The problem addressed in this thesis is the nature of design expertise and the role of visual mental imagery in design. The problem is addressed firstly, by examining the nature of problems, including design problems. It is argued that design problems are complex and ill-defined and can be distinguished from non-design problems. Secondly, design expertise is examined. It was found that design experts have a large store of design knowledge in a form that is readily accessible, and engage in extensive problem-finding prior to generating design solutions. Thirdly, the role of visual mental images as a component of design problem-solving and design expertise is examined. It is argued that visual mental images are important features of both design expertise and the transition from novice to expert. A number of case studies are designed and conducted. The findings of these studies are interpreted as supporting the theoretical ideas developed in the thesis. The introduction of design-based technology programs into Australian high schools has created the need for teachers to be able to assist students to generate creative solutions to design problems. Currently, technology teachers are experiencing difficulty in helping students to generate creative solutions to design problems. Hence a better understanding of design process may help to shape teaching and learning in design-based subjects. Furthermore, many complex everyday problems share similar properties with design problems. The research may therefore contribute to the understanding of the way people solve problems that have some characteristics in common with design problems. It is argued in this thesis that existing theories and models explaining the nature of problems and of the processes of solving problems are adequate in explaining many categories of problems and problem-solving but are inadequate in explaining the process of solving design problems. A new model of a problem space is proposed and justified. It is argued that design problems occur within a problem space that consists of a problem zone, a search and construction space and a satisficing zone. To establish, theoretically, the role of visual mental imagery in designing, two bodies of cognitive research literature are employed. Firstly, research into the utility of sketches in problem-solving are examined. This research indicates that external images assist problem-solving. Secondly, research into the relationship between perception and imagery is examined. This research suggests that visual mental images are functionally equivalent to perceived images. Thirdly, by combining the findings on sketches in problem-solving with the findings on imagery and perception, it is then possible to argue that visual mental images can assist problem-solving, and may play an important role in the resolution of complex design problems. The cognitive theory explaining the role of visual mental imagery in problem-solving in design is used to develop predictions for testing in two practical studies. Designers use visual imagery to represent and transform complex design problems within the problem space, and visual images are theorised as capable of providing more efficient representations for solving design problems than other forms of representation such as propositions. In the two studies undertaken in this thesis, a case study methodology was employed. The findings of the two studies support the arguments developed in this thesis that expert designers are able to form more complete and more detailed images of design problems and solutions than novices. Expert designers have a large store of previous solutions that can be retrieved from long-term memory as visual mental images. Expert designers are able to recognise when their existing solutions can be used, how they might be modified for use, and where something new is required. The study examined designing in terms of the deployment of procedures and the relationship among these procedures, and with images usage. It was found that designers traverse the design problem space using generative and exploratory procedures and that these procedures are facilitated by and facilitate, the production of visual mental images. The study provides a model of a problem space that can be used to explain the process of solving complex ill-defined problems, the cognitive processing involved in creative thinking and the role of mental imagery in an information processing theory of problem-solving. Conceptualising the problem space as containing a problem zone, search and construction space and satisficing zone makes it possible to apply the concept of a problem space to problems that do not contain well specified problem and goal states and with a limited number of operators. Integrating imagery theories with information processing theories provides an account of the process of solving complex design problems and the generation of novel solutions.