Cognitive Foundation of Innovative Design 1 Summary of Proposed Research

Transcription

1 Cognitive Foundation of Innovative Design 1 Summary of Proposed Research Design is a universal human activity. The ultimate goal of design activity is to enhance the quality of human life in response to material and spiritual needs through innovative artifacts. This happens in many domains. For example, a government designs its social and economic programs to promote the wellbeing of its citizens. Artists design artistic works to express her/his inner inspirations. Industrial designers design products to fulfill certain functions specified by human users. Education professionals design instructional methods to train students. Mechanical engineers design machines to transform force, motion, signal, or energy. The widespread influence of design on human life has determined the critical role of design in a nation s economy and politics. It is now a well understood fact that a nation can achieve and maintain a high social and economic status for its citizens only through innovations in a highly efficient and quality manner. Over the last four decades, many design tools have been developed to reduce the time to market and to assure the quality of product development. However, few of existing design tools promotes innovation in design. Most of them oblige designers to follow a series of structured procedures in completing a design task. Such procedures add a degree of rigidity to the designer s thinking process. This is unfortunate since a designer s thinking process relies heavily on free thinking. Therefore, the long-term objective of this research program is to develop novel design tools that naturally accommodate flexibility and freedom for individual designer's style, innovation, and creativity. As a first step, it needs to be ascertained what factors block and what factors stimulate innovation in the design process. These factors can be described as part of the cognitive loading in the designer s thinking process. Cognitive loading is how much of a person's attention is being used up in completing a task. The design tool that is best for designers should generate the cognitive loading that stimulates rather than blocks their innovative thinking in design. Therefore, the short-term objective of this research program is to understand the designer s cognitive loading and its relationship to design innovation. This research will enhance our understanding of a most challenging and interesting phenomenon: design innovation. This greater understanding may help us improve not only government and business practices but also educational and artistic quality. Page 10

2 Page 11 2 Detailed Description of the Proposed Research Program 2.1 Objectives Design is a general intellectual activity that accompanies every part of human life. Design not only influences our everyday lives, it is also an integral part of a nation s economy, cultural identity and future. Innovative solutions in planning, materials and manufacturing are enriching our lives and leading designers into new realms of possibilities, through today s major design disciplines such as urban design, industrial design, graphic design, engineering design, interactive media design, and social design. The ultimate goal of this research is to develop robust design tools to assist human designers in making innovative design decisions in a time-and-cost effective manner. Most existing design tools such as design methodology and computer-aided design systems require designers to follow a series of structured procedures in completing a design task. These structured procedures usually add certain degrees of rigidity to the designer s thinking process. Design, however, is a creative act, which relies heavily on designer s free thinking. The rigidity underlying the existing design tools goes against the flexibility and freedom required by innovative and creative design. As a result, existing design tools have had difficulty in being widely accepted in innovative design practices. This difficulty implies a new research opportunity in that efforts should be made to integrate designer's freedom of thinking into the existing design tools. Therefore, the long-term objective of this research program is to develop novel design tools that naturally accommodate flexibility and freedom for individual designer's style, innovation, and creativity. As a first step towards the long-term objective, it needs to be understood what factors block and what factors stimulate innovation in the design process. These factors can be seen as part of the cognitive loading in the designer s thinking process. Cognitive loading is how much of a person's attention is being used up in completing a task. The design tool that is best for designers should stimulate rather than block their innovative thinking in design. The systematic understanding of the cognitive loading and its relationship to design innovation leads to a designer's cognitive model. The short-term objective of this research program is to develop a designer s cognitive model during the innovative design process. The specific aims and their corresponding assumptions are listed in the following: Specific aim 1: measure and quantify designer s cognitive loading. Assumption 1: designer s cognitive loading is reflected in his/her body movement and the materials generated during the design process. Specific aim 2: investigate the relationship between designer's cognitive loading and innovation. Assumption 2: cognitive loading may play both positive and negative roles in generating innovative design solutions. Specific aim 3: test the cognitive model, obtained from specific aims 1 and 2, by using it to develop a sketching tool for conceptual design. Assumption 3: the usability of a design tool depends on the designer s cognitive model. To the best of the applicant s knowledge, this proposed research is the first research of its kind in the field of design. This research would increase the usability of design tools so that innovation may be maintained while quality, expense, and time-to-market are becoming more and more of a concern for companies and businesses. As a result, an innovative design may be brought to realization in a time-andcost-effective manner.

3 2.2 Context Literature Review Design is a complex activity, involving artifacts, people, tools, processes, organizations, and the environment in which the design takes place. Design research aims at increasing our understanding of the phenomena of design in all their complexity and at the development and validation of knowledge, methods and tools to improve the observed situation in design. The history of design research is much shorter than that of design activities. Design research started to become a formal field of enquiry in the 1960's when the so-called design method movement attempted to develop a systematic design methodology [1]. This movement intended to establish objectivism and rationality as the foundations of both the process of design and its products, by applying new concepts and computational methods from domains like operations research. Design research can be considered to have passed through three overlapping phases: experiential, intellectual, and experimental. Typical results in experiential research include the systematic design methodology [2, 3], the theory of inventive problem-solving [4], and axiomatic design [5]. They are based on their respective authors individual design experience and retrospection. Meanwhile, intellectual explorations have been made to understand the design thinking [6], including its form [7-10], its capability and limitations [11], and its psychological foundation [12]. Protocol analysis is a prevalent method in current research to identify the role of designer s thinking and reasoning in a design process. This method studies subjects' mental processes in the performance of tasks by recording their spontaneous thinking aloud and subsequently segmenting the running commentary into the discrete atomic mental operations that the subjects have used in the accomplishment of the tasks [13]. Protocol analysis method has been widely used since the beginning of the 1980s to investigate the process of designing and understand how designers design [14-16]. These studies on protocol analysis have extraordinarily advanced our understanding of design. However the experimental setup and the interpretation of protocol data heavily influence the outcome of the protocol studies. The different setups of all these variables in the studies make it very hard to compare and learn from each other s work. In addition to protocol analysis, other experimental studies include the development and applications of various design tools in different disciplines. But in all three of these phases, due to a lack of theoretical framework, most results never find their way into practice, neither directly nor indirectly. This deficiency, together with a fast growing number of researchers in the field of design, has led to increasing concerns about the efficiency of design research and the effectiveness of its outcome. It is essential to establish a methodology that covers both the study of the phenomena of design as well as the development of design tools for design research to enter its next phase: the theoretical phase Theoretical Framework of Design Research Design Practice Axioms Deduction Design Theory Induction Design Activities Phenomena of Design Fig. 1. General methodology of design research In an effort to establish a theoretical framework for design research, the applicant has adopted two strategies: bottom-up and top-down, as is shown in Fig. 1. Research with the topdown strategy attempts to derive design theories from axioms developed from philosophical retrospection. Based on the logic of design that the applicant proposed together with Cheng [7], the applicant has developed an axiomatic theory of design modelling, which includes two axioms about objects and three axioms about human thought [17]. Models for particular design problems have been deductively derived from this axiomatic theory, such as the formal structure of and the source of design requirements [18, 19], environment-based design process model [20], and dynamic mechanism of design Page 12

4 creativity [21]. These models constitute a generic formal design theory. Compared to similar efforts in the literature [22-24], the axiomatic theory of design modelling developed by the applicant is the only theory that can support the reasoning about design activities in a logical manner. Other theories remain simply as formal representations of design. In continuing effort of the top-down research strategy, the applicant s research group is developing a formal design science and a formal design methodology based on the axiomatic theory of design modelling. These research efforts are being supported by the Canada Research Chair program and NSERC Discovery Grant. However, the research results from the top-down strategy cannot be accepted as a complete scientific design theory until they have been tested through experiments, case studies, and applications. This test is the main task of the bottom-up research strategy. This proposed research program will focus on the development of a designer s cognitive model using both the top-down and bottom-up strategies. Through the top-down strategy, a formal conjectural designer s cognitive model will be derived. The bottom-up approach will test the derived conjectural model through psychological experiments. Different from the conventional experimental studies in pure natural sciences, engineering sciences, or social sciences, the scope of this research involves both human being and products made by the human being. Methods and principles from the three categories of sciences will have to be used. A few questions need to be answered through his research: 1) is design innovation researchable? 2) can design innovation be automated? 3) how to quantity the cognitive process behind the design innovation? The answers to the first two questions are philosophical. Instead of proposing a methodology to answer these two questions, the applicant would investigate formally only the third question. The partial answers to the first two questions may come out along with the investigation of the third question. The logic of design [7] and the axiomatic theory of design modelling [17] have been proposed when the applicant was investigating more tangible aspects of design. Therefore, this proposed research program aims to develop a designer s cognitive model during the innovative design process. The philosophical foundation of this effort will also be explored informally throughout this investigation process though no goals and methodology are specified in this proposal. The multi-disciplinary nature of this research will be further discussed in Section Error! Reference source not found Importance, originality and anticipated contribution to knowledge As an integral part of a theoretical framework of design research, this research program completes a science of design by complementing the efforts in the axiomatic theory of design modelling. The research results will greatly enhance the usability of currently available design tools. The methodology proposed in this research can be applied to general design practices such as instructional design and organization design. The greatest challenge in this research program is the quantification of the designer s cognitive loading. This will also be its major original contribution. It is anticipated that the following outcomes will be generated from this research program: 1. a method to quantify the cognitive loading of designers in particular and operators in general, 2. a designer s cognitive model during the innovative design process, and 3. a sketching interface based on the developed designer s cognitive model. 2.3 Methodology The specific aims proposed in this research program will be achieved by applying the general strategy given in Fig. 1. The following addresses the research methods, activities, and required instruments for the three specific aims introduced earlier. More specific details are given about Specific Aim 1 than about the other two aims because it represents the greatest challenge of the three. Page 13

5 2.3.1 Specific aim 1: measure and quantify designer s cognitive loading Background. It can generally be assumed that the designer s state of mind will be reflected by his/her body movement and materials generated during the design process. In neuroscience, many studies identify the link between the human mind and brain structure [25-28]. These researchers are trying to understand a person s mental state by examining the internal behavior of the cognitive process, which they are examining by observing any change of sensitive regions of the brain [26, 27]. Parallel to the efforts in neuroscience, ergonomics attempts to understand human mental state through physiological indices such as heart rate, skin condition, eye movement, breathing rate, blood pressure, and limb movements [29]. Human physiological responses during cognitive activities are identified by using these numeric indices. However, even for the similar cognitive activity, these physiological responses may vary from person to person. Hence, it is a challenge and maybe problematic to relate directly these physiological indices to human thoughts. On the other hand, cognitive scientists attempt to study the cognitive process based on psychological experiments [25, 26]. As was indicated in Section of this proposal, protocol analysis has been used to understand the phenomena of design. The results from protocol analysis are mostly qualitative. There is a lack of depth in its observations on design practices. In addition, since protocol analysis is a kind of uncontrolled experiment, the experimental setup and the interpretation of protocol data heavily influence the outcome of the protocol studies. The different setups of all these variables in the studies make it hard to compare and learn from each other s work. Furthermore, protocol analysis requires the participants to think aloud during the design process. This will interfere strongly with the innovative design process. In this proposed research, a general model of a designer s cognitive loading will be developed through the quantification of the conceptual design process by combining ergonomic experiments with a modified protocol analysis. The modified protocol analysis does not require the designer to think aloud while designing. Instead it asks the designer to recount her/his thinking process after the task is completed. Methods, Activities and Instruments. One group of experienced designers and one group of novice designers will be invited from the areas of architecture, mechanical engineering, electrical engineering, and software engineering to participate in the experiment. To obtain the experimental data about an innovative design process, the participant designers will be presented with an industrial design problem that is new to all of them such that innovative solutions are needed from each designer. To conduct the experiment, my research group will: 1) record each designer s entire design process. Two instruments will be used: a video camera and a non-intrusive eye-gaze tracking device. The eye-gaze tracking device will be used to record the designer s eye movements during the entire design process. The video camera will be used to record the designer s sketches. 2) record the designer s description of his/her thinking during the design process. After the design has been completed, the designer, while watching the recorded video, recounts what was going on in her/his mind. The recounting will be video-taped and eye-gaze movements will be recorded as well. 3) quantify the designer s cognitive loading. The three sets of data obtained from the previous two activities capture different aspects of the design process. The eye-gaze movements reflect the patterns of the designer s attention during the design process. The recounting data approximately reflects the designer s reasoning and feeling underlying the design process. This information will be used to relate each individual s patterns of eye-gaze movements to the cognitive loading. The video taped design sketches show the intermediate design solutions and design conflicts at each stage of the design. They will be used to calibrate the experimental data obtained from the different designers. The concept of entropy in information theory will be used to quantify the cognitive loading according to the amount of information produced in the design process. Justification. 1) the applicant has secured a video camera (Sony Handycam) and an eye-gaze tracking device through a CFI award and an NSERC-RTI grant. 2) in working on a $2M industrial project entitled Intelligent sketching system for conceptual design, the applicant proposed to use the protocol Page 14

6 analysis method to develop a design process model and analyzed the video taped design process of a senior design engineer in a company [30]. Similar experiments have been on-going in the applicant s research group. Enough experience and expertise has been acquired in my research group to conduct the experiment and to analyze the data. 3) the greatest challenge in the proposed method is the quantification of cognitive loading. Generally, there are two difficulties: a) because of the direct relationship between eye movement and the use of a system, eye-gaze movement is usually taken as a critical index for user interface design [31]. During the thinking process, however, similar patterns of eye-gaze movement may mean totally different things to different people. In this proposed research, a quantitative index of cognitive loading will be developed by identifying the meaning of different patterns of eye-gaze movement through the data obtained from the designer s recounting. b) Different designers have different cognitive capacities. Their respective cognitive loading may appear different even for the same design problem. This variance will be taken into account by comparing the design conflicts at each stage of the design process. 4) it could be argued that the cognitive loading should be quantified by including more physiological indices. The choice of eye-gaze movement is mainly based on the fact that a great amount of research in cognitive psychology has shown positive eye-mind links in the areas such as reading [28]. This proposed research can be initiated by using the methods and results from the existing models of eye-mind links. With the momentum that may be built up through this exploration, my research group will be able to take into consideration more physiological factors. In summary, the proposed methods are feasible Specific aim 2: investigate the relationship between designer's cognitive loading and innovation Background. One of the main goals of current design tool development is to release the designer s cognitive loading so that designers can concentrate more on innovative activities [32, 33]. However, the applicant has observed in his research that certain types of cognitive loading may be necessary for the burst of a designer s innovative ideas. Two types of cognitive loading may be assumed: one positive and one negative. Negative cognitive loading distracts designers from innovative work whereas the positive cognitive loading stimulates and/or prepares the generation of innovative ideas. It is therefore essential to investigate the relationship between both types of cognitive loading and design innovation. Methods, Activities and Instruments. This specific aim will be achieved by using the experimental data from Specific Aim 1. My research group will: 1) invite a group of experienced industrial designers and repeat the same experiment described in Specific Aim 1. 2) establish a set of criteria from the data obtained from Step 1 for the evaluation of the degrees of innovation. 3) evaluate the degrees of innovation of designs generated from activities in Specific Aim 1 by using the established criteria in Step 2. 4) derive a formal designer s cognitive model from the axiomatic theory of design modelling. 5) identify design conflicts at each stage of the design process recorded in the video using the derived cognitive model in Step 4. 6) establish the relationship between cognitive loading and degrees of design innovation by using regression analysis and artificial neural networks. Justification. The major difficulty of this task is to quantify design innovation, which does not seem to be possible based on the current research in the literature. Based on the literature, there are two types of design creativity: social creativity and individual creativity [34]. The former is judged in terms of the social value of a generated product while the latter in terms of the individual s existing knowledge. In this research program, the concept of the individual creativity will be adopted since it focuses more on the mechanism of creativity. Based on the observation that a creative solution comes from design conflicts, a profile of design conflicts can be constructed using both the derived cognitive model and the protocol data. This profile would roughly reflect the profile of creativity throughout the design process. Using the dynamic mechanism of design creativity that has been developed by my research group [21], design conflicts from different stages can be related. By combining the relationship between cognitive Page 15

7 loading and protocol data obtained from Specific Aim 1, the relationship between innovation and cognitive loading can be modeled by regression analysis, artificial neural networks, or other data fitting methods Specific aim 3: Test the cognitive model developed from specific aims 1 and 2 Background. Based on the research framework in Fig. 1, the validation of a design theory can be done through applications. An important application of a cognitive model is the development of user-machine interface. the developed cognitive model will be tested by using a sketching interface for product conceptual design, which has already been developed in my research group. Methods, Activities and Instruments. My research group will: 1) redesign the existing sketch interface based on the developed designer's cognitive model from this proposed research program. 2) design an experiment for testing the usability of the sketching interface. The experiment plan includes the parameters to measure, the experimental procedures, and the data analysis method. The eye-gaze movement tracking device will be used in this experiment. 3) collect the experimental data by recording invited designers' eye-gaze movement in using the sketching system. 4) analyze the experimental data and relate the data back to the effectiveness of cognitive model. Justification. The sketching system is to be developed in Tablet PC, which is secured through my CFI grant. My Master's student L. Kong has developed a sketching interface based on a cognitive model summarized from the literature [35]. In addition, the applicant was a major participant in an industry project entitled Intelligent sketching system for conceptual design. These explorations have generated insights into the problem. The further development of this work is rather routine. The application of eyegaze movement tracking devices to the usability test of interface has been proved successful on a few occasions [29, 31]. In summary, this aim can be achieved without incurring any major risks. Page 16

8 Page 17 3 List of References [1] Cross, N., ed. Developments in Design Methodology. 1984, Wiley, Chichester. [2] Pahl, G. and W. Beitz, Engineering Design: A Systematic Approach. 1988: Spinger-Verlag. [3] Hubka, V. and W. Eder, Theory of Technical Systems. 1988: Springer-Verlag. [4] Altshuller, G.S., Creativity as an Exact Science: the Theory of the Solutions of Inventive Problems. Studies in Cyberbetics, ed. F.H. George. 1984: Gordon and Breach Science Publishers. [5] Suh, N., The Principles of Design. 1990: Oxford University Press. [6] Cross, N., K. Dorst, and N.F.M. Roozenburg, eds. Research in Design Thinking. 1992, Delft University Press. [7] Zeng, Y. and G.D. Cheng, On the logic of design. Design Studies, (3): p [8] Roozenburg, N.F.M., On the logic of innovative design, in Research in Design Thinking, N. Cross, K. Dorst, and N.F.M. Roozenburg, Editors. 1992, Delft University Press. p [9] March, L.J., The logic of design, in Developments in design methodology, N. Cross, Editor. 1984, Wiley, Chichester. [10] Galle, P., Design rationalization and the logic of design: a case study. Design Studies, : p [11] Simon, H., The Sciences of the Artificial. Third ed. 1996: The MIT Press. [12] Goel, V., Sketches of Thought. 1995, Cambridge, MA.: Bradford-MIT Press. [13] [14] Suwa, M. and B. Tversky, What do architects and students perceive in their design sketches? A protocol analysis. Design Studies, (4): p [15] Gero, J.S. and T. McNeill, An approach to the analysis of design protocols. Design Studies, (1): p [16] Cross, N., H. Christiaans, and K. Dorst, (eds) Analysing Design Activity. 1996: Wiley, Chichester, UK (1996). [17] Zeng, Y., Axiomatic theory of design modeling. Transaction of SDPS: Journal of Integrated Design and Process Science, (3): p [18] Chen, Z.Y., S.J. Yao, and Y. Zeng. A systematic approach for the specification of user requirements. in The Inaugural CDEN Design Conference Montreal, Canada. [19] Chen, Z.Y. and Y. Zeng, Classification of product requirements based on product environment. Concurrent Engineering: Research and Applications, An International Journal, (in press). [20] Zeng, Y., Environment-based formulation of design problem. Transaction of SDPS: Journal of Integrated Design and Process Science, (4): p [21] Zeng, Y., B.Q. Yan, B. Chen, and S.J. Yao. A theoretical and experimental study on design creativity. in The Inaugural CDEN Design Conference Montreal, Canada. [22] Yoshikawa, H., General design theory and CAD system, in Man-Machine Communication in CAD/CAM, Proceedings IFIP WG 5.2 Working Conference, T. Sata and E.A. Warman, Editors. 1981, North Holland: Amsterdam. p [23] Salustri, F.A. and R.D. Venter, An axiomatic theory of engineering design information. Engineering with Computers, (4): p [24] Braha, D. and O. Maimon, A Mathematical Theory of Design. 1998: Kluwer Academic Publishers. [25] Calder, A.J., A.D. lawrence, J. Keane, S.K. Scott, A.M. Owen, I. Christoffels, and A.W. Woung, Reading the mind from eye gaze. Neuropsychologia, : p [26] Hooker, C.I., K.A. Paller, D.R. Gitelman, T.B. Parrish, M.M. Mesulam, and P.J. Reber, Brain networks for analyzing eye gaze. Cognitive Brain Research, : p

9 [27] Krauzlis, R.J. and L.S. Stone, Tracking with the mind's eye. Trends in Neurosciences, (12): p [28] Reichle, E.D., A. Pollatsek, and K. Rayner, E-Z Reader: A cognitive-control, serial-attention model of eye-movement behavior during reading. Cognitive Systems Research (In press), [29] Lin, Y., A novel interface design framework for complex work domain: function-behavior-state paradigm. 2003, University of Saskatchewan. [30] Zeng, Y., A. Pardasani, H. Antunes, Z. Li, J. Dickinson, V. Gupta, and D. Baulier, Mathematical foundation for modeling conceptual design sketches. Transactions of the ASME: Journal of Computing and Information Science in Engineering, (2): p [31] Kotval, X.P., Eye movement based evaluation of human-computer interfaces. 1998, The Pennsylvania State University. [32] Tovey, M., S. Porter, and R. Newman, Sketching, concept development and automotive design. Design Studies, : p [33] McGown, A., G. Green, and P.A. Rodgers, Visible ideas: information patterns of conceptual sketch activity. Design Studies, : p [34] Akin, O., Necessary conditions for design expertise and creativity. Design Studies, (2): p [35] Kong, L., User Interface for Computer-Aided Product Conceptual Design, in Department of Electrical and Computer Engineering. 2005, Master's thesis, Concordia University: Montreal. p Page 18