Skip Internal Navigation
HBB Research - Young Drivers
Teen Driving Risk Studies
This series of observational studies are designed to improve understanding of the nature of teen driving risk.
The 40-Teen Naturalistic Teenage Driving Study (NTDS)
The purpose of NTDS is to evaluate the effects of experience on teen driving performance under various driving conditions. Researchers recruited 40 novice teen drivers at the time of licensure and equipped the teens' vehicles with cameras, g-force meters, Global Positioning System (GPS), and other equipment to provide detailed information about driving performance during the first 18 months of driving experience. Of particular interest is driving performance in the presence or absence of teen passengers and at night versus during the day. Twenty of these teens share a vehicle with a parent, and these parents have agreed to be in the study to provide a comparison group.
Supervised Practice Driving Study
This study seeks to examine the nature of parent-supervised practice driving, and determine the extent to which the amount of practice is associated with driving performance after licensure. A sample of 100 families was recruited and will be observed and surveyed during a period of 18 months.
Experimental Research on the Influence of Peer Passengers on Simulated Risky Driving Among Teenage Males
The objective of this study is to understand the effects and mechanisms of the influence of peer passengers on male teen drivers, specifically the effects of peer personality stereotypes on the risk taking behavior of male teenage drivers. A series of experimental studies will be conducted in a driving simulator with various manipulations of passenger type and presence. Driving simulator and eye tracking data are collected to analyze driving behavior, risk taking behavior, and attention/distraction related variables.
The Effect of Teenage Passengers on Teenage Simulated Driving Performance (Teen Passenger Study)
The presence of teenage passengers has been shown to increase crash risk. Notably, Ouimet et al. (2010) reported that male teenage passengers increased fatal crash risk not only among teenage but also among young adult drivers, particularly male drivers. In previous research we observed vehicles exiting high school parking lots and found that teenage drivers with male teenage passengers drove faster and closer to the lead vehicle than other drivers (Simons-Morton, Lerner, Singer, 2005). However, in the NTDS we found that teen passengers (including males and females) provided a slightly protective effect on crash/near crash and risky driving compared to the no passenger condition. A series of simulation studies is being conducted to learn more about the nature of teen passenger influences in collaboration with the University of Michigan Transportation Research Institute (UMTRI; Ray Bingham, PI). One study will be completed each year over a 4-5 year period, incorporating what is learned from each study into the next study.
The Teen Passenger Study 1 (TPS1), completed in the spring of 2012, was designed to ascertain the effect of a risk-accepting or risk-averse teenage passenger on teenage risky driving. We recruited 66 newly licensed male teenage drivers and randomized them to risk-accepting or risk-averse passenger conditions. The passenger was a trained, male confederate. We were interested in the effect of social norms on driving behavior, so we employed a pre-drive priming task in which the participant and confederate passenger watched a video of risky driving and the confederate passenger verbalized that he would or would not, depending on the role he was playing, ever ride with that driver. We used a randomized block design with 2 conditions (passenger: risk-accepting vs. risk-averse) X 2 drive orders (driving alone first vs. driving with the passenger first). T-test comparisons of difference scores (passenger minus alone) were in the expected direction favoring greater driving risk in the risk-accepting passenger group. We concluded that teenage drivers exposed to a risk-accepting teenage passenger were less likely to stop at red lights (p=0.04) while driving in a simulator and this risky behavior was greater in the presence of a risk-accepting than a risk-averse peer passenger (Simons-Morton et al., Health Psychology, in press).
In other analyses of TPS1 neuroimaging data we found participants who were sensitive to social exclusion, measured by the Cyberball task, in which confederate peers play cyber catch with the participant while he is being imaged. Gradually, the confederates exclude the participant and the imaging indicates painful exclusion. We found that participants who were sensitive to social exclusion according to neuro-imaging data were also sensitive to passenger presence when driving the simulator a week later (Falk et al., Journal of Adolescent Health, in press).
The TPS2 tested the effect of teenage peer pressure on teenage risky driving performance. The study design is similar to TPS1, except we put the drivers under pressure by instructing them to reach a particular destination within a limited time without error. The confederate passenger served as the navigator and at key points in the drive verbally encourages the driver to hurry (in the role of a risk-accepting teen) or make no errors (in the role of a risk-averse teen). Assessment of fMRI and psycho-social tasks will also be conducted. Preliminary analyses indicated that the study participant drove in a more risky manner in the presence of a peer exerting mild pressure to engage in risk compared with in the presence of a confederate passenger who exerted mild pressure not to take risk.
TPS3, which is underway, evaluates the effect of pre-drive mood on risky simulated driving in the presence of a peer passenger. Participants are randomized to play a mood enhancing guitar game with the confederate passenger prior to driving in the simulator, or to sit with a confederate peer listening to quiet music.
The Uniform Naturalistic Driving Study (UNDS)
One of the limitations of naturalistic research to date has been small sample size. Larger samples are needed for analyses of risk by driving conditions and among subgroups. Toward this end the UNDS will obtain data from the SHRP2 Naturalistic Driving Study, which used the same instrumentation as the Naturalistic Teenage Driving Study and Supervised Practice Driving Study. SHRP2 obtained driving data from over 200 novices and another 600 adult drivers, which we plan to combine with the data from the NICHD Naturalistic Teenage Driving Study (n-42) and Supervised Practice Driving Study (N=90). The large combined data set will allow subgroup analyses and will allow us to answer key questions such as: 1) What are individual level predictors of risky driving? 2) Does crash risk and risky driving vary according to driving conditions? 3) What is the effect on driving outcomes of the type of passengers and driving context? 4) What is the relationship between risky driving behavior and crash risk? 5) To what extent does a small proportion of high-risk drivers account for the overall high crash risk of young drivers? The study is in the planning stages, developing the protocol and obtaining institutional approvals. SHRP2 data will be available starting in April and data from SHRP2 and SPD will be ready for analyses by December, 2014.
NEXT Naturalistic Driving Study (NEXT NDS)
Little is known about how driving behavior varies over time, particularly among young drivers. Naturalistic driving methods lend themselves to longitudinal assessment, but to date most studies have included few study participants and have been of short duration. This study assesses the driving performance of a sample of 150 young drivers starting in the 12th grade (ages 17-18) and ending when the participants are ages 21-22. Assessment is done using the DriveCam driving assessment device. The sample is drawn from the NEXT Study, and will also have completed 7-years of annual assessments on their health behaviors. The research questions of interest include: 1) What is the variability within the sample and over time in driving performance (elevated g-force events and crash/near crash)? And 2) What individual and driving condition factors are associated with risky driving and crashes/near crashes? Data collection is ongoing and study participants will be followed for a period of 4 years, concluding in 2016.
Bruce Simons-Morton, Ed.D., M.P.H.
- Simons-Morton, BG, Zhang, Z, Jackson, JC, Albert, PS. Do elevated gravitational-force events while driving predict crashes and near crashes? Am J Epidemiol 175(10): 1075-1079, 2012. doi:1093/aje/kwr440, PMID:22271924
- Zhang, Z, Albert, PS, Simons-Morton, BG. Marginal analysis of longitudinal count data in long sequences: Methods and application to a driving study. Annals of Applied Statistics 6(1): 27–54, 2012. doi: 10.1214/11-AOAS507
- Jackson, J, Albert, PS, Zhang, Z, and Simons-Morton, B. Ordinal latent variable models and their application in the study of newly licensed teenage drivers. Journal of the Royal Statistical Society-Series C, Applied Statistics 62(3): 435-450, 2012. doi: 10.1111/j.1467-9876.2012.01065.x
- Simons-Morton, BG, Ouimet, MC, Wang, J, Chen, R, Klauer, SG, Lee, SE, Dingus, T. Peer influence predicts speeding prevalence among teenage drivers. J Safety Res 43(5-6): 397-403, 2012. doi: 10.1016/j.jsr.2012.10.002 PMID:23206513
- Simons-Morton, BG, Cheon, K, Guo, F, & Albert, P. Trajectories of kinematic risky driving among novice teenagers. Accident Analysis & Prevention 51C: 27-32, 2012. doi: 10.1016/j.aap.2012.10.011 PMID:23182780.
- Zakrajsek, JS, Shope, JT, Greenspan, AI, Wang, J, Bingham, CR, Simons-Morton, BG. Effectiveness of a brief parent-directed teen driver safety intervention (Checkpoints) delivered by driver education instructors. Journal of Adolescent Health 53: 27-33, 2012. doi: 10.1016/j.jadohealth.2012.12.010 PMID:23481298
- Simons-Morton, BG, Bingham, R, Ouimet, MC, Pradhan, A, Chen, R, Barretto, A, Shope, J. (2013). The effect on teenage risky driving of feedback from a safety monitoring system: A randomized controlled trial. Journal of Adolescent Health 53: 21-26, 2013. doi: 10.1016/j.jadohealth.2012.11.008 PMID:23375825