A study on the role of physical models in the mitigation of design fixation

AbstractDesigners implement a variety of models and representations during the design process, yet little is known about the cognitive impacts of various representations. This study focuses on how physical models can assist novices in mitigating design fixation on undesirable features. During idea generation, designers tend to fixate on examples they encounter or on their own initial ideas. The first hypothesis states that designers tend to duplicate features of provided examples. The second hypothesis states that this fixation can be mitigated with appropriate warnings. The last hypothesis is that building and testing physical models can help designers in mitigating fixation. To investigate these theories, a quasi-experiment is conducted as part of a freshman class project. Students design, build and test stunt cars in three different experimental conditions, each receiving a different pictorial example: an effective example, a flawed example and a flawed example with warnings about the flaws. The results show that in all the conditions, designers duplicate undesirable features from their examples, even when they received warnings about the flawed features. Copying these flawed features creates more complicated and less effective designs. However, through the physical testing of their designs, participants identify and fix the design flaws. These results indicate that existing designs and experiences have the potential to limit innovation and that designers need to be trained with effective methods for mitigating design fixation. Building prototypes can help designers in identifying the flawed features and in reducing design fixation; hence, the use of physical models in engineering design needs to be encouraged.Keywords: design fixationengineering designphysical modelsprototypingidea generation AcknowledgementsSupport for this study was provided by the National Science Foundation CMMI-1000954 and CMMI-1322335. Any opinions, findings, conclusions or recommendations expressed in this paper are those of the authors and do not necessarily reflect the views of the National Science Foundation. The authors would like to acknowledge the support from the instructors of the ENGR 111 class at Texas A&M University, Dr Natela Ostrovskaya, Mr Jacob McFarland and Mr Joshua Bittle, for their support in gathering the data for this study. Partial results of this study have been presented in the 2012 ASEE Annual Conference and Exposition.