Feasibility analysis is the process of determining if a business idea is viable.
As shown in Figure 3.1, the most effective businesses emerge from a process
that includes (1) recognizing a business idea, (2) testing the feasibility of the
idea, (3) writing a business plan, and (4) launching the business. If a business
idea falls short on one or more of the four components of feasibility analysis, it
should be dropped or rethought, as shown in the figure. Many entrepreneurs
make the mistake of identifying a business idea and then jumping directly to
writing a business plan to describe and gain support for the idea. This
sequence often omits or provides little time for the important step of testing the
feasibility of a business idea before the business plan is written.
A mental transition must be made when completing a feasibility analysis
from thinking of a business idea as just an idea to thinking of it as a business.
A feasibility analysis is an assessment of a potential business rather than
strictly a product or service idea. The sequential nature of the steps shown in
Figure 3.1 cleanly separates the investigative portion of thinking through the
merits of a business idea from the planning and selling portion of the process.
Feasibility analysis is investigative in nature and is designed to critique the
merits of a proposed business. A business plan is more focused on planning
and selling. The reason it’s important to complete the entire process, according
to John W. Mullins, the author of the highly regarded book The New Business
Road Test, is to avoid falling into the “everything about my opportunity is
wonder” mode. In Mullins’s view, failure to properly investigate the merits of a
business idea before the business plan is written runs the risk of blinding an
entrepreneur to inherent risks associated with the potential business and
results in too positive of a plan.2 This scenario may explain the large number of
eBay drop-off stores that were opened in the mid-2000s only to quickly fail as
described in the “What Went Wrong?” feature.
This chapter provides a methodology for conducting a feasibility analysis by
describing its four key areas: product/service feasibility, industry/market
feasibility, organizational feasibility, and financial feasibility. We introduce supplemental
material in two appendixes to the chapter. Appendix 3.1 contains a
tool called First Screen, which is a template for completing a feasibility analysis.
Appendix 3.2 contains an Internet Resource Table that provides information on
Internet resources that are helpful in completing First Screen.
An outline for the approach to feasibility analysis depicted in this chapter is
provided in Table 3.1. Completing a feasibility analysis requires both primary
and secondary research. Primary research is research that is collected by the person or persons completing the analysis. It normally includes talking to
industry experts, obtaining feedback from prospective customers, conducting
focus groups, and administering surveys. Secondary research probes data
that is already collected. The data generally includes industry studies, Census
Bureau data, analyst forecasts, and other pertinent information gleaned
through library and Internet research. The Internet Resource Table in
Appendix 3.2 is useful for conducting secondary research.
It should be emphasized that while a feasibility analysis tests the merits of
a specific idea, it allows ample opportunity for the idea to be revised, altered,
and changed as a result of the feedback that is obtained and the analysis that
is conducted. The key objective behind feasibility analysis is to put an idea to
the test—by talking to industry experts, surveying prospective customers,
studying industry trends, thinking through the financials, and scrutinizing it
in other ways. These types of activities not only help determine whether an idea
is feasible but also help shape and mold the idea.
Now let’s turn our attention to the four areas of feasibility analysis. The
first area we’ll discuss is product/service feasibility.

Cognition, creativity, and entrepreneurship

Thomas B. Ward

Center for Creative Media, University of Alabama, Tuscaloosa, AL, 35487-0172, USA

Abstract
Cognitive approaches to creativity are discussed as they relate to an important task of entrepreneurs:
generating novel and useful ideas for business ventures. Attention is given to the paradoxical role of
knowledge, which can either enhance of inhibit creativity, as well as to the properties of knowledge and a
selected set of processes that influence the originality of newly generated ideas. Experimental findings
are discussed along with suggestions about how those findings might be translated to practical
applications.

Keywords: Cognition; Creativity; Entrepreneurship

1. Executive summary
Entrepreneurs face many significant challenges, not the least of which is generating or
recognizing ideas that have the potential to be developed into appealing goods or services.
Successful ideas are often a balance between novelty and familiarity: new and different
enough to capture consumers’ attention, but familiar enough to not be misunderstood or
rejected out of hand as too radically different. Because the creative cognition approach
provides a theoretical framework for understanding the thought processes involved, it has the
potential to serve as guide to more effective idea development.
The creative cognition approach views creative ideas as being the natural result of
applying basic mental operations to existing knowledge structures. The originality of a
given idea, that is, the balance between its novelty and familiarity, will be determined by
the processes employed and the way in which existing knowledge is accessed. Processes
examined in this paper include conceptual combination, analogy, and initial problem
formulation.
Work on conceptual combination reveals that when two previously separate concepts or
images are merged into a single new unit, novel properties can emerge that were not
obviously present in either of the separate components, and that the effect is particularly
strong for dissimilar or divergent concepts. Such novelty can be exploited to develop new
product ideas or market niches.
Analogy, or the mapping of knowledge from a familiar domain to a less familiar one, is
central to creative developments in science, art, music, and literature and may also have
applicability to entrepreneurship, as when a new successful venture is based on the principles
that operate in other current successful ventures. Analogies can also vary in terms of their
conceptual distance (e.g., an atom is similar to a solar system versus a washing machine is
similar to a dishwasher) and main purpose (generating an idea versus explaining an idea to
someone else), and understanding these distinctions can help with the multiple tasks of
entrepreneurship.
Problem formulation influences the approaches people adopt to solving problems and
thereby has a lot to do with their ultimate success in solving the problems. Problems can be
defined very concretely or very abstractly (e.g., develop a new disc brake system versus
develop a new device for transforming the kinetic energy of a moving vehicle), with the
former leading to less novelty but more familiarity. Knowing whether one’s goal is more or
less novelty can help to determine the most productive approach.
The work presented in this paper is largely empirical research on fundamental processes,
but suggestions are made along the way about how to translate that work into entrepreneurial
applications. In contrast to many contemporary approaches to creativity, the advice here is not
to throw out everything you already know and start over. Instead, it is to exploit everything
you know by the judicious application of basic mental operations.

2. The entrepreneurial challenge
Novel and useful ideas are the lifeblood of entrepreneurship. To be successful, entrepreneurs
must generate valuable ideas for new goods or services that will appeal to some
identifiable market, and having identified those potential opportunities, they must figure out
how to bring the project to fruition. Depending on the need for capital to develop the new
venture, entrepreneurs may even need to craft ideas for how to convince others of the value of
the project. Because novelty and usefulness are the hallmarks of creative ideas, it is not
surprising that the possible connections between creativity and entrepreneurship have been of
interest for some time (e.g., Gilad, 1984; Whiting, 1988).
The present article examines the nature and origins of novel ideas with a particular focus
on how existing knowledge shapes those ideas and on the cognitive processes by which
people access and manipulate their knowledge. Although theoretical concerns are primary,
practical suggestions for improving creative functioning are also discussed.

3. A creative paradox and a cognitive perspective
An examination of human endeavors yields an interesting paradox. Throughout history,
coexisting with a cornucopia of creative accomplishments in art, music, science and
technology, we also find stunning examples of needlessly constrained thinking. Considering
creativity first, in addition to historically noted undertakings that have radically altered the
way we live, new discoveries and advances are being made on a daily basis in a variety of
fields, and as readers of this journal will no doubt be aware, new products and services are
constantly being brought to the market by creative entrepreneurs. Clearly, humans have the
capacity to move beyond what currently exists to generate and implement new ideas.
It is also clear, however, that people’s attempts at creativity often reveal unnecessarily
limited thinking. To give just one example, Barker (1993) has claimed that Sony nearly
missed a golden opportunity when they temporarily abandoned work on developing music
CDs in the mid-1970s because they judged that putting 18 hours of music on a single CD
would not be commercially viable. Why 18 hours? According to Barker’s account, it was
because they used the size and shape of LP record albums (12 in. diameter circles) as their
starting point, and a CD of that size would presumably hold a huge amount of music. The
designers ultimately overcame the constraint, but initially their knowledge about LPs limited
their thinking. Barker and others (e.g., Ward et al., 1995) have provided many additional
anecdotes regarding constrained thinking, and there is an inestimable number of creative
failures that are lost to history and remain undocumented simply because they led nowhere.
Thus, along with our enormous capacity to create, humans appear to have an equally
impressive capacity to become stuck in the past.
How are we to resolve this paradox of creativity alongside constraint? How can people
who have the capacity to be so innovative also be so unoriginal? From the perspective of
cognitive psychology, these seemingly contradictory tendencies can be seen as alternative
manifestations of the more general propensity of people to store information about their
experiences in organized knowledge structures and then access that knowledge for use in
subsequent tasks. Sometimes knowledge provides a bridge to the next new development and
sometimes it becomes a fence that blocks our path. By examining cognitive processes and
structures, it is hoped that individuals and groups can better understand how to access and use
existing knowledge in more creative ways—to build more bridges and fewer fences. What are
the aspects of cognitive structures that constrain or direct the shape of newly formed ideas?
What are the best ways of accessing, manipulating, and transforming knowledge to produce
novel and useful entities?
In a more general sense, models that are primarily cognitive in nature view creative
products as resulting from the application of mental operations to stored information. For
example, Runco and Chand (1994, 1995) have described a model that includes processes of
problem finding, ideation, and evaluation that interact with one another and with knowledge
and motivation to determine creative outcomes. An alternative, though not incompatible
framework, is the Geneplore model (Finke et al., 1992; Ward et al., 1999b), which
characterizes creative endeavors as an interplay between generative processes that produce
candidate ideas possessing varying degrees of creative potential, and exploratory processes
that extend or modify those initially generated ideas judged to be most promising. A crucial
strength of these and other cognitive models is that the broad groupings of processes they
highlight (e.g., ideation, generation) are really just umbrella terms for more specific processes
that have received considerable investigation by cognitive psychologists. It is only by
examining the details of those specific processes and how they operate on particular aspects
of knowledge that a clear understanding of the emergence of creative ideas will be obtained.

4. Conceptual combination, other transformational processes and the origins of novel
ideas
So what specific processes might a would-be entrepreneur use to come up with a new idea
for a product or service? To gain a perspective on this question, consider the following
obstacle to novelty: the truism that one cannot produce something from nothing—ex nihilo
nihil fit—applies to ideas as well as to tangible things. Creative ideas do not appear, ex nihilo,
full-blown in the minds of their originators, but rather must be crafted from the person’s
existing knowledge. But if new ideas are rooted in old ones, how does novelty emerge? How
does the innovator move beyond the constraints imposed by that old knowledge to
differentiate the new from the old, that is, to develop something original?

4.1. Conceptual combination
A variety of processes exist by which people can modify, extend, or otherwise transform
their stored knowledge, but one of particular interest is conceptual combination, a process
whereby previously separate ideas, concepts, or other forms are mentally merged. Conceptual
combination bears a special relationship to creativity, having been mentioned frequently in
historical accounts of creative accomplishments (e.g., Rothenberg, 1979; Thagard, 1984;
Ward, 2001; Ward et al., 1995). Rothenberg (1979), in particular, has argued that simultaneously
entertaining or integrating two opposing ideas, a process termed Janusian thinking,
underlies creative acts as diverse as the paintings of da Vinci, the symphonies of Mozart, and
the scientific reasoning of Einstein. In addition, combining concepts is a crucial component in
several process models of creative functioning (e.g., Davidson, 1995; Mumford et al., 1991;
Sternberg, 1988), and because the capacity to interpret and produce combinations is a
fundamental one that underlies our use of language, it has been the focus of intense scrutiny
by cognitive psychologists (e.g., Costello and Keane, 2000; Gagne, 2000; Hampton, 1987,
1997; Murphy, 1988; Wisniewski, 1997a,b). Conceptual combination also appears to be
directly relevant to the needs of entrepreneurs in search of new ideas to pursue.
Combining concepts may be an attribute of the eminently creative, but it is also a basic
capacity available to all of us, and thus is a procedure that may be harnessed to enhance everyday
creativity. For example, even a brief consideration of new word combinations that have entered
our collective lexicon recently (e.g., web hosting, mouse pad, sport drink, personal trainer, and
yes, the infamous ‘‘hanging chad’’ from the contested 2000 presidential election) confirms that
people are quite capable of creating and interpreting novel pairings. More importantly, it
highlights that the creative potential of conceptual combination resides, at least in part, in the
fact that combinations are not mere summations of the concepts being merged. Hanging chad,
for instance, carried with it the notion of ambiguity with respect to a voter’s intent, a property not
generally thought of as being true of either hanging things or chad considered separately.
Properties that are apparent in people’s interpretations of a combination but not in their
representation of either of its components, are referred to as emergent features. Although
emergent features from simple word pairings are not as dramatically impressive as the kinds
of real-world creative advances described by Rothenberg (1979), they are nevertheless novel,
and they reveal that even the simplest merging of two previously separate concepts can lead
to ideas that are substantially different from either of those separate concepts.
Carefully controlled laboratory research also confirms that this most basic form of
conceptual combination can be the source of novel ideas, and it supports Rothenberg’s
(1979) contention that combining concepts with opposing meanings is particularly evocative.
For example, Estes and Ward (2002) had a sample of college students interpret various types
of adjective-noun combinations. Of most interest, when the adjectives and nouns were
opposing in meaning (e.g., healthy illness) the participants’ interpretations contained more
emergent properties than when the terms represented more typical pairings (e.g., harmful
illness). A healthy illness, for example, might be one that temporarily incapacitates its victim,
thereby preventing the person from engaging in some activity that could have resulted in
more harm (e.g., taking a fateful trip). A harmful illness, by contrast, is just one that causes
some harm to the body—not a particularly novel construct.
Other laboratory studies also provide evidence that discrepancy or dissimilarity of components
can yield novelty. Kunda et al. (1990), for example, asked people to describe members of
a somewhat surprising combination of social categories (e.g., Harvard-educated carpenters) and
found that people included properties (e.g., nonmaterialistic) that were not part of their
descriptions of either social category considered separately (e.g., Harvard-educated people or
carpenters). Likewise, Wilkenfeld and Ward (2001) had people interpret novel noun–noun
combinations and found that dissimilar combinations (e.g., motorcycle carpet) led to more
emergent properties than did similar combinations (e.g., sled ski). Evidently, when people
attempt to make sense of novel combinations and, in particular, when they reason out how two
discrepant concepts can fit together, the process can yield emergent properties that do not come
to mind in considering either concept in isolation. Something new emerges from the mix.
From a practical standpoint, the laboratory findings suggest that novel ideas for useful
products or marketing appeals might be generated by mentally combining opposing concepts
(e.g., Ward et al., 1995). Consider, for example, the appeal to the ‘‘affordable luxury’’ of the
Nissan Altima. The concepts of affordability and luxury would ordinarily be thought of as
being in opposition to one another, with many affordable but basic vehicles as well as many
luxurious but expensive ones. Thinking about the combination of affordability and luxury
opens a new set of possibilities in the space of vehicles. As another oppositional example,
‘‘nonalcoholic beer’’ might be expected to appeal to health conscious or socially responsible
individuals. More generally, then, thinking about opposing wants of consumers (e.g.,
excitement and tranquility, solitude and companionship, and comfort and ruggedness) may
help to identify gaps in a product space and suggest a particular target audience or marketing
appeal. What about a travel agency that offered ‘‘rugged comfort’’ tour packages?Would they
appeal to those who fancy themselves as explorers, but who also like the comforts of home?1
One reason conceptual combination might be especially useful in entrepreneurial
creativity is that combinations often represent specializations of their base concepts or head
nouns, even when no opposition in meaning is present (e.g., Murphy, 1988). To use a
noncreative example, a pocket watch is a special type of watch with features that
differentiate it from other types of watches. Consequently, a whole series of minor variations
on a product can result from combining new modifying concepts with the same head noun
concept. The idea of concept specialization is clearly not lost on athletic shoe manufactures
who, at a minimum, have produced the specialized combinations of running shoes, walking
shoes, basketball shoes, court shoes, and cross training shoes. Similarly, we now have
electric scooters and in-line skates, and it would be no surprise to shortly see electric in-line
skates.
It is also important to note that novel outcomes can result from a variety of procedures
for merging concepts. People need not be limited to combining pairs of words or
interpreting their meanings to produce new ideas. For example, Mobley et al. (1992) gave
participants a set of problems in which they had to combine four exemplars of each of three
categories to generate a new category that would account for or explain the presence of all
of the exemplars in that new category. They were to label, define, and list new exemplars
of the combined category. In some problems, the three starting categories were closely
related and in others they were not. Importantly, the novel categories the participants
generated for the latter types of problems were rated as more original than those they
generated for the former types. Apparently then, as with the results of studies already
described, the need to integrate more discrepant pieces of information provided a boost to
originality.
The work of Mumford et al. (1997) is also important in that it reveals that the outcome of
conceptual combination depends on what people are instructed to consider. Considering
shared attributes across the exemplars appears to be more effective for closely related
concepts, whereas considering more metaphoric kinds of interpretations is effective with
discrepant ones (Mumford et al., 1997). This makes sense because related concepts share
many attributes, whereas discrepant ones do not and integrating them may require people to
go beyond ordinary meanings toward more metaphoric ones.
Research also shows that a combination does not have to involve verbal units at all to be
a stimulus for creativity. Merging visually presented abstract forms, for example, can also
lead to emergent new ideas. Rothenberg and Sobel (1980) showed that participants who
viewed two images superimposed on one another created metaphors that were rated as more
creative than those produced by participants who saw the same images next to one another.
Finke (1990) also showed that people who mentally combined randomly selected visual
forms were able to develop ideas for inventions and discoveries for a variety of domains

under a wide range of procedures. Although superimposed or merged images do not always
lead to more creative outcomes (e.g., Sobel and Rothenberg, 1980), the results are
suggestive that combined images can, at least under some circumstances, be a stimulus to
originality.
More broadly, combinations need not be limited to simple concepts, but might extend for
example, to musical styles (e.g., Gregorian reggae), artistic styles (e.g., Baroque cubism), and
perhaps more practically, to diverse consumer needs. Oddly enough, when considering the
latter, combinations can push ideas in the opposite direction from specialization as when one
considers how multiple needs can be satisfied by a single product. For example, people
pursuing outdoor adventures might want to ward off bugs, protect their skin from the
damaging rays of the sun, and moisturize their weather-dried skin. Although they might
purchase three separate products to meet those separate needs, a single product that combined
all of those functions might have tremendous appeal. Certainly, Avon is banking on just that
possibility with their line of bug repelling, UV blocking moisturizing creams. Recent trends
in digital products also seem to reflect an effort to combine as many functions as possible into
a single device of the smallest possible proportions.
Interestingly, one does not need special training to interpret combinations in ways that
generate ideas for new products. For example, I gave students in one of my courses the task
of interpreting unusual combinations, one of which was ‘‘computer dog.’’ They came up
with a variety of interesting interpretations, but a particularly intriguing one was the idea
that a computer dog was a peripheral device, similar to a mouse, for interacting with a
computer. The device had a track ball and four buttons arranged in the shape of a dog’s
footprint, emergent properties that would not be considered as parts of ordinary dogs or
computers.
It is important to note, however, that an idea for a new product is not the same as the
finished product itself. As noted in the Geneplore model (Finke et al., 1992; Ward et al.,
1999a,b), the ideas initially generated in response to a particular situation may not be creative
products in themselves. Rather they represent candidate ideas or preinventive forms that may
or may not lead to a creative product as they are explored, modified, transformed, extended,
or even rejected on the basis of additional exploratory thought processes. Clearly for some
combinations, much development is necessary to bring the idea to fruition, but combinations
can at least provide a starting point.
Conceptual combination can serve many different creative purposes. Some combinations
may simply represent convenient labels invented to communicate the meaning of new
concepts as they are identified or become relevant in some way (e.g., soccer mom, religious
right, and hanging chad). In these cases, the word pair is the result of a newly emergent idea
rather than vice versa. An entrepreneur using this form of conceptual combination might
expend effort to develop a catchy phrase that captures the essence of a new product or service.
Sometimes, conceptual combinations represent a problem to be solved (e.g., computer virus,
road rage, and digital divide) and when they do so, they can provoke efforts at problem
solutions (e.g., virus protection software). Finally, as noted throughout this section, combinations
can spark ideas for variations on existing products or new classes of products
altogether.

4.2. Analogical reasoning and other processes
The intense focus on conceptual combination should not be taken as an indicator that it is
the only source of novel ideas, but merely that it is a particularly generative process with
historical ties to creativity about which much has been learned. Another process with a
special link to creativity that has also undergone careful experimental examination is
analogical reasoning or transfer, the application or projection of structured knowledge from
a familiar domain to a novel or less familiar one (e.g., Gentner et al., 2001; Holyoak and
Thagard, 1995).
Commonly cited examples of analogy in creative endeavors abound, such as Rutherford’s
use of a solar system as a model for how the hydrogen atom was structured and Robbins,
Laurents, Bernstein, and Sondheim’s adaptation of Shakespeare’s Romeo and Juliet to the
context of a 1950s New York City gang conflict in West Side Story. Meticulous case studies
have also detailed the role of analogy in major creative accomplishments, such as Kepler’s
reasoning about planetary motion (Gentner et al., 1997), Edison’s development of an electric
light distribution system (Basalla, 1988; Friedel and Israel, 1986), and the Wright brother’s
efforts to craft a workable flying machine (Crouch, 1992). Not surprisingly, then, analogy has
been a key ingredient in proposals for enhancing creativity (e.g., Gordon, 1961) and has been
listed as a component process in cognitive process models of creativity (e.g., Finke et al.,
1992).
The transformational power of analogies derives, at least in part, from the fact that good
analogies connect the familiar and novel domains at very deep levels, not merely at the
surface (e.g., Gentner, 1983, 1989; Gentner and Toupin, 1986). Consider the solar system/
atom analogy. It means that, just as planets orbit around a more massive central body, the sun,
electrons may orbit around a more massive central body, the nucleus. But the nucleus and
electrons do not resemble the sun and planets in any superficial way. The nucleus of an atom
does not appear yellow like the sun, nor does it have a high surface temperature. The
electrons are not as big as planets. What matters is that there are corresponding objects that
bear particular relations to one another. Likewise, New York City of the 1950s did not have to
resemble Verona of centuries earlier, and Maria did not have to look or dress like Juliet. What
mattered is that two young people were in love, but were also connected to larger groups that
were in conflict with one another.
As with conceptual combination, there are various manifestations of analogy and multiple
purposes to which analogies might be put. The most obvious purpose is applying the
knowledge from one domain as a kind of model to help in understanding or developing ideas
in another domain, but another purpose is to communicate a new idea to others in a concise,
understandable way. Dunbar’s (1997) on-line observations of the reasoning of intact
molecular biology lab groups, for example, led him to conclude that analogies between
distant domains (e.g., solar system/atom) are quite rare and that many creative advances are
instead the result of analogies between close conceptual domains (e.g., between two different
viruses). Dunbar went on to argue that distant analogies may be developed subsequent to a
major discovery and serve as a means of communicating the new concept to others. So for
example, Rutherford may not have gotten the idea for how an atom might be structured by
considering the structure of the solar system. Rather, he may have chosen that analogy as way
of describing his idea, which had its origins in some other source. This focus on analogy as a
means of communicating a new idea seems especially relevant to entrepreneurs, who must
eventually be able to sell other people on their idea. The right analogy can be very persuasive,
as for example when proponents of intervention in the Gulf War compared the situation to the
early days of World War II, and warned about the dangers of appeasement.
Near analogies appear to have great potential as sources of ideas for new products.
Consider, for example, the line of variations on the phenomenally popular game of Monopoly,
such as the windy city variant, Chicago in a Box. The streets and landmarks are different, but
the underlying structure of the game is the same. Similarly, other popular items, such as
Trivial Pursuit and Magnetic Poetry (itself an interesting conceptual combination) have
served as models in the development of variations for particular target audiences. A
potentially productive strategy, then, is to use successful products as the source domain
and project the important underlying relations among the elements of that product onto a new
situation.
A host of other processes that have been investigated by cognitive psychologists also have
the potential to serve creative purposes. These include the reorganization of existing category
knowledge to form ad hoc or goal-derived categories to meet a particular need (e.g., Barsalou,
1983, 1991; Mumford et al., 1994), metaphoric interpretation, which can yield emergent
properties (e.g., Tourangeau and Rips, 1991), reasoning from unexpected observations
(Dunbar, 1997), and the constructive forgetting of interfering information during incubation
(e.g., Smith, 1995). These and other processes described in more application-oriented sources
(e.g., Perkins, 2000; Ward et al., 1995; Weber, 1992) may well be of use to entrepreneurs in
generating and developing creative solutions.

5. Abstraction, problem formulation, and originality
There is, of course, more to being creative than combining concepts, using analogies, and
applying other transformational processes. At least since the groundbreaking work of
Csizkzentmihalyi and Getzels (1971) showing a link between the exploratory activities of
artists and the quality of their subsequent creations, creativity researchers have been sensitive
to the idea that the way people formulate problems or tasks is an important component of the
creative process. In addition to the Runco and Chand (1994, 1995) model described earlier,
several other creativity models include steps such as problem construction, problem
definition, and problem discovery (e.g., Basadur, 1994, 1997; Mumford et al., 1991;
Sternberg, 1988; Treffinger et al., 1994). Implicit or explicit in these models is the belief
that the way in which people conceptualize a problem strongly influences their likelihood of
achieving an original or creative solution.
By distinguishing between processes associated with initial problem formulation and
subsequent procedures, such models draw attention to that fact that creativity may be more
than just problem solving. Particularly in real world settings, in which people are confronted
with ill-defined tasks, creative behavior requires several steps. Generally, innovators are not
simply handed clearly delineated problems, which they then begin to solve. Instead, doing
something creative often requires people to construct, formulate, or otherwise define the
problem or task to be accomplished, to retrieve from memory or seek out relevant
information, and to generate and evaluate potential courses of action.
Mumford et al. (1994) provided experimental evidence that engaging in problem
formulation increases the quality and originality of problem solutions. They had college
students perform a creative generation task in which they were to develop a marketing survey
and advertisements for a fictitious product. Students in a problem construction condition were
instructed to (a) list important factors to consider, and (b) restate the problem prior to
engaging in the task, whereas those in the no problem construction condition were not.
Importantly, those in the former condition produced ideas that were higher in quality and
originality than those in the latter condition. Mumford et al. suggested that problem
construction activities allowed students to consider a range of options rather than jump at
the first idea that came to mind.
Aside from the issue of engaging in some amount of problem construction is the question
of the form that construction takes and how it either enhances or inhibits creativity. Although
retrieval of existing knowledge is necessary for crafting successful innovations, the form in
which that knowledge is accessed can vary. Those variations can serve as indicators of how
people have formulated or defined their creative task, and they can have a major effect on the
originality of resulting ideas. Consider again Barker’s (1993) example of Sony’s initial efforts
to develop music CDs. Although members of the development team might have accessed all
sorts of specific and abstract information about methods of storing musical information, by
Barker’s account, they seem to have retrieved and relied on a very specific entity as a model,
namely, the LP record album. Put differently, they seem to have formulated their problem as
one of developing a sort of ‘‘digital LP record album,’’ and that formulation limited their
thinking.
Another example of the same phenomenon is that railway passenger cars were initially
patterned directly after stagecoaches, including external seating for the conductor, with
dangerous and even deadly consequences for the conductors who fell from the vehicles (e.g.,
Ward, 1995). Passenger cars were important innovations contributing to the railroad’s radical
transformation of travel in America, but retaining certain properties from the previous
stagecoach model resulted in initial designs that were nonoptimal for the new situation.
Although the designers of early railroad passenger cars could have retrieved abstract
information about transportation in general, they nevertheless relied heavily on a highly
specific type of vehicle, the stagecoach, and modified it only slightly for the new situation.
Put differently, they seem to have formulated or defined their problem as one of adapting
stagecoaches for use on railroad tracks instead of, for example, devising a completely new
type of vehicle that would have the properties needed to meet the new demands of rail
transportation.
There is a multitude of pressures that operate on individuals who attempt to devise new
real world products, and often there are compelling external constraints that can lead
designers to adopt particular approaches to the task. However, there are also internal,
cognitive constraints related to the structure of categorical knowledge and how it is typically
accessed that contribute to the tendency to base new ideas on specific instances of a given
category. It is those more cognitive factors that are of most interest in the present analysis.
Much knowledge can be thought of as organized into taxonomic categories with a definite
hierarchical structure. LP record albums, for example, can be thought of as instances of the
more general category of records, which in turn are instances of music storage devices, which
in turn are instances of storage devices, and so on. Perhaps more important than the existence
of hierarchies is the fact that a particular level of specificity within a given hierarchy, termed
the basic level, seems to dominate over others (e.g., Rosch et al., 1976). The basic level
resides between the most specific and most general levels of a hierarchy and it is the level at
which people most characteristically conceptualize a given object. To illustrate this point,
when shown a picture of an orange tabby cat, most observers will call it a cat rather than an
orange tabby, a tabby, a feline, a mammal, an animal, a living thing, or a tangible thing. Any
of those more specific or more general characterizations would be accurate and could be
accessed, but under most circumstances, they are not used as readily or as often as the basic
level representation of ‘‘cat.’’
A wealth of research documents the dominance of the basic level in less creative forms of
cognition, and there is reason to believe that it plays a powerful role in creative cognition as
well. If people’s most characteristic mode of representing information is at the basic level,
then it stands to reason that that tendency would influence the way they mentally construct
situations and the information they retrieve to deal with those situations. More formally, Ward
(1994, 1995)and Ward et al. (2000) have proposed the path-of-least-resistance model, which
states that when people approach the task of developing a new idea for a particular domain,
they tend to retrieve fairly specific, basic level exemplars from that domain and select one or
more of those retrieved instances as a starting point for their own creation. Having done so,
they then project many of the stored properties of those retrieved instances onto the novel
ideas they are developing. Consequently, the new creation can be expected to resemble the
old exemplar closely. In other words, retrieving basic level exemplars and basing new entities
on their properties can constrain the form of the new idea and reduce its apparent originality.
The results of laboratory studies provide direct evidence on this tendency to retrieve basic
level category exemplars as starting points for developing creative products. For example,
Ward (1994) had college students imagine and draw animals that might live on other planets,
and found that the dominant approach to the task was to retrieve one or more specific
instances of Earth animals (e.g., dogs and cats) and to use those as models for the imaginary
extraterrestrial. Not too surprisingly, given such an approach, the imagined creatures bore a
striking resemblance to Earth animals, including bilateral symmetry, legs, eyes located in
heads at the tops of bodies, and so on, and this tendency has been found to remain even when
people are instructed to develop creatures that are wildly different from Earth animals (Ward
and Sifonis, 1997).
This tendency to base novel entities on specific, basic level exemplars has also been shown
for the domains of fruit and tools (Ward et al., 2002). In addition, although investigators have
not always assessed their participants’ approaches to creative idea generation, the tendency of
novel ideas to be structured in predictable ways by existing conceptual frameworks is a robust
one that has also been observed in young children (Cacciari et al., 1997; Karmiloff-Smith,
1990), gifted adolescents (Ward et al., 1999a), science fiction authors (Ward, 1994), design
engineers (Condoor et al., 1993), and other creative individuals (Ward, 1995; Ward et al.,
1995). The phenomenon has also been shown to extend to a variety of conceptual domains,
such as imaginary coins (Rubin and Kontis, 1983), faces (Bredart et al., 1998), and
restaurants (Sifonis, 1995). Thus, it is reasonable to assume that the tendency to retrieve
and rely upon basic level domain instances is a general one underlying this broad range of
structured imagination phenomena (Ward, 1994, 1995).
More important than the overall findings of these creative generation studies, however, is
the fact that the originality of people’s creations was related to their approach to the task.
Specifically, individuals who reported retrieving and using specific examples of Earth
animals generated imaginary extraterrestrials that were rated as significantly less original
than those of individuals who did not use specific examples of Earth animals (Ward, 1994).
Although the strategies of the latter individuals were varied, they tended to take the form of
considering more abstract information, such as the properties a creature might need to survive
on a planet that had certain characteristics. Similarly, in a study that asked participants to
imagine and depict imaginary fruit, gifted adolescents were less likely than college students to
retrieve and use specific instances of Earth fruit (e.g., oranges) and their creations were found
to be significantly more original (Ward et al., 1999a,b).
Why should retrieving specific domain exemplars lead to less original products than
accessing domain information at more abstract levels? Again the answer may lie in the nature
of the stored information. The information stored about specific instances is, by definition,
more specific and consequently may be more constraining than more abstract information
regarding the same types of properties. For example, the representation of the basic level
exemplar ‘‘cat’’ might be expected to include properties such as ‘‘two eyes’’ and ‘‘four legs,’’
whereas the more abstract representation of ‘‘living thing’’ might include the parallel
properties of ‘‘means of sensing the environment’’ and ‘‘means of movement.’’ Assuming
that people are equally likely to project properties of the information they retrieve onto their
creations, regardless of the abstractness of that information, an individual relying on ‘‘cat’’
would be more likely than one relying on ‘‘living thing’’ to imagine an extraterrestrial that
had eyes and legs, which in turn would lead to lower judged originality. The more abstract
‘‘living thing’’ representation would afford a greater range of possibilities (beyond eyes and
legs) for implementing the general idea that creatures should have some way of sensing their
environment and moving around.
It is important to note, however, that there is nothing wrong, in principle, with the approach
of retrieving a specific known instance of a category or a specific previous problem solution
as way of gaining some purchase on a new problem. Indeed, there may be circumstances
under which it is preferable to a more abstract approach. For one thing, it can be highly
efficient and lead to the rapid development of new products. If a perfectly good model already
exists, then using and modifying it only slightly may be the most expedient and appropriate
course of action, as in the near analogy examples described in the previous section. It might
also help to make new products more acceptable to their target audience by keeping them
from deviating too far from the familiar. An extraterrestrial that deviated greatly from known
Earth animals might not be recognized as an animal at all, and by analogy, a new product that
deviated too greatly from other members of its product class might not be accepted by
consumers at all.
The problems associated with patterning new ideas directly on specific old ones become
apparent primarily when irrelevant, unnecessary properties of the old ideas are retained in the
new ideas and result in either delayed development or flawed designs, as in the CD and
passenger car examples noted earlier. The argument being made here is that a more abstract
approach might have helped in those types of situations. It might also be expected to
contribute to the success of business ventures, at least under some circumstances. For
example, an idea to improve on existing kennels might lead an innovator to add new features
to a basic kennel structure, but a consideration of more abstract ideas about why people use
kennels (e.g., to provide a safe and comfortable environment for pets while on vacation)
might lead to a new venture altogether (e.g., a service that makes ‘‘house calls’’ and cares for
pets in the owner’s home).

6. A broader perspective
There is little doubt that creativity is a complex enough phenomenon that the structures and
processes underlying novel idea generation will not be enough to explain it fully (e.g.,
contributions to Sternberg, 1999). Clearly, interactive models that include knowledge,
cognitive processes and skills, motivation, personality factors, and environmental influences
are needed to provide a complete theoretical account (e.g., Amabile, 1983; Sternberg and
Lubart, 1991). This is no less true for entrepreneurial creativity than for artistic or scientific
creativity. If we view successful entrepreneurs as those ‘‘individuals who identify opportunities
and start new companies to develop them’’ (Baron, 2000, p. 15), then they will need to be able to
do more than simply generate useful new ideas. Likewise, if we view entrepreneurial creativity
as ‘‘the generation and implementation of novel, appropriate ideas to establish a new venture’’
(Amabile, 1997, p. 20), then a range of internal and external factors become relevant to the task.
It is one thing, for example, to envision some desirable new Internet application and quite
another to implement the idea, convince others that it is worth pursuing, and then market the
application successfully. In addition to being able to generate ideas and recognize good ones
when they see them, entrepreneurs presumably ought to have high levels of intrinsic
motivation, belief enough in their ideas to push them even in the face of negative feedback,
at least some expectation of external rewards, and a capacity to persuade others of their worth.
They need the requisite ingredients to invest in ideas that are currently unknown, unpopular,
or otherwise low in value and to develop and sell those ideas to others at a higher value as in
Sternberg and Lubart’s (1991) investment model.
By focusing more narrowly on cognitive structures and processes, the intent of this paper is
not to diminish the importance of other factors. It is simply to draw attention to constructs
from mainstream cognitive psychology and how those constructs can be applied to an
important component of the entrepreneur’s task, namely, the generation and exploitation of
novel and useful ideas. It is clear that knowledge plays a paradoxical role in creative
endeavors. It supplies the raw materials from which creative new ideas are forged, but also
carries with it the potential to inhibit creativity. By careful application of a variety of basic
cognitive processes, it is possible to put knowledge to more effective use and improve
entrepreneurial creativity.

Acknowledgements
This material is based upon work supported by the National Science Foundation under
Grant No. BCS-9983424.