The physiologically attentive user interface: Towards a physiological model of interruptability

This thesis addresses the problem of managing interruptions from computing devices and appliances in a Ubiquitous Computing (UbiComp) environment. One way of approaching this problem is through the use of Attentive User Interfaces (AUIs). AUIs sense the users' attention for objects and people in their environment, in order to avoid today's ubiquitous patterns of interruption. However, current AUIs depend on overt measurements of user attention, such as eye contact, which may not always accurately indicate a user's availability for notifications or interruptions. Although overt measures of user attention may tell us that a user is performing a given task, they do not necessarily indicate the covert state of mind of the user. This thesis presents an experimental study to develop a novel physiological model of interruptability for outside communications that is based upon the user's internal state which is independent of the user's current task. These models of attention are the basis of Physiologically Attentive User Interfaces (PAUIs), novel interfaces that are physiologically aware of the user's state of mind. PAUIs mediate interruptions reaching the user to automate the regulation of mental load. A number of PAUIs were developed, including the Physiological Weblog, or 'Plog, a PAUI web application which uses the novel interruptability model to automate the communication of attentive information to others.