Analyzing Distraction while Driving using Driving Simulation, Physiological Sensors, and Eye Tracking

Road safety remains an issue of global relevance as road crashes are believed to be a leading
cause of death, particularly for children and young adults. In this talk, I will describe a set of driving
behavior studies we have conducted at the American University of Beirut, with a focus on studying driving
distraction which is an important source of driving error.
In one study, we designed an experimental protocol that induces auditory-vocal mental workload. Using
driving simulator measures of driving performance and physiological measures of cognitive workload and
stress such as heart rate, we modeled the dynamic impact of stress on driving performance (such as speed,
reaction time, and decision to violate a red light). Such models can be used in traffic microsimulators to
enhance the fidelity of driving behavior models used in these simulators, as well as within vehicular
technologies to warn drivers when their performance significantly deteriorates. In subsequent studies, we
investigated the impact of browsing social media and voice messaging while driving on performance and
attention using driving simulation and eye tracking, and showed that both could be as detrimental to
attention as texting while driving. I will discuss the policy and technology implications of findings from these
studies for enhancing road safety. I will conclude by describing an ongoing study on human-machine
interaction in the context of conditionally automated driving that aims at increasing trust between the
human driver and the automated car.