Skin Conductance Responses of Learner and Licensed Drivers During a Hazard Perception Task

Theresa J Chirles; Johnathon P Ehsani; Neale Kinnear; Karen E Seymour

doi:10.3389/fpsyg.2021.619104

Skin Conductance Responses of Learner and Licensed Drivers During a Hazard Perception Task

Front Psychol. 2021 Mar 25:12:619104. doi: 10.3389/fpsyg.2021.619104. eCollection 2021.

Authors

Theresa J Chirles¹, Johnathon P Ehsani¹, Neale Kinnear², Karen E Seymour^{3

4

5}

Affiliations

¹ Center for Injury Research and Prevention, Bloomberg School of Public Health, Health Policy and Management Department, Johns Hopkins University, Baltimore, MD, United States.
² Transportation Research Laboratory, Wokingham, United Kingdom.
³ School of Medicine, Department of Psychiatry, Johns Hopkins University, Baltimore, MD, United States.
⁴ Bloomberg School of Public Health, Department of Mental Health, Johns Hopkins University, Baltimore, MD, United States.
⁵ National Institutes of Health, Center for Scientific Review, Bethesda, MD, United States.

Abstract

Background: While advanced driver assistance technologies have the potential to increase safety, there is concern that driver inattention resulting from overreliance on these features may result in crashes. Driver monitoring technologies to assess a driver's state may be one solution. The purpose of this study was to replicate and extend the research on physiological responses to common driving hazards and examine how these may differ based on driving experience. Methods: Learner and Licensed drivers viewed a Driving Hazard Perception Task while electrodermal activity (EDA) was measured. The task presented 30 Event (hazard develops) and 30 Non-Event (routine driving) videos. A skin conductance response (SCR) score was calculated for each participant based on the percentage of videos that elicited an SCR. Results: Analysis of the SCR score during Event videos revealed a medium effect (d = 0.61) of group differences, whereby Licensed drivers were more likely to have an SCR than Learner drivers. Interaction effects revealed Licensed drivers were more likely to have an SCR earlier in the Event videos compared to the end, and the Learner drivers were more likely to have an SCR earlier in the Non-Event videos compared to the end. Conclusion: Our results support the viability of using SCR during driving videos as a marker of hazard anticipation differing based on experience. The interaction effects may illustrate situational awareness in licensed drivers and deficiencies in sustained vigilance among learner drivers. The findings demand further examination if physiological measures are to be validated as a tool to inform driver potential performance in an increasingly automated driving environment.

Keywords: autonomous vehicles; driving experience; electrodermal activity; hazard perception; young drivers.