Investigating Drivers’ Compensatory Behavior when Using a Mobile Device
The purpose of this study was to investigate driver performance and risk associated with mobile device use (MDU) from previously collected naturalistic driving data. There were two primary objectives: (1) to investigate commercial motor vehicle (CMV) driver adaptation when conversing on a cell phone; and (2) to investigate the relationship between drowsiness and the safety-critical event (SCE) risk associated with MDU.
The first goal was to investigate whether CMV and light vehicle (LV) drivers alter the way they drive when conversing on a cell phone. It was hypothesized that drivers may increase their safety margin when conversing on a mobile device by slowing down and increasing their headway to a lead vehicle, thus compensating for the increased workload.
Analysis addressing the first goal provided no indication that CMV or LV drivers increased their longitudinal safety margins when conversing on a cell phone. CMV drivers’ headway to a lead vehicle did not differ despite the fact that they significantly increased their speed by 4 km/h when conversing on a cell phone. However, CMV drivers changed lanes significantly less when conversing on a handheld cell phone. These changes suggest that CMV drivers slightly reduced the driving demands when conversing on a cell phone.
The second goal was to investigate the relationship between drowsiness and the SCE risk associated with MDU in CMV drivers. Research has shown that drivers become more alert when conversing on a mobile device (Jellentrup, Metz, & Rothe, 2011). It was thus hypothesized that CMV drivers were at a decreased risk of an SCE when conversing on a hands-free cell phone because the conversation served to stave off drowsiness.
The CMV NDS data set used in Olson et al. (2009) was analyzed to address the second goal. Drivers’ driving time and time on duty were used to assess their fatigue level, while the time of day and the amount of sleep they obtained in the previous 24 hours (measured via actigraphy) were used to indirectly assess their drowsiness level. Odds ratios computed the SCE risk for MDU subtasks across binned levels of fatigue and drowsiness. Generalized linear mixed models and chi-squared tests were used to assess changes in MDU frequency across bins. It was found that there was an increase in SCE risk for visual-manual subtasks for all bins in which analyses were possible. CMV drivers had a higher proportion of MDU from 2:00 a.m. to 3:59 a.m. (circadian low period) than for the other times of day that were analyzed. Overall, the research shows that LV and CMV drivers did not increase their longitudinal safety margins when talking on a cell phone. However, it was found that both groups of drivers looked forward more frequently when conversing on a cell phone. This study also found that CMV drivers used their cell phones more frequently at times when they would be drowsy. The increased visual attention to the road as well as the increased use during the early hours of the morning may be reasons why some studies have found that conversing on a cell phone was not associated with an increased SCE risk.