Joshua B. Hurwitz

Using Driver Performance Measures to Estimate Workload

This study evaluated the impact of the modality of distractor tasks and the difficulty of the driving task on driver control over a vehicle. Driving is a task with a close link between visual information processing and driver performance variables such as lane keeping. Consequently, compared to auditory distractors, visual distractors should have a larger negative effect on vehicle control, but mainly when the driving task also adds significant workload for the driver. In this study, drivers drove around either a curvy or a straight track while occasionally performing either an auditory or a visual secondary monitoring task. Performance of the secondary task was associated with deterioration in steering wheel control and lane keeping, but only on the curvy track and mainly when the secondary task required visual monitoring. These results are discussed in terms of their implication for developing workload managers in in-vehicle driver support systems.

Driver Choice of Headway with Auditory Warnings

Two major causes of rear-end collisions are driver distraction and the tendency for some drivers to maintain insufficient headways with other vehicles. Previous research into in-vehicle collision avoidance systems has focused on warning drivers in near-collision situations with short headways (e.g., less than 1 second). The current study focused on mid-range headways and on warning distracted drivers to maintain safe following distances. Subjects performed a simulated car-following task in which the lead vehicle periodically decelerated. The lead vehicle was otherwise programmed to maintain a minimum following distance of 2 seconds—although the subjects could drive more slowly to leave a longer headway. On trips with an auditory headway warning, the subjects tended to leave larger headways, particularly on trips in which they performed a secondary task as well.

User Choice, Privacy Sensitivity, and Acceptance of Personal Information Collection

One issue in mobile privacy is that most mobile applications do not provide accessible disclosures and opportunities for meaningful consent. While such capabilities could increase user satisfaction with applications’ privacy protections, they might also produce greater rejection of data collection. The current study investigated these issues using a mobile application that provided simplified disclosures and some user choice in data collection. Results showed that having such choice tended to increase satisfaction with privacy-relevant aspects of the application. Furthermore, when asked to rate the application if it collected various types of information, subjects who reported taking more privacy risks and trusting authorities to protect their privacy tended to give higher ratings when the information collected was both highly relevant and sensitive, or of low relevance and sensitivity. The results suggest that companies might garner greater acceptance of data collection if they understand consumers’ privacy sensitivities and reactions to data collection requests.

The Relationship between User Location History and Interests in Products and Services

This study evaluated the use of location history as a predictor of user interests in products and services. Over a 1-month time period, subjects used a voicemail or email diary to report their visits to various establishments, such as shops and restaurants. At the end of the study, they completed questionnaires asking about their demographic characteristics, as well as their use of advanced mobile services and involvement in making decisions about the purchase or use of various products and services. A series of stepwise linear regressions showed that parameters derived from the diary data, when combined with demographic and mobile usage parameters, significantly improved predictions of product/service involvement, when compared to using the demographic and mobile usage predictors alone. These results suggest that location history measures could potentially be valuable components of algorithms for targeting commercial content to end users.

Acceptance of Advertising and Collection of Personal Information

The current study investigated the relationship between acceptance of advertising and of services that collect personal information. Survey respondents were presented with a questionnaire asking about their agreement with various forms of advertising, and were also presented with scenarios describing mobile and TV-based services. The results showed that adding collection of personal information to the services had less of an impact on the ratings of respondents who were more accepting of advertisements inserted into applications (software, games and MP3s) or presented on mobile devices. These results suggest that acceptance of advertising could be an indicator of privacy sensitivity, and could be used to personalize privacy management in part based on individual differences in sensitivity.