Vimal Viswanathan

Design Fixation in Physical Modeling: An Investigation on the Role of Sunk Cost

Physical models are very commonly used as tools for engineering idea generation, yet the guidelines in literature about their implementation are conflicting. A prior study has shown that physical models have the potential to supplement designers’ erroneous mental models; whereas a few observational studies have shown that physical models can cause a high degree of fixation under certain circumstances. At the same time, a previous controlled study fails to show the presence of fixation in idea generation with physical models. This study hypothesizes that prior observed fixation in physical modeling is due to Sunk Cost Effect, which is the reluctance to choose a different path of action once significant money, time or effort is invested in present one. Consistent with the prior study, this study also hypothesizes that physical models supplement designers’ mental models. These hypotheses are investigated through a controlled between-subject experiment. The results show that cost of building plays a vital role in fixation and fixation is not likely inherent in physical representations. Results also show that physical models supplement designers’ mental models and lead them to higher quality ideas.Copyright

Physical Models in Idea Generation: Hindrance or Help?

Engineering idea generation is a critical part of new product development and physical models are one tool used in this phase of design. Unfortunately, few guidelines about the effective use of physical models to support idea generation exist. The advantages and disadvantages of physical models need to be clarified so that engineers know when and where to implement them effectively. Previous literature indicates there is potential for design fixation on physical prototypes. This limits the solutions considered. In contrast, other recommendations encourage the extensive use of physical models and the psychological literature indicates that physical representations have the potential to lead to more feasible design by supporting designers’ mental models of physical phenomena. This study evaluates these questions with a between-subjects experiment with four conditions, sketching only, building, building & testing, and constrained sketching. No evidence for design fixation is observed. The results show that physical models supplement designers’ mental models, thereby leading to higher quality ideas (fraction of functional ideas). This result shows a potential way of improving designer’s innovation by strategically implementing fast and cheap prototyping methods.Copyright

Enhancing student innovation: Physical models in the idea generation process

Innovation and engineering creativity are highly sought after skills. Unfortunately, little guidance exists for developing and teaching these critical skills. Physical representations ranging from simple, very rough models to full working prototypes are common within the design process and are a likely tool for supporting innovation. There are varied and conflicting recommendations on when physical models should be employed. Some studies suggest physical models cause design fixation. Design fixation is when designers think of a particular concept and it limits the ideas they are able to generate. The effects of physical models in the idea generation process were explored through a controlled experiment comparing participants who sketched, built, or built & tested their concepts. Results show that physical models overcome gaps in designers mental models thus producing a greater number of functional ideas that solve the design problem. In addition, physical models appear to not cause design fixation but a larger sample size is required to verify this result. Similar levels of novelty and variety are observed across the conditions.

A Study on the Role of Expertise in Design Fixation and its Mitigation

Engineering idea generation plays a vital role in the development of novel products. Prior studies have shown that designers fixate to the features of example solutions and replicate these features in their ideas. This is a major hindrance in idea generation as it restricts the solution space where designers search for their ideas. This study hypothesizes that though expert designers fixate to example features, they still can outperform novices in terms of quantity of ideas as they have a larger set of knowledge acquired through their experience. To investigate this, the experimental by Linsey et al. is replicated for novice designers. Novices generate ideas for a design problem in three groups: one group working with a fixating example, a second group working with the same example along with alternate representations for the design problem and a control group only presented with the problem and no additional materials. The obtained results support the hypothesis. Both novice and expert designers are fixated to the example features, but the expert designers generated more nonredundant ideas. The alternate representations of the design problem help experts in mitigating their fixation, whereas in novices, these have no effect.Copyright

Understanding physical models in design cognition: A triangulation of qualitative and laboratory studies

Designers use various kinds of physical models throughout their design process to enhance creativity. The existing literature provides conflicting guidelines about their implementation. The effects of physical models on design cognition remains largely unknown. Prior laboratory studies show that physical models supplement designers erroneous mental models and thereby lead to higher quality ideas. These prior studies fail to demonstrate any design fixation associated with the use of physical models. In contrast, a few prior observational studies on practicing designers show that the use of physical models causes design fixation. Based on these conflicting results, this study investigates the role of physical models in industry-sponsored projects and in the development of award-winning products through a qualitative research approach. This study explores two hypotheses: The Mental Models Hypothesis - physical models supplement designers mental models and the Fixation Hypothesis - physical models cause design fixation during the idea generation process. The data are coded qualitatively and then tested quantitatively. The results are triangulated with the results from the prior controlled study. The results provide support to the hypotheses. The differences observed between current and prior studies point to the potential role of the Sunk Cost Effect in engineering idea generation with physical models.

The influence of design problem complexity on the attainment of design skills and student perceptions

One of the current educational challenges is how do we educate engineers to systematically solve open-ended real-world design problems? The capstone design course often plays a critical role in this, but there are numerous questions on how best to teach design and what are the characteristics of realistic design problems which provide excellent learning opportunities. This paper reports on a controlled evaluation of the effects of design problem complexity on students ability to functionally abstract a design problem and its effect on their perceived value for variety of design methods. It is important for students to learn a systematic approach to the design process and to perceive its effectiveness. Students perceptions and functional modeling skill are measured. Results, while preliminary due to limited sample size, indicate the complexity of the design problem is a critical factor in teaching design methods. There is a statistical interaction between the complexity of the design problem and opinion of functional modeling on their ability. Results also indicate that students who work on more complex design problems are more likely to expect to use functional modeling in the future. More complex design problems lead to a more positive student opinion. Further development of the functional modeling quiz is needed as is a larger sample size. Overall, results indicate that more complex design problem demonstrate to the students the effectiveness of the methods.

Physical modeling in design projects: Development and testing of a new design method

Physical models are widely used as idea generation tools by industrial designers, engineers, engineering educators and government agencies. Many schools promote the use of physical models in their engineering curricula. Despite the apparent popularity of physical models, little is known about their cognitive impacts and when they should be implemented in the design process. A few studies have explored physical models and their use as idea generation tools; however the guidelines from them are conflicting. Based upon these conflicting guidelines, a series of controlled and qualitative studies are conducted by the authors to understand the cognitive impacts of physical models in engineering idea generation. In addition to the insights from these studies, data are collected from a project-based graduate design course. The reports from design teams prototyping their ideas as a part of a class project are studied. These reports provide insights about the conceptual errors that student designers make as they build and test physical models of their designs. To reduce the two most critical errors, a design method (Model Error Reduction Method) is formulated. This design method forces the designers to think about two potential conceptual errors in their designs and provides guidance to rectify the issues. The two conceptual errors that the method addresses are: failure to account for critical loads and failure to design connections. This paper presents a controlled experiment evaluating the effectiveness of the method. The preliminary results show that novice designers find the design method extremely useful; moreover, the method seems to help eliminate, to a large extent, said conceptual errors. These findings suggest that the method might augment existing engineering design curricula.

A Study on the Representation of Examples in Learning Engineering Concepts.

The use of examples in engineering curricula is a commonly used means to teach engineering students new concepts and ideas; these examples play an important role in teaching engineering students how to become technically competent engineers and designers. Being able to learn from examples and avoid fixation to those examples is an important task in that process. Design fixation is a major constraint in design thinking as it constrains the solution space where designers search for their ideas. The experiments described in this paper aim to investigate how students fixate to different types of representations given to them. A pilot study comparing sketched and physical representations of examples show that students are less likely to fixate to the design specifications of examples provided in the form of physical model, this suggests that they are able to better understand the design limitations of examples presented in the form of a physical model. A second experiment is described which will explore this trend further and will compare how students fixate on and derive information between sketched and computer-aided design representations. INTRODUCTION: The use of examples in engineering education is a common practice for engineering educators. The use of visual or physical examples allow students to get a clearer picture of the topic being taught and can enable them to have a better understanding of the concepts.

Teaching capstone design: The influence of problem complexity

Many capstone courses include design methodology but the characteristics of design problems that provide the best learning opportunities need to be defined and the effect on student perceptions measured. In an industrial setting, it is generally up to the engineer to decide which design methods should be applied in order to obtain the desired outcome. Thus if students do not believe the design methods to be effective, they will not choose to use them in the future. Functional modeling is a technique for systematically breaking down a design problem into its sub-functions. Groups were given either a simple or complex problem to solve for their senior design project. At the end of the semester, a quiz over the functional modeling concepts was administered, and a questionnaire was answered by the students to measure their perceptions of the design methods. An ANOVA shows an interaction between student opinions and the complexity of the design problem on the students functional modeling ability. Results indicate the complexity of the problem and perceptions of the design methods are significant factors when determining the students functional modeling abilities. Additionally, the complexity of the design problem likely affects other design skills.

Work in progress — Understanding design fixation: A sunk cost perspective on innovation

In the current competitive world, engineers must be innovative to be competitive and these skills must be taught. Physical representation is a likely tool to assist in being innovative. Guidelines do not exist for when physical representation should be implemented in the design process. Past studies provide conflicting results on the effects of physical representations on cognition. An observational study of an industrial team shows that physical representations could cause design fixation. In contract, a controlled lab study conducted by the authors failed to reproduce this result. One possible explanation is the sunk cost theory from behavioral economics, which states that people show the tendency to continue with the chosen path based on the amount of time, money and effort already put in. The controlled lab experiment used a fairly simple design problem whereas the other study contained a more complex design problem, thus the observed differences in apparent design fixation could be due to the sunk cost effect. This work in progress paper investigates sunk cost as an explanation for the result differences. This paper proposes an experiment to test the sunk cost effect with physical representation and discusses the future work along with possible implications of the results.