Predicting time-sharing in mobile interaction

Abstract

The era of modern personal and ubiquitous computers is beset with the problem of fragmentation of the user's time between multiple tasks. Several adaptations have been envisioned that would support the performance of the user in the dynamically changing contexts in which interactions with mobile devices take place. This paper assesses the feasibility of sensor-based prediction of time-sharing, operationalized in terms of the number of glances, the duration of the longest glance, and the total and average durations of the glances to the interaction task. The data used for constructing and validating the predictive models was acquired from a field study (N = 28), in which subjects performing mobile browsing tasks were observed for approximately 1 h in a variety of environments and situations. The predictive accuracy achieved in binary classification tasks was about 70% (about 20% above default), and the most informative sensors were related to the environment and interactions with the mobile device. Implications to the feasibility of different kinds of adaptations are discussed.

Forecasting user attention during everyday mobile interactions using device-integrated and wearable sensors

“…Oulasvirta et al investigated how different environments affected attention while users waited for a web page to load on a mobile phone [32]. In a follow-up work, the same authors used a Wizard-of-Oz paradigm with simulated sensors to assess the feasibility of predicting time-sharing of attention, including prediction of the number of glances, the duration of the longest glance, and the total and average durations of the glances to the mobile phone [27].…”

“…Prior work mainly focused on estimating the point of gaze on the device screen using the integrated front-facing camera [15,52] or on using inertial sensors or application usage logs [7,10] to predict user engagement [26,47] or boredom [35]. In contrast, allocation of user attention across the device and environment has rarely been studied, and only using simulated sensors [27]. Most importantly, existing attentive user interfaces are only capable to adapt after the fact, i.e.…”

Layered evaluation of interactive adaptive systems: framework and formative methods

“…For example, Maulsby et al (1993) used wizard-of-Oz to prototype an intelligent agent. Several recent UMUAI papers report on wizard-of-Oz studies (Miettinen and Oulasvirta 2007;Batliner et al 2008;Damiano et al 2008;Conati and MacLaren 2009). For example, Miettinen and Oulasvirta (2007) used wizard-of-Oz to simulate the system functionality that corresponds to the CID/ID layers: sensors were simulated by human codings of data.…”

“…Several recent UMUAI papers report on wizard-of-Oz studies (Miettinen and Oulasvirta 2007;Batliner et al 2008;Damiano et al 2008;Conati and MacLaren 2009). For example, Miettinen and Oulasvirta (2007) used wizard-of-Oz to simulate the system functionality that corresponds to the CID/ID layers: sensors were simulated by human codings of data. In a layered evaluation, wizard-of-Oz can also be useful to simulate layers preceding the one being evaluated, to ensure these work perfectly and to enable the evaluation of a layer in isolation.…”

Cognitive resource-aware unobtrusive service provisioning in ambient intelligence environments

“…The side effects of this competition on the user's performance are termed as cognitive resource depletion (CRD) in the area of cognitive psychology [63]. This leads to distractions; increases errors, stress, and frustration; and reduces the ability to perform mental planning, problem solving, and decision making [20,42,45]. In particular, recent studies have shown how annoying technostress is to users [38,39,48].…”