3.1. Knowledge transformers in inventions and discovery

The purpose of this section is to show some evidence that the discoveries in science and inventions in technology can be explained by knowledge transformers described in Section 2. This task is not without difficulties, as the inferences made in support of the transformers or the lack of these are dependent on the details and clarity by which the accounts of the discoveries or inventions were given by the authors themselves or as recorded by the historians. The second difficulty is that the terms or the phrases used to describe the inventions may have different meaning(s) to that of knowledge transformers given in Section 2. Recognizing these limitations, the purpose here is to glean through published books or journals that recorded these inventions and scientific discoveries that made these inventions possible. As this is a mammoth undertaking, what is elicited here cannot be claimed to be exhaustive.

Watson and Crick's discovery of the double-helical structure of DNA, perhaps the most important discovery of biology of the 20th century, occurred through analogical reasoning, hypotheses formation, and substantiation by experiments (Wallace & Gruber, 1992; Strathern, 1997). The molecular modeling orientation that attracted Watson and Crick had as its most well-known chemist and practitioner Linus Pauling, who had scored a great triumph with a helical model that he proposed for the structure of protein α-keratin, which forms many structures, including hair, horn, and fingernails. Pauling's working methods were of particular interest to Watson and Crick, because proteins are in many ways analogous to DNA: both are long-chain molecules, composed of building blocks, each joined to the next through chemical bonds. The building blocks of protein are the amino acids, while those of DNA are the nucleotides: a sugar, a phosphate, and one of the four nitrogen-rich bases (adenine, cytosine, guanine, and thymine). The double helix of DNA is constructed like a spiral staircase, with two sugar–phosphate chains connected by rungs made up of pair of bases.

The analogy in structure between protein and DNA leads to the possibility that Pauling's methods might be applicable to the analysis of DNA, and in the fall of 1951, soon after Watson arrived in Cambridge, he and Crick decided that they would try to build a helical model of DNA. Two specific problems arose in constructing the model: how many helical strands it should contain, and where the bases should be located. The available information did not make clear the quantities or locations of the parts of the molecule, although it was known that the molecule was wider than a single-stranded helix. In Watson and Crick's initial model, there were three strands rather than the correct two, and the bases projected outward from the backbones rather than being between them. Three strands were held together by bonds between magnesium ions. The initial insight that Watson and Crick had for the structure of the DNA can be explained by the cognitive process of analogical transfer. Analogical transfer depicts situations in which information from a previous situation is transferred to a new one that is analogous to the old; the new situation is similar in structure to the old one. The knowledge transformer that described the analogical transfer is similarity comparison.

Another example of similarity comparison is provided by Design Continuum, which used an old idea in a new way to develop an innovative medical product for cleansing wounds (Sutton, 2002). It was to be used in emergency rooms to clean wounds with a pulsating flow of saline solution. The new product, called a pulsed lavage, had to meet strict guidelines for cleanliness and safety. It had to be low cost and disposable. The Design Continuum engineers recognized similarities between the pulsed lavage and a battery-powered squirt gun. On the surface, an emergency-room tool and the children's toy seem unrelated. However, once these engineers recognized the similarities between the two products, they were inspired to modify the inexpensive electric pump and battery of the squirt gun to meet the guidelines for the new medical product.

Table 2 summarizes other important inventions that have changed our lives over the past century and the knowledge transformers that are inferred from the published accounts of these inventions.

Inference of knowledge transformers through an account of selected inventions

4.1. Bisociation

Bisociation is the name given by Arthur Koestler (1964) in his book The Act of Creation to the notion that the essential ingredient in ideation is the effective combination of possible disparate or unconnected ideas: “elements drawn from domains which are far apart,” in Poincare's words.¹

The knowledge transformers that can be inferred here are that of association/disassociation that Koestler (1964) referred to as bisociation in which unrelated or incompatible concepts are linked together.

4.2. The Darwinian model

Donald Campbell proposed a Darwinian theory of the generation and growth of knowledge (Campbell, 1960). The application of Darwinian ideas to the resolution of particular problem has its foundation on the principle of natural selection. Keith Simonton (1988) proposed a Darwinian model of psychology of scientific discovery while Walter Vincenti (1990), the variation-selection theory. The evolutionary algorithms (e.g., genetic algorithms, genetic programming), based on the principle of natural selection, have become a well-established approach in solving problems involving multiobjective optimization.

The Darwinian perspective shed light on creativity, which can be summarized by its three mechanisms (Campbell, 1960). First, the production of genuinely new knowledge or ideas that are original demands the generation of variations that are blind in the sense that the consequences of these variations go beyond what can be foreseen or anticipated. Second, the variations are then subjected to a selection process that prunes out all variations that do not demonstrate a fit with the problem at hand. In the context of creative designs, this implies that one or more criteria are used to judge or rank alternative solutions, resulting in the rejection of those that fail to meet the criteria. Third, there must be a mechanism for the retention and propagation of the selected variation. This is because, without retention, a selected or successful variation can hardly make a permanent contribution to the acquisition and growth of knowledge. There is no knowledge transformer identified for the third mechanism, because it is more akin to that of storage (i.e., memory) of the selected variation.

The knowledge transformer inferred based on the first mechanism is that of randomization, which ensures generation of variations that are blind. The knowledge transformer inferred in the second mechanism is that of similarity comparison/dissimilarity comparison to prune variations that do not fit the problem under investigation.

4.3. Geneplore model

The Geneplore model consists of two distinct processing components: a generative phase, followed by an exploratory phase (Finke et al., 1995). In the initial, generative phase, one constructs mental representations called preinventive structures, having various properties that promote creative discovery. These properties are then exploited during an exploratory phase in which one seeks to interpret the preinventive structures in meaningful ways. These preinventive structures can be thought of as internal precursors to the final, externalized creative products, and would be generated, regenerated, and modified throughout the course of creative exploration.

4.4. Li's theory of conceptual intelligence

Li (1996) developed a theory of intelligence based on thinking and learning. He defines intelligence as follows: “Intelligence is the mental capacity for higher-order conceptual activities of thinking and the acquisition of knowledge.” Based on the assumption that intelligence has a strong link with symbolic systems, his conjecture is that intelligence is the unintended consequence of using symbol systems that facilitate thinking and learning. Recognizing that human beings use various symbolic vehicles (e.g., language, music, gesture, mathematics, and pictures; Gardner, 1983) in expressing and communicating meaning, Li suggests that it is language that has led to the creation of concepts that further enhance thinking and learning. The human language has led to the creation of a symbolic world, while the mental occurrences have brought about a mental world. The interactions of both worlds have transformed rudimentary thought into concepts. Once concepts are formed, thinking and higher order conceptual activities come into being, hence the emergence of intelligence (see Fig. 1).

The links between language and intelligence (adapted from Li, 1996).

Figure 1 suggests that thinking can be abstracted at three levels: rudimentary thought, conceptual thought, and creative thought (Li, 1996). Li distinguishes between rudimentary and conceptual thought by suggesting that the former is carried out without the medium of language while the latter is. Rudimentary thought is very much dependent on the visual presence of physical objects, and hence limited to operation on physical objects and actions. Conceptual thought is made possible by the creation of concepts that are the interaction between the symbolic world and the mental world. Through the medium of language, higher order concepts such as relations between objects and qualities extracted from generalization of objects are possible through conceptual thinking.

Li (1996) considers conceptual thinking and conceptual learning as the two major unique characteristics of human intelligence. Although Li has developed a theory of conceptual intelligence, what is unique about the theory is the link between conceptual thought and conceptual learning to creativity. In fact, Li suggests that learning precedes creativity in that the discovery of new knowledge (i.e., the creative process) cannot possibly exist out of the blue; it can always be traced back to its roots in existing knowledge. He posits that in a problem-solving situation, a person is making use of a number of existing conventional frameworks of knowledge to bear on the problem. Li suggests that a transformation may take place leading to the creation of new knowledge, but he did not suggest how this transformation takes place.

In Li's (1996) example of a creative work by Jaynes in developing a theory to explain consciousness, he alluded to the fact that Jaynes combined all the subbranches of knowledge (i.e., philosophy, psychology, language, and history) that were configured and reorganized to the point of proposing a unifying theme in a meaningful way. He suggested that it is the important insight of the connection/similarity between the description of hallucinating schizophrenics and the mind of the ancient people as obedient to the voices of God that led him to gather evidence from different knowledge domains for the development of his theory. Although Li did not suggest the nature of the transformation, here in this study we infer that the knowledge transformer inferred is that of a similarity comparison.

4.5. Summary

Table 3 shows a summary of the identification of the knowledge transformers in the various models and theory of creativity.

Identification of knowledge transformers in models or theories of creativity