Hazelton, T. H., "Investigating, designing, and validating a haptic-affect interaction loop using three experimental methods,", M.Sc. Thesis, University of British Columbia, 2010.
Computer interfaces commonly make large demands on our visual and auditory attention, which can make multi-tasking with multiple systems difficult. In cases where a primary task demands constant, unbroken attention from the user, it is often implausible for such a user to employ a system for a secondary task, even when desirable. The haptic modality has been suggested as a conduit for the appropriately-intrusive delivery of information from computer systems. Furthermore, physiological signals can be used to infer the affective state of a user without requiring attention. Combining these underexplored channels for implicit system command, control and display, we envision an automated, intelligent and emotionally aware interaction paradigm. We call this paradigm the Haptic-Affect Loop (HALO). This work investigates the potential for the HALO paradigm in a specific use case (portable audio consumption). It uses three experimental techniques to gather requirements for the paradigm, validate its technological feasibility, and develop the feedback-supported language of interaction with a HALO-enabled portable audio system. A focus group is first conducted to identify the perceived utility of the paradigm with a diverse – albeit technologically conservative – group of portable audio users, and to narrow its scope. Results of this focus group indicate that participants are sceptical of its technological feasibility (in particular, context resolution) and are unwilling to relinquish control over their players. This scepticism was alleviated somewhat by the conclusion of the sessions. Next, technological validation of online affect classification is undertaken via an exploratory, but formally controlled, experiment. Galvanic skin response measures provided a means to make introductory measures of interruption and, in some cases, musical engagement. A richer signal array is necessary to make the full array of required affect identifications for this paradigm, and is under development. The final phase of work involves an iterative participatory design process with a single participant who was enthusiastic but practical about technology to better define system requirements and to evaluate input and output mechanisms using a variety of devices and signals. The outcome of this design effort was a functioning prototype, a set of initial system requirements and an exemplar interaction language for HALO.