Interaction. Please do not cite or circulate without permission from the authors. A computer system does not itself express the motivations that initiated its design, the user requirements it was intended to address, the discussions, debates and negotiations that determined its organization, the reasons for its particular features, the reasons against features it does not have, the weighing of tradeoffs, and so forth. This information comprises the design rationale of the system. It can be critical to the many stakeholders in a design process: customers, users, service providers, and marketers, as well as designers who want to build upon the system and the ideas it embodies. Design rationale can contribute to theory development in HCI in three ways. First, it provides a foundation for ecological science in HCI by describing the decisions and implicit causal relationships embodied in HCI artifacts. Second, it provides a foundation for action science in HCI by integrating activities directed at description and understanding with those directed at design and development. Finally, it provides a framework for a synthetic science of HCI in which the insights and predictions of diverse technical theories can be integrated. The preceding chapters illustrated how theoretical concepts and methods from a variety of sciences are used to gain insight in the context of HCI design and evaluation. This chapter inverts that train of thought to some extent. It shows how reflective HCI design practices — involving design rationale documentation and analysis — can be used (1) to closely couple theoretical concepts and methods with the designed artifacts that instantiate them, (2) to more closely integrate theory application and theory development in design work, and (3) to more broadly integrate the insights of different technical theories. 1.

Introduction Agents are increasingly upon us. Although opposition is rare, \intelligent agents" have been attacked for user interface problems, and on larger social issues. Agent supporters have countered these arguments and raised doubts about alternative technologies. We place this in historical, social, and ethical contexts, noting the cyclic nature of such debates. One conclusion is that many problems with arti cial agents arise from a poor understanding of social aspects of human agents. 2. Historical Perspectives The history of technology has seen many movements call for human-like systems, and use anthropomorphic terminology to generate understanding and support. Such movements often make excessive claims, perhaps misled by their own rhetoric or their (sometimes impressive) partial success. This raises unrealistic expectations, which often leads to disappointment, which is surprisingly often followed by rebirth with similar goals, and somewhat improved terminology and techno

Faculté de psychologie et des sciences de l’éducation de l’Université de Genève pour obtenir le grade de Docteur en sciences de l’éducation par

Abstract not found

Abstract Design patterns are recognized in the software engineering community as useful solutions to recurring design problems that improve the quality of programs. They are more and more used by developers in the design and implementation of their programs. Therefore, the visualization of the design patterns used in a program could be useful to efficiently understand how it works. Currently, a common representation to visualize design patterns is the UML collaboration notation. Previous work noticed some limitations in the UML representation and proposed new representations to tackle these limitations. However, none of these pieces of work conducted empirical studies to compare their new representations with the UML representation. We designed and conducted an empirical study to collect data on the performance of developers on basic tasks related to design pattern comprehension (i.e., identifying composition, role, participation) to evaluate the impact of three visual representations and to compare them with the UML one. We used eye-trackers to measure the developers ’ effort during the execution of the study. Collected data and their analyses show that stereotype-enhanced UML diagrams are more efficient for identifying composition and role than the UML collaboration notation. The UML representation and the pattern-enhanced class diagrams are more efficient for locating the classes participating in a design pattern (i.e., identifying participation). Keywords Eye-tracking ⋅ Design Patterns ⋅ Visualization ⋅ Empirical study ⋅ UML

Interruptions are frequent in many work domains. Researchers in health care have started to study interruptions extensively, but their studies usually do not use a theoretically guided approach. Conversely, researchers conducting theoretically rich laboratory studies on interruptions have not usually investigated how effectively their findings account for humans working in complex systems such as intensive care units. In the current study, we use the memory for goals theory and prospective memory theory to investigate which properties of an interruption influence how long it takes nurses to resume interrupted critical care tasks. We collected data with a mobile eye tracker in an intensive care unit and developed multiple regression models to predict resumption times. In 55.8 % of all interruptions there was a finite—and therefore analyzable—resumption lag. For these cases, the main regression model explained 30.9 % (adjusted R 2) of the variance. Longer interruptions ( � �.36, p �.001) and changes in physical location due to interruptions ( � �.40, p �.001) lengthened the resumption lag. We also calculated regression models on subsets of the data to investigate the generality of the above findings across different situations. In a further 37.6 % of all interruptions, nurses used behavioral strategies that greatly diminished or eliminated individual prospective memory demands caused by interruptions, resulting in no analyzable resumption lag. We introduce a descriptive model that accounts for how nurses ’ behaviors affect the cognitive demand of resuming an interrupted task. Finally, we discuss how the disruptive effects of interruptions in the intensive care unit could be diminished or prevented.