1

Problem Solving and Creativity

CHAPTER OUTLINE

The Problem-Solving Cycle

Types of Problems
Well-Structured Problems
Isomorphic Problems
Problem Representation Does Matter!
Ill-Structured Problems and the Role of Insight
Early Gestaltist Views
The Neo-Gestaltist View
Insights into Insight
Neuroscience and Insight
Obstacles and Aids to Problem Solving
Mental Sets, Entrenchment, and Fixation
Negative and Positive Transfer
Transfer of Analogies
Intentional Transfer: Searching for Analogies
Incubation
Neuroscience and Planning during Problem Solving
Intelligence and Complex Problem Solving
Expertise: Knowledge and Problem Solving
Organization of Knowledge
Reflections on Problem Solving
Automatic Expert Processes
Innate Talent and Acquired Skill
Artificial Intelligence and Expertise
Can a Computer Be Intelligent?
The Turing Test
Expert Systems
Creativity
What Are the Characteristics of Creative People?
Neuroscience and Creativity
Key Themes
Thinking about Thinking: Analytical, Creative, and Practical Questions
Key Terms
 2

The Problem -Solving Cycle

The steps of the problem-solving cycle include problem identification, problem definition,
strategy formulation, organization of information, allocation of resources, monitoring, and
evaluation.

We may go back and forth through the steps. We can change their order, or even skip or add
steps when it seems appropriate. Following is a description of each part of the problem -solving
cycle.

1. Problem identification: Do we have a problem?

2. Problem definition and representation: What exactly is our problem?
3. Strategy formulation: How can we solve the problem? The strategy may involve analysis—
breaking down the whole of a complex problem into manageable elements. Instead, or perhaps
in addition, it may involve the complementary process of synthesis—putting together various
elements to arrange them into something useful.
Another pair of complementary strategies involves divergent and convergent thinking. In
divergent thinking, you try to generate a diverse assortment of possible alternative solutions to a
problem. Once you have considered a variety of possibilities, however, you must engage in
convergent thinking to narrow down the multiple possibilities to converge on a single best
answer.
4. Organization of information: How do the various pieces of information in the problem fit
together?
5. Resource allocation: How much time, effort, money, etc., should I put into this problem?
6. Monitoring: Am I on track as I proceed to solve the problem?
7. Evaluation: Did I solve the problem correctly?

Our emotions can influence how we implement the problem-solving cycle (Schwarz & Skurnik, 2003).
 3

Types of Problems

Well-structured problems have clear paths to solutions. These problems also are termed well-
defined problems. An example would be, “How do you find the area of a parallelogram?”

Move Problem: “hobbits” and “orcs”

People seem to make three main kinds of errors when trying to solve well -structured problems
(Greeno, 1974; Simon & Reed, 1976; Thomas, 1974). These errors are:
[1] Inadvertently moving backwards
[2] Making illegal moves
[3] Not realizing the nature of the next legal move

A problem space is the universe of all possible actions that can be applied to solving a problem,
given any constraints that apply to the solution of the problem.

Algorithms are sequences of operations (in a problem space) that may be repeated and that, in
theory, guarantee the solution to a problem (Hunt, 1975; Sternberg, 2000). Generally, an
algorithm continues until it satisfies a condition determined by a program.

Unlike computers, however, the human mind does not specialize in high-speed computations of
numerous possible combinations. The limits of our working memory prohibit us from considering
more than just a few possible operations at one time (Hambrick & Engle, 2003; Kintsch et al.,
1999; see also Chapter 5).
 4

Newell and Simon recognized these limits and observed that humans must use mental shortcuts
for solving problems. These mental shortcuts are termed heuristics—informal, intuitive,
speculative strategies that sometimes lead to an effective solution and sometimes do not (see
Chapter 12 for more on heuristics; Gilovich et al., 2002; Stanovich, 2003; Sternberg, 2000).

Isomorphic Problems. Sometimes, two problems are isomorphic; that is, their formal structure is
the same, and only their content differs.

Problem Representation Does Matter. Consider the various versions of a problem known as the
Tower of Hanoi. In this problem, the problem
solver must use a series of moves to transfer a
set of rings (usually three) from the first of
three pegs to the third of the three pegs,
using as few moves as possible.
Problems such as the Tower of Hanoi
challenge problem-solving skills, in part
through their demands on working memory.
The brain areas that seem most involved in
the Tower of Hanoi task are the prefrontal
cortex, bilateral parietal cortex, and bilateral
premotor cortex (Fincham et al., 2002).

Ill-Structured Problems and the Role of Insight

Lack clear paths to solutions (Shin et al., 2003). These problems are also termed ill-defined
problems. The preceding ill-structured problems are insight problems because you need to see
the problem in a novel way. You need to see it differently from how you would probably see the
problem at first, and differently from how you would probably solve problems in general.
 5

Insight is a distinctive and sometimes seemingly sudden understanding of a problem or of a

strategy that aids in solving the problem. Often, an insight involves reconceptualizing a problem
or a strategy in a totally new way. Insight often involves detecting and combining relevant old
and new information to gain a novel view of the problem or of its solution.

Early Gestaltist Views

Gestalt psychologists emphasized the importance of the whole as more than a collection of
parts. Gestalt psychologist Max Wertheimer (1945/1959) wrote about productive thinking, which
involves insights that go beyond the bounds of existing associations. Gestalt psychologists held
that insight problems require problem solvers to perceive the problem. He distinguished it from
reproductive thinking, which is based on existing associations involving what is already known.

The Neo-Gestaltist View

Some researchers have found that insightful problem solving can be distinguished from non-
insightful problem solving in two ways (Metcalfe, 1986; Metcalfe & Wiebe, 1987). Problem
solvers show remarkable accuracy in their ability to predict their own success in solving a
problem prior to any attempt to solve it. In contrast, when given insight problems, problem
solvers show poor ability to predict their own success prior to trying to solve the problems. Not
only were successful problem solvers pessimistic about their ability to solve insight problems, but
unsuccessful problem solvers were often optimistic about their ability to solve them.

Insights into Insight

According to Smith (1995a), insights need not be sudden “a-ha” experiences. They may and
often do occur gradually and incrementally over time.

Neuroscience and Insight

Neuroimaging studies suggest that the activity of our brain during rest can be divided up into
several different networks. Some of these networks are also active when we engage in problem
solving (Andreasen et al., 1995; Christoff et al., 2004; Damoiseaux et al., 2006; Kounios et al.,
2008).
 6

The right hippocampus is critical in the formation of an insightful solution (Luo & Niki, 2003). The
hippocampus is integral to the formation of new memories. Therefore, it makes sense that the
hippocampus would be involved in the formation of an insightful solution, as this pr ocess
involves combining relevant information stored in memory.
In one study, during the preparation prior to viewing of a problem, participants who would later
generate an insightful solution had substantial activation in the frontal lobes, whereas those who
would not generate an insightful solution had comparable activation in the occipital lobes
(Kounios et al., 2006).

Obstacles and Aids to Problem Solving. Several factors can hinder or enhance problem solving.
Among them are mental sets as well as positive and negative transfer.

Mental Sets, Entrenchment, and Fixation

One factor that can hinder problem solving is mental set—a frame of mind involving an existing
model for representing a problem, a problem context, or a procedure for problem solving.
Another term for mental set is entrenchment. When problem solvers have an entrenched mental
set, they fixate on a strategy that normally works well in solving many problems but that does
not work well in solving this problem.

For example, in the two-string problem, you may fixate on strategies that involve moving
yourself toward the string, rather than moving the string toward you. In the oft-marrying
minister problem, you may fixate on the notion that to marry someone is to become wedded to
the person.

Another type of mental set involves fixation on a particular use (function) for an object.
Specifically, functional fixedness is the inability to realize that something known to have a
particular use may also be used for performing other functions (German & Barrett, 2005;
Rakoczy et al., 2009). An additional type of mental set is considered an aspect of social cognition.
Stereotypes are beliefs that members of a social group tend uniformly to have types of
characteristics.

Negative and Positive Transfer

Transfer is any carryover of knowledge or skills from one problem situation to another
(Detterman & Sternberg, 1993; Gentile, 2000). Transfer can be either negative or positive.

Negative transfer occurs when solving an earlier problem makes it harder to solve a later one.

Positive transfer occurs when the solution of an earlier problem makes it easier to solve a new
problem. That is, sometimes the transfer of a mental set can be an aid to problem solving.

Transfer of Analogies. Researchers designed some elegant studies of positive transfer involving
analogies (Gick & Holyoak, 1980, 1983).
 7

Intentional Transfer: Searching for Analogies

To find analogies between two problems, one must perceive the relationships between them
(Gentner, 1983, 2000). The actual content attributes of the problems are irrelevant. In other
words, what matters in analogies is not the similarity of the content but how closely their
structural systems of relationships match. The opposite phenomenon is transparency, in which
people see analogies where they do not exist because of similarity of content.

Incubation—putting the problem aside for a while without consciously thinking about it—offers
one way in which to minimize negative transfer. It involves taking a pause from the stages of
problem solving. During incubation, you must not consciously think about the problem. You do,
however, allow for the possibility that the problem will be processed subconsciously.

Incubation is an essential stage of the problem-solving process (e.g., Cattell, 1971; von
Helmholtz, 1896). Others have failed to find experimental support for the phenomenon of
incubation (e.g., Baron, 1988).

Neuroscience and Planning during Problem Solving

One way to invest enough initial time in a problem is through the formation of a plan of action
for the problem. Recall from Chapter 2 that the frontal lobes are involved in high-level cognitive
processes. It is therefore not surprising that the frontal lobes and in particular the prefrontal
cortex are essential for planning for complex problem-solving tasks (Unterrainer & Owen, 2006).

Additionally, both the left and right prefrontal areas are active during the planning stage of
complex problem solving (Newman et al., 2003). Further evidence for the importance of the
prefrontal regions in problem solving can be seen in cases of traumatic brain injury. Both
problem solving and planning ability decline following traumatic brain injury (Catroppa &
Anderson, 2006). In fact, regarding the problem-solving ability of patients with traumatic brain
injury, those patients who performed best were ones with limited damage to the left prefrontal
regions (Cazalis et al., 2006).

Intelligence and Complex Problem Solving

Components are the mental processes used in performing these tasks, such as translating a
sensory input into a mental representation, transforming one conceptual representation into
another, or translating a conceptual representation into a motor output (Sternberg, 1982).

Participants who score higher on traditional intelligence tests take longer to encode the terms of
the problem than do less intelligent participants. But they make up for the extra time by taking
less time to perform the remaining components of the task. In general, more intelligent
participants take longer during global planning — encoding the problem and formulating a
general strategy for attacking the problem (or set of problems). But they take less time for local
planning —forming and implementing strategies for the details of the task (Sternberg, 1981).
 8

Expertise: Knowledge and Problem Solving

Expertise is superior skills or achievement reflecting a well-developed and well-organized
knowledge base.

Organization of Knowledge
Knowledge can interact with understanding in problem solving as well (Whitten & Graesser,
2003). Consider a study investigating how knowledge interacts with coherence of a text. This
finding suggests that, in general, learners do better when they are presented new material in a
coherent way.

Elaboration of Knowledge
What differentiated the experts from the novices was the
amount, organization, and use of knowledge. For both
chess tasks, the experts used heuristics for storing and
retrieving information about the positions of the pieces on
the chessboard. The novices, to the contrary, had not
stored significant knowledge about positions.

The key difference, therefore, was that chess experts had

stored and organized in memory tens of thousands of
board positions. When they saw sensible board positions,
they could use.
For random scatterings of pieces on the board, however,
the knowledge of the experts was of no use. The experts
had no advantage over the novices. Like the novices, they
 9

had to try to memorize the distinctive interrelations among many discrete pieces and positions.
This memorization requires the storage of many more items, thus taxing one’s memory abilities.

Reflections on Problem Solving

Another difference between experts and novices can be observed by asking problem solvers to
report aloud what they are thinking as they are attempting to solve various problems (Bilalic,
2008; Dew et al., 2009). Statements made by problem solvers are called verbal protocols. An
interesting effect of verbal protocols is that they can lead to increased problem -solving ability.

Another difference between expert and novice problem solvers is the time spent on various
aspects of problems, and the relationship between problem-solving strategies and the solutions
reached (Lesgold, 1988; Lesgold et al., 1988).

Automatic Expert Processes

Through practice in applying strategies, experts may automatize various operations. They can
retrieve and execute these operations easily while working forward. They use two important
processes: One is schematization, which involves developing rich, highly organized schemas; the
other is automatization, which involves consolidating sequences of steps into unified routines
that require little or no conscious control.

Through these two processes, experts may shift the burden of solving problems from limited-
capacity working memory to infinite capacity long-term memory.

Innate Talent and Acquired Skill

Many espouse the “practice makes perfect” point of view (Ericsson, 2003). The practice should
be deliberate, or focused. It should emphasize acquisition of new skills and applications rather
than mindless repetition of what the developing expert already knows how to do.

However, some take an alternative approach. This approach acknowledges the importance of
practice in building a knowledge and skill base. It also underscores the importance of something
like talent. Indeed, the interaction between innate abilities modified by experience is widely
accepted in the domain of language acquisition as well as other domains.

Certainly, some skill domains are heavily dependent on nurture. For example, wisdom is partly
knowledge based. The knowledge one uses to make wise judgments is necessarily a result of
experience (Baltes & Smith, 1990).
10
 11

Artificial Intelligence and Expertise

Computer programs have been developed both to simulate human intelligence and to exceed it.
In many ways, computer programs have been created with the intention of solving problems
faster and more efficiently than humans.

Can a Computer Be Intelligent?

Much of early information-processing research centered on work based on computer
simulations of human intelligence as well as computer systems that use optimal methods to
solve tasks.

Programs of both kinds can be classified as examples of artificial intelligence (AI), or intelligence
in symbol-processing systems such as computers (see Schank & Towle, 2000). Computers cannot
think; they must be programmed to behave as though they are thinking.
That is, they must be programmed to simulate cognitive processes. In this way, they give us
insight into the details of how people process information cognitively.

Turing Test
Probably the first serious attempt to deal with the issue of whether a computer program can be
intelligent was made by Alan Turing (1963). The basic idea behind the Turing Test is whether an
observer can distinguish the performance of a computer from that of a human. The test is
conducted with a computer, a human respondent, and an interrogator.

Often, what researchers are interested in when assessing the “intelligence” of computers is not
their reaction time, which is often much faster than that of humans. They are interested instead
in patterns of reaction time, that is, whether the problems that take the computer relatively
longer to solve also take human participants relatively longer.

Expert Systems are computer programs that can perform the way an expert does in a specific
domain. They are not developed to model human intelligence, but to simulate performance in
just one domain, often a narrow one. They are mostly based on rules that are followed and
worked down like a decision tree.

CREATIVITY
Broadly defined as the process of producing something that is both original and worthwhile
(Csikszentmihalyi, 1999, 2000; Kozbelt, Beghetto, & Runco, 2010; Lubart & Mouchiroud, 2003;
Sternberg & Lubart, 1996). The something could take many forms. It might be a theory, a dance,
a chemical, a process or procedure, a story, a symphony, or almost anything else.

What Are the Characteristics of Creative People?

✓ divergent production—the generation of a diverse assortment of appropriate responses,
an approach originated by Guilford (1950) (see Runco & Albert, 2010, for a history of the
field, and Plucker & Makel, 2010, for a discussion of assessment of creativity)
 12

✓ other psychological researchers have focused on creativity as a cognitive process by

studying problem solving and insight (Finke, 1995; Ward & Kolomyts, 2010; Weisberg,
1988, 2009); work long and hard
✓ some computer programs, such as those composing music or rediscovering scientific
principles, can be viewed as creative
✓ Personality and motivation play important roles in creativity (Barron, 1988; Feist, 2010;
Hennessey, 2010; Runco, 2010).
✓ One may differentiate intrinsic motivation, which is internal to the individual, from
extrinsic motivation, which is external to the individual.
✓ tend to be more open to new experiences, self-confident, self-accepting, impulsive,
ambitious, driven, dominant, and hostile than less creative individuals; less conventional

One can seek to understand creativity by going beyond the immediate social, intellectual, and
cultural context to embrace the entire sweep of history (Simonton, 1988, 1994, 1997, 1999,
2010).
Evolutionary thinking also can be used to study creativity (Cziko, 1998; Gabora & Kaufman, 2010;
Simonton, 2010).

Neuroscience and Creativity

The prefrontal regions are especially active during the creative process, regardless of whether
the creative thought is effortful or spontaneous (Dietrich, 2004).

Previous research has indicated that this and related Brodmann’s areas are involved in verbal
working memory, task switching, and imagination (Blackwood et al., 2000; Collette et al., 2001;
Sohn et al., 2000; Zurowski et al., 2002).

A selective thinning of cortical areas seems to correlate with intelligence and creativity. A
thinning of the left frontal lobe, lingual, cuneus, angular, inferior parietal, and fusiform gyri
relates to high scores on creativity measures. These areas include several Brodmann’s areas,
including BA 39. Additionally, a relative thickness of the right posterior cingulate gyrus and right
angular gyrus was related to higher creativity as well. These variations in cortical thickness, and
especially a thinning in various areas, probably influence information flow within the brain (Jung
et al., 2010).

KEY THEMES
Domain generality versus domain specificity . Early work on problem solving, such as that by Allen
Newell and Herbert Simon and their colleagues, emphasized the domain generality of problem
solving. These investigators sought to write computer routines, such as the General Problem
Solver, that would solve a broad array of problems. Later theorists have emphasized domain
specificity in problem solving. They have especially called attention to the need for a broad
knowledge base to solve problems successfully.
 13

Validity of causal inference versus ecological validity. Most studies of creativity have occurred in
laboratory settings. For example, Paul Torrance gave students paper -and-pencil tests of creative
thinking administered in classrooms. In contrast, Howard Gruber has been interested only in
creativity as it occurred in natural settings, such as when Darwin generated his many ideas
behind the theory of evolution.

Applied versus basic research. The field of creativity has generated many insights regarding
fundamental processes used in creative thought. But the field has also spawned a large industry
of “creativity enhancement”—programs designed to make people more creative. Some of these
programs use insights of basic research. Others represent little more than the intuitions of their
inventors. When possible, training should be based on psychological theory and research, rather
than guesswork.