Exploring the relationship between the driving behavior questionnair and hiway driving behavior

The Driver Behavior Questionnair (DBQ) is a well documented instument for obtaining selfreport information on aberrant driving behaviors. The DBQ is comprised of three subscales: (a) errors – the failure of planned actions, which could result in unsafe driving; (b) violations – deliberate contravention of behaviors which are considered necessary for safe operation, and: (c) lapses – absentminded behaviors which are assumed unlikely to impact driving safety. A substantive body of research has demonstrated a relationship between DBQ scores and both retrospective and prospective accident involvement. In addition, DBQ subscale scores have been shown to have relationships with drivers’ attitudes, personality characteristics, psychological wellbeing, demographics, etc. However, to the best of our knowledge, there is little or no published information on the relationship between DBQ scores and driving performance variables under actual driving conditions that may bear some relationship to accident risk. The present study focuses on the relationship between DBQ subscales and highway driving behaviors. A sample of 108 drivers in selfreported good health and having a safe recent driving history (no accidents in the previous year) was balanced by gender and across three age groups (2029, 4049, 6069). Prior to driving, participants completed a 24item U.S. version of the DBQ. After approximately 30 minutes of driving an instrumented Volvo XC90, driving behaviors were assessed over an 18 minute period. During this time, participants were engaged in both periods of single task driving and driving while engaged in a structured working memory task requiring a division of attention. A dischotomous breakdoown of subscales of the DBQ (above or below the median) were independently examinded as predictors of driving behavior (average velocity, standard deviation of velovity, standard deviation of steering wheel position, hard braking, rapid throttle acceleration, and sudden unidirectional acceleration (the vector sum of longitudinal and lateral acceleration)). Signifi cant relationships were found to exist between subscales of the DBQ and actual driving behaviors. Drivers with high violations scores drove faster, had poorer lateral control (higher standard deviation of wheel position) and more sudden unidirectional accelerations. Among these factors, higher driving speed is known to infl uence the probability of accidents. The relationship between lateral control and sudden acceleration to accidents is less established but appears reasonable. High lapses scores were related to less consistent gas pedal control (larger standard deviation of velovity and more frequent periods of rapid throttle acceleration). It is interesting to speculate as to whether there is an attentinal factor that links the lapses score and these consistency of control measures. There were no main effects observed between errors scores and any of the driving behavior measures. Signifi cant interactions in errors×gender on rapid Short reports, Traffic accident and injury, FORENSIC SCI SEM, 1 (2013) 74-78 throttle acceleration and violations×age on hard braking appeared. Male drivers with high errors scores speed up more abruptly than those with low scores while female drivers did not. High violations drivers in their 60s more frequently engaged in periods of hard braking than those with low violations scores. The relationships observed here are likely conservative since drivers having had accidents in the past year were excluded.