In the design process, which also requires creative thinking, prior knowledge and experiences play a major role. Previous studies in cognitive psychology support the important role of prior knowledge and experiences in creative thinking.
Prior knowledge and experiences in design
Successive studies conducted by a group of researchers (A. Purcell, Williams, Gero, & Colbron, 1993; A. T. Purcell & Gero) reported fixation in both mechanical and industrial designers, although the degree of fixation varied across disciplines. .
Organization of design knowledge
When these episodic experiences are stored in the memory system, a process of generalization accompanies the abstraction of the specific details of the experiences. Although Oxman (1990) has suggested the role of prior knowledge in the design process, the theory was supported by an understanding of the memory system rather than the cognitive process itself.
Participants
Design briefs with different constraints
The Influence of constraints on the cognitive process
In terms of constraints, many researchers describe the planning process as a constraint management process (Bonnardel, 2000; Chandrasekaran, 1990; Gero, 1990; M. In the “Geneplore model” proposed by Ward et al (Ward, Smith, & Finke, 1999 ), constraints were described as an important part of the planning process.
Differences between two design briefs
As shown in Figure 1, the design constraint model consists of three dimensions – generators, domain, and functions of the design problem. In the case of design 1, the participants were able to define four different functions themselves, except for the radical limitation set by the client - the folding chair.
Experimental setting and procedure
Depending on the functions of the constraints, Lawson (2005) defined four different functions – radical, practical, formal and symbolic. These protocols were used as a main source for the analysis, while being supplemented by sketches of each participant.
Identifying a think flow
15 were combined into a thinking flow, which was defined as a coherent thought process with a single topic or coherent perspective. Four different types of behavior were used as an indicator of the beginning of a new thought flow.
Coding procedure
18 primary encoding process, a new category was developed, the definition of each category was refined, and the final coding scheme was developed. There was at least a month's time gap between the first encoding and the second encoding.
The coding scheme
- Ideas
- Precedents - prior knowledge and experiences
- Interpreters
- Relationships among encoded items
The recent understanding of memory system was adopted in this study to develop the first level of the coding category - episodic and semantic precedents. The second level of the coding categories considered the role of each precedent in the creative process. A generalized model of the design process was adopted which described it as an iteration of analysis, synthesis and evaluation (Jones, 1963; Lawson, 2005; Wynn & Clarkson, 2005).
The analysis of the problem space included the redefinition of the problem and the examination of existing or similar solutions. These were called Interpreters, because they helped interpret the meaning of the design brief, and were involved in manipulating the problem space.
Constructing a cognitive map
27 The design brief, visualized as a gray star, formed the starting point of each stream of thought. The first element of a thought flow was related to the design brief, and the following elements were connected in sequential order. Although most thought streams began with the design brief, certain thought streams are sequentially connected to other streams because the beginnings of new streams were derived from the elements of other thought streams.
Analysis of cognitive maps
Thus, the concept of node corresponds to the coded item of a protocol in this study. Link data that defines the link between nodes corresponds to the relationship between coded items in this study. The number of thought streams and ideas are factors related to participants' creative activities.
Thought streams unrelated to idea generation Thought streams related to. Used for problem space analysis Used for solution space analysis Used for synthesis.
Effect of constraints on the creative process
This indicates that the level of constraints of a design problem does not have a significant impact on the creative output of designers. The result of the t-test showed that the proportion of episodic precedents is significantly higher in the more limited problem, t p = 0.04. The analysis shows that the proportion of precedents for problem space analysis was significantly lower in the less constrained brief.
In summary, the degree of constraint did not have a significant effect on the number of ideas and streams of thought, but it did have a significant effect on the use of precedents. However, this study shows that designers can change the types of precedents and their use depending on the design task, while they generated the same number of ideas regardless of the design task.
Episodic knowledge - dominant and essential
This phenomenon can also be related to the theory of prior knowledge organization suggested by Oxman (1990), and it proposes a further explanation of the relationship between episodic and semantic precedents. In most cases, episodic and semantic memories were associated with each other, and this helps the theory. However, several instances were observed in the protocols of this study that a semantic precedent triggers a series of episodic precedents.
It could thus be concluded that episodic knowledge is essential for the utilization of previous knowledge, but both episodic and semantic knowledge contribute to the search and retrieval of memories. Based on the interpretation of cognitive maps, three phases of the cognitive process were identified - exploration, generation and development.
Phases of the creative process
Unlike the two previous phases – exploration and generation – the development phase was observed from a group of participants. The essential insight in this division is the distinction between retrieval of prior knowledge and generation of new ideas. Although different models of the creative process have been proposed, they considered the process of exploring relevant knowledge as part of the generation phase or a preparatory step of it.
In the case of the Geneplore model proposed by Ward, Smith & Finke (1999), the retrieval of prior knowledge is described as one of the elements that constructed the generation phase. Although several researches have proposed the distinction between idea generation and retrieval of relevant knowledge, there have been few attempts to investigate the different cognitive aspects of the two phases (Amabile, 1996; Getzels & Csikszentmihalyi, 1976).
Cognitive styles in each phase
In the exploration phase
In this study, the proximity centrality score close to 1 indicated a cognitive process consisting of different streams of thought with a similar level of progress. 42 A hierarchical cluster analysis was performed on the number of total thought streams and the closeness centrality index score using the centroid linkage method. The six participants of cluster 1 showed limited exploration, consisting of relatively fewer numbers of thought streams – on average about thirteen thought streams.
The diverse characteristics of the participants in Cluster 3 were supported by the average number of thought streams – approximately 32, which is double that of the cluster. That is, they remained in the midst of limited and divergent explorations in terms of the number of thought streams and the centrality of their cognitive maps.
In the generation phase
As depicted in Table 11, ten of all participants generated a limited number of initial ideas. Participants classified as group 1 used 1.7 thought streams to generate an average of 1.7 initial ideas. Both in number of thought streams and ideas, these participants were extremely limited in generating new concepts.
Unlike group 1, two participants in group 3 showed a strong divergence in generating initial ideas. Multiple streams of thought and initial ideas show that both participants generated different ideas that are discrete from each other.
In the development phase
Half of the participants belong to cluster 2, which is a medium between clusters 1 and 3 in terms of the diversity of ideas created. The difference between cluster 2 and cluster 3 is less clear in the average number of ideas developed. The participants in cluster 3 developed a single idea further, and the level of development is more progressive than the participants in cluster 2.
In cases of the exploration and generation phases, each participant showed cognitive activities related to the two phases, regardless of differences in the degree of divergence. However, in the case of development phase, nine participants made no effort to develop an idea.
Four cognitive styles in the creative process
- Focused probers
- Treasure hunters
- Selectors
- Explorers
This is not the first attempt to classify the cognitive styles of the creative process. It provided a more elaborate explanation than the results of previous studies which defined cognitive style without considering the different steps of the creative process (Cross, 1985; Kvan & Jia, 2005; Tovey, 1984). Furthermore, this study examined the cognitive process itself to identify cognitive styles.
I analyzed and used the cognitive activities of designers as an indicator of cognitive style. In this section, we will discuss the differences between cognitive styles in terms of generating ideas and using precedents throughout the creative process.
Differences of cognitive styles in idea generation
A Tukey post-hoc test revealed that the number of thought flows of Focused Probers was significantly less compared to Selectors. In the case of the number of initial ideas, a Tukey post-hoc test reported that Focused Probers and Treasure Hunters generated statistically significantly fewer ideas compared to the Selectors and Explorers. There were no statistically significant differences between Focused Probers and Treasure Hunters, as well as between Selectors and Explorers.
As described in the method to determine the cognitive style in the generation phase, Focused Probers and Treasure Hunters were participants who showed a limited diversity in the generation of ideas. 58 In terms of the number of ideas developed, four cognitive styles were significantly different from each other, except for the Focused Probers and Treasure Hunters.
Differences of cognitive styles in utilization of precedents
This showed that participants could generate a different number of ideas while using the same type of precedents for the same purpose. Thus, we can conclude that assigning designers appropriate cognitive styles is much more effective for achieving desired results than providing appropriate design information. As shown in Table 18, the level of restrictions did not have a significant effect on the results of creative activities.
On the other hand, the cognitive styles showed no differences in the use of precedents, while the level of constraints distinguished it. However, the cognitive style is unchanged, while they use different types of precedents for different purposes.
Implications on design education and practice
The interaction of cognitive styles between teachers and students can cause significant differences in learning outcomes (Felder & Silverman, 1988; Pask & Scott, 1972). Depending on which phase the designer is involved in, the design tasks change and the preferred cognitive styles also differ. If a company had identified the designer's cognitive styles and abilities in advance, the company can allocate appropriate designers to appropriate design tasks.
The identification of cognitive styles can also provide useful information to construct a better design team. Further, statistical analysis revealed that these four cognitive styles had significant differences in the creative outcomes, but non-significant difference in the use of examples. Characteristics and differences in the four cognitive styles were also reviewed and discussed.
Further study should be done on the method that also helps the identification of cognitive styles in an easier way.
