4 chp2 Intro Design Cognition

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/301979017
Introduction of Design Cognition
Chapter · February 2015

DOI: 10.1007/978-3-319-14017-9_2
CITATIONS READS
5 1,904
1 author:
Chiu-Shui Chan
Iowa State University
79 PUBLICATIONS 630 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Work on the VR historical models for headmount display. View project
All content following this page was uploaded by Chiu-Shui Chan on 04 January 2019.
The user has requested enhancement of the downloaded file.

Chapter 2
Introduction of Design Cognition
2.1 Definition of Design
Design is the human conception and planning of virtually everything in the world.
All man-made design has as its fundamental essence that everything is driven by
certain intentions and is accomplished by a series of actions to generate results.
Design is process, artifact, and discipline. As explained in the American Heritage
dictionary, to design, seen from the action point of view, is: (1) to conceive or fash-
ion in the mind, invent; (2) to have as a goal or purpose, intend; and (3) to create or
contrive for a particular purpose or effect. Synthesizing these definitions into an
integrated conceptual framework, design can be described as to conceive a purpose,
contrive a goal, and formulate a plan for a purposeful intention in the mind. On the
other hand, design seen from the perspective of an entity, is: (1) a drawing or sketch;
(2) a graphic representation; (3) a particular plan or method; (4) a reasoned purpose;
or (5) a deliberate intention (American Heritage Dictionary 2013). Here, a purpose
and an intention are both treated as entities, for they are the products of creative
actions. Thus, design is a created object, a generated method, a developed purpose,
or a conceived intention for everyday routines conducted through mental efforts.
Design should be recognized as an essential part of human life—in fact, a critical
component of human intelligence—and, as such, deserves critical discourse.
If looking at design from the fields of making artifacts (e.g., architectural,
graphic, fashion, industrial, interior, and engineering design, to name just a few),
design can be defined specifically as all human creative endeavors of shaping
objects to meet purposes or constructing a structure adapted to objectives, which
require professional consideration on aesthetic beauty, functional uses, social sym-
bols, and market demands and supplies. All these endeavors can also be seen as
solving certain problems with satisfactory solutions. Thus, to make the long defini-
tion short, design can be seen as human endeavors of creating satisfactory solutions
or beautiful artifacts to fulfill certain functions, which explains the essence of design
© Springer International Publishing Switzerland 2015 9

C.-S. Chan, Style and Creativity in Design, Studies in Applied Philosophy,
Epistemology and Rational Ethics 17, DOI 10.1007/978-3-319-14017-9_2
10 2 Introduction of Design Cognition
activities—the intellectual creation of feasible solutions or aesthetic artifacts—and

these creation efforts must be devoted to meeting functional requirements. This
explains a part of design thinking.
In fact, an astounding array of design products are created by designers and
shared and used every day by the public. Constant use of any product has dramatic
impacts on many aspects of human life. These product impacts are causal effects
from design and by design. For instance, usability provided in household products
will impact users’ productivity (Norman 2002). Phenomenon of patterns created in
building forms will yield different perceptions in the built environment to the
beholders (Holl 2006). Color and materials used in working environments will
affect human cognitive performance (Chan 2007). Since the impact of design
products influences our well being and everyday performance, a comprehensive
understanding of design, especially design thinking, would help to explain how
design is generated to improve design ability and enhance design products. Indeed,
design thinking is a part of problem solving activities and is an ability that can be
utilized in the creation of artifacts and for other everyday tasks. Techniques used in
formulating and solving design problems can also be used to solve relevant problems
in our own lives.
2.2 Design Thinking and Cognition
If we see design as a series of mental activities generating entities, then we should

see these activities as intellectual processes, the parts of human thinking operated by
consciousness. In cognitive psychology, thinking is defined as the phenomena of
human cognitive operations. Thus, design activities can be defined as thinking activ-
ities executed by cognitive operations. On the other hand, design products could
also be seen as the physical results coming from the design thinking process oper-
ated by cognition. Thus, whether interpreting design from the process or from the
product point of view, the bottom line is that design is created by human cognition.
For decades, scholars have studied the nature of and attempted to define design
thinking, which has been considered a “way of thinking.” Peter Rowe was the first
one who used the term in the title of his book, Design Thinking, to explain problem
solving procedures used by architects and urban planners (Rowe 1987). From
combining numerous scientific studies conducted in various fields, a clearer picture
has emerged and the term design cognition used to categorize the activities that
occur in the design process. Design cognition has gradually been developed into a
discipline. For instance, the nature of design cognition has been studied through
looking at design from the computational (Cross 1999) or problem solving perspec-
tive (Cross 2001). Chuck Eastman used the term to refer to the study of human
information processing by using different theoretical and empirical paradigms to
describe the process of design information (Eastman 2001).
Given the rapid development of new technologies and new study tools available,
it is prudent to re-examine and rethink this topic to incorporate new scientific
2.3 Development of Design Studies 11
developments occurring in other related fields. This chapter takes this new
approach by looking at the conventional studies, rethinking the methodologies
applied, reviewing the directions, and suggesting a new approach, particularly
from the architectural design perspective. Once design cognition is well defined
and fully understood, it is appropriate to analyze how designers stylistically and
creatively do design, and identify the causes of style and creativity that occur in
the design process.
2.3 Development of Design Studies
If design research is defined as a scientific approach to design or regarded as a study

approached from the scientific perspective, as Herbert Simon explained in his
theory of the “Science of Artifact” in 1969, then all the works done through
methodological and rigorous sequences by artists in their creative processes or by
scholars in their studies should be recognized as design research. In his book, Simon
proposed that a science of design is a body of intellectually tough, analytic, formal-
izable, empirical, observable, and teachable doctrine about the design process
(Simon 1969). Thus, all design works performed by a system of inquiry procedures
can be taught, recorded, and studied. Artists, in their creation processes, are research-
ing the ways of creating artifacts through implementation and execution, which can
be duplicated. Scholars, in studying artists’ creative processes, explore the various
methods used for creation and expect to summarize the knowledge of creation that
can be learned and replicated. Therefore, both artists working systematically on
creating a work of art or a craftwork, and scholars on studying how a designer thinks
or creates a design, should be recognized as conducting design research.
Following this premise, the geometrical perspective drawing method, developed
in the Fifteenth Century in Florence as Quattrocento Art (Hartt 1994), is a result of
design research. The use of a physical model to represent design, such as the scale
model of the Basilica of St. Peter’s, founded by the Emperor Constantine, should
also be recognized as design research. However, the modern development of design
research with a systematic approach only traces back to the works done by the De
Stijl group in the beginning of the twentieth century (Cross 2000). De Stijl (Dutch
term for “The Style”) was a movement led by architect Theo van Doesburg in the
early 1920s. Their design principles were based on functionalism with the following
characteristics: (1) using rectilinear planes in the way that is similar to slide the
planes across one another; (2) eliminating decoration on surfaces; and (3) limiting
color schemes to pure primary hues together with black and white. Their works had
clear principles and procedural routines to follow, which were the tendency of meth-
odological approaches of the time on artistic creation. Because these rules and pro-
cedures were followed by a group of artists and designers in various design fields,
their creation endeavors are recognized as the beginnings of design research that
applies scientific principles to systematically arrange design composition.
Bauhaus, the school of Art and Craft, also incorporated concepts around 1920
from the Arts and Crafts Movement in England to develop structural concepts in
their design curriculum. By integrating arts and science and combining crafts and
practice, Bauhaus, led by Walter Gropius, created a new direction for architectural
design education. They emphasized the honest and direct use of materials as the
most functional path to design. At Bauhaus, students were required to take craft
courses as well as painting, drawing, and theoretical studies in design and color.
Faculty at Bauhaus also conducted research on mathematical and representational
analysis of the concept of how sensations were organized into a unified perception,
which was the “field” theory developed by Gestalt psychology1 (Wertheimer 1923).
Paul Klee, the Swiss artist who joined Bauhaus from 1921 to 1931, utilized the
rules of organizing patterns to explore sensation and perception in his course mate-
rials (Teuber 1973). In the period from 1920 to 1930, scientific methods for explor-
ing design emerged, which became a discipline, and was termed design studies.
During the two world wars (1914–1918, 1939–1945), the war activities drove the
need to automate production. They resulted in mass production of bombs, ships, and
weaponry. Equipment in factories used by the war industries were thus of the highest
possible level of sophistication at that time. All the rest, for example, housing, food, or
transportation was in need of mass automation after two world wars. So mass produc-
tion during the world wars was set as an example to drive automation in civil indus-
tries. Thus, studies in design started to focus on mechanical efficiency to improve the
performance of products. However, an individual designer had limited ability to han-
dle the increasing complexity of the industrial manufacturing of products. Thus, meth-
ods of production through systematic design procedures became the research focus. In
the post war era, the shortage of labor required technological development on mass
production and product automation. Labor issues, together with the swift transforma-
tion from military wartime equipment to civilian demand for consumer products
required the development of new production methods in manufacturing. For instance,
in industrial design, studies began to utilize systematic methodologies in design
around 1950 to make processes more efficient and effective.
From 1950 to 1960, design studies were mainly influenced by system theory and
systems analysis on design, which set up the basis for a “design method movement”
(Cross 1984). The system theory and analysis were coming from well-developed
information processing theory and operations research, which inspired scholars to
study design methodology. At that time, design methodology was regarded as a
prescribed and rigid approach to design. This movement called designers’ attention
1
Gestalt Psychology involves recognizing how sensations are organized into a united perception
by human beings. Its major concentration is, in vision, on understanding how the whole is differ-
ent than the sum of its parts while we are perceiving things. The major concepts of this theory are
that the perceiving behavior of an object is determined by the spatial-temporal configuration of
the objects in our visual sense. Because of the good understanding of the phenomenon of percep-
tion, it has big influence to the fields of industrial design, painting, graphic design, and typogra-
phy after it is developed from 1920 to 1930. Details of the notions applied in design are explained
in Chan (2008).
towards exploring systematic design procedures, and proposing systematic methods

for designers to apply. Two classical examples presented in the first Conference on
Design Methods, held in London in 1962, attracted designers’ interests.
The first classic example is the works of J. C. Jones. Jones, as an industrial
designer for a manufacturer of large electrical products in the 1950s, applied
ergonomics as a means to understand the design process of engineers on designing
electrical equipment that better responded to user requirements. When his ergonomic
studies found that the engineer had not considered user behavior, he redesigned the
engineer’s design processes to respond to the human requirement first, and the
machine requirement second. A procedural sequence applied in design to allow
intuition and rationality to co-exist was called design methods. Other than generat-
ing and utilizing design methods in the design process, Jones proposed that the
design process has cycles, or stages of analysis, synthesis, and evaluation. The anal-
ysis component is the problem analysis stage that requires making all design
requirements on a list, and carefully studying their interactions to make a set of
logically related performance specifications. The synthesis stage is to find solutions
for each specification and build up an integrated design solution with the least
conflict among the specifications. The evaluation stage is to test the accuracy of
each alternative solution in order to select the final one. The processes could be
recycled and repeated whenever necessary (Jones 1963). He described these three
essential stages of design in simple words as “breaking the problems into pieces
(analysis) … putting the pieces together in a new way (synthesis) … and testing
to discover the consequences of putting the new arrangement into practice (evalua-
tion).” The purposes of his study were: (1) to reduce the amount of design error,
re-design and delay; and (2) to make possible more imaginative and advanced
designs (Jones 1963, p. 63).
In architecture and urban design, Christopher Alexander (1964) developed a
“pattern language” methodology, which serves as a second classic example from
the 1960s. The fundamental notion of pattern language was to solve design prob-
lems through combinations of selected design patterns. To Alexander, a language
had vocabulary, a collection of named solutions to problems, called design patterns.
The conceptual framework was that our built environment included the geometry of
physical objects, which was a 3D embodiment of a culture, and an organization of
social institutions defined by human activities. The human activities were used to
categorize the social institutions and be anchored in spaces. Thus, spaces catego-
rized social institutions that attributed to human activities. For instance, categories
of space inside a house embodied the culture of its family; whereas the categories of
space in a city embodied the culture of the people in the city. Different environments
had their own morphological laws given by the acts made by builders, and the for-
mations of acts were guided by the combination of the builder’s mental images.
Such combinational systems of images were exactly, to Alexander, like human lan-
guages. These systems allowed a person to produce various combinations, and
every environment was formed from such combinatorial systems of images repre-
senting patterns, much as in combining words in languages. That is why Alexander
called such systems “Pattern Languages” (Alexander et al. 1977).
The patterns in pattern language correspond to the rules of grammar in natural

language. Each pattern is a fluid image that can be combined with other patterns. In
all, patterns can be described as the overall layout of a building, ecology, large-scale
social aspects of urban planning, regional economics, building components, struc-
tural engineering, or building construction. The rules of grammar and their meaning
in pattern language systems are not given predigested. Every designer can generate
and share his or her own language. Then, shared languages gradually evolve towards
greater and greater wholeness. To Alexander, the good patterns would spread
widely; the bad patterns would eventually drop out. Therefore, the environment so
created by designers, though changing constantly, were as coherent, whole, and
real, as a totality. In 1996 at an Association for Computing Machinery (ACM)
Conference, he reinforced that the purpose of a pattern language is to create mor-
phological coherence in the things that are made with it.
The definition of each pattern in the language, explained by Alexander, includes
a problem that occurs over and over again in the environment. A solution to this
problem is described so that it can be used many times without solving it the same
way twice (Alexander et al. 1977, p. X). Following this definition, each pattern has
the following standardized format:
1. A picture shows an architectural example of that pattern.
2. An introductory paragraph sets the macro context of the pattern by explaining
how it helps to complete certain larger patterns.
3. A short format of a headline with less than two sentences explains the essence of
the problem.
*** (Three diamonds mark the beginning of the problem)
4. The body of the problem describes the empirical background of the pattern, the
evidence for its validity, and the range of different ways the pattern can be mani-
fested in a building.
5. The solution part describes the field of physical and social relations that are
required to solve the stated problem, in the stated context. The solution is stated
instructionally on what is needed to do to build the pattern.
6. A diagram shows the solution with labels to indicate its main components.
*** (Three diamonds show that the main body of the pattern is finished.)
7. A paragraph ties the pattern to all smaller patterns needed to complete this pat-
tern, to embellish it and to fill it out.
Therefore, pattern language consists of problem statements explaining the prob-
lem context, solutions of the problem with rules embedded, a sketch diagram show-
ing the solution schematically, and contexts to combine the solution with other
patterns globally and locally. Table 2.1 is a general outline of patterns. An example
given in A Pattern Language (Alexander et al., p. 854, pattern 184) on a cooking
layout that could be located in a kitchen (pattern 139, p. 663) is simplified in
Table 2.2. Figure 2.1 is a diagram showing the cooking layout solution (p. 856).
As shown in the format, we can present the problem statement component as X
describing the context of the pattern, Y is the solution, and Z is the problem to be
Table 2.1 Format of patterns Format of a pattern (problem):

1. Example picture
2. Introductory paragraph, global context
***
3. Problem headline
4. Problem body
5. Solution grammar
6. Sketch diagram
***
7. Links with other patterns, local context
Table 2.2 A cooking layout example

Name: Cooking layout pattern #184:
Pattern context: Within the Farmhouse Kitchen (#139), or any other kind of kitchen. The
character of a good kitchen comes from the arrangement of the stove and
food and counter
Problem: Cooking is uncomfortable if the kitchen counter is short and also
if it is too long
Problem The best arrangement for a kitchen design is the one that saves the most
statements: steps; and this led to compact kitchens. These compact layouts save steps
but don't have enough counter space. Studies have shown that there is
insufficient counter space in many kitchens
Solutions: To strike the balance between the kitchen which is too small, and the
kitchen which is too spread out, place the stove, sink, and food storage and
counter in such a way that:
(Rules:) 1. No two of the four are more than 10′ apart
2. The total length of the counter – excluding sink, stove, and refrigerator –
is at least 12′
3. No one section of the counter is less than 4′ long
Solution context: Put the pattern with Thick Walls (#197), Sunny Counter (#199), Open
Shelves (#200)
Fig. 2.1 A diagram of

cooking layout (Source:
Image was adapted from
A Pattern Language, 1977,
page 856 by Alexander et al.,
Oxford University Press)
resolved. Yet, the most essential nature of the language is the various contexts
associated with the problem Z and solutions Y. Contexts X are not only paired with
problems, but also are required to be considered in the solution syntax. For instance,
the larger context of the cooking layout pattern is in the farmhouse kitchen (pattern
#139) or any kitchen area, whereas the solution contexts are that it will be in the
working surface of sunny counter pattern (#199), thick walls pattern (#197), and/or
open shelves pattern (#200). While designers apply the patterns in design, various
combinations can be generated due to the flexible rules provided. In execution, the
format of the language can be seen as an hierarchy of parts, linked together by pat-
terns that have a carefully defined set of design rules for solving generic recurring
problems associated with the parts.
A famous example given by Alexander to explain the method of running the
language is the Indian village design, published in 1963. In that example, he
observed the problem context, made a list of 141 design requirements, then studied
these requirements by pairs and determined by graph theory through computer pro-
grams whether they were dependent or not. The interacted pairs of requirements
were further connected by links and grouped into twelve subsystems, which were
combined further again into four major subsystems. The set of requirements and the
set of links together defined a linear graph that served as a complete structural
description of the village. Of course, each subsystem had its diagrammatic concept
to show their images (Alexander 1963).
Pattern language has had a big influence on design education. Many scholars
have discussed its applications in studio teaching, but there also are concerns that:
(1) the problem situation encountered by the designer might be different from the
given context (X), but the designer might pay attention to the factors of the problem
situation that are similar to X and assume the same solution Y, rather than to dig
into the problem further; and (2) the context (X) may not be an accurate represen-
tation of the situation reflecting the relationship between the solution (Y) and
problem (Z), thus solutions might be misguided (Lang et al. 1974). Despite con-
cerns expressed in the field of design, pattern language concepts have influenced
other fields, because the notion of pattern languages helps problem solvers to
tackle the complexity of systems through patterns; for example, it shares the same
notion in object-oriented work on patterns in computer science. Each object in the
object-oriented programming (OOP) paradigm has context, problem, and solu-
tion, many of which can be shared and evolved as entities. Thus, after the original
publications in design fields, notions of pattern language have since been applied
to the fields of software engineering (Gamma et al. 1994), computer science
(Buschmann et al. 1996), software design (Fowler 2002), and user interaction
design (Tidwell 2005).
Similar studies done in the 1960s and 1970s on developing procedural tech-
niques to be used in design to systematically manage the progress of the design
processes include work by Bruce Archer. Archer developed a model to illustrate a
systematic method for solving industrial design problems (Archer 1965). He indicated
that a design had six sequential stages of: (1) receive brief, analyze problem, prepare
detailed program and estimate; (2) collect data, prepare performance specification,
reappraise proposed program and estimate; (3) prepare outline design proposal(s);
(4) develop prototype design(s); (5) prepare and execute validation studies; and (6)
prepare manufacturing documentation. These stages logically covered the design
from the beginning to the ending stage. Within the process, there were 227 activity
items formulated in a completed checklist for product designers to consider. These
sequential stages, to him, were sometimes overlapping, sometimes confusing, and
sometimes required returning to early stages when difficulties were found. He
explained that the art of industrial designing was essentially the art of reconciling a
wide range of factors drawn from function, manufacturing, and marketing. In design
practice, some projects were quite complicated, and involved contrasting skills and
a wide variety of disciplines. Therefore, designers must make some assumptions or
judgments, which might not be supported by collected data. Under such circum-
stances, any proposed design solution that constituted a hypothesis based upon
imperfect evidence had to go through tests of the marketplace, or an indirect feasi-
bility analysis, before it was completed (Archer 1965). Archer also used manage-
ment science and operations research with logic operations to frame the overall
structure of the design process that fit textile, clothing, jewelry, ceramics and inte-
rior design. The essence of the overall structure of the design process was in apply-
ing a reiterative problem-solving operational model to a goal driven system that was
appropriate to the design problem (Archer 1970). Archer’s studies have shown the
influences of problem solving theory and operational research, in addition to the
characteristics of systematic methodology.
This 1960s–1970s period, categorized as the methodological approach to
design (Cross 1984), generated a number of studies on design methodology,
yielding significant results on exploring the steps to complete a design. A num-
ber of special groups were also formed. These groups, including the Design
Research Society (founded in London, 1966), Design Methods Group (in
Berkeley, 1967), and Environmental Design Research Association (in North
Carolina, 1968), have hosted conferences and published conference proceed-
ings or newsletters since 1962 (Cross 1984; Bayazit 2004). Researchers in this
period mainly concentrated on observing design experience and phenomena
occurring in the process. Through observing these phenomena and from apply-
ing simple mathematical diagrams and flow-chart type models, certain patterns
and structures in the design processes were analyzed and reported descriptively.
However, these approaches were criticized as merely studies on the general
interpretation of design logic or on general explanations of processes. These
methodological studies disenchanted professional practitioners. Scholars were
also aware of the weaknesses of these approaches (Archer 1979). Interestingly,
Alexander indicated that design methodology would not help design because it
lacked the motivation for making better buildings (Alexander 1971). Similarly,
Jones also rejected design methods and changed his research direction (Jones
1977). Thus, first generation design methods were not capable of handling the
complexity of real-world problems.
At the same time, problem solving theory guided studies to understand the nature
of the complexity of design problems, particularly the structure of “ill-defined” prob-
lems (Reitman 1964; Rittel and Webber 1973; Simon 1973). Moving forward to the
early 1980s, architectural research on “design methodology” had been changed by
the inclusion of three influential factors to the study of the nature of design activity
and “design thinking.” The first factor, scientific research method, was inspired by
empiricism (or empirical studies) and positivism. Empiricism posits that the origin
of all knowledge comes from sense experience. This theory requires running obser-
vations or experiments to test all hypotheses and theories for obtaining knowledge.
Parallel to empiricism is positivism. Positivism, similar to empirical study and
inspired by natural science, says that knowledge is basically constructed by observ-
able and measurable facts. These facts exist around us and are waiting for us to sys-
tematically and objectively discover their nature. Scholars must define the cause-effect
relationships among the variables to synthesize the relationships into general princi-
ples that can be applied by others for further prediction and developments.
The second factor was based on the notion that designers’ thinking had close
connections to psychological sequences, and design was the psychological process
of solving problems. Scholars had mutually recognized the basic premise that in
everyday life, the ways of design—treated as solving design problems—came from
human nature. For the purpose of solving complex design problems and meeting
user requirements, design was considered as a problem-solving and decision-
making activity. Thus, studies on design started to center on fundamental human
intellectual capacities and cognitive faculties.
The third factor was on the adequacy of applying systematic theory in design.
Systematic design as studied from 1950 to 1980 had the tendency of applying sys-
tematic analysis-synthesis procedures, which do not fit the character of design as
ill-defined problems. Systematic design and its associated systematic procedures
could not adequately cover the entire complexity of modern design problems.
Particularly, systematic procedures are mostly linear sequences. Thus, studies on
systematic procedures and methodologies changed to understand the nature of
design activities through studying the design processes. On the other hand, if design-
ers needed to improve on their design procedures, their design thinking also needed
to be explored more.
Following this trend and based on positivism and empirical research methods
prevailing after World War II, studies in architectural design thinking in the 1970s
started to move from using flow-chart diagrams to analyze designing patterns to
concentrate on designers’ ways of designing and decision making. They identified
their research, conducted psychological experiments, collected data for analyses,
built abstract and simple models, generalized complex psychological activities, and
investigated cognitive phenomenon in the design processes. That led to the new
scientific exploration of “thinking processes” in design, and studies in this regard
have investigated what designers do when they design, from open-ended interviews
to controlled laboratory experiments. Methods applied were very scientific in nature
with several techniques for data collection, including interviewing designers with
specific questions asked, performing case studies of designers’ individual projects,
and asking designers to think aloud and record their verbal data. Details of each
technique are explained in the section on study methods in this chapter.
After the late 1980s, studies in design thinking focused on understanding the
cognitive mechanisms applied for and involved with implementing design activi-
ties. A number of conferences have been organized since then. The most significant
one is Design Thinking Research Symposium (DTRS), which was initially founded
in 1991 at Delft University of Technology by Cross et al. (1992). Because of the
success of its first meeting, it was hosted sequentially in following years to study the
nature of design ability, how designers think, and ways of knowing design in gen-
eral. Each symposium has its own theme: the application of protocol analysis in
1994 at Delft (Cross et al. 1996); descriptive models of design in 1996 at Istanbul
Technology University (Akin 1997); representation in 1999 at MIT (Goldschmidt
and Porter 2004); interdisciplinary approach in 2001 at Delft (Lloyd and Christiaans
2001); expert designers’ design nature in 2003 at University of Technology, Sydney
(Cross and Edmonds 2003); analyzing design in 2007 at University of the Arts,
London (McDonnell and Lloyd 2009); interpret design thinking in different disci-
plines in 2010 at University of Technology, Sydney (Stewart 2011); and analyses of
different responses to a given design task in 2012 at the University of Northumbria,
England (Rodgers 2012). These studies have brought the research in design think-
ing to a new age and given credibility to design thinking research as a distinct field
of inquiry across different disciplinary backgrounds.2
Looking back through history, the first generation of design research was influ-
enced by mathematics, operational research, and system theory around the 1960s to
posit a condition with a set of actions to follow. When the condition that matched
the problem situation were met, consequences followed to accomplish the design.
This first modern approach after World War II, however, was recognized by scholars
as being unable to solve complicated design problems. Then, in the 1980s, research
directions changed to a second generation on exploring the nature of design. This
change adopted notions from cognitive science to set up hypotheses about some
design phenomenon and behavior, and then conducted psychological experiments
to test, prove, and revise the hypotheses. This second movement of design study has
been a major research trend that prevails to the present time and generates a lot of
insight on how designers think.
When information technologies bloomed in the 1990s, design practice changed
from using traditional pencil-and-paper to the applications of computer systems.
Scholars realized that conventional graphic thinking mode and drawing representa-
tion would eventually be replaced by digital representation; studies must be per-
formed on the impact of the change. They also realized that psychological
experiments applied in the second generation of design research could provide data
to develop computational models. Thus, a new conference series, Design Computing
and Cognition (DCC), was initiated in 2004 at MIT (Gero 2004) and hosted in the
following years (Gero 2006, 2010; Gero and Goel 2008). DCC has focused on
2
Details can be seen on the Design Group, Open University Web pages: http://design.open.ac.uk/
cross/DesignThinkingResearchSymposia.htm
computational theories and systems that enact design. Research efforts have focused
on predicting design processes, developing computational models of these pro-
cesses, and examining processes from their computational results. Expectations of
combining computation and cognition in DCC conferences were not on exploring
the essential relationship between aspects of human cognition as computation, but
on how computation could inspire cognition, which was influenced by theories of
artificial intelligence. The Deep Blue II chess-playing machine is a good example to
describe the connections between artificial intelligence on computing and natural
intelligence on thinking (Cross 1999). As concluded by Casti (1998) on the chess-
playing program, nothing has been learned about human cognitive capabilities from
it or about methods from the construction of the machine point of view. But the fact
that the program defeated world chess champion Garry Kasparov in 1997, has
taught us something of the cognitive strategies of human chess players. Similarly, it
is possible to also learn the nature of design cognition through looking at design
from the computational perspective.
After studies gained some understanding on mental activities in design, scholars
started to explore particular areas to improve human cognition. For instance, the
ACM Creativity & Cognition (C&C) conference brings artists, scientists, designers,
and researchers from various fields together to understand human creativity in vari-
ous aspects. Their program3 explains that they “seek to understand human creativity
in its many manifestations, to design new interactive techniques and tools to aug-
ment and amplify human creativity, and to use computational media technologies to
explore new creative processes and artifacts in all human endeavors ranging from
the arts to science, from design to education.”
Similarly, spatial cognition is another interdisciplinary research field with orga-
nized symposia. Spatial cognition concentrates on the acquisition, organization,
modification, and revision of our knowledge learned from perception about spatial
environments. It has been considered an imperative building block of general cogni-
tion for the reason that it is the procedure by which we obtain, modify, memorize,
and retrieve information from memory in relation to spatial images. In 2003, the
“Spatial Cognition: Reasoning, Action, Interaction program” was established by
German Research Foundation (DFG) at the Universities Bremen and Freiburg. The
concept of spatial cognition defined by the group is that “Spatial Cognition is con-
cerned with the acquisition, organization, utilization and revision of knowledge
about spatial environments, be it real or abstract, human or machine. Research
issues range from the investigation of human spatial cognition to mobile robot navi-
gation. The goal of the SFB/TR 8 is to investigate the cognitive foundations for
human-centered spatial assistance systems.”4 Their research issues range from the
investigation of human spatial cognition to mobile robot navigation, which provide
understanding of human perception in space.
Through these years of design studies, starting from the systematic design
method movement initiated in the 1960s, to the utilization of theories applied from
3
ACM Creativity & Cognition Conference: http://dl.acm.org/event.cfm?id=RE326
4
URL of SFB/TR 8 Spatial Cognition: http://www.sfbtr8.spatial-cognition.de/
2.4 Cognitive Mechanisms and Functions 21
cognitive science to explore the design nature in the 1980s, and further to the
computational approach developed in the dawn of twenty-first century, theories of
human cognition in design have been developed to describe essential patterns of
design activities. Some theories, demonstrated by experiments, did explain aspects
of cognitive operations applied, factors involved, and mechanisms utilized in the
design processes. Thus, summarized from the cognitive science perspective, all cog-
nitive mechanisms and cognitive functions as reported in published literatures can
be categorized into patterns: (1) design is problem-solving; (2) design is making
associations; (3) design is goal and constraint driven; (4) design is reflective and
problem structuring; (5) design is looking for representation; (6) design utilizes
cognitive strategies; (7) design uses some reasoning; and (8) design applies repeti-
tion for design generation. These patterns of activities rely on the use of cognitive
mechanisms to fulfill cognitive functions.
2.4 Cognitive Mechanisms and Functions
By organizing patterns of theories into categories, the characteristics of design

thinking can be better described and well presented. By doing so, an overall picture
of design cognition would be further developed, much like the “State, Operator And
Result (SOAR)” system proposed. The SOAR was a candidate unified theory of
general cognition developed by Newell (1990) in the field of artificial intelligence
to handle the full range of capabilities of an intelligent agent. It has been used by
computer science and AI researchers to model aspects of human behavior. Similarly,
if a unified theory of design cognition could be developed, then it could help design-
ers to understand more on design thinking, style, creativity, and how to do digital
simulation better. In fact, these categories of concepts, as discussed in the following
sections, describe cognitive aspects of how design processes are moved forward,
knowledge acquired, solutions searched, ideas obtained, and forms created across
many design fields. These notions explain the involved cognitive factors that are the
fundamental elements of style and creativity in design.
2.4.1 Design Is Problem Solving
The new movement of scientifically studying design thinking in the 1970s was
influenced by Herbert Simon, who proposed to discover design processes that would
provide the means by which mental phenomena, including human design behav-
ior and cognitive patterns, might systematically be predicted and explained
(Simon 1969).5 Since then, research shifted from developing systematic design
5
In his book of the “Sciences of the artificial” published in 1969, Herbert Simon gave a very good
definition of the knowledge of design.
methodologies for managing overall design processes (Alexander 1964; Broadbent

1969; Jones 1970) to understanding how designers tackle problems with conven-
tional procedures (Wade 1977; Cross 1984; Akin 1986). With this change, the
research trend started to treat design activities as ways of solving problems (Simon
1969), which was led by problem-solving theory.
Problem solving theory was developed by Newell and Simon around 1956. In
their view, a human is a processor of information, and a computer is also an infor-
mation processor. Thus, a man could be modeled as a digital computer that pro-
cesses information. Similarly, human problem solving activities could be modeled
by an information processing system executed in computers to explain how humans
process task information (Newell and Simon 1972). Later on, such an information
processing approach to study the human factor combined the areas of computer sci-
ence and psychology into the field of cognitive psychology. The theory studied the
ability to solve problems, which is one of the important psychological behaviors
associated with intelligence. A problem exists when a person is confronted with a
situation where he or she wants to accomplish something but does not know imme-
diately what series of actions to take or how to find the means to achieve the ends.
Thus, a problem occurs and the process of solving the problem involves high-level
cognitive processes.
Historically, the study of problem solving was first approached in the 1930s by
Gestalt psychologists conducting experiments to study visual behavior. Their stud-
ies created interesting results on perception (Kohler 1930; Koffka 1935). In 1960s,
Newell et al. (1958) generated systematic outlines on how humans responded when
they were challenged by unfamiliar tasks. Their studies on problem solving applied
experiments conducted in a laboratory setting using problems that could be solved
in short periods of time and sought maximum data on the solution processes.
Experimental tasks involved asking subjects to solve structured, puzzle-like prob-
lems that were also simulated by computer programs written at that time to clearly
explore the subjects’ problem-solving strategies. These problems included the
Missionaries and Cannibals problem,6 Tower of Hanoi puzzle7 chess moves,8 proofs
for theories in Euclidean geometry, and crypto-arithmetic problems.9
6
Three missionaries and three cannibals come to a river and find a boat that holds two. If the can-
nibals ever outnumber the missionaries on either bank, the missionaries will be eaten. How can all
six get across the river? See URL: http://www.learn4good.com/games/puzzle/boat.htm or URL:
http://en.wikipedia.org/wiki/Missionaries_and_cannibals_problem
7
The Tower of Hanoi was invented by Edouard Lucas, a French mathematician in 1883. The original
tower has eight disks stacked in increasing size on three pegs. The puzzle is to move the entire stack
to another peg. The rules (or constraints) are to move only one disk at a time and never a larger one
onto a smaller. The number of disks could range from three to eight. See URL: http://www.cut-the-
knot.org/recurrence/hanoi.shtml or URL: http://en.wikipedia.org/wiki/Tower_of_Hanoi
8
In 1957, Irving Chernev wrote the famous book “Logical Chess: Move by Move” which uses
unique style of presentation to explain every move of 33 games from the masters.
9
A crypto-arithmetic puzzle is made of adding two integer expressions to an equation, and num-
bers in which are replaced by letters. Famous examples are “SEND + MORE = MONEY” and
“GERALD + DONALD = ROBERT.” The puzzle is to find the integers. See URL: http://en.wikipe-
dia.org/wiki/User:Paolo_Liberatore/Crypto-arithmetic_puzzle
In the 1970s, research on problem solving further classified the types of problems
that human beings encountered, and explored the natures of those problems. Defined
by task domains and their complexity, problems could be categorized to five types.
One was the everyday routine problem that could be solved by common sense or
educated guess. The second type was procedural problems with certain rules or that
had fixed steps to resolve (accounting, math, or engineering problems). The third
type was an open-ended problem that had many satisfactory solutions available
(design problems). The fourth type was insight problems, where a problem solver
must discover a critical element, and once this element was found, all the other ele-
ments would fall into place and the problem would be resolved (scientific discover-
ies). The fifth type was the problem that has a changing nature over time, and the
solution given not only has its own consequences, but might also need to be revis-
ited again and again (social policy problems).
After studies on the nature of problems generated interesting reports, research
started to evolve in two directions. One was finding methods for working with large
bodies of semantic information10—for instance, medical diagnoses and interpret-
ing mass spectrogram data. This research direction focused on tasks that were well
structured, with clear-cut goals and constraints. With this level of understanding of
problem solving related to real-world problems, an expert system emerged that
combined computer science and artificial intelligence. The other research direction
focused on understanding problem-solving tasks when goals were complex and
sometimes ill defined, and when the very nature of the problem was successively
transformed in the course of exploration. The processes of architectural design were
one area of interest. Especially, the nature of architecture design problems may
change from time to time in the designers’ design processes. During that time, the
term ill-structured problem (also called ill-defined) was used to differentiate it from
a well-structured (well-defined) problem.
Problems in the arts, humanities, social sciences, and engineering areas were
categorized as ill-structured problems (Reitman 1964; Newell 1969). An ill-
structured problem had a large problem space and no fixed problem-solving
sequences or definite goal sequences; any procedure could lead to a possible satis-
factory solution, but not an optimal one. On the other hand, well-defined problems
were those existing in crossword puzzles, chess-playing, standard calculations, or
the fields of natural sciences or mathematics. Well-defined problems had clear goals
specified, which could be resolved sequentially by a series of fixed rationales. These
rationales responded to certain formulas or rules. The steps to solve well-defined
problems were limited, with limited solutions. When the correct formulas and rules
were found and fixed routines applied, problems were resolved (Simon 1973).
One example of a well-defined problem solving technique is troubleshooting.
Troubleshooting involves a systematic approach, usually a checklist application, to
10
Data is original raw information that has not been organized nor analyzed. Semantic information
is organized information that is processed with certain methods or procedures, which relate to the
semantic considerations, to turn the data into a certain format that can be used for decision
making.
locate and correct problems in existing engineering systems. By the same token, an
ill-structured problem can be resolved similarly by dividing the problem into sub-
problems, which can again be divided further to other levels of manageable sub-
problems until the problem is solved.
Regarding the nature of a problem, Newell and Simon clearly defined that a
problem has a starting stage and a final goal stage with a number of in-between
stages. In order to achieve the final solution stage, actions on selecting operators
must be taken step by step. The available information and actions to be executed for
each problem constituted its unique universe, which was called the problem space
(Newell and Simon 1972). Regarding the nature of solving a problem, a number of
studies were conducted through experiments to explore the characteristics (Newell
et al. 1958; Newell and Simon 1972; Eastman 1970; Akin 1979; Chan 1990). In
general, problem solving activities could be outlined as: identify the problem, pos-
tulate possible solutions, test the best solution, and determine if the problem is
solved. More specifically, the problem solving activities included the following
eight cognitive sequences: (1) identify and select problem; (2) analyze the selected
problem; (3) generate possible solutions; (4) select and plan the solution; (5) imple-
ment the solution; (6) evaluate the solution; (7) determine if the problem is solved;
and (8) memorize the final solution as a knowledge schema for future use. These
eight cognitive stages are common thinking sequences for addressing a problem. In
design, each of these stages has unique procedures that characterize design cogni-
tion, which are elaborated in the following on knowledge and human intelligence.
In the 1970s, studies in social policy by Horst Rittel and Melvin Webber
posted a good question on whether social scientists could solve social policy
problems in the ways scientists and engineers solve their sorts of problems
(Rittel and Webber 1973). The question is based on the reasons that: (1) policy
problems cannot be definitively described; (2) the classical paradigm of science
and engineering problem solving is not applicable to resolve problems of open
societal systems; and (3) the cognitive and occupational style of the social pol-
icy professions could not work on a much wider array of social problems. They
used the term wicked problems to differentiate tame problems occurring in the
areas of natural sciences and engineering.
To Rittel and Webber, natural science problems are definable and solutions are
also findable, thus these problems mostly are tame ones. Governmental planning
problems are ill-defined, and the resolution relies on elusive political judgment,
which might lead to the situation of solving the problem over and over again. For
instance, what should we do to reduce street crime? Street crime is a problem that
has infinite associated variables. Any solution proposed might trigger other prob-
lems. Thus, social policy problems are inherently wicked. Unlike well-defined
problems, which have true or correct solutions, solutions to wicked problems are
not true-or-false, but good-or-bad or better-or-worse. In comparison to well-defined
problems that have many applicable and potentially satisfactory solutions, wicked prob-
lems have the characteristic that the problem might be a symptom to another problem
and it may reoccur. The solution to be implemented is obligated according to law
and consequential in action. It is also true that the implemented solution might have
bigger impacts to a larger population. As such, one cannot build a subway to see
how it works and correct it after its unsatisfactory performance is found. As Rittel
and Webber argued, every attempt of the solution counts significantly even though
the problem may re-occur and the solution may be revisited. This explains the spe-
cial type of social policy problems that differ from scientific discovery, natural sci-
ence, and design problems. They concluded that the kinds of problems that planners
deal with—societal problems—are wicked ones, for they are liable for the conse-
quences of the actions they generate; they have no right to be wrong. For govern-
mental public policy problems, it has been argued that policy problems are defined,
and not identified or discovered; they are rather the products of imposing certain
frames of reference on reality (Dery 1984).
Differences between problem solving activities of well-defined and ill-defined
problems are that in the well-defined domain, the problem most of the time is
known, solutions are limited, steps used to evaluate the solutions are limited and
goal state is clear. Legal impacts to the public on solving well-defined problems are
not big in scope. Wicked problems are very much the same as ill-defined ones, but
actions executed in wicked problems are liable and solutions executed should not be
wrong. Table 2.3 briefly summarizes the sequential steps of the thinking activities
that occur in the three different kinds of problems. In the table, a checkmark indi-
cates the type of activities that would occur in the category of problems. For instance,
analyzing the selected problem in the second row of the table would definitely
occur, as checked, across all problems. Design problems, classified as ill-defined
problems, are similar to wicked problems to some extent that design solutions
should also be right or correct. Otherwise, it shall go through a number of revisions.
For instance, the bridge design on the Canal Grande in Venice by Santiago Calatrava
in 1996 had gone through numerous structural changes, because of the mechanical
instability of the structure and the excessive weight of the bridge, which could cause
the banks of the canal to fail.11
In studying design thinking, problem-solving theory has inspired scholars to
work on psychological modeling of the way designers think and do design. In fact,
Table 2.3 Cognitive sequences and characters of three major problem solving activities
Problem solving stages: Well-defined Ill-defined Wicked
Identify and select problem Known Unclear Unknown
Analyze the selected problem √ √ √
Generate possible solutions Limited Unlimited Unlimited
Select and plan the solution √ √ √
Implement the solution √ √ √
Evaluate the solution √ √ √
Determine if the problem is solved Clear Unclear Unclear
Develop a schema for future use √ √ √
Consequence of actions Limited Limited Liable
Problem reoccurrence No No Yes
11
See the criticism of Santiago Calatrava’s bridge design on Wikipedia Web page, URL: http://
en.wikipedia.org/wiki/Santiago_Calatrava
the problem solving theory, with its connection to cognitive psychology, provides a
good structure for scientifically explaining how humans learn, process, store, and
search for information for design. Because of the research methods developed in the
field since the 1950s on modeling human thinking patterns, problem solving theory
provides an essential research platform for exploring style and creativity. For exam-
ple, one early working hypothesis on creative thinking is that creative thinking is
simply a special kind of problem-solving behavior (Simon et al. 1962), and it applies
that creative design thinking is also a special type of problem solving behavior.
Research methods used to explore design thinking are to identify the characteris-
tics of human information processing. Particularly, it is critical to recognize mecha-
nisms and operators that designers use to process information mentally while they
are working on design, and to have appropriate conceptual representations used to
model the process. Conceptual representation is to use some means to appropri-
ately represent the ideas created in the design process. A number of studies have been
generated with this approach. For example, Chuck Eastman (1970) observed the
cognitive processes in a space-planning task and explained the characteristics of
“generate-and-test” processes. Foz (1972) observed design behavior in developing a
parti design. Omer Akin (1979) developed a number of processing models to describe
the cognitive aspects that occurred in design processes. Jane Darke (1979) intended
to find whether architects had in mind an image or expectations about users during
their design period. Stiny and March (1981) developed a design language starting
from the information-processing theory to the creation of shape grammar. Other
research studied Frank Lloyd Wright’s design process by observing Wright’s works
from 1901 to 1910 and theorizing the resulting Prairie House style (Chan 1992).
Most studies in this trend focused on exploring design actions with the purpose
of finding an appropriate model or procedure that could explain the processes and
the phenomena generated by the actions. The efforts of these researchers are based
on treating design activities as mental processes of problem-solving, which could be
further applied to explore: (1) what kinds of external information are attended to in
the design problem solving processes? (2) how is internal information (knowledge)
applied efficiently on solving problems? (3) how is design knowledge structured in
memory? (4) what are the cognitive sequences on the use of design representations
in drawing, digital models, or physical models? and (5) what design tactics have
been used in the processes? These inquiries, as explained in the following sections,
relate to the cognitive processes and the internal knowledge structures stored in
memory, which are specifically addressing the essential nature of how knowledge is
created, maintained, utilized, processed, and presented in handling design tasks.
2.4.2 Design Is Making Associations: The Nature

of Knowledge
In design education, when a project is assigned as a studio exercise, instructors

encourage students to search for related and similar precedents and learn from
examples to solve the design problems at hand. This “learn by example” is a part
of the theory of making associations. But how could the learning processes be
scientifically explored to verify the theory? In the 1950s, Alan Turing proposed
the famous brain-as-computer metaphor, that the computer is an information-
processing unit, and the human brain also is an information-processing unit for
they both share similar mechanisms of input, output and central processing unit
(Turing 1950). The acquisition, storage, retrieval, and utilization of knowledge in
the human mind could be treated as a number of separated processing stages. If
knowledge could be represented numerically and programmed in computers by
sequences, the processes could be simulated to let machines execute intelligent
human tasks by implementing step-wise algorithms. Thus, digital computers
could perform the same tasks humans do. This notion relates to earlier studies
on understanding how knowledge is processed in the brain and controlled by
consciousness.
How human beings process knowledge is the study of human cognition.
Cognition is recognized as the “mental process of transforming, reducing, elaborat-
ing, storing, retrieving, and utilizing information” (Chan 2008). Information, in
this regard, is knowledge. The ways of organizing knowledge and combining the
knowledge with reality is cognition, which defines intelligence. In psychology,
knowledge has long been classified into two major types—declarative knowledge
and procedural knowledge. Declarative knowledge is concerned with knowledge as
static information that comprises the facts and concepts we know; whereas proce-
dural knowledge is knowing how to perform a task that comprises the knowledge of
procedures and methods of performance (Posner 1973; Winograd 1975; Anderson
1980). After procedural knowledge is practiced or performed for a while, it becomes
an automatic skill. On the other hand, when performing certain tasks, declarative
information will be transformed into a procedural form. Drawing is one example.
After basic drawing skills are learned and practiced, they become self-acting or self-
regulating under fixed conditions. When thoughts (declarative knowledge) are put
into drawing routines (procedural knowledge), higher levels of expression are cre-
ated. For instance, Picasso’s cubism paintings, Dali’s surrealistic paintings, or per-
spective drawing used to highlight certain aspects in architectural design are results
of adding declarative knowledge into procedural.
How is knowledge established? Jean Piaget, a major figure in developmental
psychology, explained the learning process of establishing knowledge from explor-
ing the cognitive development of children. Piaget theorized that children are con-
stantly creating and re-creating their own model of reality, achieving mental growth
by integrating simpler concepts into higher-level concepts at each stage. Intellectual
development has two processes (Piaget 1953). One is assimilation. Piaget assumed
that whenever one transforms the world to meet individual needs or conceptions,
one is assimilating it. It occurs when a child responds to a new event in a way that
is consistent with an organized pattern of thoughts already stored in memory. The
other process is accommodation, where one would modify some of the saved mental
structure of patterns to meet the demands of the environment. A child would either
modify an existing pattern, or form an entirely new pattern to deal with a new object
or event. Piaget argued that intelligence develops in a series of stages because one
stage must be accomplished before the next can occur. For the last stage, the child
would form a view of reality for that period of age. At the next stage, the child must
keep up with the earlier view by reconstructing or revising it to form a new version
of the concept from an older stage (Piaget 1967).
Such knowledge generation is a part of the evolution process of human intelli-
gence. Putting this theory into a larger picture, knowledge is accumulated from
experience and learned knowledge is connected to a-priori ones. Such learning that
involves relations between events, for instance, between a stimulus and a response,
or between a response and its consequence, is called associative learning (Gluck
et al. 2007). Associative learning, as explained in neuropsychology (Frith 2007), is
an important mechanism for acquiring knowledge about the world. The concept is
that “what is learned is an association between an arbitrary stimulus and a reward-
ing stimulus or a punishing stimulus (Frith 2007, p. 88). An example given by Frith
to explain such a phenomenon of learning is the color of a fruit signifying whether
it is ripe. As a fruit ripens, it tends to be less green or redder, and people prefer nice
ripe (redder) fruit over the nasty unripe (greener) fruit. Therefore we learn about
which are the nice fruit from their color. The association between color and ripe
(arbitrary stimulus) and nice fruit (rewarding stimulus) provide us with enough
knowledge for decision making on selecting nice fruit.
In the professional practice of design fields, procedural knowledge also relates to
the professional knowing that is engaged by practitioners in everyday practice over
various projects, different from learning the knowledge presented in textbook or
journal papers. Designers could learn more through reflection in the midst of action
and get intuitive knowing from this reflection. This reflection-in-action notion
(Schon 1983) also explains the cognitive interaction between procedural knowledge
and declarative knowledge, which is what designers could learn from executing
procedural knowledge to earn new knowledge. And, the new knowledge learned
from professional practice would be the new facts or concepts generated from
actions of doing, which would become a part of the designers’ declarative knowl-
edge. Often times, the newly generated knowledge that is inspired by the old knowl-
edge would have some connections between the two, which also relates to how
knowledge is structured in mind and stored in memory.
How is knowledge structured in memory? In psychology, the structure of
knowledge in memory was theorized as the mind composed of a network of ele-
ments—usually referred to as sensations and idea—that are organized by various
associations. The concept of associations between events or elements are formu-
lated by chronological contiguity, frequency of connection, similarity and contrast,
cause and effect, or by associated meanings or experiences formed through learn-
ing. Philosophers have discussed the theory in detail. But, Hermann Ebbinghaus
(1850–1909) was the first one who scientifically studied the association of memory
through applying a set of nonsense syllables to explore how memory is formed and
recalled. He theorized that participants would not have previously associated
meanings and experiences with the provided stimuli, and when asked to remember
them, they would be forming new memories. The goal of his study was to explore
how associations between these nonsense syllables could be created without the
use of previous knowledge, learning, and experiences that are normally available
to humans (Davis and Palladino 2002). Results confirmed in his experiments that
the most commonly accepted law of association is by contiguity (the items next to
one another are associated and memorized). Ebbinghaus argued that memory was
a process of receiving experiences and storing them away to be recalled later. The
processes of memorizing something involved the formation of new associations,
and these associations could be strengthened through repetition. More repetitions
of exposure to the nonsense syllables generated a stronger memory of them. Thus,
Ebbinghaus proposed that after an association is established between two ele-
ments, these elements are linked, stored, and could be recalled through the built
associations (Ebbinghaus 1913). Together with other associations on meaning and
experience between memorized events, the concept of human memory structure
was further developed into semantic network theory (Collins and Quillian 1969;
Collins and Loftus 1975), schema theory (Rumelhart and Ortony 1977) and net-
work theory (Anderson 1980). All these theories propose that memory is treated
as a hierarchical network of nodes or chunks. Chunks are connected by various
associations. Personal knowledge is stored in information chunks or schemata that
compose the mental constructs for ideas.
How is knowledge stored in the brain? Metaphorically speaking, the brain could
be seen as the hardware and the mind as the software representing the conscious-
ness that manages information stored in the brain. When information comes into
our memory through sensory input, it needs to be changed into a form that the
memory system can store, which is the encoding process. Details of the encoding
process are unknown, but formats of information code stored in the brain have been
explored in psychology. For instance, music input will have acoustic coding,
whereas pictures have visual coding and concepts semantic verbal coding. Thus,
after information is encoded and passed through short-term memory, it will be
stored in long-term memory by chunks. People could remember the pictures,
sounds, or words and present them immediately after they are recalled. In cognitive
psychology, Paivio (1971, 1986) proposed a dual-coding theory, which posits that
an encounter with either a picture or a word may develop both a “visual” and a
“verbal” code. But the visual codes are formed for pictures and verbal codes are for
words. This suggests that knowledge is coded by certain format and humans could
handle various codes developed during the process of perception.
The dual-coding theory has strong implications for design visual thinking. This
is due to the fact that the use of mental images to help thinking is not new to most
designers. Designers often mentally manipulate the spatial relationships among fun-
damental geometric shapes and then represent results of this manipulation through
graphic drawings or physical models. Sketches are tried out in the mind’s eye before
they emerge onto the drawing board. The concepts of visual coding were explored
in a study of mental images of fireplaces, which applied dual-coding theory to detect
the internal knowledge representation and further investigated the differences of
knowledge structure between expert and novice designers (Chan 1997, 2008). That
study discovered that architects seem to organize their architectural knowledge by
architectural functions. In the experiments conducted on drawing fireplaces located
in different environmental contexts, subjects were asked to mentally recall a
fireplace image they had seen before and draw on paper. The theory was that the
sequences of the fireplace drawing suggest the sequences of recalling the fireplace
coding from memory, which will reveal how the information chunks are stored in
memory. If the sequences between two consecutively retrieved image components
have architectural functional connections, then the chunks stored in memory would
also be associated with architectural functions. The collected experimental data
showed that expert architects had 75 % of the links that were functional, whereas
52 % of the novices’ links were functional. The study also found that expert archi-
tects have larger chunks of architectural symbols (M = 73) than do novices (M = 40.5)
in the collected memory data. Results of the study suggest that architectural knowl-
edge in memory is hierarchically built up through functional associations that link
architectural components together.
Expert designers seem to organize their design information in a functional way
such that they are able to store more information, retrieve it more rapidly, and more
capably fit known objects into new contexts. Studies in chess (Chase and Simon
1973), painting (Gombrich 1960), and architectural design (Chan 1997) demon-
strated that expert chess players, painters, and architects tend to develop large sets
of domain-specific knowledge chunks during years of practice, which are stored in
long-term memory. Among these knowledge chunks, some are the patterns of the
solutions generated before, termed pre-solution models or cases, which could be
used to solve new problems. All these knowledge chunks shaped their chess-playing
strategies, painting skills, and design expertise into a set of repertoire in memory,
which forms master chess player, painter, and architect resources of creativity. This
body of knowledge repertoire or pre-solution is the first notion of design creativity
(DC1) explained from the cognitive psychology perspective.
Similar notions that knowledge is stored by chunks in memory have also been
demonstrated in the field of neuroscience.12 In neuroscience, scholars have catego-
rized knowledge storage through memory types. For instance, semantic memory
refers to the memory of meanings, understanding, and acquired general knowledge.
Episodic memory refers to the memory of events, times, places, emotions, and
knowledge that relates to an experience. Procedural memory refers to the memory
of skills and procedures. Combining semantic and episodic memory, declarative
memory is formed in the brain. Neuroscientific research suggests that procedural and
episodic memory use different parts of the brain working independently. Some
believe semantic memory resides in the temporal neocortex, which is involved in
auditory processing and is home to the primary auditory cortex. The temporal lobe
(see Fig. 2.2) in the brain is also heavily involved in semantics, both in speech and
vision. It is suggested by evidence that the temporal lobe contains the hippocampus,
12
Design thinking is the information processing (IP) sequences in our brain and mind.
Metaphorically speaking, the brain is the hardware and the mind (consciousness) is the software of
the human IP system. At this point of time, neuroscience studies the hardware (brain) and cognitive
science focuses on the software (mind). In the future, we have to explore how the two components
align together. The brief introductions provided in this chapter intend to explain the connections
between the brain and the mind during thinking. More details could be found in the well written
book entitled “Making up the Mind, How the Brain Creates our Mental World” by Frith (2007).
Fig. 2.2 The temporal lobe

in the brain (Source: Image
was adapted from Wikimedia
public domain, http://en.
wikipedia.org/wiki/Human_
brain#mediaviewer/
File:Gray728.svg.
Accessed 8 July 2014)
which is essential in the formulation of new memories about episodic and general
declarative memory, and it is used for storing and processing spatial information.
Therefore, the temporal lobe that contains the hippocampus should be involved in
formulating memory as well. This brief explanation describes the fundamental struc-
ture of how knowledge is stored by types in different brain areas, and different areas
of the brain process different types of intelligence.
How is knowledge retrieved from memory? In semantic network theory (the first
modern theory developed and the most influential one), each node is a symbol that
can be interpreted as representing a specific concept, a word, or a feature. The link
between nodes is a relation or “association” that can hold between any two nodes.
Processing in a semantic network is the form of spreading activation or search via
associations. Spreading activation suggests that the activation of a concept can lead
to the activation of other related concepts as the activation spreads along the paths
of the network. This explains the notion that knowledge can be learned by associa-
tion, stored in memory by groups, and better retrieved by association as well. For
instance, in the psychological study of language, some psychologists have explored
that humans can learn language by learning the associations between adjacent words
in a sentence. Blank and Foss (1978) studied how an appropriate semantic context
facilitated the comprehension of sentences by providing relations. In other psycho-
logical experiments, studies on organized and random words recall have shown that
words linked together with associations are recalled better than randomly linked
ones (Bower 1970). Recent experiments also show that people are biased by past
experience to make new decisions between options that were never previously
rewarded (Wimmer and Shohamy 2012). In neuropsychology, it has been proposed
that the hippocampus encodes relationships between items and events that appear
together, forming an associative link between them. It is because of the built
associative link, when a person encounters one item; the hippocampus completes
the pattern and reactivates the neural representation of the other item to integrate the
old memories with new ones. This explains the association aspects of memory con-
struction, storage, and recall in both the hardware (brain) and software (mind) sides
of human cognition.
Lastly, how is knowledge presented in the mind when we recall and process it?
This relates to the ways of describing the nature of ideas and concepts appearing in
thinking, and the format of internal knowledge representations. Here, the format
relates to the structure of a code, as defined by the nature of the elements and the
nature of the relations between them (Kosslyn and Pomerantz 1977). It has been
argued that mental objects exist that are abstract structures of propositions express-
ing precise relations between entities, and they have evaluable semantic properties
of content, reference, and true-value (Minsky and Papert 1972; Anderson and Bower
1973). This propositional format of knowledge explains that material in memory is
represented by networks of propositions (or called nodes), which are entirely
abstract objects (Pylyshyn 1973). On the other hand, visual modality is another
significant set of information in our mind, and visual image has been described as a
picture in the head, stored in memory and organized into meaningful parts, which in
turn are remembered in terms of spatial relations among them (Reed 1974). Mental
images, in fact, are easy to recall and utilize to help solve certain types of problems
requiring the use of images.
Beyond the network theories, studies on knowledge representation in the 1980s
turned to the notion of “connectionism,” which proposed another network modeled
after neural networks in the human nervous system. The concept is analogous to
neural processing in the human brain. Every node in the network is a neuron-like
(see Fig. 2.3) processing unit with weights that measure the strength of connections
between the neuron units.13 Connectionism also is called the Parallel Distributed
Processing (PDP) approach, as the processing takes place in a parallel format and
the output is distributed across many units (McClelland et al. 1986). This approach
generated another conceptual framework for understanding the nature of the mind
and its relation to the brain, which also applies to the study of design thinking.
In sum, we organize our knowledge in memory by association and retrieve infor-
mation by association as well. Association is inferred from the tendency of one item
to evoke another, which is triggered by resemblance, by contiguity in place or time,
by frequency of connection, and by cause and effect between two items. The theory
of association has served as the central core in the history of publications on mem-
ory, and has become a theoretical view embracing the whole of psychology.14 But
the associative properties among chunks (or nodes) in regard to the linguistic infor-
mation stored in both semantic network and network theory have been criticized, as
they might not adequately encompass all aspects of meaning of the information
(word or feature) in the chunk (Harley 1995), and the search for information through
the network hierarchy is not economical (Rosch 1978). Similar considerations were
13
A neuron is a functional unit of the nervous system. Each neuron could be seen as a powerful
microprocessor, for it has three components analogous to a computer. One component of cell body
(soma) contains information for managing cellular functions. One of the cellular functions is
receiving and sending messages. The second component is dentrites that serves as the reception
sites for incoming messages. The third component is the axon that is the mechanism for sending
messages on to other neurons.
14
Concepts of association could be found in the Encyclopedia Britannica online edition. See URL:
http://www.britannica.com/EBchecked/topic/39421/association
Fig. 2.3 A diagram of neuron (Source: Wikimedia public domain, http://commons.wikimedia.org/

wiki/File:Complete_neuron_cell_diagram_en.svg. Accessed 10 Oct 2013)
posted to the schema theory as well. However, association has long been recognized
as the major theory that the mind is composed of elements of knowledge chunks
organized by means of various associations. Thus, association is a major cognitive
factor in human intelligence operation. Therefore, if a designer makes an unusual
link to a chunk of knowledge in memory, retrieves it, and applies in design, then the
result might be unconventional, and suggests one of the causes of creativity. This
association of diversified knowledge information in design is the second notion of
design creativity (DC2).
2.4.3 Design Is Goal and Constraint Driven
Problem solving is a sequence of moving the problem forward stepwise to reach the
solution. If the problem solver does not have goals in mind to accomplish, he/she
would not be able to strategically move the problem ahead to complete the mission.
Real-world problems tend to be complex, which cannot be solved with a simple
one-step solution. Problem solvers must set up sub-goals that must be accomplished
on the way to achieve the overall goal (Thevenot and Oakhill 2006). Therefore,
problem solving is goal driven.
Similarly, designing is proactive problem solving (Baker and Dugger 1986). It

includes defining a concept, refining the concept through research, experimentation,
and necessary development to prepare the product for production. When designers
develop the concept, they have to understand the nature of the problem, explore the
issues to address, formulate the problem, and utilize strategies to come up with a
workable scenario. These activities usually consist of various formats of thinking
patterns of setting up design hypothesis, deciding issues, providing intentions, and
putting up proposals. If occurring at the beginning of the problem solving processes,
the stage of having these cognitive activities activated could be called the design
intentions development stage, which is the fundamental stage determining the out-
line of the solution. At the end of this stage, a concept is created and a series of
following stages would be further developed to complete the concept. Thus, the
consequential following stages represent goal stages. As such, design activities are
driven by sequences of achieving certain goals (Akin 1986, p. 20), and each goal has
constraints attached to narrow down the scope of search for solutions (Simon 1969).
The goal sequences appearing in the design processes can be seen as a game plan
for solving a design problem. A study on exploring the cognitive processes in archi-
tectural design problem solving has proven that a goal plan does appear in a design
course (Chan 1990). In that study, a designer was asked to design a three-bedroom
dwelling for a single family on a large site. The total floor area was limited to 2,200
square feet. The owner was a professional architectural perspective draftsman.
Design requirements were set to minimal to explore how the designer would generate
the design and what personal design knowledge had been applied within the inten-
sive period of design creation. Protocol data collected in that nearly four hour
(232 min) lab experiment shows a total of 22 goals that were achieved. Among them,
14 were major goals, and eight subordinate goals were developed to help solve their
primary major goals. Data also showed that when a goal was achieved, the subject
had a clear idea about what the next goal state was and gave keywords to describe the
next move. The new state was usually discrete from its previous state. That study
explained that at the first stage of problem solving, the problem understanding stage,
a general goal sequence would have been prepared to guide the sequence of actions,
and then executed one after another. If a goal cannot be achieved and the problem is
not solved, a sub-goal is strategically developed immediately (Chan 1990). This phe-
nomenon of setting up goals in the beginning is the process of understanding, formu-
lating, and structuring the problem. During design, the structure needs sub-structures
through developing sub-goals to find a satisfactory solution.
Another design thinking study found that a designer knows the general method
of approaching a design, for that designer applies his own algorithms skillfully on
solving the problem (Chan 2001a). In that experiment studying an architect’s
individual style, protocol data showed that the architect used the average construc-
tional cost per square feet as a hypothetical assumption for determining the overall
building size. He divided the given budget by the unit cost; a gross total floor area
was created and used to generate a list of suitable rooms and their dimensions.
Again, from modifying the gross area through adjusting room sizes, the final
building size and footprint were determined. After that, the architect developed a
scenario that summarized the design assumptions and major considerations into a
conceptual framework that served as a potential solution (Chan 1993). Such steps of
sequences were used in his other designs in the experiment. Thus, a general method
or an algorithm was developed, and a design routine employed, formulated from
years of practice working on a particular building type (Chan 1990, 1993). When
designers encounter a new problem that has not been solved before, a new plan will
be tried, new approach will be explored and memorized. As indicated by Heath
(1984), such a new method developed might be individual and personal. It is also
possible that every architectural problem requires a method specific to its building
type. For instance, techniques used for generating a hotel design are different for
designing a residential building or an airport terminal.
In accomplishing a goal in the design process, there usually is some type of
information applied as a guide to limit the search efforts for solutions, or as opera-
tors used for creating solutions. These pieces of information, as add-on informa-
tion called constraints, imposed in the design processes externally and internally,
become limiting parameters. For instance, external constraints are the design
requirements given to the designers by the clients, users, regulations, and the envi-
ronmental context as parts of the problem components. In practice, clients present
their expected ideas to designers when they commission the designers to work on
the design project. Users, sometimes neither the clients nor the designers, have dif-
ferent perspectives or functional requirements on the use of the products than the
designers, and designers must receive information relating to user-friendly and
accessibility issues. In industrial design, legislators pose considerable restrictions
to the health issues on the products that industry might create. In architecture
design, the context of the site environment of the building provides forces to influ-
ence the orientation, circulation, and façade designs. Material specification, build-
ing code, and urban ordinances have certain regulations applied to the building
design as well. All these requirements and expectations are part of the information
given externally to the designers for design considerations, and, thus, are termed
external constraints.
When the given external information makes the problem more complicated,
problem solvers have to sort out the complexity and search for the appropriate solu-
tion among a huge number of alternatives. In order to effectively solve the problem,
designers formulate the problem, identify key issues to consider, address only
related information to guide the search effectively, and propose possible solutions.
In this framework, some information from knowledge stored in memory or informa-
tion accumulated from research is retrieved and collected to analyze the problem,
shape the structure of the problem, develop a solution scenario, and generate alter-
native solutions. Such information composed from knowledge in memory is the
internal constraint. Internal constraints are imposed by problem solvers resulting
from cognitive mechanisms of reasoning operating in the processes. In architecture
design, the internal constraints frequently comprise the majority of the design pro-
gram (Lawson 2006). Architects generate internal constraints in the early stages to
determine the number and sizes of spaces of various kinds and qualifications from
the clients’ given external constraints. In the same study of cognitive processes in
architectural design (Chan 1990), the collected protocol data had clearly shown that
the designer called from his experience to develop a set of internal constraints to
respond to the given external constraints in the beginning of the design stage. These
established constraints became global constraints that were constantly addressed in
the entire processes. Based on the set of constraints, a conceptual framework was
formulated; solutions were generated as design conventions do, through bubble dia-
grams and flow charts to represent the relationships of the spaces. Thus, internal
constraints are the pieces of information retrieved or gathered from knowledge
internally by designers for solution generations.
Regarding the characteristics of constraints applied in design, Lawson has cate-
gorized four types of radical, practical, formal, and symbolic constraints (Lawson
2006). Radical constraints are the primary and fundamental purpose of the object or
system being designed. Practical constraints are the technical problems of making
the design in reality. Formal constraints are the rules of proportion, form, color and
texture applied to organize the form and appearance of the design objects. Symbolic
constraints are symbolic meanings used to generate the object. These categorized
constraints could be external or internal ones utilized in the design process, depend-
ing upon who provides the information. If the constraints are initiated and imposed
by designers personally, then they are internal ones and external otherwise. In sum,
it is clear that design problems are constructed by external constraints given by
external forces, formulated by internal constraints imposed by the designers, and
proceed from executing the goal sequences of actions to generate a solution that
satisfies all these constraints. If designers can either utilize the external constraints
uniquely, impose internal constraints innovatively, or take extraordinary actions in
goals to sequentially create a solution that meets the constraints unconventionally,
then a creative design emerges. This unique development of goal sequences, or
utilization of external and internal constraints explains the third notion of design
creativity (DC3).
As explained, designers apply internal constraints either to minimize their
decision making, or to narrow down the space for searching for solutions. From the
design strategy point of view, the utilization of internal constraints would make the
design processes more efficient and effective. However, external constraints gener-
ate different effects. More given external constraints set up more limitations for
decision making, for solution search, and for knowledge application, and are less
flexible for considering creative components for solution generation and on mak-
ing special links for innovative associations. For example, museums and public
buildings might have fewer constraints than commercial buildings. Most of the
classical designs are public buildings of theaters, museums, cultural, or convention
centers done by master architects. In designing public buildings, architects may
have more internal constraints to utilize and less external constraints to restrict
their imagination. The Pritzker Prize laureates, for instance, have designed more
public than commercial buildings, which make their designs outstanding. It is due
to the fact that most commercial buildings, particularly shopping mall designs,
have many constraints on the number of stores, the operational format, the business
types, and the type of the image to be provided to the developer. These constraints
are mostly imposed by clients and are more critical than the constraints given in
public building designs. Thus, public building types provide better opportunities
for creative innovation. Another example of how fewer external constraints may
promote creativity is the Beijing MOMA project designed by Steven Holl. Holl
explained that his client (Wanker) would accept any ideas he provided and increase
the building budget to cope with all the concepts proposed.15 The real estate devel-
oper had indicated to him, as explained in an interview, “We are going to build all
of these. We know we can sell all these apartments. What is important is the spiri-
tual dimension.” Thus, creativity appears more frequently with less extrinsic con-
straint circumstances. This quantity of design environment constraint is the fourth
notion of design creativity (DC4).
2.4.4 Design Is Reflective and Problem Structuring
In life, people have to think to resolve problems that might happen every day. In
some cases, there might not be problems waiting for them except in their daily life
routines. Under this situation, some people would ask questions to either find out
what they should be doing to continue the life routine, or to identify potential ,solu-
tions that might exist to make their life better. In other cases, people might have
problems waiting, and they would also self-ask questions about the importance,
critical state, and deadline of each problem to decide a priority list of the problems
to be resolved. Considerations of the criticality and importance of problems are, in
fact, setting up constraints to determine the goal sequences before handling the
problems. Such a self-asking question phenomenon is a part of cognitive procedures
utilized to generate actions for life routines, to find new problems, or to determine a
problem priority list.
After a problem is taken into consideration, the problem solver needs to under-
stand the problem, find related information, decide the requirements, recognize
expected outcomes of the problem, formulate the problem in the mind, and find
actions to implement the outcome until a solution is generated. This series of proce-
dures could be seen as standardized problem solving routines for handling general
problems. Yet, during the entire process, people would keep asking hypothetical
questions as a technique to narrow the search efforts for best results. This phenom-
enon of asking questions to understand the problem, to set up sequences of moves,
to impose constraints strategically for solution generation are the same as what hap-
pens in design thinking.
In design, a design project commissioned by a client has many sub-problems
involved, which are more complicated in nature than well-defined problems.
Designers handle sub-problems by managing the relationships among sub-problems.
15
In the interview by Charlie Rose on July 23, 2007, Steven Holl indicated the real estate developer
of the MOMA complex had handled the design ideas proposed by Holl with fewer constraints. The
developer would take everything proposed by Holl and build them.
The relationships among sub-problems and the overall framework that holds the
design project together define the structure of the problem. Explained from the cog-
nition point of view, a problem structure can be seen as the format that consists of
goal sequences, knowledge representation used to make design components, and
design constraints developed to achieve goals. Such a format, formulated in the
design processes, is the result of overall problem structuring (Chan 1990, p. 69).
The problem structure signifies the problem context that characterizes the situation
of a problem. When the problem structure is established, a schematic solution shall
be reached. Usually, the primitive problem structure is formulated in the very begin-
ning of the problem solving (Dunn 2004; Chan 2008), which is also called problem
definition (Bardach 1981, 2000) or problem delimitation (Vesely 2007). At the stage
of initial problem structuring, the problem solver would go through a series of ques-
tions to better understand the problem and to establish scenarios. After the problems
are clearly understood and acted on, the basic structure would be refined or modi-
fied as needed. Analyzed from the design point of view, a problem structure is
developed through the uses of drawings, sketches, scribbles, bobble diagrams, or
any other representations; designers would mostly either keep working on the same
structure or some would redefine the structure until a product is generated. Logan
and Smithers (1993) have described that the formulation of the problem at any stage
is not final. Whenever new aspects or inconsistencies are revealed in the formulation
of the problem, the problem and the solution are redefined. To them, the process of
designing is the process of exploring the formulation of the problem, or the explora-
tion of structuring the problem. Nonetheless, while working on the problem struc-
turing, designers would post hypothetical questions for implementing, maintaining,
and modifying the problem structure from time to time.
Since the problem structure consists of a number of sub-problems, the corre-
sponding solutions for sub-problems also generate a relationship structure, which
can be seen as a solution structure. A solution structure is the sequential order of
creating solutions. Design is an accumulation of a series of solutions to yield a big
solution product. The timing for solution generation might affect the problem struc-
ture or affect the solution for the next sub-problem in line. This notion of problem
and solution contexts can be identified through a problem behavior graph (Newell
and Simon 1972; Chan 1990, 2008). Both the problem structure and solution struc-
ture compose the complete problem situation. Cognitively, designers would con-
stantly check the problem situation on both problem and solution structures for
accomplishing a final solution. Thus, the thinking activities in a design have the
following patterns:
1. Defining problem stage: The initially presented set of external constraints by the
clients for a design is translated into a set of design problems, issues and ques-
tions sufficiently enough to well define the project situation to allow specific
actions on research (Woolley and Pidd 1981). At this initial stage of problem
solving, designers would review the context of the problem and ask questions to
clarify the problem context. The original problem statement would have been
considerably widened and recast. In some cases, it might have the whole set of
problems defined.
2. Creating solution stage: When the problem solving is moving ahead, the designer
would ask what-if questions to simulate the problem situation to generate a num-
ber of alternative solutions. This is similar to the play in chess. Master chess
players would consider a number of moves in advance and select the most win-
nable move. In design, designers would also ask what-if questions, responding to
the problem context for simulating a series of hypothetical moves, and generat-
ing a number of alternative solutions.
3. Deciding and evaluating solution stage: When decisions on solutions must be
made, designers would think of all possible questions to evaluate the situation of
the problem context. They will simulate the future aspect and perceive any con-
flicts or potential problems that might arise for selecting the best one to fit the
solution structure. Designers must be aware of the problem situation from time
to time by asking questions.
These cognitive phenomena of asking questions for communicating the problem
situation in the processes of designing has been explained by Donald Schon (1983)
in his theory of reflection-in-action. Schon explained that design processes are
actions of reflection, and designers learn from asking “what-if” questions in design.
Schon indicated that there are more variables than can be represented in a finite
model to simulate the design process. Any of the designer’s moves tend to produce
consequences other than expected and intended. When it happens, the designer may
think on the unexpected changes by forming new understandings, new apprecia-
tions, and by making new moves. In this case, when a designer makes a move, the
original problem situation changes and talks back to the designer, and he must
respond to the situation’s back talk. In design, the conversation with the situation is
reflective. Whatever moves the designer made, he must reflect on the unexpected
consequences and implications, listen to the situation’s back talk, form new appre-
ciations and reframe the problem to guide his further moves. While answering to the
situation’s back talk, the designer makes a reflection-in-action on the construction
of the problem, shifts stance as they do so from “what if?” to recognition of implica-
tions, from involvement in the unit to consideration of the total, and from explora-
tion to commitment, which are the characteristics of reflective conversation with the
problem situation (Schon 1983, p. 103). Therefore, Schon concluded that our think-
ing serves to reshape what we are doing while we are doing it (Schön 1987). This
also explains the nature of problem structuring and restructuring in the processes.
Lawson further explained the concept of reflection in the design processes as
reflection in action and reflection on action. Reflection in action relates to the
activities to formulate the problem, move the problem ahead, and evaluate the
solutions. The designer continually reflects on the current understanding of the
problem and the validity of the solutions (Lawson 2006). Reflection on action
relates to the asking of whether the process is moving correctly, which problems
have been examined, and the processes involved in representing, formulating and
moving. These activities are acting upon the checking of the problem situation. In
a design process, there might be the chance that a new solution created would
totally change the problem situation, which is then called a critical problem situa-
tion. In a study on design problem solving, it was found that when the designer
generated a solution (in that study, the problem was to solve a bay window for a
façade design) and put the solution on a drawing, a critical problem situation was
recognized and forced him to make design decisions. The critical problem situation
was defined as having the possibility of changing the problem structure or solution
structure, and is the state of affairs that will lead to a possible restructuring of the
problem (Chan 2008). In that study, the designer perceived that the solution would
cause the change of the interior form, material, structure and the character of space;
thus, he abandoned the solution and changed to a different one (Chan 1990). This
explains the phenomenon of reflection on action that relates to the activities in the
process of problem structuring.
In summary, the processes of actions in reflection would have the potential of
finding new problems in the context of designing, or redefining the problem struc-
ture while evaluating the solution structure. In industrial design, Dorst and Cross
have explained in their studies that the restructuring of the problem in the design
processes do lead to the generation of creative solutions. From the protocol data of
asking nine industrial designers with 5 or more years of professional experience to
create a concept for a “litter disposal system” in a Netherlands train, they identified
the aspects of creative solution in design. In their studies, they concluded that cre-
ative design does relate to the formulation of the design problems and to the concept
of originality. Creative design seems more focused on developing and refining
together both the formulation of a problem and ideas for a solution (Dorst and Cross
2001). Eberhard (1970) gave an example on designing his client’s office door.
During the design process, the designer asked whether a doorknob was the best way
of opening and closing the door, and further questioned whether the office needed a
door. Eberhard indicated the phenomenon as regression, and it could be explained
that after the problem was analyzed further and more information obtained about it,
the designer became aware of the necessity to change the problem structure for a
creative solution (Eberhard 1970). These aspects of the cognitive activities of
problem structuring (in the beginning of the problem solving) and reconstructing,
(from what-if reflection) attributing to creative leap, do explain the aspect of possi-
ble resources of creativity in design thinking. Thus, the problem structuring and
restructuring is the fifth notion of design creativity (DC5).
In the fields other than design, problem structuring, or what Schon called prob-
lem framing, has been recognized as an important problem solving process. Many
studies have explored the techniques of problem structuring for systematically
developing the relationship among sub-problems to make the problem solving
more effective (Vesely 2007). In operational research (OR), problem structuring
methods (PSM) have techniques developed to provide decision makers with sys-
tematic help in identifying an agreeable framework for their problem (Rosenhead
1996). In civil engineering, the processes of problem re-structuring to match with
the development of solution structure have been applied as the basis of computer
algorithms for the co-evolution of a design problem with the potential solutions
(Maher and Poon 1996).
2.4.5 Design Is Looking for Representation
Knowledge, as explained in the previous section, is saved in memory by chunks.

When knowledge is accumulated to a shelfful, a repertoire of skills is developed as
well. The notion of repertoire symbolizes the chunks of specialized knowledge in
memory. In order to effectively utilize these chunks of knowledge for solving a
problem, chunks must be transformed into certain symbols and expressed outward
through some kind of representation. In this regard, the intangible knowledge would
become tangible. Thus, representation is to have something standing in for some-
thing else, and is the means for representing the things that happened in reality
(Echenique 1972; Hesse 1966). Here, representation could be either the act of rep-
resenting or the product of representing (Greco 1995).
When solving a problem, representations are developed internally and externally
by adding information, deleting information as irrelevant, and interpreting informa-
tion to create means for generating solutions. The means, created internally, is the
knowledge, and its structure relating to the task in memory is called internal repre-
sentation. Internal representation is the knowledge in the form of proposition,
schema, image, or some sort of isomorph. For solving a problem efficiently, it is
critical to build an adequate internal representation at the very beginning of every
problem solving process, and then constantly modify the representation to match
the problem situation until the solution is finalized. In some cases, an internal rep-
resentation is a medium representing the external task being confronted. In other
cases, it is an abstract structure of the problem. In design cases, the internal repre-
sentation developed in the mind shall be presented externally by certain means,
which is called external representation. External representations are the knowledge
and structure of the task in the environment used to construct physical objects, sym-
bols, graphs, scripts, programs, and as external rules, constraints, relations or logics
embedded in physical configurations (Zhang 1997; Chan 2009, 2011). There are
differences in the cognitive procedures applied to solving a well-defined problem
versus an ill-defined problem.
For well-defined problems, external representations are usually given, and solv-
ing the problem requires mentally constructing one or more internal representa-
tions, and finding appropriate rules to execute the tasks required in the external
representation. The mutilated checkerboard (MC) problem is a good example
(Newell 1965; Wickelgren 1974; Anderson 1980; Korf 1980), which applies a
standard 8 by 8 square checkerboard with two diagonally opposite corners removed
(see Fig. 2.4). The problem is to cover the 62 remaining squares using 31 dominos,
each domino covering two adjacent squares. In this example, the checkerboard is an
external representation, and internal representations must be developed to map the
external one for processing solutions. Data from an experiment shows that most
subjects solve the MC problem by applying dominos, the number of squares, and
their geometrical arrangement as their representations (Kaplan and Simon 1990).
Since each of the 31 dominos covers two squares, a covering initially seems possible.
But, after covering 30 black and white pairs with 30 dominos, the problem situation
Fig. 2.4 The mutilated

checkerboard problem
example
is always left with the impossible situation of having to cover two same-colored
squares with one domino. However, switching from the domino representation to
partition the squares into two equivalence classes of black squares and white squares
would allow subjects to reason about the numbers of squares of each type and to
make the crucial inferences needed to prove logically why such a covering (30 ≠ 32)
is impossible (Korf 1980). Thus, finding out the right internal representation or
changing representation is critical for solving well-defined problems.
For ill-defined problems faced by writers, painters, computer programmers, law-
yers, planners, policy makers and designers, there is no clear external representation
given to show the expected final goal state. Problem solvers must construct internal
representations first and use external representations to visualize the internal ones to
move the process forward (Hayes 1981; Eastman 2001). In design, designers must
generate some design concepts in the mind, which are intangible and conceptual in
essence, and transform the concepts to internal representations to re-present design
concepts, schemes, knowledge chunks, visual images, or mental pictures. Then, the
designer would apply some media to implement the concept. The available media
for generating external representations include traditional paper-and-pencil mode of
sketch drawings, handmade physical models used since the Renaissance Period in
1400, digital models created in digital computers since 1937, and objects generated
by digital fabrication. In results, external representations are created, which are
visible and tangible, representing the forms to be constructed or to be built. These
forms are results of design ideas conceptualized and formulated through a series of
mental activities done by the designer. For example, in a building design, the archi-
tect must generate some mental image internally while he is designing, and use
drawings or models to display the image externally. During the process, the architect
must recursively modify both the internal image and external drawing to come up
with a product. Since the internal and external representations are both constructed
by designers, who have enormous and diversified categories of representations to
choose from, ill-defined problems provide more opportunities for designers to

develop creative products. As explained by Kosslyn, it is possible for human beings
to visualize the objects in question and mentally represent the image to solve it
(Kosslyn 1975). If designers make unusual mental images and apply them to solve
the design problem, then some creative forms are generated and creativity shows.
In the study of how humans search for representation on solving the MC prob-
lem, Kaplan and Simon (1990) showed that cues, hints, heuristics, and prior expe-
rience could guide the problem solver to a particular representation, which could
then be explored for a solution. Design is the process of making and searching for
representation. Designers must select some types of visible external representa-
tions to present their invisible internal representation, which could be mentally
constructed ideas, concepts, images, symbols, or graphs from experience, to the
world. During design, mapping activities between external and internal representa-
tions are constantly modified to correspond to the attempted design intentions;
dialogs are shaping constantly, and interactions are occurring. In complex prob-
lems, without recognizing how any of the two representations are mapping to each
other, or if they miss the interactions between the two representations, designers
cannot visualize how the design concepts are feasible and how they have been
implemented. In some cases, design conflicts happen and designers should look for
different representations to resolve the conflict, which might provide opportunities
to create innovative solutions. In sum, if a designer has generated a new representa-
tion that has not been used before or applied a conventional representation uncon-
ventionally to solve the same problem, then the outcome would be creatively
different. Thus, representation is a medium used to create and present a design
concept, a cognitive mechanism for solving a design problem, and one of the fac-
tors that causes creativity. These internal and external representation used in design
is the sixth notion of design creativity (DC6).
2.4.6 Design Utilizes Cognitive Strategy
In well-defined problems, there are limited procedures required to solve the problem.
For instance, algorithms in mathematics provide a series of steps. Attention paid to the
steps results in successful completion of the problem. However, complex tasks in an
ill-defined domain cannot be completed by just executing a fixed series of steps;
instead, a number of cognitive mechanisms are utilized strategically to deliberately
calculate a vast amount of design factors. These cognitive strategies are the use of
cognition strategically or methodologically to solve complex problems. A strategy is
not a direct procedure or an algorithm. It is a heuristic that supports humans to develop
internal procedures to perform higher level operations. For instance, in the field of
education, cognitive strategies applied could include the use of guiding procedures in
the process of learning, plus practicing basic academic skills simultaneously in a sys-
tematic way. Results of utilizing such strategies are to understand the task at hand and
learn the procedures to complete the task. A good example of a cognitive strategy
developed in teaching reading comprehension is to teach students to generate
questions about their reading. Generating questions does not lead to comprehension.
But, in generating questions, students have to search the text and combine informa-
tion, which helps students comprehend what they read (Rosenshine 1997).
Cognitive strategies can also be explained as the cognitive procedures that aid in
the performance of specific cognitive tasks. Studies have been done on exploring the
nature of the interaction between a problem’s intrinsic structure and a problem solv-
er’s strategies or behaviors. In design, cognitive strategies are the mental activities
of strategically developing a problem structure and a solution structure, which are
important in guiding the problem-solving process. It is critical to the designer to
have confident control of both the problem structure and the solution structure to
maintain a balance between the two. For instance, Akin et al. (1988) investigated the
correlation between problem structuring and problem solving, and reported that
problem structuring could serve as a mechanism to see how problem parameters
were identified by a designer during a design process. This is how designers will
handle design problems at the cognitive level. Other strategies used by designers to
solve design problems include top-down, bottom-up, inside-out, outside-in, and
case-based reasoning strategies.
The top-down strategy is to approach the design first from the macro level, then
narrow down to details, whereas bottom-up is to focus on details and individual base
elements before taking care of the macro level. The inside-out strategy is to handle
design by resolving internal functionalities first, and then consider the overall form
afterwards. Outside-in strategy is to start thinking from the external site forces
impacting interior functionality and overall form before addressing the interior
arrangements. Case-based reasoning is to apply previous experiences, situations, and
solutions for handling similar cases as references for solving new problems. Another
commonly used design strategy is user-centered method. This method considers the
users’ behavior, activities, and needs in the building, and the design would be cen-
tered on these considerations and arrangements. Then, based on the considered
issues, which could either be hypothesized and predicted by the designer or requested
from users, a design program is generated. Following the design program, functional
arrangements are made to create building forms. Such a set of sequences is a general
and conventional strategic operation applied in design studio education.
However, in generating a form that solves design problems, abstract design
problems require designers to turn the abstraction into a physical form that meets
functional, structural, and material requirements. Designers need methods to
accomplish such creations. These design methods include the use of design parti,
analogy, metaphor, deformation, as well as other methods not yet reported. These
methods have been used by many famous architects, and they are utilized conven-
tionally at early design stages of most design projects, for they are the thinking
mechanisms used for design generation. Next, we briefly explore the concepts
found within these methods.
A design parti is a basic scheme or main organizing concept of an architectural
design formulated in the conceptual design stage. It is an individual module, which
could be an abstract concept presented in a simple statement or a fundamental form
presented in a basic diagram. As Ching puts it, a design parti is the chief organizing
thought or decision behind an architect’s design presented in the form of a basic

diagram and/or a simple statement (Ching 1995). Designers use a diagram or a
statement to formulate an initial concept that might basically resolve most of the
design problems. It also has the potential for future growth, and could include the
use of one shape or one concept multiple times to create a larger sum. A design parti
forms the basis of preliminary design philosophy, which leads to the final solution.
Therefore, a parti can be the big idea, the central concept, the essence of the design,
or the core element of the project. For example, in urban design, urban fabric is the
combination of discrete architecture and traffic network that have been formed and
combined by a certain revolution process. During the process, there are basic rules
and basic patterns applied to shape the creation. The uses of the rules and patterns
are attributed to design methodologies used anytime in the design process, and such
rules and patterns are used to implement the parti. Similarly, other than design parti,
other methods of analogy, metaphor and deformation could also be used at any
stages during the process for form generation.
Analogy is the use of a similar concept to solve a similar problem. The concept
could be a solution produced by the designer in previous projects and saved in
memory, or a solution done by someone else and applied. Usually, a form created in
a design is applied later in several other similar design projects. The previously
generated solution is also termed a presolution model (Foz 1972; Chan 1990, 1993),
which is similar to the notion of case-based reasoning (Eastman 2001). Analogy is
commonly used to generate a solution idea. In design education, students are
encouraged to study a number of master cases for design learning. In professional
practice, designers would easily recall a number of solutions and test one after the
other until the solution is reached (Chan 1993, 2001a).
Metaphor has been recognized as a structure of our conceptual system. Collected
from linguistic evidence, Lakoff and Johnson indicated that metaphor is pervasive in
everyday life, not just in language but also in thought and action. Using the metaphori-
cal concept of “argument is war” as an example, when we talk about arguments, we
think of winning or losing arguments. We see the person we are arguing with as our
opponent and we attack their positions. We gain or lose our ground by using strategies.
As such, it is clear that we use the notion of war to structure what we do and how we
understand what we are doing when we argue (Lakoff and Johnson 2003). Therefore,
human thinking processes are metaphorical, which means that the human conceptual
system is metaphorically structured and defined. Our ordinary conceptual system is
fundamentally metaphorical in nature. Design is a part of thinking and we inevitably
apply metaphor to think about design verbally and through imagery as well.
The essential nature of metaphor is to understand and experience one kind of
thing in terms of another (Lakoff 1987, 1993). In design, it is the use of one thing to
represent something else. The concept of something for another thing could be a
substitution of one idea, concept, word, or phrase with a shape, geometry, or object
for the purpose of transforming an abstract idea into a physical form. The substitute
“form” or “shape” could be a symbol or a sign used to represent a design notion or
consideration. For example, the World Expo 70, held in Osaka, used air filled
structures for most pavilions. The Japanese pavilion was composed of five circular
Fig. 2.5 Aerial view of Osaka World Expo 1970 site (Source: Davis-Brody, http://www.columbia.
edu/cu/gsapp/BT/DOMES/OSAKA/0489-84.jpg. Accessed 10 Oct 2013)
exhibitions halls. These five circles symbolized the five leaves of sakura (see
Fig. 2.5), the national flower of Japan. This application of using five circles to rep-
resent sakura, which represents the nation, indicates that the constructed artifacts
symbolized the nation. This also exemplifies the method of using metaphor for cre-
ating a design. Another example is the famous gate that sits in the sea off the shore
of Miyajima Island near Hiroshima, Japan. It was built as the main approach, which
is the concept of gate, to its Itsukushima Shrine (see Fig. 2.6). This example, similar
to Eero Saarinen’s big arch design in St. Louis as the gate to the western frontier
(see Fig. 2.7) is another example of using an object (arch) that implies a metaphori-
cal form (symbol of a gate) to implement a function (monument). This explains the
visual use of metaphor.
Another example of using metaphor approached from a metaphorical object
point of view is found in the famous fish designed by Frank Gehry. His notion of
fish was to create a representation of the origin of our universe. He said that “Three
hundred million years before man was fish” (see http://en.wikipedia.org/wiki/
Frank_Gehry). Thus, in looking back at the history, he created a standing glass fish
in the Minneapolis Sculpture Garden (1986, see Fig. 2.8) to make a statement. For
a few years, he repeatedly used the form of fish in several of his proposed designs
and built buildings, including the Richard B. Fisher Center for the Performing Arts
in New York’s Hudson Valley (2003) and the Port Olympic in Barcelona Spain
(1992, see Fig. 2.9). He also generated a number of household items of lamps and
sculptures based on the motif of fish.
On the other hand, metaphor could also be used more conceptually than visually.
For instance, the California Academy of Science in San Francisco, designed by Renzo
Fig. 2.6 Gate at Itsukushima Shrine in Miyajima, Japan (Source: Bo-deh/Wikimedia Commons,
http://commons.wikimedia.org/wiki/File:Miyajima_Itsukushima_Shrine_Portal.jpg. Accessed
10 Oct 2013)
Fig. 2.7 The gateway arch

in St. Louis (Source: Matt
Kozlowski/Wikimedia
Commons, http://en.
wikipedia.org/wiki/
File:Gateway_Arch_edit2.
jpg. Accessed 10 Oct 2013)
Piano, provides a good example (see Fig. 2.10). The roof of the design is a metaphor
for the entire project, seen by Piano as topography (Pearson 2009). The idea was to cut
a piece of the site, which is a part of the Golden Gate Park, and push it up 35 ft and
then put whatever was needed underneath. The green roof is an extension of the park
and serves as a thermal buffer for the spaces below. It is clear in this design, the object
that represents the metaphorical concept of roof is the Golden Gate Park’s topography
Fig. 2.8 The Glass Fish by Frank Gehry in Minneapolis, 1986 (Source: Photo was taken by author)
to shelter the function of the Academy of Science. Other than using metaphor as a big
concept in design, metaphor could also be used as a philosophical statement, as pro-
posed by Mies van der Rohe in his modern architecture. He proposed the influential
motto of “less is more” to symbolize the economic consideration of spatial arrange-
ment, structural layout, and functional organization. Effects of this phase became a
design classic on reducing spatial dimension to the minimal, eliminating unnecessary
materials, and fully utilizing extreme simplicity to accomplish beauty.
The last cognitive strategy applied by designers is the method of deformation.
Deformation is the application of external force or stress to change the shape, form,
or dimensions of an object, which could be utilized at any time in the design process
for the purposes of creating new forms. For example, Peter Eisenman uses
Fig. 2.9 The Fish by Frank Gehry in the Port Olympic, Barcelona, Spain, 1992 (Source: Till
Niermann/Wikimedia Commons, http://commons.wikimedia.org/wiki/File:Barcelona_Gehry_fish.
jpg. Accessed 10 Oct 2013
transformation as a step-by-step procedure by which one element is substituted for

another, a volume is divided into planes, parallel planes turned to a grid, a grid is
rotated and scaled, or even to twist, fold, or bend objects. In a video showing the
Guardiola House design in Spain (see Fig. 2.11), he explained through animation
the methods of utilizing movement of subtracting, intersecting, and uniting volumes
for creating a novel and fragmental form.16 These design processes can be seen as
exercises of deconstructive architecture, which relate to the idea of generating a
form with certain linguistic context associated. Such methods of linguistic associa-
tion and form deformation are popular in the areas of digital architecture and rapid
prototyping. Even though the deformation method has no direct relation to cogni-
tive processes, it is a thinking phenomenon resulting from applying a set of particu-
lar procedures yielding particular forms, which should be identified as a phenomenon
of cognitive strategy for form generation.
These examples explain briefly the use of certain strategies and methods in think-
ing for generating a design project that solves embedded design problems. The major
difference between design strategy and design method is that design strategy is to use
a particular method adopted in an early stage to develop an overall plan, or a scenario,
or a goal sequence. Such generated schematic plan, scenario, or goal plan would be
used to control the design sequences from the beginning and persistently utilized until
the design is completed (Chan 1990, 2008). Design methods relate to the procedural
16
The design concept by Peter Eisenman had been recorded in an animation video, which could be
found in the following Web page: http://www.youtube.com/watch?v=JAH97LcQ2Cs
Fig. 2.10 The green roof of the California Academy of Science in San Francisco (Sources:
Wolfman SF/Wikimedia Commons, http://commons.wikimedia.org/wiki/File:California_
Academy_of_Sciences_pano.jpg, and Wikimedia public domain, http://en.wikipedia.org/wiki/
File:CalifAcadSciRoof_0820.JPG. Accessed 19, July 2014)
Fig. 2.11 The Guardiola House by Peter Eisenman in Spain (Source: Peter Eisenman, A + U,
89:01, No. 220, Jan 1989, p. 21, Fig. 15)
operations for implementing functional, structural, or constructional requirements.

A design method could be used at any time in the process, and many methods could
be used in a design for the sake of creating forms. Thus, a design method is a part of
a design strategy. If a designer could creatively apply a strategy, be it an analogy,
a parti, a metaphor, or a deformation to create a new form, such thinking should be
recognized as a creative one. The major point is not the use of the method, but the
thinking that strategically activates the application of the method. This application of
unique design strategy is the seventh notion of design creativity (DC7).
2.4.7 Design Uses Certain Reasoning
How do human beings make judgments and decisions in thinking? In life, human
beings use information presented to them as observation data, generate some
hypotheses or premises as the case, apply experience recalled from memory as ref-
erence, and utilize logical operations to set up inference for decision making. The
commonly used logical operations, described by philosophers, are deductive and
inductive reasoning. Deductive reasoning is to find specific fact from general data,
find cause to effect, or start with a general principle and deduce that it applies to a
specific situation through which a conclusion can be drawn, or start from reasons
and look for consequences (Magnani 2009, p. 10). It does derive from the conse-
quences of the assumed proposition. For instance, when deriving B from A, if the
assumptions in A are true, a valid deduction would guarantee a true conclusion of
B. Inductive reasoning is to draw from specific to general conclusions, from data to
come up with general principles, or to generalize from a sample to a general prop-
erty, and reasoning from data to a causal hypothesis (Aliseda 2006, p. 33). For
example, if inferring B from A, we might have good reason to believe the conclu-
sion of B drawn from the premise of A, but the conclusion of B is not always guar-
anteed to be true. Inductive reasoning is a commonly used type of reasoning in
physics and philology.
These two logical methods of deduction and induction are most popularly dis-
cussed theories on human reasoning addressed before the mid-nineteenth century
(Fischer 2001, p. 365). They are based on valid premises and therefore yield valid
conclusions, which are seen as formal logic reasoning. However, in some cases,
some data in the observation set might not be appropriately inferred or deducted,
thus the third reasoning mode of abductive reasoning was introduced or re-
discovered17 by Charles Sanders Peirce (1839–1914) in the late nineteenth
century.18 At first, Peirce emphasized abduction as an evidencing process with a
17
Aristotle (384–322 BC) dealt with deduction and induction but did not analyze abduction in
detail; thus, abduction remained absent from the history of logic until it was re-discovered by
Peirce (Fischer 2001, p. 365).
18
In fact, the type of inference called abduction had been studied by Aristotelian syllogistics. But,
Peirce had further interpreted abduction essentially as an inferential creative process of generating
new hypotheses in broad philosophical and semiotic perspective (Magnani 2009, p. 9). It later is
used by scholars to explain the reasoning in scientific discovery.
syllogistic interpretation of rule, case, and result. The famous example used by
Peirce was a bag of beans. All the beans in the bag were black (rule), and these
beans are from this bag (case), so these beans are black (result). Later on, he
compared the relationship of abduction to a guessing instinct. To him, abductive
reasoning was guessing through a person’s experience to infer the best explana-
tions. He wrote, “the abductive suggestion comes to us like a flash. It is an act of
insight, although of extremely fallible insight” (Peirce 1997, p. 242). Abduction is
inferring causes from effect. It represents an explanatory principle that even though
logically it is invalid, it may still be confirmed inductively. For example, the road
is wet because it has rained. The rain is the cause inferred from the consequential
effect. In Pierce’s perspective, all inferential thinking is to discover something we
do not know and thus enlarge our knowledge by considering something we do
know. The discovery is done by ways of guessing explanations or by logical leaps
of the mind. In scientific discovery, abductive reasoning is seen as the operation of
adopting explanatory hypotheses.
The purpose of using abductive reasoning is not to conclude whether something
is true or false, but to posit what could possibly be true. The conclusions drawn in this
mode are based on probabilities and mostly used within fields of science and research.
While abductive reasoning possesses all the elements of true premises and conclu-
sions in formal reasoning, it also consists of the element of probabilities, and thus is
classified as a type of informal logic reasoning. As Fischer indicated, philosophically
our knowledge of the world might not necessarily come from our logically true infer-
ences, but all our inferences are invented hypotheses, the adequacy of which cannot
be proved within logic, but only pragmatically. Therefore, making hypotheses is
another form of “think rationally and infer irrationally.” As such, knowing represents
inferring, inferring means rule-governed interpreting, and interpreting is an act that
proves adequate of our knowledge (Fischer 2001). Similarly, scholars use hypotheti-
cal reasoning to describe such a form of thinking and knowing, which is giving an
excuse or reason for something that has not happened or could possibly happen.
In design, design reasoning is sometimes different from formal logic reasoning,
which concentrates on the true or false of premises and conclusions with resulting
abstract forms. There is no true or false in aesthetic value or in judging the amount
of beauty in architectural forms. Design reasoning leans more towards abductive
reasoning, which provides more hypotheses of what may be (March 1984; Martin
2009; Cross 2011). Lorenzo Magnani has described clearly that “abduction is the
process of inferring certain facts and/or laws and hypotheses that render some sen-
tences plausible, that explain (and also sometimes discover) some (eventually new)
phenomenon or observation; it is the process of reasoning in which explanatory
hypotheses are formed and evaluated” (Magnani 2009, p. 8). As such, design activ-
ity involves the processes of making hypothetical statements and evaluations from
the beginning to the end of the course of a design.
In fact, design reasoning is not totally a formal logical operation with specific
algorithms in place, but a certain way of informal reasoning guiding the mode of
conduct. It should be seen as all the mental operators that move design actions.
Rittel explained that reasoning consists of more or less orderly trains of thought,
which include deliberating, pondering, arguing, and occasional logical inferences.

But, design reasoning is a process of argumentation based on certain reasons to
make decisions on making moves (Rittel 1988). Designers are actively formulating
hypotheses inductively from observing certain events, looking for new data points,
challenging accepted explanations, inferring possible new forms and new function-
alities, and contemplating consequences (Martin 2009).
On the other hand, design requires a lot of effort on handling graphics. Thus,
spatial reasoning is a part of design cognition, specifically spatial cognition. Spatial
reasoning requires mental operations to visualize spatial patterns, and human
rationales to manipulate the perceived patterns for accomplishing certain concep-
tual intentions over a time-ordered sequence of spatial transformations to generate
results. Spatial reasoning is a special form of thinking in pictures, versus a form of
thinking in words. Both formats of pictures and words are internal representations
of the tasks that designers are facing or the problems that designers are solving. The
methods used in spatial reasoning are to think about the problem deeply and express
the understanding of the problem pictorially, or to relate learned experiences to a
given problem situation through identifications, or to connect perceived information
with certain concepts and project into the problem situation through association.
The pictures applied could be symbols, sketches, or doodles that certain information
deems necessary.
All these concepts of design reasoning explain the fundamental operators that
motivate and drive the design process ahead. In summary, if a designer applied cer-
tain unique reasons that led to the use of certain design strategies yielding an unpre-
dictable result of a novel product, then the conjecture, inference, and hypothetical
reasoning could be the driving source of creativity in thinking. It is possible that the
reasoning can strategically activate some unique design strategies. Thus, the reason-
ing used to strategically activate unique design strategies is the eighth notion of
design creativity (DC8).
2.4.8 Design Applies Repetition for Design Generation
The last cognitive factor applied in the design processes is repetition. Repetition is
to repeat the same thinking or action subconsciously or consciously. It is a part of
human cognition utilized in language, writing, learning, and design thinking. Details
of the nature of repetition and the resulting phenomenon in design can be found in
a number of publications (Ross 1933; Ching 1979; Goodridge 1998; Mithen 2005;
Chan 2012). This section emphasizes the aspect of repetition as a part of human
cognition and its impact in design.
In everyday life, people often repeat the same life routine subconsciously. We
repeatedly drink the same brand of beverage, navigate the same route to the same
destination, apply the same schedule every day to run errands, go to the same gro-
cery store for shopping, and use the same methods solving similar problems. After
the same procedures on a particular task are exercised for a long time, the knowledge
of executing them are turned into automatic skills, such as how we learned to ride a
bicycle. Therefore, repetition causes automatic execution of skills that becomes a
default cognitive procedure for accomplishing tasks. Research done in various fields
has shown that repetition is the cognitive strategy used in rhetoric to affect or coor-
dinate attitudes, especially when terms are used repeatedly in question-begging-
form to persuade belief (Boisvert 2011); in literature to emphasize a notion and to
generate emotional inspiration (Boisvert 2011); in music to generate a piece of
music by repeating a fixed rhyme, beat, or melody (Yeston 1976); in psychology to
improve learning through rehearsal (Waugh and Norman 1965; Atkinson and
Shiffrin 1968); and is a human cognitive operation applied every day.
In the fields of design, repetition is one of the cognitive mechanisms used by
applying a basic and simple feature as a module with a set of rules repeatedly to
create a pattern. Reasons of causing repetition in design could either be the inten-
tional plan or well-practiced procedural knowledge applied recursively to create
characteristic patterns that generate the effect of rhythm. As defined, if a regularity
of changes in a regularity of measures causes the effect of movement upon our
minds, then rhythm occurs (Ross 1933). Rhythm has been seen, in arts and perform-
ing arts, as patterned recurrence, repetition, or movement in actions or in artifacts.
It is the regular, harmonious recurrence of a specific element, often a single particu-
lar entity of line, shape, form, color, light, shadow, and sound. If a designer chooses
an entity from these elements, creates a composition, and repeats the composition
with a movement or in time, then an order to the whole design is generated. Such an
order is the character of rhythmic phenomena. In fact, rhythm generates some
regularity, simplicity, balance, and hierarchical order of composition in the design
product that possesses a nature of consistency that is well perceived and easily com-
prehended by beholders.
A good example in design of the rhythm created by repetition is found in work
done by Alvar Aalto, the famous Finland architect. Aalto would use the same shape
of volume proportionally seven times and rotate the module slightly along the con-
tour of a hill (see Fig. 2.12), use the same shape of balcony seven times on the
façade (Fig. 2.13), the same configuration of curved beams four times with skylight
to create a patterned shadow on the wall (Fig. 2.14), and the same curved roof shape
four times (Fig. 2.15) in his designs. Similar methods can be found in his other
building projects. Results of using the same form at least four times generated a
wonderful visual order with consistency. These examples demonstrate that Aalto is
one of many architects who skillfully applies repetition, which creates the beautiful
effect of rhythm.
The same idea of using repetition methodologically (Chan 2012) to generate
rhythm went under experimentation by a young designer, Jasmine Brown, in her
research. She selected her third year design studio project as the base (see Fig. 2.16).
She applied the same design program on the same site, but revised the design to test
whether rhythm did occur through repetition. She repeated functional units with
regular change of proportion, exposed structural elements with regular intervals in
space, and unified skylight units on roof regularly for the purpose of satisfying func-
tional, structural, and energy requirements. Results of the experiments, as shown in
Fig. 2.12 The Shiraz Art Museum by Aalto, 1969–1970 (Source: © Alvar Aalto Museum with
permission)
Fig. 2.13 The Paimio Sanatorium design by Aalto (Source: Leon Liao/Wikimedia Commons,
http://commons.wikimedia.org/wiki/File:Paimio_Sanatorium2.jpg. Accessed 10 July 2014)
Fig. 2.14 Interior of the Riola Church by Aalto (Source: http://snyfarvu.farmingdale.edu/~straaw/

design2/project3/alvaraalto.html. Accessed 10 Oct 2013)
Fig. 2.15 The roof of the Riola Church in Riola di Vergato, Italy by Aalto, 1966, 1969, 1975–1980
(Source: Maija Holma © Alvar Aalto Museum with permission)
Fig. 2.16 Jasmine Brown’s previous design (Source: Renderings by Jasmine Brown, fall 2009)
Fig. 2.17, have systematic spatial orders occurred and rhythmic visual effects
revealed. She was pleased with the quality of the revised design.
Figure 2.18 is the interior rendering that has the rhythmic natural light casting
rhythmic shadow patterns on the floor to create a pleasant visual image. Such a
pleasant image would provide a comfortable perception to the users of the environ-
ment. Even though she cannot show that repetition is a part of a cognitive strategy
that occurs in design, the results of purposefully utilizing reoccurrence in design did
testify to the causal-effect of repetition for rhythm. Some designers consciously use
the same features in a design project to generate rhythm, and some would repeat the
Fig. 2.17 Revised results of applying repetition (Source: Design by Jasmine Brown, spring 2012)
same features (or sub-solutions from previous design) across design projects to
signify an individual style, which will be discussed in the following chapters. For
instance, Frank Lloyd Wright reused his elevation grammar in his Prairie House
design. Thus, rhythm is the cognitive function of repetition, and repetition is the
driving force that causes rhythm and style. Since rhythm is ingrained in the human
conscience of repetition, it should be a key component of design applied universally.
As a summary, design repetition is the last but the most important notion that serves
as the definition of individual style.
2.5 Overview and Operational Definition of Design Cognition 59
Fig. 2.18 Interior rendering of the revised design with rhythm (Source: Rendering by Jasmine
Brown, spring 2012)
2.5 Overview and Operational Definition of Design

Cognition
All these described cognitive mechanisms and strategies used in design could be
summarized to clearly explain what is design cognition? and what are the cognitive
factors causing style and creativity? Cognition, in general, is the process of perceiv-
ing, receiving, analyzing, storing, retrieving, and utilizing information. Information
represents knowledge. Intelligence is a way to manage knowledge. The mental
operations used to utilize intelligence include cognition. Various operations com-
pose our thinking and design thinking as well. However, design thinking is different
from other thinking processes of solving accounting, financial, statistic, software
engineering, or medical problems. Solving design problems requires special knowl-
edge with special strategies to operate the methods, plus special logic reasoning
with special procedures to process domain-specific knowledge for creating three-
dimensional tectonics to fulfill architectural functions. Skillful designers have cer-
tain ways of solving design problems that creates beautiful forms, and which
involves the use of strategy, methodology, reasoning, logic, and representation in
the entire thinking process to complete the design (Chan 1990, 2001a, 2008).
What does the picture of design cognition look like? Design cognition could be
portrayed by including all cognitive activities that may occur and cognitive
mechanisms applied in the design process to show a describable picture. The picture
could be illustrated by major sequential activities that happened in the mind—similar
to the items listed in Table 2.3—that a design is problem solving, with the problem
understood and the tasks structured first, goals generated and constraints organized,
then use of cognitive strategies of analogy, metaphor, previous cases, or association
to generate a scenario (or parti). These major activities comprise the initial stage of
conceptual design. Further, the following activities would recursively occur: looking
for appropriate representation, using reasoning to operate constraints for accom-
plishing goals, applying strategies (of analogy, metaphor, deformation, cases, and
association) to generate elements as solutions, reflecting the problem structure and
solution structure to keep the solution on track, and operating repetition to create
rhythm and style. All these activities of association, problem structuring, representa-
tion, reasoning, goal sequences, and constraint developments may be operated con-
ventionally to solve the design problems. In some cases, they are uniquely utilized to
generate a creative solution. As explained previously, as long as these mechanisms
are operated in a novel way that generates a novel form, creativity occurs. However,
it is repetition that causes style.
In most cases, designers use the logic of induction to find general conclusions
from particular data, or from general data to find particular principles (deduction),
or even to use educated guesses to explain things in the data (abduction) for making
design decisions. Other than these patterns of reasoning, humans also make hypoth-
eses for a case, and find from experience of heuristic to solve the design problem.
As soon as the found experience has appropriate associations linking to the problem
task, then it is applied immediately. This pop-up-and-apply phenomenon is the
aspect of intuition, which also has some connections with divergent thinking styles.
It has been proposed that design problem solving should involve a divergent as
opposed to a convergent thinking process (Hatch 1988). In theory, divergent think-
ing usually happens in a free-flow fashion, and more ideas are generated through
building up more associations with knowledge chunks in memory. It is because
more links provide more potential solutions than less links, and chances of yielding
unexpected surprising results are higher. In terms of logical operation, after the
divergent thinking is completed, concepts and information will be structured and
arranged by convergent thinking, which is to follow a set of procedures for achiev-
ing a solution. This is the fashion of logical operation in design thinking.
What is the operational definition of design cognition? Design activities consist
of the operation of reasoning to achieve goals for satisfying constraints, using asso-
ciations to retrieve knowledge, applying representations to communicate, and utiliz-
ing strategies to create form. Thus, the essence of thinking relates to the processing
of information in our mind, and design cognition could be operationally defined as
the processes of logic and reasoning on operating design knowledge through a
specific representation for conceptualizing, planning, and implementing certain
intentions to make artifacts. Different representations used would need different
2.6 Other Affecting Design Thinking Factors 61
logic procedures.19 Design representation is the medium used to communicate

design, which could be sketch drawing, flow chart diagram, digital model, physical
model, or even scripting. Studies in design cognition should concentrate on how
designers use graphic thinking, how they execute reasoning, what representations
have been used, and on exploring the patterns of operating logics for generating
solutions in the design thinking process.
2.6 Other Affecting Design Thinking Factors
Reasoning and logic operations have connections with the design methods and strat-
egies described earlier. But, different design media require different cognitive pro-
cedures that definitely yield different reasoning phenomena. For instance, design
processed in the conventional paper-and-pencil mode versus the use of a digital
system, such as Building Information Modeling (BIM), might have different design
methods used in different phases of schematic design, design development, and
constructional design stages. The conventional pencil-and-paper method is used to
conceptualize design ideas. After ideas are developed, sketch drawings can be
scanned and imported into the digital system to serve as visual guides for three-
dimensional digital modeling. Alternatively, concepts could also be drawn two-
dimensionally in a drawing system and imported to the BIM system for 3D
modeling, which is related to the top-down or conceptual driven method. On the
other hand, if a design is approached from the digital system, a grid system would
usually be generated in the system and used as reference, desired details of struc-
tural components, corresponding walls, and standard architectonic elements would
be added to the grid patterns with bottom up or data driven methods, or by applying
a massing tool to create representative masses and put details afterwards as with
top-down conceptual driven methods. However, design approached from digital
systems is not popularly accepted in design education.
Regardless of whether conceptual driven (top-down) or data driven (bottom-up)
thinking methods are applied in the BIM system, the flexibility of the software system
cannot match the information processing speed of human brain, and the needed mul-
tiple functions and long procedures for execution in the digital system would conse-
quently affect the human design vision and delay the progress of thinking. Thus, the
various media utilized impacts design thinking. This is due to the fact that during the
problem solving process, the mind creates similar mental representations to match up
the external representation of the media used at hand for communication and design
dialog. If the media used were sketch drawings, then mental pictures would exist in the
mind (Chan 1997). If the media were physical models or digital models, then similar
19
An operational definition is a definition of a measure that is detailed and concise, and is used to
identify one or more specific observable conditions or events.
abstracted three-dimensional images would be generated to match the external models

for continuing the design dialogues. And, if the task at hand were mathematic calcula-
tions, such as is applied in parametric modeling, then the internal representation
would be numeric symbols (Chan 2009). In this situation, without having a monitor
displaying executions, it is difficult to mentally predict design results. Yet, in order to
understand the process of thinking and to explore the impact of design media to think-
ing results, more in-depth studies on design representations are needed. In Chap. 7, the
effect of representation on creativity will be further explained.
2.7 Study Methods
Studies in exploring design thinking have applied various methods of conducting

experiments to observe design activities, interviewing designers to collect needed
information, concentrating on one designer’s works and investigating the patterns
and behaviors displayed as case study, asking designers to think aloud while work-
ing and recording the design processes for data collection, or mixing these methods
for the purposes of accomplishing research objectives. These methods can be cate-
gorized by the following types.
2.7.1 Controlled Experiment Type
Controlled experiments have been used to observe certain aspects of design behav-
ior. Thomas and Carroll studied various tasks, i.e. restaurant design, letter writing,
and software design etc., to explore the cognitive processes (Thomas and Carroll
1979). By analyzing videotaped interactions between designer and client, and the
subject's introspection, they showed that design problems were structured in terms
of subproblems, which were dynamically produced during design. This was proven
by the evidence that the overall attack on a design problem was organized into rela-
tively smaller and simpler cycles. This finding could be interpreted as partial solu-
tions generated locally by trial-and-error (generate-and-test) search. They reported
that right representations suitable for the problem in question helped in solving the
problem. They also pointed out that the goal structure the designer brought to the
design situation would drastically alter the design activity and the design product.
Thus, a crucial aspect in design is the specification of design goals.
Similar methods by running controlled experiments for observing responding
activities has been done by Carroll et al. (1980) to study the representation and
presentation in design problem solving. Their experiments were to ask 81 subjects
to work on two design problems that were isomorphic in structure—an office lay-
out and a manufacturing schedule—and to observe the design behavior. Their find-
ings indicated that graphic representation helped solve the well-defined problem
(manufacturing schedule) as well as the ill-defined problem (the office layout).
2.7 Study Methods 63
Moreover, certain presentations encouraged the utilization of graphic representa-

tion and therefore enhanced problem-solving ability. The controlled observation
methods allowed experimenters to study a particular aspect by using special exper-
imental tasks. However, the subject matter in these experimental tasks were not
suitable for studying the overall cognitive behavior in architectural design pro-
cesses. For instance, the office layout, the restaurant interior design, the letter writ-
ing, the software design, and the manufacturing schedule used in their experiments
were not architectural design tasks and were not able to yield evidence explaining
architectural problem-solving behavior. As indicated by Carroll et al., their study
could not detect the subjects' goal structure because the design problems used in
their experiments were too artificial and the subjects were just manipulating the
arrangements of design blocks (Carroll et al. 1980).
2.7.2 Interview Type
Interview method used interviews with questionnaires, and asked designers to recall
their design activities and report back to the interviewers (Krauss and Myer 1970;
Darke 1979; Lawson 1994; Cross and Clayburn Cross 1996). This method has been
used by Krauss and Myer (1970) to study the preliminary design process of a nurs-
ery school, which was a real project. They used charts to reconstruct and to record
the architect’s design activities. Such a method mainly relied on observations and
designer’s retrospections from reviewing field documents for data collection. They
reported that designers carried out two activities that were considered as the essence
of design. The first one was to make forms (solutions) relating to the pertinent data
and analysis. The second one was to reevaluate the problem and possible solutions
(forms), emphasizing one set of criteria (constraint) after another. These activities
were performed in a continuous cycle throughout the whole design. This aspect
was analogous to the generate-and-test concept. Another observation they made was
that the order in which designers changed constraints could affect the final solution,
and the design decision revealed a designer’s characteristic judgments.
In studying built architecture design, Darke was the first one who reported this
method in 1979 as the retrospective process to interview architects on their housing
designs. Her study intended to find if during the design period the architectural
designers have in mind an image or expectations about users (Darke 1979). She
used the technique of interviewing the architects of five housing projects, and tape
recorded the interviews. She relied on observations of sketched and written outputs,
and on asking designers to recall their own processes, a retrospection and introspec-
tion method, for data collection. Darke’s analysis of interviews showed that archi-
tects use a few simple objectives (an architectural term which refers to the intended
to be achieved issues, or global constraints) to reach an initial concept (also called a
design scenario). This initial concept further generated a visual image. Darke did
not explain how an initial concept generated a visual image; she indicated that this
visual image might come early in the design process, or it might appear after a
certain amount of preliminary analysis, which was a conjecture or conceptualiza-

tion of a possible solution. She called the initial concept and the objective the pri-
mary generator. She pointed out that, in most cases, the design concept was arrived
at before the requirements had been worked out in detail. Once the concept was
generated, it gave rise to a conjecture and an analysis followed. Thus design pro-
gressed in a generate-conjecture-analysis cyclic fashion, which differed from the
analysis-synthesis model developed in the early 1960s.
From her study, Darke also reported that such an interview might not get per-
fectly accurate data, and there were problems. For example, architects might not be
able to get the correct and original memories back, might interpret the activities
from a different perspective, and might not be able to explain details of some graphic
output (Darke 1979). Similarly, Ericsson and Simon (1996) indicated that subjects
might provide incomplete reports due to the mental processes of forgetfulness or
selective reporting of thoughts. Durkin (1994) posited that interviews and question-
naires provided potentially erroneous or distorted and therefore inaccurate reports.
In some cases a person’s recollection of thinking might be totally different from the
actual thoughts made at the time.
Other interview examples are the studies by Cross (2011) of Gordon Murray’s
famous Formula One race car design, and on Kenneth Grange, a product designer.
In his interviews with Murray, he had enough information to clearly explain the
winning aerodynamic feature to lower the car for gaining speed, the innovative
devices of a hydro-pneumatic suspension and the generated “stop pit” in the racing
processes, and the working methods of managing a team applied by the designer to
achieve winning records. Similarly, in his interview with Grange on the prototype of
the sewing machine and the nose of the high speed train, Cross explained that the
designer’s works are not based on styling or re-styling of a product, but from a fun-
damental reassessment of the purpose, function and use of the product. From inter-
viewing designers and collecting enough data, he could explain what these two
designers had learned from failure and the working methods utilized by them to
generate good products. The methods used by Cross are asking designers questions
and request for retrospection on a few products to study the character of each design,
which also is a type of case study.
2.7.3 Case Study Type
A case study is a careful analysis of a person, group, event, or a phenomenon on fac-

tors in relation to context. It can be used to explore causation for finding underlying
principles. The method on conducting a case study is to generate hypotheses, then
provide a systematic way of looking at events, collecting needed data, analyzing
related information, and reporting the results to test the hypotheses. The case study
has been used as a teaching method at the Harvard Business School through inter-
viewing leading practitioners of business and writing detailed accounts of what these
managers were doing, to serve as textbooks (Barnes et al. 1994; Dul and Hak 2008).
In Psychology, it is used as a descriptive research tool to provide detailed descrip-

tions of rare events and specific conditions (Christensen 1994). One interesting
approach is the critical case study, which has strategic importance in relation to a
general problem that applies to a larger population. For instance, if just one observa-
tion does not fit with the proposition, it is considered not valid generally and must
therefore be either revised or rejected. Thus, for a scientific proposition, if it is false,
then observation or experiment will at some point demonstrate its falsehood.
In medicine and psychology, case study is an in-depth study of one person on the
subject’s life to find patterns and causes for behavior. In management, it is conducted
over an extended period of time to explore the progression of an event, or used to
study historical information of events or organizations. For gathering data, case study
typically combines data collection methods such as archives, interviews through
structured survey-type questions or open-ended questions, questionnaires, docu-
ments, and direct observations. The evidence collected may be qualitative (e.g. texts)
and quantitative (e.g. numbers), or both. Because of the quantitative nature, models
can be generated and tested through case studies, and new variables identified in case
studies can be formalized in models (Bennett and Elman 2006).
In applications, case studies can be: exploratory to serve as a pilot study before
developing further hypotheses, explanatory to investigate causal effect and test the-
ory (Pinfield 1986; Anderson 1983), descriptively to generate a descriptive theory
and provide description test (Kidder 1982), collectively to study a group of indi-
viduals, or generate theory (Gersick 1988; Harris and Sutton 1986). These studies
are usually expected to be generalized to many others. But, such studies are
sometimes subjective and not able to cover a larger population. In order to use it
more effectively, scientific methods of gathering data and a well planned research
agenda should be fully constructed. For instance, Eisenhardt (1989) outlined a rig-
orous set of eight procedures with associated criteria for using case studies to
sequentially build up a ground theory. These steps include getting started (to define
research question), selecting cases, crafting instruments and protocols (for data col-
lection), entering the field, analyzing data, shaping hypotheses, enfolding literature,
and reaching closure. They do set up a roadmap for building theory. Most empirical
studies lead from theory to data. However, the accumulation of knowledge involves
a continual cycling between theory and data, thus, the use of case studies does allow
a researcher to go from data to theory (Eisenhardt 1989).
As a summary, the case study method allows us: to understand the way new ideas
arise and come to fruition in an innovative product (Candy and Edmonds 1996); to
sequentially observe the true and false side of the proposition for testing the hypoth-
eses (Flyvbjerg 2006); to study the reasoning involved in justifying a decision chain
from a given design brief to a given solution (Galle 1996); and to develop tools and
guidelines for improving team design practice in product designs (Valkenburg and
Dorst 1998). In art and architecture, case studies have been used to explain a design-
er’s design method that generates some characteristics in design products. For
instance, Frank Lloyd Wright used similar grid systems to fit the module of five
casement windows in his prairie houses (Chan 1992, 2000), and Alvar Aalto used
repetition to create the phenomenon of rhythm (Chan 2012). In this book, the caus-
ing factors of creativity will be explained through case studies in later chapters.
2.7.4 Think Aloud Type
Among all the methods used and reported, the most suitable one for closely studying
thinking is protocol analysis, which involves videotaping the verbalization of
thoughts while designing in real time as a case and analyzing the case. The method is
to ask subjects to “think aloud” while working on a given task in an experiment, with
the collected data to be systematically analyzed to find insight on cognitive processes
in solving problems. The earliest documented analysis was done by Watson (1920) to
illustrate the characteristics of cognitive process in problem solving. De Groot (1978)
used protocol analysis, based on the concept of thinking as a hierarchically organized
linear series of operations, to study chess players selecting moves in games. Even
though de Groot’s protocols were recorded by hand and were incomplete, the data
provided enough information to compose a rather detailed illustration of problem
solving processes of novice and expert chess players. After tape and video recorders
were developed in the 1950s, their utility for instantly recording sounds made the
thinking data transparent. Particularly, a video recorder could also record drawing
while designers talked. Thus, concurrent verbal reports were recognized as major data
source on cognitive processes in specific tasks (Anderson 1987).
Studies on applying “think aloud” and “protocol analysis” techniques had been
done in various fields, including: accounting to discover human decision making
(Belkaoui 1989); architecture to analyze design behaviors (Eastman 1970; Akin
1979; Chan 1990; Lloyd and Scott 1994); artificial intelligence to show problem
solving strategies (Conati and Vanlehn 2000); computer simulation to find out
search methods applied in problem solving (Newell and Simon 1956); decision
making to explore reasoning (Montgomery and Svenson 1989); education to under-
stand human learning methods (Chi et al. 1989); human-computer interaction to
describe user group interaction (Howard 1997); second-language learning to study
knowledge formulation (Faerch and Kasper 1987); software engineering research to
reveal engineers’ behaviors and strategies on the usage of methodologies (Guindon
1990; Davies and Castell 1992; Hughes and Parkes 2003); industrial design (Dorst
1995; Valkenburg and Dorst 1998); mechanical engineering design (Lloyd et al.
1996; Atman et al. 1999); electronic engineering design (McNeill et al. 1998); and
others in exploring the character of problem space (Goel and Pirolli 1992).
In testing architectural thinking, studies have been done in running design experi-
ments to identify the mental operators and representations applied in the design pro-
cesses (Eastman 1970; Akin 1978; Chan 1989). When exploring cognitive behavior,
various cognitive models have been developed to best fit the experimental data (Akin
1986; Chan 1990). While analyzing design activities on parti, design strategies, and
graphic thinking, protocol analyses are used as the study tool to describe characters
of design activity (Foz 1972; Chan 1990; Goldschmidt 1991; Valkenburg and Dorst
1998) and patterns of behavior (Chan 1993; Suwa et al. 1998). In fact, through ana-
lyzing the protocol data, researchers can effectively simulate designers’ cognitive
processes through computer programs. Detailed explanations on methods and proce-
dures of using protocol analysis for studying architectural design have been provided
(Chan 1990, 2003) and discussed (Chan 2008).
Regarding the data efficiency on studying cognitive tasks, we can learn about
what we are thinking by verbally describing what is going through our mind while
performing the tasks. Such verbal description is called verbal protocol. However, to
learn the mental tasks of designing, which requires the development of mental
images, is more complicated than other problem solving tasks. Because the activi-
ties of design require sketching in designing, there are concerns regarding the limi-
tations of using the think aloud method to collect data for studying design thinking.
The following three items are the major concerns:
1. Gaps in verbal protocol data: Human beings think faster than they can talk and
draw. People might not be able to quickly catch the marvels that flash in their
mind before they are gone.
2. Cognitive actions cannot be verbalized: Some actions are automatic skills that
are not that easy to describe through language; particularly, some drawings
might have multiple meanings.
3. Cognitive overload: It might consume more energy to process information while
four tasks of talking, drawing, perceiving, and processing information are occur-
ring simultaneously.
Although there are deficiencies in and criticism of this method, some researchers
have addressed the issues of existing gaps in verbal data, unable to be verbalized
cognition actions, and cognitive overload (Akin 1979; Eastman 2001; Cross 2001).
Regardless, this method is still justified for capturing first-hand data for cognitive
studies. For instance, a workshop on “analyzing design activity” held in 1994 at the
Delft University of Technology to study the virtues of protocol analysis, demon-
strated the value of its data analysis. In that workshop, the same sets of videotapes
and transcribed protocols were sent to a group of researchers to perform the analysis
in any form they chose (Dorst 1995). The workshop yielded many interesting results
based on the diverse array of approaches researchers used. Their findings varied, all
arising from the same data set, from the study of social activity in a group design
process (Cross and Cross 1995) and episodic knowledge used in the design activity
(Visser 1995) to verbal and visual coding applied in the design process (Akin and
Lin 1995), among others. This demonstrates that verbal data are a rich data mine
that has potential for various explorations. In fact, a single set of data generated 20
different papers selectively published in the journal Design Studies, which showed
clearly that protocol data is a rich set of original data bank and protocol analysis is
a rigorous research tool for empirical studies on design activity.
Even though the protocol data cannot catch 100 % of the thinking phenomenon
and there are limitations, this method has been improved by filling the gap in the
data through retrospective post-task interviews and questionnaires (Ericsson and
Simon 1980, 1996; Chan 2008). Methods of coding and decoding protocol data
have also been carefully studied and developed to ensure the reliability of data anal-
ysis (Chan 1990, 2008; Gero and McNeill 1998). In fact, any verbalization produced
by a subject while problem solving does clearly represent the contents of the sub-
ject’s working memory (Ericsson and Simon 1996). Thus, it is regarded as the best
method for studying and understanding human thinking, and is used to study indi-
vidual style in this book. Interestingly, the term protocol analysis has also been used
in computer science, but it relates to the concept of employing proper software and
hardware tools to operate the package of data for transiting a network’s media, and
thus specifically called network protocol analysis.
2.7.5 Other Study Types
Other than the discussed methods, virtual reality is another means for studying
design thinking in processes. For example, in 1997, a Virtual Architectural
Design Tool (VADeT) was created in an immersive virtual environment (CAVE).
This VADeT system had a number of metaphorical icons (see Fig. 2.19) serving
as design tools for generating, modifying, and editing three-dimensional objects
of architectural elements (see Fig. 2.20), and other tools for defining dimensions,
materiality, and colors of the objects. By using these tools in the system, users
could create a design and observe how designers design in virtual space.
Experiments having architecture students use the system to design a kitchen were
conducted right after the system was built (see Figs. 2.21 and 2.22). Through
Fig. 2.19 The icons used in the VADeT system (Source: “Collaborative Design in Virtual
Environments”, 2011, pp. 34–35, “Virtual Representation and Perception in Virtual Environments”
by CS Chan, Fig. 1, p. 34; with permission from Springer Science + Business Media B.V.)
Fig. 2.20 The menu selections used in VADeT (Source: “Collaborative Design in Virtual
Fig. 2.21 A design example A done in the VADeT system (Source: “Collaborative Design in Virtual
Fig. 2.22 A design example B done in the VADeT system (Source: “Collaborative Design in
Virtual Environments”, 2011, pp. 34–35, “Virtual Representation and Perception in Virtual
Environments” by CS Chan, Fig. 3, p. 35; with permission from Springer Science + Business
Media B.V.)
compiling the protocol analysis on the collected verbal data, it was found that the
overwhelming sense of immersion and projection in the virtual reality environ-
ment altered design behavior and thinking routines. Several interesting findings
were discovered in the experiments.
1. In conventional design, designers heavily rely on using scale to get accurate
measurements for spatial layouts. In the full scale environment and in this sys-
tem, there was no physical scale available in the virtual environment; designers
used their bodies as scale references.
2. Subjects focused a great deal of attention on the proportions of each object, their
spatial relationships with adjacent objects, and their locations in the space that
fulfilled functional links and visual connections with other objects in the scene.
3. Design processes in the virtual environment were almost purely visual, with con-
siderable attention devoted to the sizing, texturing, and coloring of the details of
the objects, and less on reasoning and logical problem solving.
4. No alternative design solutions were created or considered in the experiments
and the entire design process was linear progression. This might have been due
to the large work-load consumed in the environment.
In short, the subjects’ design processes seemed more intuitive rather than deliber-
ate. It is possible that their perception was overwhelmed by 3D images and their
thinking processes were driven mainly by geometric (visual) thinking instead of
Table 2.4 Summary of cognitive actions causing design creativity (DC) and individual style (IS)
conventional logical reasoning (Chan et al. 1999; Chan 2001b). It could also be due
to the fact that the full scale of the immersive virtual environment created a very
strong sense of presence (Chan and Weng 2005). Therefore, designers focused their
attention on the objects created and ignored overall functional layouts. As such,
design cognition, under this new context, may be adjusted automatically to accom-
modate the new sensations created by this exotic visual world. This example
explains the influences of design representation, which might have changed the
design strategies used, to thinking patterns. It also correlates with the concept that
design could be characterized as a construction of representations (Visser 2006).
In this chapter, all cognitive factors that constitute design cognition operating in
the design process are explained. Factors causing design creativity and individual
style are highlighted and summarized in Table 2.4. For the items as explained in
Sects. 2.4.3, 2.4.4, and 2.4.7 of: (1) having more of extrinsic constraints would
decrease the chances of creativity (DC4), (2) the shaping and changing of problem
structures would cause creativity (DC5), and (3) reasoning used that affects creativ-
ity (DC8), they do need a longer period of observation time to trace their causal-
effects. Thus, they are highlighted in the table as a reference.
These included cognitive factors are mechanisms used to process cognition, and
each of them has different levels of significance on the formulation of design products.
Some cognitive factors are more influential and dominant than others. In the discus-
sion provided in this chapter, the major cognitive factor or mechanism dominating
or influencing design is design representation, either internal or external. The secondary
factor is the design strategy used. Domain knowledge (e.g., environmental issues,
comprehension of spatial arts, and principle of compositions etc.) and design inten-
tion (e.g., goals and constraints used to construct the problem structure) are the third
and fourth factors contributing to the completion of a solution, regardless whether
the solution is a creative one or not.
Future studies in this area should concentrate on exploring the relationships
between the level of significance of these cognitive factors and design creativity.
A measurement scale should be further developed to evaluate their levels of signifi-
cance to design outcome. Other studies could also be conducted to explore the new
thinking patterns that might emerge for adjusting the cognitive capacity to cope
with the evolution of information technology and digital architecture. For instance,
the design of the virtual scene and representation of avatars in the Internet-based
virtual world has an impact to the issues of reasoning in virtual reality versus rea-
soning in reality.
It is clear that design thinking logic and reasoning in the digital architecture area
would have been changed cognitively due to the change of design representations.
In the following chapters, how design cognition affects designers’ individual style
will be introduced by conducting experiments for protocol analyses and by case
studies to provide evidence for explaining the causes of creativity. Thus, concepts
defined in this chapter are intended to provide readers with enough information for
conducting future studies and a feasible starting point for chapters to follow.
References
Akin O (1978) How do architects design? In: Latombe JC (ed) Artificial intelligence and pattern
recognition in computer aided design. North-Holland, New York, pp 65–104
Akin O (1979) An exploration of the design process. Des Methods Theor 13(3/4):115–119
Akin O (1986) Psychology of architectural design. Pion, London
Akin Ö (ed) (1997) Descriptive models of design. Des Stud (special issue) 18:4
Akin O, Lin C (1995) Design protocol data and novel design decisions. Des Stud 16:211–236
Akin O, Dave B, Pithavadian S (1988) Heuristic generation of layouts (HeGeL). In: Gero J (ed)
Artificial intelligence in engineering design. Computational Mechanics Publications,
Southampton, pp 413–444
Alexander C (1963) The determination of components for an Indian village. In: Jones JC, Thornley
DG (eds) Conference on design methods. Pergamon Press, Oxford, pp 83–114
Alexander C (1964) Notes on the synthesis of form. Harvard University Press, Cambridge, MA
Alexander C (1971) State of art in design methodology: interview with C. Alexander. DMG
Newsletter, pp 3–7
Alexander C, Ishikawa S, Silverstein M et al (1977) A pattern language, towns, buildings, con-
structions. Oxford University Press, New York
Aliseda A (2006) Abductive reasoning: logical investigations into discovery and explanation, vol
330. Springer, The Netherlands
American Heritage Dictionary (2013) http://ahdictionary.com/. Accessed 5 Mar 2014
Anderson JR (1980) Cognitive psychology and its implications. W. H. Freeman, San Francisco
Anderson P (1983) Decision making by objection and the Cuban missile crisis. Adm Sci Q
28:201–222
Anderson JR (1987) Methodologies for studying human knowledge. Behav Brain Sci
10:467–505
Anderson JR, Bower GH (1973) Human associative memory. V. H. Winston & Sons, New York
Archer B (1965) Systematic method for designers. In: Cross N (ed) Developments in design meth-
odology. Wiley, New York, pp 57–82
Archer B (1970) An overview of the structure of the design process. In: Moore GT (ed)
Emerging methods in environmental design and planning. MIT Press, Cambridge, MA,
pp 285–307
Archer B (1979) Whatever became of design methodology? Des Stud 1(1):17–18
Atkinson RC, Shiffrin RM (1968) Human memory: a proposed system and its control processes.
In: Spence KW, Spence JT (eds) The psychology of learning and motivation, vol 2. Academic,
New York, pp 89–195
Atman C, Chimka J, Bursic KM, Nachtmann HL (1999) A comparison of freshman and senior
engineering design processes. Des Stud 20(2):131–152
References 73
Baker GE, Dugger JC (1986) Helping students develop problem solving skills. Technol Teach
45(4):10–13
Bardach E (1981) Problems of problem definition in policy analysis. Res Publ Pol Anal Manag
1:161–171
Bardach E (2000) A practical guides for policy analysis. The eightfold path to more effective prob-
lem solving. Chatham House Publishers, New York
Barnes L, Christensen C, Hansen A (1994) Teaching and the case method: text, cases, and read-
ings. Harvard Business School Press, Boston
Bayazit N (2004) Investigating design: a review of forty years of design research. Des Issues
20(1):16–29
Belkaoui A (1989) Human information processing in accounting. Quorum Books, New York
Bennett A, Elman C (2006) Qualitative research: recent developments in case study methods.
Annu Rev Polit Sci 9:455–476
Blank MS, Foss DJ (1978) Semantic facilitation and lexical access during sentence processing.
Mem Cognit 6:644–652
Boisvert DR (2011) Charles Leslie Stevenson. In: Zalta EN (ed) The Stanford Encyclopedia of
Philosophy. http://plato.stanford.edu/archives/sum2011/entries/stevenson/. Accessed 4 May 2014
Bower GH (1970) Organizational factors in memory. Cogn Psychol 1:18–46
Broadbent G (1969) Design method in architecture. In: Broadbent G, Ward A (eds) Design meth-
ods in architecture. George Wittenborn, New York, pp 15–21
Buschmann F, Meunier R, Rohnert H, Sommerlad P, Stal M (1996) Pattern-oriented software
architecture: a system of patterns. Wiley, New York
Candy L, Edmonds E (1996) Creative design of the Lotus bicycle: implications for knowledge
support systems research. Des Stud 17(1):71–90
Carroll JM, Thomas JC, Malhotra A (1980) Presentation and representation in design problem-
solving. Br J Psychol 71:143–153
Casti J (1998) The Cambridge Quintet: a work of scientific speculation. Addison-Wesley, Reading
Chan CS (1989) Cognition in design process. In Proceedings of the 11th Annual Conference of the
Cognitive Science Society. Lawrence Erlbaum, Hillsdale, pp 291–298
Chan CS (1990) Cognitive processes in architectural design problem solving. Des Stud
11(2):60–80
Chan CS (1992) Exploring individual style through Wright’s design. J Architect Plann Res
9(3):207–238
Chan CS (1993) How an individual style is generated? Environ Plann B 20(4):391–423
Chan CS (1997) Mental image and internal representation. J Architect Plann Res 14(1):52–77
Chan CS (2000) Can style be measured? Des Stud 21(3):277–291
Chan CS (2001a) An examination of the forces that generate a style. Des Stud 22(4):319–346
Chan CS (2001b) Design in a full-scale immersive environment. In: Proceedings of the 2da.
Conferencia Venezolana sobre Aplicaciones de Computadoras en Arquitectura, pp 36–53
Chan CS (2003) Thoughts of Herbert A. Simon – on cognitive science in design. In: Chiu ML (ed)
CAAD talks 2: dimensions of design computation. Garden City Press, Taipei, pp 34–43
Chan CS (2007) Does color have weaker impact on human cognition than material? CAAD future
07. Springer, Amsterdam, pp 373–384
Chan CS (2008) Design cognition: cognitive science in design. China Architecture & Building
Press, Beijing
Chan CS (2009) The impact of design representation to design thinking. New Architect 3:88–90
Chan CS (2011) Design representation and perception in virtual environments. In: Wang XY, Tsai
J (eds) Collaborative design virtual environments. Springer, Amsterdam, pp 29–40
Chan CS (2012) Phenomenology of rhythm in design. J Front Architect Res 1(3):253–258
Chan CS, Weng C (2005) How real is the sense of presence in a virtual environment? In: Bhatt A
(ed) Proceedings of the 10th international conference on computer aided architectural design
research in Asia. TVB School of Habitat Studies, New Delhi, pp 188–197
Chan CS, Hill L, Cruz-Neira C (1999) Can design be done in full-scale representation? In:
Proceedings of the 4th design thinking research symposium – design representation, vol
II. MIT, Boston, pp 139–148
Chase WG, Simon HA (1973) Perception in chess. Cogn Psychol 4:55–81

Chi MTH, Hassok M, Lewis MW, Reimann P, Glaser R (1989) Self-explanations: how students
study and use examples in learning to solve problems. Cognit Sci 13:145–182
Ching FDK (1979) Architecture: form, space & order. Van Nostrand Reinhold, New York
Ching FDK (1995) A visual dictionary of architecture. Van Nostrand Reinhold, New York
Christensen LB (1994) Experimental methodology, 6th edn. Simon & Schuster, Needham Heights
Collins AM, Loftus EF (1975) Spreading activation theory of semantic processing. Psychol Rev
82:407–428
Collins AM, Quillian MR (1969) Retrieval time from semantic memory. J Verbal Learn Verbal
Behav 8:240–248
Conati C, Vanlehn K (2000) Toward computer-based support of meta-cognitive skills: a computa-
tional framework to coach self-explanation. Int J Artif Intell Educ 11:398–415
Cross N (1984) Developments in design methodology. Wiley, New York
Cross N (1999) Natural intelligence in design. Des Stud 20(1):25–39
Cross N (2000) Designerly ways of knowing: design discipline versus design science. In: Picazzaro
S, Arruda A, De Morales D (eds) Design plus research, proceedings of the Politenico di Milano
conference, pp 43–48
Cross N (2001) Design cognition: results from protocol and other empirical studies of deign activ-
ity. In: Eastman E, McCracken M, Newstetter W (eds) Design knowing and learning: cognition
in design education. Elsevier, Amsterdam, pp 79–103
Cross N (2011) Design thinking: understanding how designers think and work. Berg, Oxford
Cross N, Clayburn Cross A (1996) Winning by design: the methods of Gordon Murray, racing car
designer. Des Stud 17(1):91–107
Cross N, Cross AC (1995) Observations of teamwork and social processes in design. Des Stud
16:143–170
Cross N, Edmonds E (eds) (2003) Expertise in design. University of Technology, Australia
Creativity and Cognition Press, Sydney
Cross N, Dorst K, Roosenburg N (1992) Research in design thinking. Delft University
Press, Delft
Cross N, Christiaans H, Dorst K (eds) (1996) Analysing design activity. Wiley, Chichester
Darke J (1979) The primary generator and the design process. Des Stud 1(1):36–44
Davies S, Castell A (1992) Contextualizing design: narratives and rationalization in empirical stud-
ies of software design. Des Stud 13(4):379–392
Davis SF, Palladino JJ (2002) Psychology, 3rd edn. Prentice-Hall, Upper Saddle River
De Groot AD (1978) Thought and choice in chess. Mouton, The Hague
Dery D (1984) Problem definition in policy analysis. University Press of Kansas, Lawrence
Dorst K (1995) Analyzing design activity: new directions in protocol analysis. Des Stud
16(2):139–142
Dorst K, Cross N (2001) Creativity in the design process: co-evolution of problem-solution. Des
Stud 22(5):425–437
Dul J, Hak T (2008) Case study methodology in business research. Butterworth-Heinemann,
Oxford
Dunn WN (2004) Public policy analysis: an introduction. Prentice Hall, Upper Saddle River
Durkin J (1994) Expert systems: design and development. Prentice-Hall, Upper Saddle River
Eastman C (1970) On the analysis of intuitive design processes. In: Moore GT (ed) Emerging
methods in environmental design and planning. MIT Press, Cambridge, MA, pp 21–37
Eastman C (2001) New directions in design cognition: studies of representation and recall. In:
Eastman C, McCracken M, Newsteller W (eds) Design knowing and learning. Elsevier,
Amsterdam, pp 147–198
Ebbinghaus H (1913) A contribution to experimental psychology. Teachers College, Columbia
University, New York
Eberhard JP (1970) We ought to know the difference. In: Moore GT (ed) Emerging methods in
environmental design and planning. MIT Press, Cambridge, MA, pp 363–367
References 75
Echenique M (1972) Models: a discussion. In: Martin L, March L (eds) Urban space and struc-
tures. Cambridge University Press, Cambridge, pp 164–174
Eisenhardt K (1989) Building theories from case study research. Acad Manag Rev
14(4):532–550
Ericsson KA, Simon HA (1980) Verbal reports as data. Psychol Rev 87:215–251
Ericsson KA, Simon HA (1996) Protocol analysis: verbal reports as data. MIT Press, Cambridge,
MA, pp 48–62
Faerch C, Kasper G (1987) Introspection in second language research. Multilingual Matters,
Clevedon
Fischer HR (2001) Abductive reasoning as a way of worldmaking. Found Sci 6(4):361–381
Flyvbjerg B (2006) Five misunderstandings about case study research. Qual Inq 12(2):219–245
Fowler M (2002) Patterns of enterprise application architecture. Addison-Wesley, Boston
Foz ATK (1972) Observations on designer behavior in the parti. Master’s thesis, Massachusetts
Institute of Technology
Frith C (2007) Making up the mind: how the brain creates our mental world. Blackwell Publishing,
London
Galle P (1996) Design rationalization and the logic of design: a case study. Des Stud
17(3):253–275
Gamma E, Helm R, Johnson R, Vlissides J (1994) Design patterns: elements of reusable object-
oriented software. Addison-Wesley, Boston
Gero JS (ed) (2004) Design computing and cognition’04. Kluwer Academic Publishers, Dordrecht
Gero JS (ed) (2006) Design computing and cognition’06. Springer, Dordrecht
Gero JS (ed) (2010) Design computing and cognition’10. Springer, Dordrecht
Gero JS, Goel AK (eds) (2008) Design computing and cognition’08. Springer, Dordrecht
Gero JS, McNeill T (1998) An approach to the analysis of design protocols. Des Stud
19(1):21–61
Gersick C (1988) Time and transition in work teams: toward a new model of group development.
Acad Manage J 31:9–41
Gluck MA, Mercado E, Myers CE (2007) Learning and memory: from brain to behavior. Worth,
New York
Goel V, Pirolli P (1992) The structure of design problem spaces. Cognit Sci 16:395–429
Goldschmidt G (1991) The dialectics of sketching. Creativ Res 4(2):123–143
Goldschmidt G, Porter W (eds) (2004) Design representation. Springer, London
Gombrich EH (1960) Art and illusion: a study in the psychology of pictorial representation.
Pantheon, New York
Goodridge J (1998) Rhythm and timing of movement in performance drama, dance and ceremony.
Jessica Kingsley, London
Greco A (1995) The concept of representation in psychology. Cogn Syst 4(2):247–256
Guindon R (1990) Designing the design process: exploiting opportunistic thoughts. Hum Comput
Interact 5:305–344
Harley TA (1995) The psychology of language: from data to theory. Psychology Press, East
Sussex, pp 175–205
Harris S, Sutton R (1986) Functions of parting ceremonies in dying organizations. Acad Manage J
29:5–30
Hartt F (1994) History of Italian Renaissance. Abrams, New York
Hatch L (1988) Problem solving approach. In: Kemp WH, Schwaller AE (eds) Instructional strate-
gies for technology education. Glencoe, Mission Hills
Hayes JR (1981) The complete problem solver. The Franklin Institute, Philadelphia, pp 51–69
Heath T (1984) Method in architecture. Wiley, New York
Hesse M (1966) Models and analogies in science. University of Notre Dame Press, Indiana
Holl S (2006) Questions of perception: phenomenology of architecture. William Stout Publishers,
San Francisco
Howard S (1997) Trade-off decision making in user interface design. Behav Inform Technol
16(2):98–109
Hughes J, Parkes S (2003) Trends in the use of verbal protocol analysis in software engineering
research. Behav Inform Technol 22:127–140
Jones JC (1963) A method of systematic design. In: Jones JC, Thornley DG (eds) Conference on
design methods. Pergamon Press, Oxford, pp 53–73
Jones JC (1970) Design methods: seeds of human futures. Wiley, London
Jones JC (1977) How my thoughts about design methods have changed during the years. Des
Methods Theor 11:1
Kaplan C, Simon HA (1990) In search of insight. Cogn Psychol 22:374–419
Kidder T (1982) Soul of a new machine. Avon, New York
Koffka K (1935) Principles of gestalt psychology. Harcourt Brace, New York
Kohler W (1930) Grouping in visual perception. In: Murchison C (ed) Psychologies of 1930. Clark
University Press, Worcester, pp 143–147
Korf RE (1980) Toward a model of representational changes. Artif Intell 14:41–78
Kosslyn SM (1975) Information representation in visual images. Cogn Psychol 7:341–370
Kosslyn SM, Pomerantz JR (1977) Imagery, propositions, and the form of internal representations.
Cogn Psychol 9:52–76
Krauss RI, Myer JR (1970) Design: a case history. In: Moore GT (ed) Emerging methods in envi-
ronmental design and planning. MIT Press, Cambridge, MA, pp 11–20
Lakoff G (1987) Women, fire and dangerous things: what categories reveal about the mind.
University of Chicago Press, Chicago
Lakoff G (1993) The contemporary theory of metaphor. In: Ortony A (ed) Metaphor and thought.
Cambridge University Press, New York, pp 202–251
Lakoff G, Johnson M (2003) Metaphors we live by. University of Chicago Press, London
Lang J, Burnette C, Moleski W, Vachon D (1974) Emerging issues in architecture. In: Lang J,
Burnette C, Moleski W, Vachon D (eds) Designing for human behavior: architecture and the
behavioral sciences. Dowden, Hutchinson & Ross, Inc., Stroudsburg, pp 3–14
Lawson B (1994) Design in mind. Butterworth-Heinemann, Oxford
Lawson B (2006) How designers think, the design process demystified. Elsevier, Amsterdam
Lloyd P, Christiaans H (eds) (2001) Designing in context. Delft University Press, Delft
Lloyd P, Scott P (1994) Discovering the design problem. Des Stud 15(2):125–140
Lloyd P, Lawson B, Scott P (1996) Can concurrent verbalization reveal design cognition? Des Stud
16(2):237–259
Logan B, Smithers T (1993) Creativity and design as exploration. In: Gero JS, Maher ML (eds)
Modelling creativity and knowledge-based creative design. Lawrence Erlbaum Associates,
Hillsdale, pp 139–175
Magnani L (2009) Abductive cognition. The epistemological and eco-cognitive dimensions of
hypothetical reasoning. Springer, Berlin/Heidelberg
Maher ML, Poon J (1996) Modeling design exploration as co‐evolution. Comput Aided Civ
Infrastruct Eng 11(3):195–209
March LJ (1984) The logic of design. In: Cross N (ed) Developments in design methodology.
Wiley, New York, pp 265–276
Martin RL (2009) The design of business: why design thinking is the next competitive advantage.
Harvard Business Press, Boston
McClelland J, Rumelhart D et al (1986) Parallel distributed processing, vol II. MIT Press,
Cambridge, MA
McDonnell J, Lloyd P (eds) (2009) About: designing – analysing design meetings. Taylor &
Francis, London
McNeill T, Gero J et al (1998) Understanding conceptual electronic design using protocol analysis.
Res Eng Des 10(3):129–140
Minsky M, Papert S (1972) Research at the laboratory in vision, language, and other problems of
intelligence. Artificial Intelligence Progress Report, Memo, 252
Mithen S (2005) The singing Neanderthals: the origins of music, language, mind and body.
Weidenfeld & Nicolson, London
References 77
Montgomery H, Svenson O (1989) Process and structure in human decision making. Wiley,
Chichester
Newell A (1965) Limitations of the current stock of ideas about problem solving. In: Kent A,
Taulbee OE (eds) Electronic information handling. Spartan Books, Washington, DC,
pp 195–208
Newell A (1969) Heuristic programming: ill-structured problems. In: Aronofsky J (ed) Progress in
operations research. Wiley, New York, pp 360–414
Newell A (1990) Unified theories of cognition. Harvard University Press, Cambridge, MA
Newell A, Simon HA (1956) The logic theory machine – a complex information processing sys-
tem. IRE Trans Inf Theory 2(3):61–79
Newell A, Simon HA (1972) Human problem solving. Prentice-Hall, Englewood Cliffs
Newell A, Shaw JC, Simon HA (1958) Elements of a theory of human problem solving. Psychol
Rev 85:151–166
Norman D (2002) The design of everyday things. Basic Books, New York
Paivio A (1971) Imagery and verbal processes. Holt, Rinehart and Winston, New York
Paivio A (1986) Mental representations: a dual coding approach. Oxford University Press,
New York
Pearson C (2009) California academy of science. Archit Rec 2009–01:58–69
Peirce CS (1997) Pragmatism as a principle and method of right thinking: the 1903 Harvard lec-
tures on pragmatism. SUNY Press, Albany
Piaget J (1953) The origin of intelligence in the child new fetter lane. Routledge & Kegan Paul,
New York
Piaget J (1967) The psychology of intelligence. Routledge & Kegan Paul, London
Pinfield L (1986) A field evaluation of perspectives on organizational decision making. Adm Sci Q
31:365–388
Posner MI (1973) Cognition: an introduction. Scott Foresman, Glenview
Pylyshyn ZW (1973) What the mind’s eye tells the mind’s brain: a critique of mental imagery.
Psychol Bull 80:1–24
Reed SK (1974) Structural descriptions and the limitations of visual images. Mem Cogn
2:329–336
Reitman WR (1964) Heuristic decision procedures, open constraints, and the structure of ill-
defined problems. In: Shelley MW, Bryan GL (eds) Human judgments and optimality. Wiley,
Rittel H (1988) The reasoning of designers. Arbeitspapier A-88-4. Institut für Grundlagen der
Planung. Universität Stuttgart, Stuttgart
Rittel H, Webber M (1973) Dilemmas in a general theory of planning. Policy Sci 4(2):155–169
Rodgers P (ed) (2012) Articulating design thinking. Libri Publishing, Faringdon
Rosch E (1978) Principles of categorization. In: Rosch E, Lloyds B (eds) Cognition and categori-
zation. Laurence Erlbaum Associates, Hillsdale, pp 189–206
Rosenhead J (1996) What’s the problem? An introduction to problem structuring methods.
Interfaces 26(6):117–131
Rosenshine B (1997) Advances in research on instruction. In: Lloyd JW, Kameenui EJ, Chard D
(eds) Issues in educating students with disabilities. Lawrence Erlbaum Associates, Mahwah,
pp 197–220
Ross DW (1933) A theory of pure design; harmony, balance, rhythm. Peter Smith, New York
Rowe P (1987) Design thinking. MIT Press, Cambridge, MA
Rumelhart DE, Ortony A (1977) The representation of knowledge in memory. In: Anderson RC,
Sprio RJ, Montague WE (eds) Schooling and the acquisition of knowledge. Lawrence Erlbaum
Associates, Hillsdale, pp 99–135
Schon DA (1983) The reflective practitioner. Temple-Smith, London
Schön DA (1987) Educating the reflective practitioner: toward a new design for teaching and learn-
ing in the professions. Jossey-Bass, San Francisco
Simon HA (1969) Sciences of the artificial. MIT Press, Cambridge, MA
Simon HA (1973) The structured of ill-structured problem. Artif Intell 4:181–201

Simon HA, Newell A, Shaw JC (1962) The processes of creative thinking. In: Gruber HE, Terrell
G, Wertheimer M (eds) Contemporary approaches to creative thinking. Lieber-Atherton, Inc,
Stewart S (ed) (2011) Interpreting design thinking. Des Stud (special issue) 32:6
Stiny G, March L (1981) Design machines. Environ Plann Plann Des 8:245–255
Suwa M, Purcell T, Gero J (1998) Macroscopic analysis of design processes based on a scheme for
coding designers’ cognitive actions. Des Stud 19:455–483
Teuber ML (1973) New aspects of Paul Klee’s Bauhaus style. In: Teuber ML (ed) Paul Klee: paint-
ings and watercolors from the Bauhaus years, 1921–1931. Des Moines Art Center, Des Moines,
Iowa, pp 6–17
Thevenot C, Oakhill J (2006) Representations and strategies for solving dynamic and static arith-
metic word problems: the role of working memory capacities. Eur J Cogn Psychol
18:756–775
Thomas JC, Carroll JM (1979) The psychological study of design. Des Stud 1:5–11
Tidwell J (2005) Designing interfaces: patterns for effective interaction design. O’Reilly Media,
Inc., Sebastopol
Turing AA (1950) Computing machinery and intelligence. Mind LIX(236):433–460
Valkenburg R, Dorst K (1998) The reflective practice of design teams. Des Stud 19(3):249–271
Vesely A (2007) Problem delimitation in public policy analysis. Central Eur J Publ Pol
1(1):80–100
Visser W (1995) Use of episodic knowledge and information in design problem solving. Des Stud
16:171–187
Visser W (2006) Designing as construction of representations: a dynamic viewpoint in cognitive
design research. Hum Comput Interact 21:103–152
Wade J (1977) Architecture, problems, and purposes. Wiley, New York
Watson JB (1920) Is thinking merely the action of language mechanisms? Br J Psychol
11:87–104
Waugh NC, Norman DA (1965) Primary memory. Psychol Rev 72:89–104
Wertheimer M (1923) Untersuchungen zur Lehre von der Gestalt II. In: Psycologische Forschung,
4, 301–350. Translation, entitled laws of organization in perceptual forms, published in Ellis,
W. (1999) A source book of gestalt psychology. Routledge, London, pp 71–88
Wickelgren WA (1974) How to solve problems. Freeman, San Francisco
Wimmer GE, Shohamy D (2012) Preference by association: how memory mechanisms in the hip-
pocampus bias decisions. Science 338(6104):270–273
Winograd TW (1975) Frame representations and the declarative-procedural controversy. In:
Bobrow DG, Collins A (eds) Representation and understanding: studies in cognitive science.
Academic, New York, pp 185–200
Woolley RN, Pidd M (1981) Problem structuring – a literature review. J Oper Res Soc
32(3):197–206
Yeston M (1976) The stratification of musical rhythm. Yale University Press, New Heaven
Zhang J (1997) The nature of external representations in problem solving. Cognit Sci
21(2):179–217
View publication stats

4 chp2 Intro Design Cognition

Uploaded by

Copyright:

Available Formats

4 chp2 Intro Design Cognition

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

4 chp2 Intro Design Cognition

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Introduction of Design Cognition

Chapter · February 2015

Work on the VR historical models for headmount display. View project

The user has requested enhancement of the downloaded file.

2.1 Definition of Design

© Springer International Publishing Switzerland 2015 9

activities—the intellectual creation of feasible solutions or aesthetic artifacts—and

2.2 Design Thinking and Cognition

If we see design as a series of mental activities generating entities, then we should

2.3 Development of Design Studies

If design research is defined as a scientific approach to design or regarded as a study

towards exploring systematic design procedures, and proposing systematic methods

The patterns in pattern language correspond to the rules of grammar in natural

Table 2.1 Format of patterns Format of a pattern (problem):

Table 2.2 A cooking layout example

Fig. 2.1 A diagram of

2.4 Cognitive Mechanisms and Functions

By organizing patterns of theories into categories, the characteristics of design

2.4.1 Design Is Problem Solving

methodologies for managing overall design processes (Alexander 1964; Broadbent

2.4.2 Design Is Making Associations: The Nature

In design education, when a project is assigned as a studio exercise, instructors

Fig. 2.2 The temporal lobe

Fig. 2.3 A diagram of neuron (Source: Wikimedia public domain, http://commons.wikimedia.org/

2.4.3 Design Is Goal and Constraint Driven

Similarly, designing is proactive problem solving (Baker and Dugger 1986). It

2.4.4 Design Is Reflective and Problem Structuring

2.4.5 Design Is Looking for Representation

Knowledge, as explained in the previous section, is saved in memory by chunks.

Fig. 2.4 The mutilated

choose from, ill-defined problems provide more opportunities for designers to

2.4.6 Design Utilizes Cognitive Strategy

thought or decision behind an architect’s design presented in the form of a basic

Fig. 2.7 The gateway arch

transformation as a step-by-step procedure by which one element is substituted for

operations for implementing functional, structural, or constructional requirements.

2.4.7 Design Uses Certain Reasoning

which include deliberating, pondering, arguing, and occasional logical inferences.

2.4.8 Design Applies Repetition for Design Generation

Fig. 2.14 Interior of the Riola Church by Aalto (Source: http://snyfarvu.farmingdale.edu/~straaw/

2.5 Overview and Operational Definition of Design

logic procedures.19 Design representation is the medium used to communicate

2.6 Other Affecting Design Thinking Factors

abstracted three-dimensional images would be generated to match the external models

2.7 Study Methods

Studies in exploring design thinking have applied various methods of conducting

2.7.1 Controlled Experiment Type

Moreover, certain presentations encouraged the utilization of graphic representa-

2.7.2 Interview Type

certain amount of preliminary analysis, which was a conjecture or conceptualiza-

2.7.3 Case Study Type

A case study is a careful analysis of a person, group, event, or a phenomenon on fac-

In Psychology, it is used as a descriptive research tool to provide detailed descrip-

2.7.4 Think Aloud Type

2.7.5 Other Study Types

Chase WG, Simon HA (1973) Perception in chess. Cogn Psychol 4:55–81

Simon HA (1973) The structured of ill-structured problem. Artif Intell 4:181–201

View publication stats