problems that are potentially present but not yet manifest. They can become manifest if there are certain changes in the environment. Manifest problems are problems that are actually apparent.
- structural and incidental problems
'Structural' and 'incidental' problems refer to types of presented problems. Structural problems are problems that interfere with the structure of some (part) of society. Structural problems involve many societal phenomena. Unless a more fundamental change in society occurs, these problems remain. Structural problems require a more fundamental approach and a more permanent solution than incidental problems. Incidental problems are problems that occur by a certain concurrence of circumstances, often by coincidence, and often occur onceonly.
the definition of the problem
- well-defined, well-described and well-structured problems and undefined,
ill-defined, unstructured and ill-structured problems
In research about problem solving one encounters problems defined as well-defined, well-described and well-structured problems and others as undefined, ill-defined, unstructured and ill-structured problems.
A well-defined problem according to McCarthy (1956, p. 73) is:
"A problem proposed to an information processing system is well defined if a test exists, performable by the system, that will determine whether an object proposed as a solution is in fact a solution. By performable we mean, more specifically, performable with a relatively small amount of processing effort."
Crombag
(1984, p. 7) states that well-defined problems:"are problems of which there is one good solution upon which experts agree or can agree, and there is at least one good way of solving the problem upon which experts agree or can agree."
By the concept 'well-defined', these authors indicate those problems for which it can be proved that the solution is the right solution. Duyne (1983, p. 279) states it this way:
"Well-defined problems ... are problems of which the correct solution can be tested by the problem solver by means of an irrefutable test."
In addition to 'well-defined', other words are used to indicate more or less the same phenomenon (e.g. Reitman, 1965; Simon, 1973).
McCarthy (1956), Newell & Simon (1972), and Duyne (1983) mention the term 'well described problems'. 'Well described' problems refer also to the solution of the problem. Well-described problems are problems of which the problem solver himself or herself, based on an unambiguous and indisputable test or criterion, can see whether the solution is correct, as in the solution of logical and arithmetic questions (Newell & Simon, 1972). Other researchers use the word 'well-structured' instead of 'well-defined' or 'well-described' to indicate more or less the same thing.
Well-structured problems are also directly related with a provable solution. A well-structured problem can be defined as (Sol, 1991, p. 28):
"- the set of alternative courses of action or solutions is finite and limited;
- the solutions are consistently derived from a model system that shows a good correspondence;
- the effectiveness or the efficiency of the courses of action can be numerically evaluated."
In reports on problem solving research there are also several words to indicate the opposite of well-defined, well-structured and well-described. Terms used to indicate the opposite are: 'ill-defined', 'undefined' or 'not-defined' problems and 'ill-structured', 'unstructured' or 'not-structured' problems. These terms are often used to indicate a residual category of problems that do not fit into the definition of well-defined, well-structured and well-described problems.
According to Simon (1973) all problems represented in reality are ill-structured problems. They become well-structured problems only in the process of being prepared for the problem solvers. Wierda (1991, p. 28) agrees that many real world problems can, at best, be regarded as ill-structured.
Comment:
The above definitions of well-defined, well-structured and well-described problems make a direct connection between definition and solution. They indicate that there is one or more (good) solutions that can be proved. Using such definition, however, excludes many complex interdisciplinary societal problems from this category. What we mean by a definition of a problem in this study has a rather different connotation. The definition of a problem is explained in section 2.1.2. We do not connect the definition of a problem with its solution, not even a good definition of a problem. Even when a problem is defined, it may sometimes be impossible to find a solution, let alone a solution that can be proven. This is often the case not only with complex interdisciplinary societal problems, but also domain related well and strictly defined problems for which experts do not know a solution. An example of this would be the problem of the traveling salesman. For this reason we do not subscribe to the above definitions of well-defined, well-structured or well-described problems. We make a clear distinction between defined and solved. We use the word 'defining' solely to indicate the conceptual model of the problem.
At the start of the problem handling process, many complex interdisciplinary societal problems are undefined in the sense of not (yet) defined or ill-defined. Complex interdisciplinary societal problems are often very hard to define. Due to the complexity, the often missing data and unknown phenomena, a complex interdisciplinary societal problem is seldom completely defined. An important distinction between the problems we focus on in this study and problems studied in cognitive psychological research on problem handling, is the difference between defined and undefined problems. The complex interdisciplinary societal problems are mostly all undefined, whereas the problems that cognitive psychology focuses on are mostly already defined problems.
Whether a complex interdisciplinary societal problem is defined well enough or completely enough will depend on interpersonal judgment, based on mutual agreement. There is no objective measuring instrument. It can only be stated that in the given situation and given the interpersonal judgements, the problem is defined as well as possible.
the range of the problem
- domain problems and interdisciplinary problems
The distinction is based on a problem belonging to one domain, or to more than one domain: domain specific problems and interdisciplinary problems. For handling domain specific problems, domain specific knowledge is required. For these problems domain specific problem rules and handling techniques are applied. For handling interdisciplinary problems, knowledge from different domains is required. To analyze these kinds of problems different kinds of problem handling techniques can be used.
- non-complex and complex problems
This distinction refers to the range of the problem, as well as to the 'solution' of the problem. There are domain related non-complex and complex problems just as there are interdisciplinary non-complex and complex problems. The complexity makes it more difficult to handle the problem
the dynamics of the problem
- static and dynamic problems
A static problem does not change with time, whereas a dynamic problem changes as time progresses. Both these kinds of problems can be embedded in either a static or a dynamic environment.
the solution of the problem
- open problems
With open problems the problem handler cannot conclude, on the basis of an unambiguous and indisputable test or criterion, whether the solution is the right solution. The problem handler does not have univocal criteria or good utilization procedures that can test the solutions. Duyne defines open problems as (1983, p. 279):
".. problems for which such an (irrefutable) objective test is not available i.e. the problem solver cannot determine whether his solution is the correct one"
Open problems include such problems as making a travel schedule or composing a piece of music. Duyne (1983) points out that the group of open problems is very large
.Research on these kinds of open problems has been done by Eastman (1969) who, based on thinking-aloud protocols, analyzed the way architects handle a problem. Goor & Sommerfield (1975) analyzed creative problem solving. Bahle (1936) analyzed, based on introspection, the creation of musical compositions, Enzensberger (1965) focused on analyzing the process of writing poetry.
- easy and difficult problems
The words 'difficult' and 'easy' often refer to the amount of time, knowledge and effort it takes to handle a problem. Insolvable problems are often considered as difficult problems, because they have no solution. Solvable problems can be either easy or difficult.
An easy problem is often a problem that can be handled rather quickly and satisfactorily.
When it takes a longer time or more effort to achieve the desired situation, a problem is often referred to as a difficult problem. 'Difficult' and 'easy' are not always absolute viewpoints, it can also depend on who is dealing with the problem
- already solved problems and problems never solved before
Already solved problems can still be new problems for a particular person. In her book on creativity, Boden (1990) distinguishes two kinds of creativity, one for handling already solved problems and one for solving new problems that have never been solved before. In this regard, Boden speaks about P(sychological)-creativity and H(istorical)-creativity. For problems that have already been solved by someone, P-creativity is used and for problems that have never been solved before, H-creativity is used. Seen from a personal point of view, P-creavity is just as unique and new as H-creativity. Seen from the point of view of society, P-creativity is not unique and new, only H-creativity is.
Sometimes problems are already solved and even well-known in one field of science, while for researchers in another field the problem is new. Working on these problems these researchers may have the idea that the problem has never been solved before
Some problems have never been solved before. These may be unique or new problems, as for example a new problem in an old situation or an old problem in a new situation. Many complex interdisciplinary societal problems fit this last category.
- urgent and non urgent problems
This refers to the 'solution' of the problem. Urgent problems require direct intervention, which demands a special approach compared to problems that are not so urgent. Often the handling of urgent problems is directed at the effects, not the causes. A riot is an example of an urgent problem, whose effects are often handled rather than the cause. In order to prevent a repetition of the riot, however, a more structural intervention directed at the cause should be made.
For less urgent problems there is more time to think things over. Whether a problem is urgent or not is not only an objective fact. It can also depend on a person's view or position.
To summarize, we have discussed above distinctions between problems based on:
the presence:
- present and future problems; latent and manifest problems; structural and incidental problems
the definition:
- well-defined, well-described, well-structured and undefined, ill-defined, unstructured and ill-structured problems
the range:
- domain and interdisciplinary problems; non-complex and complex problems
the dynamics:
- static and dynamic problems in a static or dynamic environment
the solution:
- solvable, insolvable and changed problems; open problems; easy and difficult problems; already solved and problems that have never been solved before; urgent and non-urgent problems
Most of the societal problems we focus on are complex interdisciplinary dynamic problems imbedded in a dynamic environment. Society and the societal problems are intertwined and continually influence each other. Sometimes these problems are structural problems, sometimes they are incidental. They have often never been solved before, or have never been solved in this situation, and they are often undefined. These problems can mostly be solved only partly and/or temporarily, and can seldom be solved completely. Because of the complexity, the uniqueness and the interdisciplinarity, they can be considered as difficult. Sometimes the problems are open problems, sometimes the goal is defined precisely.
Complex interdisciplinary societal problems can have all kinds of different time aspects. The problem can be structural or incidental, urgent or not so urgent. It may be a present or a future problem
, a latent or manifest problem.The problems can be viewed from the micro-, meso- or macro aggregation level. The levels of aggregation will be discussed further in chapter three.
Many complex interdisciplinary societal problems create or are bound to create an unwelcome situation. In this sense these problems cause a discrepancy between the actual situation and the desired (future) situation. These problems are also problems in the second sense of the word 'problem'. There is often a discrepancy between the actual state of understanding a phenomenon or set of phenomena and the desired state. This not only refers to cognitive knowledge or technological knowledge but also to a lack of knowledge about what the problem looks like, a lack of data, a lack of know-how on how to handle the problem, a lack of knowledge on what kind of interventions can be performed and even sometimes a lack of knowledge about the desired state.
We can now alter our definition of a (general) problem to one that also includes a complex interdisciplinary societal problem:
something is called a problem when there is a discrepancy between the actual or (near) future situation and the desired future situation and/or there is a lack of knowledge and/or a lack of know-how, and/or a lack of relevant data; as for complex interdisciplinary societal problems, the problem is often undefined and the actual and the desired situation is not always clear
The distinction between kinds of complex interdisciplinary societal problems can be made in several ways. We make a distinction between problems that are primarily knowledge problems, problems in which two or more parties with different interest are involved, and urgent problems. Problems with different parties often also have a problem of knowledge. Urgent problems often include a lack of knowledge in combination with parties with different interests and a time problem. In this study, attention is mainly directed to knowledge problems, which will be regarded from a policy-making (macro) point of view.
2.3 When is a problem solved?
In research on problem handling, a distinction is made between solvable and insolvable problems, although often it is not a dichotomy but a continuum. It is not always clear in advance whether a problem is solvable, insolvable, or not yet solvable, partly solvable or only changeable. It can be stated that the problem is solved when the discrepancies between the actual state and the desired state have disappeared. Whether or not this is a fact depends on the definition of the problem and the definition of the desired situation. A problem is considered solved when people agree that the problem is solved. This can be based on scientific proof or on a subjective opinion. Defining a problem is a crucial point in problem solving. Some ways of defining a problem make it impossible to solve the problem, while another kind of definition can easily lead to a solution. A problem may appear insolvable only because it is defined in such a language that it seems insolvable
(Van Dijkum, 1988).When a problem is called insolvable it often means that the problem is not solvable within the constraints or in this situation or time period. However when the boundaries are widened then sometimes an 'insolvable' problem can be 'solved'. Knowing how a problem can be solved does not mean that the problem will be solved, however, there can be many reasons, financial and political or pragmatic why a problem whose solution is known is not solved.
For complex interdisciplinary societal problems it is often difficult and sometimes impossible to know or to implement the 'complete' solution(s) to the problem. Many interdisciplinary societal problems are not solvable in the sense that there is one or more good answer(s) and final solution(s) to the problem. These kinds of problems can at best be 'solved' only temporarily. Therefore we prefer to talk about 'handling' a problem rather than 'solving', and to say that a problem is changed instead of solved. By change we mean that the problem is not solved or has not disappeared, and yet the problem is not the same as before, some aspects of the problems are altered.
A factor that complicates the handling of problems is that there are often different parties involved. A problem changed to the satisfaction of one group, can initiate new problems for another
. It can also be the case that in a conflict only one party thinks that the problem is solved, while other parties still consider the problem unsolved.Sometimes a problem may appear solved, but subsequent changes make it clear that the problem was not completely solved and still remained potentially present. Looking at history we find that many problems which once were thought finally solved had to be reconsidered. These problems are not handled once but over and over again. For many complex interdisciplinary societal problems there are no solutions in the sense of a final and an objective answer (Rittel & Webber, 1973). Problems are, at most, resolved over and over again.
Sometimes a problem is not solved but forgotten. Nobody refers to this problem any more, it simply 'disappears' from view or is hidden or fades away due to changing circumstances. Sometimes it is not possible to solve the problem directly. Circumventing the problem can be a way to handle the problem.
Some problems are solved by coincidence. While working on some problem, or performing some other activities, suddenly, as if by coincidence, a solution or a good intervention can be found. This is called serendipity.
Most complex interdisciplinary societal problems will only be changed. Such change could make it easier to cope with the problem, but is not necessarily so.
Crombag (1984), in his research on the way physicians and judges handle their ill-structured problems, found that experts often disagree on the best solution of an ill-structured problem. Sometimes there are contradicting criteria or at least different criteria to judge whether a problem is solved or not. The disagreement among experts concerning the best solutions to these kinds of problems does not mean that all solutions of the problem are evaluated as equally possible. There is indeed an area in which there are some solutions on which the experts could have different opinions, but that area is smaller than the area on which the experts agree that the solution cannot be found there. It seems that the more important the problems are, important in the sense that they have a greater impact on society, the more these problems demand some kind of solution, the greater is the chance that they can be categorized as undefined problems (Crombag, 1984). A large number of problems that physicians or judges are confronted with daily can be categorized as ill-defined problems. Crombag (1984) states that the higher the degree of interdisciplinarity, the greater the chance that there is some disagreement on the solution.
2.4 When is something a problem
Before a problem can be handled, it must be recognized as a problem. This means awareness that there is a problem. This does not mean that the problem is already defined or labelled. In the beginning it will be enough to realize that there is something wrong, that there is a question that something happens or is going to happen that will cause a discrepancy between the actual or (near) future situation and the desired (future) situation. Sometimes this is based on a vague notion, sometimes one is already confronted with the consequences of the problem. Sometimes a problem has already been there for some time before it is noticed by someone or before it is put on the political agenda
.2.5 To whom is it a problem
Problem handling has two aspects that are closely related to each other. One aspect is the problem itself, the other aspect is the problem solver. Not all problems are everyone's problems. The same issue can be a problem for one group, while at the same time it is no problem at all for another group, although people may disagree about this.
With regard to complex interdisciplinary societal problems, those for whom it is a problem, 'the victims' of the problem, are often not the same people who are (held) responsible for handling the problem.
The question of to whom something is a problem connects a person, group, an organization or a society with a problem. Wierda (1991) speaks in this regard of a problem 'owner'. The 'owner' can be a human being, an organization, a group who 'owns' the problem.
2.6 Interventions
Interventions are deliberate changes towards the problem, with the intention of reducing the gap between the actual situation and the desired situation. When something is recognized or signalled as a problem the question arises as to whether some interventions towards the problem should be taken, or whether the problem should be ignored.
The answers to these questions depend on many things, such as the importance of the problem, the urgency of the problem, time, money and the possibilities of how a problem will be handled
.2.7 Who should handle the problem?
Once it has been decided that the problem should be handled the question arises of who should handle the problem? In politics there are many discussions about whose problem it is, who should handle the problem and who is responsible for handling the problem, varying from total governmental responsibility for many kinds of problems, which, to a large extent, was the case in large parts of the former U.S.S.R., to a society that does not want too much intervention by the government, as is the case in the U.S.A.. How these matters are approached will depend on the political construction of society.
2.8 Research question 1a
The research question we want to answer is:
1a Is handling complex interdisciplinary problems so difficult because of the special character of these problems?
In this chapter we analyzed several aspects of the character of a complex interdisciplinary societal problem. In subsequent chapters we will further aspects.
The discussion in this chapter leads us to conclude that the combination of the many aspects implicit in the special character of complex interdisciplinary societal problems make it difficult to handle these kinds of problems.
To summarize:
Once aware of a problem, the process and the 'solution' of the problem is difficult. Defining these problems is difficult, firstly because the knowledge and data necessary for definition are missing or incomplete, further because they are embedded in, and interact with, a continuously changing environment which means that the problem is continuously changing. This is the reason the problem can only be defined temporarily. The problems have seldom been 'solved' previously, and can mostly only be temporarily changed rather than permanently solved. Finding out who is responsible for handling can be a source of conflicts.
2.9 Research on problem solving
To be able to discuss the second research question we explored some of the research on problem handling in the field of cognitive psychology. The second research question is:
1b in what way do the problems that are studied in the field of cognitive psychology differ from complex interdisciplinary societal problems and what are the similarities that are relevant for analyzing and defining complex interdisciplinary societal problems
In this chapter we will discuss the first part of the research question:
1b in what way do the problems that are studied in the field of cognitive psychology differ from complex interdisciplinary societal problems
In order to answer this question we studied some of the ideas of researchers which we feel have contributed to contemporary ideas on problem handling. We selected:
- some of the researchers that influenced the development of theory on problem solving in Anglo-American countries and in Western-Europe
- some of the researchers who conduct research on undefined problems
We have selected only some of the researchers as an example of the range of ideas on the subject.
2.9.1 Selz and Bahle
Otto Selz (1881-1943) was a researcher on problem solving in the first half of the 20th century. Selz was influenced by the German phenomenological tradition in the humanities following the hermeneutic cycle in psychology. Selz considered the human being as an information processing system.
To Selz, the brain and what happens in there was a black box. The only way to know what is going on in the mind is to have people state this in retrospection. In order to find out what is happening in the brain he used the thinking-aloud method, in which people state what they are thinking.
Based on this research, Selz draws conclusions about the methods people use in problem solving. Selz (1922) demonstrated that thinking was not a passive process, as was generally believed at that time, but a continuous activity that can be described as a chain of consecutive mental operations. Selz considers problem solving as finding the missing parts of an incomplete scheme. A scheme is the idea one has about something. Finding the missing parts is a rule-bound process, whether it concerns a simple thinking process or a complex goal-oriented thinking process using all kinds of problem solving techniques. Selz distinguishes three ways of finding the fitting parts of the incomplete scheme with which a person begins the thinking process
.Within the problem solving process, Selz (1922) distinguishes reproductive thinking, productive thinking and seeing the solution based on intuition and experience.
1 In reproductive thinking, the mental operations of effective strategies of former problem solving methods are now applied to comparable problems. Most skills are based on this kind of economical problem solving methods (Snoek, 1989, p. 119). Most of them are automated problem solving methods.
2 Productive thinking is used when there is no ready-made solution at hand. Here more complex ways of thinking are involved, although Selz stated that complex problems can also be solved in a reproductive way. The concept of 'productive thinking' is close to the concept of 'creative thinking' that Boden (1990) uses.
Selz distinguishes two methods of productive thinking: the method of problem solving 'Mittelfindung' and applying the method 'Mittelanwendung'. 'Mittelfindung' can be translated as 'finding a method', 'Mittelanwendung' can be translated as 'applying a method'. 'Mittelfindung' can consist of an actualization of know-how, how to solve a problem. Which can be at the conscious and/or on the unconscious level. It can entail different ways of abstraction. The result of an abstraction can be schematically anticipated as a still unknown problem solving method that can be successfully applied.
These abstractions can recall 'old' reproductive problem solving methods, and can take place by coincidence or without purpose. 'Old' reproductive problem solving methods can be recalled when a connection between two items is known but where using the first object to influence the latter has never beem thought of. Solving by coincidence is the case when during the process of thinking about the problem, by coincidence, the solution is suddenly 'seen'.
3 'Seeing' based on intuition and experience happens especially with experts. A medical expert sees intuitively in what way the problem can be solved, as does the chess-master in the research of De Groot (1946), and as is reported in the studies of Crombag (1984) and Snoek (1989).
Selz calls this the 'unmittelbare Mittelabstraktion'. Unmittelbare Mittelabstraktion can be translated as using directly the method without 'linking'. Experts sometimes directly 'see' the solution intuitively, without applying rules.
Selz considers thinking as an active process. In this he uses the information paradigm. He states that the thinking process is supported by applying rules. Bahle (1936), a student of Selz, emphasizes the thinking process and the importance of the goal in problem solving in his research. Bahle focuses on open problems. He found that ideas about the goal may change during the problem solving process. He discovered this by analyzing the creative process of composers. When a composer compares his/her temporary results with the main idea about his/her creation, the result could be rejected or accepted but it could also end in a modification of the goal itself. Bahle discovered the possibility of the changing goal in a creative process. Like Selz, Bahle regarded the creative thinking process as a linear, sequential chain of coherent operations.
2.9.2 De Groot and problem solving in chess
De Groot, a well-known Dutch psychologist, based his ideas on the information paradigm of Selz. De Groot analyzed the problem solving methods of chess-masters by using the thinking-aloud method (De Groot, 1946). In this research he demonstrated that it is, in general, possible to reconstruct the reasoning process.
He discovered that the goal of a chess player is not to find the best solution, but to find a rational good move. The goal is not considered as a static concept, but the goal is influenced by the positive or negative effect of the problem solving methods applied. The schematic anticipation changes continuously. The conception of the total goal can, as a result of temporary solutions, be dramatically changed via a process of continuous feedback. The process of continuous feedback is also important in creative thinking. De Groot confirmed the ideas of Bahle about the changing goal in the problem solving process.
De Groot distinguishes several phases in chess problem solving:
- The first phase is the orientation phase, mostly based on observation. The chess player distinguishes:
the static situation of what this situation looks like;
the dynamic situation of what can happen;
an overall evaluation.
This results in a first conception of the problem.
- The second phase is when the player analyzes several possible moves, mostly based on two alternative plans of which one is preferred.
- In the third phase there are systematic operations to find proof of the
correctness of the choice. Here the chess player uses many automated, often unconscious problem transformations that make it possible to handle the very complex system of problems and sub-problems.
- The last phase is the phase of the proof. In this phase the chess-player checks the correctness of his move.
During these phases there is often time for reflection during which the chess player considers the situation. In these phases 'Mittelfindung' takes place. By returning to the original problem and evaluating the results, the chess-player comes to a new or modified definition of the problem.
Some comments:
Is it possible to reconstruct the thinking process a person uses in problem handling? Much cognitive psychological research on problem solving is based on the thinking-aloud method that is used to analyze the thinking process. However, the thinking-aloud method has certain constraints (Crombag
, 1984). The assumption of the thinking-aloud method is that human beings are able and willing to say all they think. It is not clear whether and to which extend this is true. In Snoek's (1989) research on how medical professionals handle problems, the thinking-aloud method was combined with video recording. Snoek concluded that the method could be used to analyze the problem solving process, although it has its limitations. De Groot (1965, p. 79) writes of the thinking aloud method referring to Selz:"This deviation from the old methodology was an important novelty. By this relatively simple technique it appeared to be possible experimentally to investigate and systematically to analyze complicated, lengthy processes - provided the experimenter is content with a somewhat rougher, more macroscopic overall picture of the thought process."
De Groot (1965, p. 79) continues by summarizing the reasons for incompleteness of data in the protocols. Some thought steps:
"......remain under the threshold;...thoughts move so quickly;.....not all thought is immediately formulatable thought;....the subject may intentionally suppress one or more steps.."
De Groot concludes:
"All four mechanisms may lead to incompleteness in a protocol...."
"'Incompleteness' refers to the possible gaps in the reasoning process as it develops. On this basis there are two criteria for (in)completeness: first the degree to which the subject is satisfied that his protocol is a rendering of his thought processes; second, to the experimenter and any other expert reader, the understandability of the protocol text, as a gradually developing argumentation leading up to a certain choice." (De Groot, 1965, p. 82-83)
In referring to his own data gathered based on the thinking-aloud method, De Groot states (1965, p. 83):
"These criteria were in fact not always met, i.e., gaps did occur here and there."
A thinking-aloud protocol often contains considerable ambiguity. The verbal reproduction of the thinking process need not automatically be the original thinking process. We believe that the verbal report of the thinking-aloud method is not identical to the way of thinking. The direct way of thinking is often very diffuse and associative and disrupted by other thoughts (Ericsson & Simon, 1984), whereas the verbal report of the thinking process is more a selected interpretation by the person of his or her thinking process, selected and interpreted towards the goal of the process: the process of problem handling. However, we consider a combination of the thinking-aloud method together with a reconstruction of the reasoning process a method which, in some cases, results in interesting material.
2.9.3 Newell & Simon's theory of problem solving
In the field of cognitive psychology research on human problem solving, the work of Newell & Simon has a central place. Many researchers have based their problem solving theory on the theory of Newell & Simon, set out in their book 'Human problem solving' (1972). We will discuss some issues of this theory:
thinking as information processing, the problem space, defining a problem, and the kind of problems Newell & Simon use in their research.
2.9.3.1 Thinking as information processing
Newell & Simon regard the process of thinking as information processing. They postulate that the human being operates as an information processing system (IPS) (Newell & Simon, 1972, p. 19):
"Our theory of human thinking and problem solving postulates that the human operates as an information processing system."
Newell & Simon consider an information processing system as:
"an abstract system to describe how man processes task-oriented symbolic information" (Newell & Simon, 1972, p. 19-51)
A definition of an information processing system (IPS) is:
"An IPS is a system consisting of a memory containing symbol structures, a processor, effectors, and receptors." (ibid, p. 20)
By defining thinking as information processing, they use the way a computer functions as a metaphor for human problem solving.
The idea of considering human thinking as information processing can be found among many cognitive psychologists, before and after Newell & Simon, for instance Selz (1922), Rumelhart (1984), Dörner (1974) and Wierda (1991). Using technical inventions, in this case a machine metaphor, as a metaphor for thinking, is a way that is often used to make a 'black box' procedure intelligible
.Newell & Simon regard problem solving as an information processing and goal seeking activity. To them, problem solving is a special way of information processing in which one tries to reach goal(s) without initially knowing how to do this.
2.9.3.2 The problem space
The problem space is a central issue in the theory of Newell & Simon. Newell & Simon consider problem solving as a search within a problem space. The problem solving behavior takes the form of a search through a space, known as the 'state-space-search paradigm'.
The problem space consists of a (defined) problem, the initial state, knowledge, a search-path (the method and the direction to solve the problem) and a goal (the intended solution to the problem). Newell & Simon explain this by means of the following example. In an experiment a problem is presented with a set of instructions and a sequence of stimuli. The person must encode these problem components - defining goals, rules, and other aspects of the situation - in some kind of space that represents the initial situation presented to him, the desired goal situation, various intermediate states, imagined or experienced, as well as any concepts the person uses to describe these situations to him or herself. This does not imply that all behavior relevant to problem solving is a search in a problem space. Newell & Simon say about this (Newell & Simon, 1972, p. 809):
"Initially, when a problem is first presented, it must be recognized and understood. Then, a problem space must be constructed or, if one already exists in the LTM, it must be evoked. Problem spaces can be changed and modified during the course of solving. These activities, crucial to problem solving, need not themselves be searches in a problem space."
The problem space is not a stable concept with fixed definition. It can change or be modified during the development. It is not a space that can be pointed to or described as an objective fact. Newell & Simon define a problem space as an number of possibilities for interventions. The problem space contains not only the actual solution but also possible solutions the problem solver might consider. Boden (1988, p. 152) describes the problem space as:
" The problem-space is defined in terms of the solver's representation of the problem, and it comprises the set of all problem-states that could possibly be reached by the available operators. An operator is a way of getting from (or transforming) one problem-state to another. Even a small set of operators can generate a problem-space too large to search exhaustively...........Rules of thumb, or heuristics, were used to guide the search in directions likely but not guaranteed to lead to success."
An example of the problem space given by Newell & Simon (1972, p. 59) is
a two-person game number scrabble with nine numbers from one to nine. The players draw pieces alternately from the set. The first player who holds any subset of exactly three pieces, from those he has drawn, with digits summing to fifteen, wins. Draw after draw the problem space becomes smaller. Another example of a problem space is the problem of the 8-puzzle (Luger & Stubblefield, 1989/1993). Here the problem space is limited to a block containing nine places. Eight places contain blocks that are numbered from one to eight; one place is open. In the initial state the blocks are in random order. The goal state is ordered blocks from one to eight. The operations allowed are moving with the blocks in the up, down, right and left directions. After one move one reaches the first intermittent state. This state can be compared with the goal state. If the goal state is not yet reached one can continue. After the second move one comes to the second intermittent state. As long as the goal state is not yet reached these operations can continue in order to arrive, via the intermittent states, at the goal state.
2.9.3.3 Defining a problem
In the view of Newell & Simon a problem is defined in terms of a set of final desired states to be reached with operators that are available. For Newell & Simon, having a problem implies (at least) that certain information is given to the problem solver: information about what is desired, under what conditions, by means of what tools and operations, starting with initial information, and with access to resources. The problem solver has an interpretation of this information, exactly that interpretation which enables the person to label some part of it as goal, another part as side conditions and so on. As a consequence, if a person provides a representation for this information (in symbol structures), and if one assumes that the interpretation of these structures is implicit in the program of the problem solving information processing systems (IPS), the problem is defined.
2.9.3.4 The kind of problems
What kind of problems do Newell & Simon deal with in their research? Newell & Simon combined cognitive psychology with computational science. In the research of Newell & Simon the objects of research were mainly proofs of theorems or simple English sentences. They use problems like game players, theorem provers, puzzle solvers, chess problems, symbolic logic and algebra-like puzzles, called crypt arithmetic puzzles, problems that, on a large scale, are used in research on problem solving in Artificial Intelligence.
Newell & Simon use problem handling tasks that:
- are short (half an hour). Thus in the example they give of a chess game, each state lasts a matter of seconds, the whole game a matter of ten minutes
- have moderately difficult problems of a symbolic nature.
Long-term activities extending over periods of days or years received no attention in their research.
The problems Newell & Simon used for analyzing the human capacity of problem solving were not new, neither were they very complicated or difficult. They had already been solved by many persons. Newell & Simon's focus is not on how to handle a particular problem, but on the way human beings handle a problem. Their focus is on the (individual) human being.
The actions Newell & Simon selected for proving their theoretical ideas are mainly manipulations of symbol structures. Crucial activities in solving these problems are symbol manipulating activities combined with physical actions.
Their theory is not verified by control groups, but by analysis of verbal protocols. However it is difficult to test dynamic, history-dependent systems.
Newell & Simon never claimed to have said the last word on problem solving. They know their limits, saying (Newell & Simon, 1972, p. 1):
"Not all things can be done at once, and even if one aspires to go far, he must start somewhere. If one aims at covering all of human thinking in a single work, the work will necessarily be superficial. If one aims at probing in depth, then many aspects of the subject, however important, will be left untouched."
Further they state that (Newell & Simon, 1972, p. 7):
"There are many kinds of thinking that one might like to study: designing a house, discovering a new scientific law, preparing a law case, arguing over political parties, creating new music, day dreaming while watching the clouds, preparing a five year economic plan, and so on. Detailed theories of these and many other kind of thinking are largely beyond the current state of art. Of course, there have been investigations into some of these areas, many of them still in midstream. Only their incomplete state and our limits of space and energy have inhibited us from including some of them in this work, since they are in fact part of the same story we like to tell."
In these quotations they defend, on the one hand, the limitations of their research, while, on the other hand it could be concluded that their ideas about problem solving are also legitimate for the way many other kinds of problems could be solved.
Support for the idea that Newell & Simon believe that their theorem is applicable to all kinds of problems can, already, be found in the development of the General Problem Solver (GPS) (Newell & Simon, 1961, 1963b)
. The General Problem Solver is a computer program developed by Newell & Simon that is supposed to be able to solve all kinds of problems. Until the end of his life, Newell was convinced that his ideas were correct. This can be illustrated by his second attempt, some decades later, to build a computer program that can solve all kinds of problems: the program SOAR (Newell, 1990). Both programs are based on the idea that there is something like a universal theory of cognition.Although Newell & Simon suggest, in the above stated quotation, that their theorem is probably universal and applicable to all kinds of problems, we believe that research must be done on all kinds of problems before we are able to prove this statement.
2.9.4 Reflecting these theories of problem solving
Reflecting on the ideas on problem solving of Selz (1922), we see that whereas Selz employs the term 'thinking' for indicating problem handling, later researchers use the more general term 'problem solving' in which they include, in addition to thinking, also other problem handling activities. In our opinion, although very important, thinking is only one of the instruments for problem handling. Problem handling may consist solely of thinking but often implies more than that. Although problem handling is an active process it can be a conscious and/or an unconscious process, in which coincidences also play a role. In accordance with Bahle and De Groot, we agree that the goal can be reformulated, modified and altered during the problem handling process. New information can make it necessary to change the goal. Although it is goal directed, the goal does not have to be identical to the solution. The goal can also be another phase in the problem handling process.
In their book 'Human problem solving' Newell & Simon (1972) do not reflect on changing goals. In the problems they use for their research, the goal of the problem is not disputable. The goal is clear, mono-interpretable and undoubtedly specified beforehand.
In the theorem of Newell & Simon, two important issues stand out. They assume, in accordance with the above stated researchers, that the human being is an information processing system and that thinking is a matter of information processing and problem solving can be a combination of thinking and activities. Secondly they assume that problem solving is a search in a problem space.
2.9.5 Our ideas about problem handling
Although information processing is a useful metaphor for the way people think, we do not subscribe to the idea that the brain functions the way a computer works according to the principle of Von Neumann, or even with the modern metaphor of parallel computing and neural networks. We think that information processing is influenced by many processes. In addition to the linear process there are many moments of thinking that are associative, non-linear, sometimes more or less parallel, circular and reflecting. The moment of thinking also includes feedback.
We subscribe to the ideas of Newell & Simon that the human being is not a tabula rasa. The person who handles a problem already has knowledge, data and knows some problem handling strategies. But there is more than knowledge, data and rules involved. The information that is going to be processed is selected, based on attention, motivation and knowledge that is already available.
The way information is to be processed depends on the internal state of the human being (hunger, anxiety, sleep), the motivation of the thinking process, and the context of the thinking process. It is an associative and selective process, which is time, circumstantial, motivational and context dependent, rational and irrational. Information processing can also occur unconsciously. Thinking does not always have to be goal directed, it can also be diffuse, undirected, direct or associative, as well as rational and irrational, on purpose or not, jumping back and forth between ideas in which the processing data can come from inside and from outside. Thinking and problem handling also contain associative thinking (Selz, 1922) creative thinking (Selz, 1922 ; Bahle, 1936), forgetting (Freud, 1901/1982) and memorizing.
Comparing the human being with a computer limits the way of conceptualizing problem handling. It influences, for instance, the selection of problems. As we have seen, the problems on which Newell & Simon base their research, are domain related, well-defined, small, more or less artificial, with a well-defined problem space. These are the kind of problems that were later used by Artificial Intelligence researchers.
Another important idea in the theory of Newell & Simon is the idea that problem solving is a search into a problem space: the state-space-search paradigm. A problem space, as defined by Newell & Simon, is the result of analyzing the instructions and stimuli, and encoding the problem components in defining the goal, deciding about the rules, describing the initial state and the desired state including various intermediate states. In the experiments of Newell & Simon the initial state is described for the subject. From this description, the subjects have to separate the relevant task aspects from the irrelevant task aspects. In the experiments of Newell & Simon, this was not so hard to do. The state-space-search paradigm suggests that the problem can be solved by a search in a space in which not only the direction but also the goal of the desired state is clear. The problem space already includes the solution. It also includes some operators that can be used to reach the goal.
Newell & Simon's ideas about problem solving are primarily valid for the small structured well-defined artificial problems they used in their research. For those problems it is easy to define the problem space. The problem space is easier to find in an artificial situation, like a chess game, or the 8-puzzle, than in real life problems.
Wierda (1991) states that the nature of problems studied by Newell & Simon (1972) in their research on human problem solving differ immensely from the design problems that are contemplated in Wierda's research on inter-organizational information systems. The problems Wierda uses are ill-structured problems about inter-relational information systems.
For most complex interdisciplinary societal problems the problem must first be defined, which can be very difficult. Defining the problem can be compared with the initial state in the theory of Newell & Simon. With many problems the problem handling 'rules' are unknown, even the language in which the problem can be formulated and the model which can represent the problem is unknown. The knowledge needed to handle the problem adequately is missing or incomplete, often even the goal - the final state, what it should look like, in which direction the problem should be handled - is uncertain.
In real life, the problem space of many complex problems is unknown. Defining the problem space is an advanced phase in the problem handling process. Even knowing the goal and the operators will not guarantee that a solution is possible. For complex interdisciplinary societal problems finding each component of the problem space is complicated. Each component can take many hours or years and many people to analyze.
Nevertheless, Simon believes that the state-search-space paradigm can also be applied to complex and ill-defined problems. Simon (1973) has a reductionist view on problem solving. He states that the difference between a complex and ill-defined problem and a well-defined problem is a matter of reducing the complexity. He claims that ill-structured large-scale problems can be well-structured when broken up into their constituent sub-problems. He demonstrates this assumption with examples from chess and from designing large constructions like buildings and battleships, in which domain strategies can be identified that reduce large ill-structured problems to many structured ones. In this way an ill-structured problem can be changed into a well-structured problem by dividing the complexity into small parts, solving these small parts and putting them back again. In the reductionist view, complexity is only more of the same thing. The method of dividing a problem into sub-problems is often used in many problem solving strategies and not always without success. Some problems can be divided into parts and then reassembled again, but, these are mostly static problems in a static environment. On the other hand, complex interdisciplinary societal problems are dynamic problems imbedded in a dynamic environment. During the time one handles the parts of the problem the whole problem may be significantly changed.
The reductionists' view neglects or underestimates the interrelations between the various well-structured sub-problems: there will remain parts of a problem that cannot be simply reduced to a number of smaller sub-problems. Complexity is more than the sum of the sub-problems. Gestalt psychology
earlier drew attention to the fact that the whole is more than the sum of the different parts. Complexity is an essential part of the problem. It is something that does make the problem difficult. We do not think that Simon's reductionist view is applicable to handling complex interdisciplinary societal problems. In our opinion, handling complex interdisciplinary problems is more than cutting the problem into sub-problems, solving these sub-problems and then putting the pieces together again later on. Dividing a complex problem into (small) sub-problems that one can handle leaves out the complexity and the interaction of the various aspects of the problem.Conclusion:
Reflecting on the above described research of Selz, Bahle, De Groot and Newell & Simon, we conclude that, in order to call an activity problem handling, whether this is motoric or mental, there must be a problem and an intention to reflect it. In this sense problems are goal directed, although the goal does not have to be the final goal. The goal can also be a more intermediate state. The concept of a problem space is too limited for complex interdisciplinary societal problems. Defining the problem space as the space in which the solution of the problem can be found can only be fruitfully applied to problems that have a(n) (artificial) space, and have strictly defined boundaries. To phrase this in terms of system theory, the system boundaries for complex interdisciplinary societal problems can only be artificially and temporarily marked out (Flood & Jackson, 1991).
For undefined complex interdisciplinary societal problems it will often not be possible to construct a problem space ín the way Newell & Simon defined it.
For many complex interdisciplinary societal problems one of the difficulties is how to define the initial state, and to know what a possible solution would be. Therefore we prefer to avoid the term 'problem space. Instead of problem space we prefer to talk about handling space. The handling space is the space in which the changes of the problem must be searched for. However, there is no guarantee that the 'solution' will be found there. The concept of the handling space will be explained in more detail in section 3.4.2.
2.10 Differences between problems dealt with in research in cognitive psychology and complex interdisciplinary societal problems
Based on the discussion in this chapter, we are now able to answer part of research question 1b. The part we can answer now is:
1b in what way do the problems that are studied in the field of cognitive psychology differ from complex interdisciplinary societal problems.....
Much of the research on problem handling in cognitive psychology is based on the ideas of Newell & Simon. Some of the differences between the problems cognitive psychologists deal with in their research and complex interdisciplinary societal problems are:
1 The kind of problems.
Most problems cognitive psychology focuses on are small artificial problems, seldom real life problems, whereas complex interdisciplinary societal problems, as focused in this study, mainly involve real life problems.
2 The start and the end of the problem handling process.
In contrast with the research problems cognitive psychology focuses on, which are clearly described, and of which the solution is known, at least to the researcher, for most of the complex interdisciplinary societal problems there is uncertainty about the beginning, the process, and the ending of the problem. For complex interdisciplinary societal problems it is often very hard to know what the solution could be. Often the problem cannot be solved but only changed.
3 The data, the description and the knowledge of the problems.
The data, the description and the knowledge of the problems studied by cognitive psychology are almost always correct and complete, and in addition the problem handling techniques are often known. Where, as with complex interdisciplinary societal problems, many factors are unknown, or there is uncertainty about them, there is uncertainty about which parties, which persons, which organizations and/or which countries are involved, and how they influence each other. There is often uncertainty about which domains are involved. There is often uncertainty about the quality of the collected data needed for handling complex interdisciplinary societal problems. The data may be unreliable, inconsistent, missing or in contradiction with each other. The knowledge of how to handle the problem is often missing or only partly available.
4 The degree to which the problem is defined.
The problem space as Newell & Simon define it, is already defined for the problems cognitive psychology researchers deal with. The problem space is not defined beforehand for complex interdisciplinary societal problems.
In contrast with the research problems in cognitive psychology, most complex interdisciplinary societal problems are undefined.
5 The uniqueness of the problem.
Research problems that cognitive psychology uses are problems that have been solved many times before. Complex interdisciplinary societal problems are often unique, and the experience of handling these problems is often lacking.
6 The implementation of the interventions.
Another difference is the implementation of the interventions. Most research problems are cognitive problems, like a puzzle or a small game. The solution found does not have to be implemented in real life. This contrasts with complex interdisciplinary societal problems. Even when a 'solution' is known on the cognitive level, there may be an impediment to implementing it. There are often political or financial, emotional or psychological constraints that prevent the problem from being handled adequately.
7 The amount of time that is needed for handling the problem.
The research problems in cognitive psychology can often be solved within a short period of time. Handling complex interdisciplinary societal
problems,
however, often takes much more time.
8 The number of people handling the problem.
A very important difference, which has major consequences for the way the problem can be handled is the number of persons needed to handle
the problem. Most of the problems cognitive research focuses on can be handled by one person alone, because the problems are small, simple and domain related. However most of the complex interdisciplinary societal problems need more than one person to be able to handle these adequately.
Some reasons for this are:
- The knowledge of different disciplines that is needed. Interdisciplinary
problems demand knowledge of different disciplines. It is not possible for one persons to have all the knowledge that is needed.
- The complexity of the problem.
Interdisciplinary societal problems on the contrary are often too complex to be handled by one person alone.
- The importance of the problem
The problem is often too important, from societal point of view, to be handled by one person alone, at least not in a democratic society.
- The different parties involved.
Different parties may have different interests. It can be very important from the point of view of the acceptation of the 'solution' that those parties are included in the problem handling process (Glasbergen, 1992).
We have now answered part of research question 1b.
Besides differences, there are also many similarities between the problems cognitive psychology deals with and complex interdisciplinary societal problems. The similarities can be found in the phases of the problem handling process. For both kind of problems the phases of the problem handling process are, in principle, the same. However, because research problems in cognitive psychology have mostly already been defined, the first part of the problem handling process is already performed by the researcher himself or herself. The phases of the problem handling process are described in section 3.2.
Comment:
As we have seen, many of the small everyday problems and most of the important societal problems are complex interdisciplinary problems (Duyne, 1983; Anderson, 1990), which differ a great deal from the kind of problems cognitive psychology focuses on
. Therefore it is a pity that much of the attention of problem solving research in cognitive psychology concentrates on small, domain related problems that have already been solved. What can be the reason for this? We suggest:- Analyzing small problems can be surveyed. It is easier to do research and
to control everything, for instance a laboratory situation, with small problems
than with broad undefined complex interdisciplinary societal problems of which no one can tell what the right answer is (Vennix, 1990).
- Research on complex interdisciplinary societal problems is much more
difficult to conduit according to the rules of the hypothetico-deductive method of De Groot. Many researchers want to do 'safe' research i.e. the researcher takes no risks (Gleick, 1987).
- Some of the problem handling research in cognitive psychology is used to
support the knowledge about learning strategies for education. Most problem solving in (regular) education concerns domain related problem solving. Therefore these kinds of problems are analyzed in research (DeTombe, 1991b).
Another question we could raise is: 'Why is this approach not seen as problematic by many researchers?'. We can only speculate on the answer. The reason could be found in a collective blind spot. One is so used to approaching problems as domain specific that one simply overlooks complex interdisciplinary societal problems. Another, and possibly more fundamental reason for this can be found in what Simon said about complexity, namely, that a complex problem can be handled by dividing the complexity into small parts, handling these small parts and reassembling the parts again. The main reason why these kinds of problems are not studied, could be that it is supposed that all problems can, in essence, be handled in the same way.
2.11 Summary and conclusions
In this chapter we have discussed some aspects of complex interdisciplinary societal problems and some aspects of human problem handling. In this discussion we have selected those matters we considered relevant for our study. We defined problems as:
something is called a problem when there is a discrepancy between the actual or (near) future situation and the desired future situation and/or there is a lack of knowledge and/or a lack of know-how, and/or a lack of relevant data; as for complex interdisciplinary societal problems, the problem is often undefined and the desired situation is not always clear
and problem handling as
the process of analyzing a problem in order to gain more insight into the problem, whether or not this leads to influencing the problem in order to reach the desired situation. This process can take place actively or passively, consciously or unconsciously, routinely or onceonly, whether it is by circumventing or by forgetting the problem, by shifting the problem to another party or by (partly) changing the problem, whether through thinking, applying tools and/or methods
We discussed the difference between problems based on the absence or the presence of certain qualities which are relevant for the exclusion or inclusion of problems in our study.
We discussed in this chapter (a part of) the first research expectation:
the character of the complex interdisciplinary societal problems is itself an important cause of the difficulty of handling these problems
A research question derived from this expectation:
1a is handling complex interdisciplinary problems so difficult because of
the special character of these problems?
Did we succeed in answering the research question? Is handling complex societal interdisciplinary problems so difficult because of the special character of these problems? What is the special character of complex interdisciplinary societal problems? A combination of many aspects makes it difficult to handle these kinds of problems, including:
The complexity and interdisciplinarity. The continuously changing environment. The missing or incomplete knowledge and data. The beginning, the process and the end of the problem which is hard to find. The problems have seldom been 'solved' before, and can mostly only be temporarily changed rather than solved. These kind of problems can generate many misunderstandings and conflicts.
All these aspects, which are part of or closely connected to the character of this kind of problem make it very difficult to handle them fruitfully. We have thus answered research question 1a positively.
In this chapter we have also answered part of the research question 1b:
1b in what way do the problems that are studied in the field of cognitive psychology differ from complex interdisciplinary societal problems....
In a brief review, we discussed some aspects of the work of some of the researchers on problem handling that we considered relevant for our study. We discussed some aspects of research by Selz, Bahle, De Groot, Newell & Simon, Crombag, Snoek, Duyne and Wierda. All these researchers regard the human being as an information processing system. This is an important concept, although one must be aware that the information will be collected, processed and reproduced, selectively, linearly and non-linearly, and influenced by former knowledge, interest and motivation of the person. We discussed the state-space-search paradigm of Newell & Simon. Newell & Simon consider problem solving as a search in a problem space. The problem space is the space where the problem solving activities take place and where the solution of the problem can be found. A problem space can be defined as knowledge about the task, the set of operators, the initial state of knowledge, a problem with an intermediate state and a desired state and the total knowledge available. The problem space can change during the course of problem handling.
We noticed that the problems on which Newell & Simon based their theoretical ideas, are small, artificial domain related, already defined and solved problems. These problems differ greatly from the undefined complex societal interdisciplinary hard to handle empirical problems we are focusing on in this research. In our opinion the state-space-search paradigm can only be applied usefully to artificial and rather small problems like those Newell & Simon use in their research on human problem solving, and for the problems Artificial Intelligence often deals with to substantiate its ideas. For handling complex interdisciplinary problems, the state-space-search paradigm is not only too limited but also incorrect. It suggests that it is possible to find a solution and that the solution can be found in the problem space. Finding a solution for a complex interdisciplinary societal problem, however, is not always possible. In section 2.10 we summarize some of the differences between the problems cognitive psychology deals with for research and the problems we focus on in this study. Some of these difference are: the range of the problem; the beginning and the end of the problem; the data, the description and the knowledge of the problem; the definition of the problem; the uniqueness of the problem; the solution of the problem; and the amount of time needed for handling the problem. Other differences will be discussed in chapter seven. In the next chapter we discuss a number of similarities. We also look at the number of people needed to handle the problem in our discussion of the phases in the problem handling process.
See for more publications of Dorien J. DeTombe
See for lectures of Dorien J. DeTombe
Ó Dorien J. DeTombe, All rights reserved, update September 2004