EFFECTS OF READING TEXT WHILE DRIVING ANALYSIS OF 200 HONOLULU TAXI DRIVERS ON A VS500M SIMULATOR

Abstract

Although 47 US states make the use of a mobile phone while driving illegal, many people still use their phone for texting and other tasks while driving. The first purpose of this research is to summarize the large literature on distracted driving and compare major outcomes with those of our study. In this study, I have focused only on distraction due to reading text because this activity is most common. For this research project, we collected simulator observations and survey of 203 professional taxi drivers (175 male, and 28 female) working at the same Honolulu taxi company, using the mid-range driving simulator VS500M by Virage. After a familiarization period, drivers were asked to read realistic text content relating to passenger pick up displayed on a 7-inch tablet affixed to the dashboard. The large sample size (N=203) of our study provides a strong statistical sample base for driving distraction investigation on a driving simulator compared to all but one of the previous studies. The comparison between regular and text-reading conditions revealed that the drivers significantly increased their headway (20.7%), lane deviations (353.9%), total time of driving blind (351.8%), maximum duration of driving blind (87.6% per glance), driving blind incidents (169.7%), driving blind distance (337.5%) and significantly decreased lane change frequency (35.1%). There was no significant effect on braking aggressiveness while reading the text. The outcomes indicate that driving performance degrades significantly by reading text while driving. Texting is a distraction to driving a vehicle. The second purpose of this study is to provide additional insights on the association of demographic characteristics (age, gender, race, education, and driving experience) with driving performance while texting. The dependent variables generated by the simulator were: Average following interval (headway), line encroachment-incidents, lane change frequency, hard braking, and total time, maximum time, incidents, and travel distance of driving distracted. Correlation, analysis of variance and regression analyses were conducted. We considered three conditions: Control (No Texting), Texting and Change due to Texting. Our sample includes a large number of Asian and Asian-American drivers including Chinese, Japanese, Korean, Filipino and Vietnamese drivers, so ethnicity was modeled with dummy variables. Gender and some ethnicity variables have significant associations with driving performance outcomes, but driver age has the most significant associations with worsened driving performance under texting conditions. Drivers with higher levels of education seem to have a driving performance that is less affected by texting conditions. There are indications that drivers of Korean dissent may be somewhat more aggressive (shorter headways) and drivers of Vietnamese dissent may be somewhat more distracted than average (longer distance driving blind.)