Reactive information displays to augment comprehension and problem solving performance

In recent years, interactive systems have increasingly become a part of people's daily lives. People want to use new systems without having to spend a lot of time learning its commands. So interface design is rapidly moving from a technology-centered approach to a human-centered approach. However, even human-centered design is not a perfect solution because people still need to give explicit commands to activate and control a system. An even more novel and useful way of interacting with a system will be one that avoids the need for explicit commands.; One particular strand of research in this direction addresses the question of how an interface can actively assist the user in information processing tasks such as decision making, planning and problem solving. In this context, the focus of our research is the design and evaluation of adaptive interfaces that are designed to help people in problem solving activities. These interfaces adapt the information they present to the user in response to his or her eye movements, and are called Reactive Information Displays. Our research also tackled the problem of providing software to support the development of such interfaces.; We developed Reactive Information Displays that employ various strategies to assist users based on a cognitive model of problem solving proposed by Narayanan and Hegarty (2002). The design and operating principles of these displays are presented here. We evaluated these displays with several users. From these empirical studies, described herein, we have strong evidence that Reactive Information Displays can increase multiple aspects of users' problem solving performance. The display designs and evaluation studies form two of the three major parts of this dissertation.; Since an eye tracker produces a very large amount of raw eye movement data, it is impossible to analyze the data manually. Therefore, we developed automatic analysis tools to extract useful information from raw eye movement data. Also, for the various reactive strategies that we investigated, we developed software design toolkits to implement the corresponding displays semi-automatically. The third part of this dissertation consists of architectural specifications of the analysis tools and design toolkits.