Individual Differences in Peripheral Physiology and Implications for the Real-Time Assessment of Driver State

The workload associated with driving involves both the physical activities of the driving task as well as the cognitive thought processes necessary to manage the physical elements. Arguments can be made that drivers are more aware of the physical demands than of the cognitive ones. While this is an issue for all drivers, with advancing age, the ability to manage all aspects of workload declines and is of particular concern. Increasing amounts of information technology coupled with efforts for advanced vehicle-to-vehicle and vehicle-to-infrastructure communications make active driver workload management an area of increasing importance. When assessing cognitive demands, studies often rely on self-report measures such as the NASA TLX to classify the difficulty of a task. While being useful research tools, self-report methods have limited applicability to real-time state detection. Both performance based and physiological measures provide relatively continuous, noninvasive methods of characterizing cognitive workload (Brookhuis & De Waard, 2001). To develop a more comprehensive system of assessing driver workload, physiological measures have been proposed as a complementary measure to driving performance-based measures (Wu & Liu, 2007). Theory suggests that physiological measures are likely to be more sensitive to the initial changes in workload than performance based measures (Brookhuis & De Waard, 1993; Lenneman, Shelley, & Backs, 2005; Wilson, 2002) and some data has been developed to support this position (Mehler, Reimer, Coughlin, & Dusek, 2009). In this project, we intend to explore further the utility of using driving performance, visual attention and other physiological measures to discriminate subtle changes in driver cognitive workload. Further, we intend to examine how the demographic characteristics such as age and gender, along with individual variability, impact the sensitivity of physiologically based detection systems. This project will conclude with the development of initial detection algorithms for driver state based upon physiological and driving performance measures.