The Effect of Cellphone Usage on Driving Performance Using an Eye Tracking System and a Driving Simulator

Abstract

In 2016 the National Highway Traffic Safety Administration (NHTSA) reported that 10% of fatal crashes, 18% of injury crashes, and 16% of all police-reported motor vehicle crashes resulted from distracted driving. Thus distraction while driving is a major risk factor for road traffic crashes in the U.S. and the State of Maryland. There are different types of distracted driving, usually categorized as those in which the source of distraction is internal (in-vehicle), such as using a mobile phone or tuning a radio, or external (out-of-vehicle) like looking at accidents, surrounding landscapes, or pedestrians. This study focuses on the different types of mobile phone distractions (hand-held, hands-free, voice commands, texting) and the effect they have on drivers’ performance while driving on different road classes, to show that the potential risk to road safety is increasing rapidly as a result of the exponential growth in the use of mobile phones in society. Different studies from different countries suggest that the proportion of drivers using mobile phones has grown over the past decade, ranging from 1% to 11%. The use of hands-free mobile phones is likely to be higher, but this figure is more difficult to ascertain. In many countries the extent of this problem remains unknown, as data on mobile phone use is not routinely collected when a crash occurs. Using a driving simulator and an eye tracking system, this study evaluates the driver’s performance (speed, steering, brake, throttle, etc.) when distracted by a cellphone in a simulated road network that includes four different road classes: urban, highway, rural, and local - school zone. Forty participants drove six scenarios sequentially with a few minutes break between scenarios. There are no cellphone distractions in the first and last scenarios to benchmark the pure effect of distraction and capture and remove the effect of learning and/or fatigue. The second to fifth scenarios have hands-free, hand-held, voice command, and texting as the distracting element, respectively. A total of over 960 simulator runs was collected and analyzed. Statistical analyses such as Kolmogorov-Smirnov test, ANOVA test, and Mann-Whitney U test were performed to find the effect of each distraction on driver performance. The first and last scenarios were specifically evaluated to examine the effect of fatigue on a driver’s performance. Since the results showed the effect of learning influences drivers’ speed, another study was conducted to examine the impact of learning on the performance of drivers. Additionally, a distraction model was also designed in this research to show the relationship between distraction and some variables.

The statistical analysis of the results indicated; impaired performance of participants due to these distractions, is affected by other driving parameters such as; speed, steering and throttle. Based on the results of this analysis, increasing the complexity of the distraction will result in decreased speed. In other words, participants decreased their speed in all scenarios, on all roads, in the presence of an external distraction. It is the author’s hope that this study’s findings will help to root out the issue of distracted driving, identify key effective factors, and ultimately identify factors associated with driving distraction to remove or mitigate this issue.