Previous theories of design decision-making have discussed how designers use analogical reasoning to quickly scope the solution space down to one viable solution. This initial analogy forms the template of a possible action plan that is then modified to suit the unique properties of that particular problem. This use of analogical reasoning allows designers to quickly engage with the problem and generate a workable solution. Our findings indicate that this initial analogy actually persists across all stages of decision-making, and does not play a role solely during the first stage of scoping. This analogical persistence leads to poorer design decisions. This thesis presents a series of studies that adopt a mixed method approach to investigating the influence of analogies on the decision-making of Interaction Designers. We employed qualitative methods such as the Critical Decision Method for Eliciting Knowledge (Klein, 1989), which aided in identifying analogical persistence as a problem that leads to poorer quality decisions. We also employed quantitative methods such as the Design Fixation paradigm (Jansson & Smith, 1991) to investigate how different types of analogies (self-generated & provided Priming Analogies) can influence the expression of analogical persistence. Finally in an attempt to mitigate the potential pitfalls of analogical persistence, this thesis attempts to control it using principles from Design Rationale (Lee & Lai, 1991) and Reflection (Schon, 1983). Rather than seeing a decrease in analogical persistence, our manipulation actually increased fixation. A follow-up study identified that designers tend to poorly appraise the weaknesses in the initial analogy, which may have led to the aforementioned unexpected result. These findings challenge the notion that greater understanding of the design space will lead to higher quality design decisions.

This research monograph describes the integration of analogical and case-based reasoning into general problem solving and planning as a method of speedup learning. The method, based on derivational analogy, has been fully implemented in PRODIGY/ANALOGY and proven in practice to be amenable to scaling up, both in terms of domain and problem complexity. In this work, the strategy-level learning process is cast for the first time as the automation of the complete cycle of construction, storing, retrieving, and flexibly reusing problem solving experience. The algorithms involved are presented in detail and numerous examples are given. Thus the book addresses researchers as well as practitioners.

The overall goal of this research program is to develop a descriptive model of decision making that can be applied to tactical and strategic decision making domains. We wish to explore variables such as time pressure, risk, and levels of expertise as they affect the use of decision strategies and situational assessment. We also want to contrast analytical and recognitional models of decision making. Year 1 was devoted to field studies and knowledge elicitation interviews with decision makers including urban fire ground commanders, wildfire incident commanders, and Army Armored Division platoon leaders. In addition a study we performed comparing the quality of chess moves under speeded and non-speeded conditions served as an analogue for real-world strategic and tactical decisions. All of these studies allow a comparison of the decision making at higher and lower-levels of expertise. These studies were designed to test the generality of a Recognition-Primed Decision model as a function of different domains, varying experience levels to group decision settings and varying levels of time pressure.