A fix for fixation? Rerepresenting and abstracting as creative processes in the design of information systems DORIS ZAHNER, 1 JEFFREY V. NICKERSON, 1 BARBARA TVERSKY, 2 JAMES E. CORTER, 2 AND JING MA 1 1 Stevens Institute of Technology, Castle Point on Hudson, Hoboken, New Jersey, USA 2 Teachers College, Columbia University, New York, New York, USA (RECEIVED April 15, 2009; ACCEPTED October 15, 2009) Abstract Fixation prevents the associations that are bridges to new designs. The inability to see alternative solutions, or even to see how to map known solutions onto current problems, is a particularly acute problem in the design of software-intensive systems. Here, we explored two related ways of liberating fixated thinking: abstracting and rerepresenting. Although both techniques helped designers generate original ideas, not all the added ideas fit the problem constraints. We discuss ways the results might be used to generate reflective design aids that help designers to first generate original ideas and later prune them. Keywords: Abstraction; Design of Information Systems; Fluency: Originality; Rerepresentation 1. INTRODUCTION How does thinking happen? At the core, thinking is associa- tionistic (e.g., Rumelhart & McClelland, 1986; Anderson, 1993), that is, one thought leads to another. The paths these thoughts can take are too numerous to count, similarity on as- pects too plentiful to list, growing exponentially with each link. Despite the abundance of associations and paths, minds get stuck in ruts, all too often not going in the right direction or not going far enough. One challenge for creativity is en- couraging thought to travel in many directions. Thought may be associationistic and unconstrained, but problems have constraints. Not all paths of thought lead to vi- able solutions. Unfortunately, the paths that fail to lead to vi- able solutions are not marked dead end. A second challenge to creativity, then, is encouraging thought in viable directions. These challenges to creativity have long been noted: the first is promoting divergent thinking, the second, convergent. The ordering of these challenges makes sense, exactly because it is not problem constraints that typically block associative paths of thoughts. One way to be creative is to first generate a broad range of ideas, then check to see if they conform to the con- straints of the problem. This is how the mind works naturally, both in memory (Rundus & Atkinson, 1970) and in judgment (Sloman, 1996; Kahneman & Frederick, 2005): first a burst of rapid associations, then a slow evaluation of them. New de- signs are often created by adapting previous ones, a process called transfer. In one set of studies, students read a problem about radiation, where the goal is to destroy a tumor in the body without destroying surrounding tissue (Dunker, 1945). They also read the solution, which is to send weak rays from many directions that converge on the tumor. After performing irrelevant tasks, students were presented with analogous prob- lems in other domains, capturing a castle or putting out a fire, where success depended on converging from many directions. Students typically failed to transfer a solution from one domain to another, especially when those domains were far or remote from the solution domain (Gick & Holyoak, 1980, 1983). Within engineering design, some research suggests that trans- fer, both near and far, can happen during the design process (Christensen & Schunn, 2004). However, most studies lead to a more pessimistic conclusion. Unfortunately, research in general problem solving, a superset of design problem solving (Simon, 1995), has repeatedly demonstrated failures to transfer known solutions to new problems (Gick & Holyoak, 1980, 1983; Ross, 1989; Novick, 1990; Ross & Kennedy, 1990; Novick & Hmelo, 1994). The typical reason for the failure to apply old solutions to solve new problems is a failure to see the similarity of the global constraints or the deep structure of the problems. The reason for this failure is, in turn, a general property of associationistic thought: it travels most easily along Reprint requests to: Doris Zahner, Teachers College, Columbia University, 525 West 120th Street, Box 118, New York, NY 10027. E-mail: dwc14@ columbia.edu Artificial Intelligence for Engineering Design, Analysis and Manufacturing (2010), 24, 231–244. # Cambridge University Press, 2010 0890-0604/10 $25.00 doi:10.1017/S0890060410000077 231