Photo of Shannon Roberts  

Shannon C. Roberts, PhD



Teenage Driving and Social Influence (Ongoing)

Teenagers' strong connection to technology combined with other unsafe behaviors contributes to the high number of teenage traffic fatalities. Given that social influence is highly prevalent during adolescence, implementing a feedback system utilizing this concept is a potentially powerful approach to curb the number of teenage traffic fatalities. In addition, sharing driving information within the teenage social network should lead to a more widespread behavior change. Therefore, the objective of this research is to examine the effect of social influence on distracted driving behavior within a teenage social network and to examine the effect of a feedback system on teenage driving behavior. This network-based perspective has far-reaching consequences for how changes in driving behavior might spread or is suppressed within a social network. An effective feedback system might encourage teenage drivers to engage in safer driving behavior and decrease the number of teenage traffic fatalities. This research has been presented at the Human Factors and Ergonomics Society Annual Meeting.

Driving Feedback and Calibration (Complete)

Many drivers believe they are better than average at handling multitasking situations. Previous studies have examined driver calibration-one's perception of performance compared to their actual performance, but these studies often use different methods and statistical tests to measure calibration. In addition to different metrics, there is also a lack of consensus on how to improve driver calibration. Performance feedback has been used in other domains and presents itself as a viable solution to combat miscalibration, but it has not been thoroughly tested in the driving domain. Therefore, the goal of this project is to apply a known set of metrics to driver calibration and to examine the effect of feedback on driver calibration. This study contributes to the development of a common protocol to assess driver calibration, which allows researchers to accurately compare results from study to study. It also highlights how feedback can be useful in some driving situations and detrimental in others. This research has been presented at the Automotive User Interface Conference.

Distraction Detection and Mitigation (Complete)

Vehicle crashes caused by driver distraction are of increasing concern, especially as ubiquitous computing introduces new devices into the car, ranging from navigation systems to social networking applications. Providing drivers with feedback might mitigate the effects of distraction on driving performance. For mitigation systems to be effective, they must accurately detect and classify instances of driver distraction. Therefore, the goal of this research is to assess the performance of algorithms that detect distraction and to examine the effect of a feedback system on driving performance and acceptance. This research identified the importance of selecting detection algorithms based on their performance in varying driving scenarios. It also provides insights into designing mitigation systems that consider user preference. This project was described in a NHTSA technical report. Results from the project were presented at the Enhanced Safety Vehicles Conference and were published in the International Journal of Human-Computer Studies.

MP3 Player Usage While Driving (Complete)

Listening to music while driving is a commonplace activity, with many drivers choosing to select music from an MP3 player. Because MP3 players are operated in vehicles, it is important to understand the demands they place on drivers. Adapters are designed to make MP3 players less distracting and more compatible with the demands of driving, but there is no testing required to verify these benefits. The goal of this experiment was to assess how different MP3 player tasks affect driving performance and to determine whether the use of an aftermarket adapter improved driving performance. Results of this study indicated that scrolling through playlists on an MP3 player has different effects on driving performance that depend on playlist length. It also uncovered the unintended consequences of using devices that claim to promote safer driving. This research was published in the Human Factors Journal and was presented during the Human Factors and Ergonomics Society Annual Meeting.

User Study of Voice Navigation System (Complete)

In-vehicle navigation systems allow drivers access to a variety of information, such as directions and points of interest. These systems also serve as a distraction from driving, as they require drivers to look away from the road and take their hands off of the wheel. As such, navigation systems have been developed that allow the user to interact with the system via voice commands, thereby eliminating the potential for visual and manual distraction. Therefore, the goal of this study was to optimize a voice navigation system for automotive use and gather user feedback on operation of the system. Results of this study provided evidence for the utility of navigation systems that rely on conversational voice commands as their primary source of input. It also highlighted the importance of considering demographic characteristics when examining user interaction. This study was the topic of my undergraduate thesis. Portions of this study have been presented at CHI-the conference for Human Factors in Computing Systems and at the Industrial Engineering Research Conference.

Copyright 2014 Shannon C. Roberts. All rights reserved.