Distraction-induced driving error: An on-road examination of the errors made by distracted and undistracted drivers
Highlights
► Driver distraction and driving error are both serious road safety issues. ► This study explores the nature of distraction-induced driving errors. ► Drivers were significantly more likely to make errors when distracted. ► The nature of distraction-induced error was similar to errors made when undistracted.
Introduction
Over the past two decades the concept of driver distraction has been the focus of intense research attention. There exists a large and expanding body of research that has documented the myriad ways in which distraction can impact on driving performance and safety. These include reduced longitudinal (Rakauskas et al., 2004, Strayer and Drews, 2004) and lateral control (Engström and Johansson, 2005, Reed and Green, 1999); reduced situation awareness (Kass et al., 2007, Owens et al., 2011); and impaired response times to roadway hazards (Lee et al., 2001, Liang and Lee, 2010). Moreover, these degradations have also been shown to translate into an increased risk of crash involvement, with estimates indicating that secondary task distraction is a contributing factor in up to 23% of crashes and near-crashes (Klauer et al., 2006).
Distraction is a complex, multifaceted phenomenon and, despite the immense research effort devoted to studying the concept over the past two decades, there is still much to understand about its mechanisms and its relationship with other aspects of human cognition and behaviour (Regan et al., 2011). One aspect of distraction for which there is limited evidence is its relationship with driving errors (Young and Salmon, 2012). Although the two phenomena are both popular areas of road safety research and are ostensibly related, there has been little exploration of how they are related and how they may interact. That is, there has been little systematic exploration of the causal role of distraction in driving errors. A critical knowledge gap in this context relates to the link between distraction and the driving errors prevalent in road traffic crashes. Although various detailed driving error taxonomies exist (e.g. Stanton and Salmon, 2009), the role that driver distraction plays in the range of driving errors documented remains unclear.
From a theoretical perspective, understanding the relationship between driver distraction and driving errors and the mechanisms by which distraction contributes to different errors is important for improving knowledge of the consequences of distracted driving and the mechanisms through which it leads to crashes. From a more practical perspective, understanding the role of distraction in driving errors can inform the development of better countermeasures to mitigate distraction induced error through training and technology (e.g., advanced driver assistance systems) and roadway design (e.g., error/distraction tolerant design such as use of tactile lane markings).
Driving error has long been a focus of road safety research and a range of methods have been developed to specifically measure this concept, including the Driver Behaviour Questionnaire (DBQ) (Reason, 1990) and the Wiener Fahrprobe method (Risser and Brandstätter, 1985). Estimates suggest that driving error is a causal factor in 75% (Hankey et al., 1999), and even up to 95% (Rumar, 1990) of road crashes and, thus, is a significant contributor to road trauma. It makes sense that being distracted can lead to driving errors. Indeed, as discussed earlier, there is a large body of research demonstrating how distraction can impact a range of driving performance measures and lead to a range of driving errors. However, the link between distraction and different error types is often not clear, particularly regarding whether distraction is viewed as a driving error in itself or one of a number of causal factors that leads to errors. A number of taxonomies will list distraction as the driving error, rather than as a causal factor (e.g., Reason et al., 1990, Sabey and Staughton, 1975, Sabey and Taylor, 1980). These taxonomies do not consider what the error type following the distraction episode is; thus, offering little insight into the nature of the errors associated with distracted driving. According to these taxonomies, if a distracted driver was unable to stop at a red traffic signal, their failure to see the traffic signal following the distraction would not be captured; rather, the distraction itself would be listed as the error. Other taxonomies list distraction as a casual factor in driver errors, but do not indicate the mechanisms by which it contributes (e.g., Staubach, 2009, Wierwille et al., 2002). Wierwille et al., for example, list internal and external distraction as one of the factors contributing to recognition errors, but do not indicate how distraction contributes to these errors.
Two recent studies that have examined the nature of errors made by drivers found evidence that distraction is one of a number of factors that contribute to drivers committing errors (Sandin, 2009, Staubach, 2009). In an in-depth examination of 474 crashes, Staubach (2009) found that a significant number of crossroads, lane departure and same direction crashes were the result of errors caused by the driver being distracted. Likewise, Sandin (2009) sought to identify the factors underlying the most common errors and violations occurring at intersections (i.e., a failure to yield, or running a traffic light or sign). Sandin found that distraction contributed to a range of the errors occurring at intersections including missing a sign or traffic signal, misjudging the timing of amber lights, and a failure to see other vehicles.
While these, and other studies (e.g., Klauer et al., 2006), indicate that distraction has a role to play in driving errors, they, along with the error taxonomies, provide little insight into the nature (number and type) of errors made by distracted drivers and how these might differ from errors made by the undistracted driver, and the mechanisms by which distraction causes errors or disrupts error recovery. This is due, in large part, to the methods typically used to examine driving errors, which have relied on the use of retrospective crash data analysis (Curry et al., 2011, Sandin, 2009) and self-report measures such as the DBQ (Özkan et al., 2006, Reason, 1990). These methods have a range of limitations that constrain our understanding of driver error; including self-report measures being both retrospective and subjective and thus open to recall and desirability bias, and retrospective accident analysis lacking enough detailed data to accurately identify and classify error types and often failing to consider the role of wider system factors in driving error causation (Salmon et al., 2010a). In terms of distraction-induced error, one of the key limitations of the traditional approaches is that they cannot accurately or objectively determine if a driver was distracted or not. Investigating distraction-induced error therefore requires a more objective approach that is capable of collecting data on driving errors made under carefully controlled distraction conditions. Recent advances in driver behaviour measurements methods, namely the advent of on-road instrumented vehicles, has made the objective, real-time recording of distracted driving errors and their underlying mechanisms possible.
This study aimed to explore, using a suite of human factors methods in an on-road context, the nature of errors made by drivers distracted by a Visual Detection Task (VDT) and how these errors may differ, in both number and kind, to those made when drivers are not distracted. Although the study was designed as a proof-of-concept study and, therefore, largely exploratory in nature, a number of predictions can be made about the nature of the errors that may be made by distracted drivers based on the existing distraction literature and theory. First, it was predicted that drivers would make a greater number of driving errors when distracted. Also, given the visual-manual nature of the distracter task used and previous evidence suggesting that visual distraction impairs drivers’ lane keeping behaviour and hazard detection (e.g., Engstrom et al., 2005, Greenberg et al., 2003), it was predicted that drivers would make a greater number of observation-based and lane keeping errors when distracted.
Section snippets
Participants
Twenty-three drivers (10 males, 13 females) aged 19–51 years (mean = 28.9, SD = 8.6) took part in the study. Seventeen of the participants held a valid Full driver's license while the remaining six held a valid Probationary (P2) license. No significant difference was found in the number of errors made across the Full and Probationary licence holders, thus licence status was not included as a covariate in the analysis. Participants had an average of 10.1 years (SD = 8.9) driving experience and drive
Number and nature of observed errors
Drivers made a total of 268 errors when distracted and 182 errors when driving undistracted. All drivers committed driving errors on each drive, with the average number of errors made per driver higher when distracted (11.7) compared to when not distracted (7.9). Results of the GEE model indicated that drivers were 48% more likely to make an error when distracted compared to when not distracted (Exp(B) = 1.478, p < .001). The driving errors observed were classified into 18 specific error types. A
Discussion
This study examined the nature of errors made by distracted drivers and how these differ, in both number and kind, to those made when drivers are not distracted. A multi-method approach was used to obtain comprehensive data on the nature of distraction-induced errors and the potential mechanisms underlying them.
As predicted, drivers made a significantly greater number of driving errors when they were distracted by the visual detection task. It is important to note, however, that even when not
Conclusion
The current study has provided unique insight into the nature of errors made by distracted drivers under real-world driving conditions. It was found that driving errors are common even under undistracted conditions, but are significantly more pronounced when drivers are distracted. It was also revealed that the profile of errors made by distracted and undistracted drivers was very similar, suggesting that, at least for drivers distracted by a low demand visual-manual task, the errors made
Acknowledgements
This study was funded by the Monash University Researcher Accelerator Program. Dr Salmon's contribution to this article was part funded by his Australian National Health and Medical Research Council Post Doctoral Fellowship. We thank Johan Engström for his advice on the Visual Detection Task, Nebojsa Tomasevic and Ashley Verdoorn for their assistance with the instrumented vehicle and its data, and Stuart Newstead for his statistical assistance.
References (38)
- et al.
Prevalence of teen driver errors leading to serious motor vehicle crashes
Accident Analysis & Prevention
(2011) Reconstructing human contributions to accidents: the new view on error and performance
Journal of Safety Research
(2002)- et al.
Effects of visual and cognitive load in real and simulated motorway driving
Transportation Research Part F: Traffic Psychology and Behaviour
(2005) - et al.
Effects of visual and cognitive load in real and simulated motorway driving
Transportation Research Part F: Traffic Psychology and Behaviour
(2005) - et al.
Reliability of drivers in urban intersections
Accident Analysis & Prevention
(2010) - et al.
Driver distraction: the effects of concurrent in-vehicle tasks, road environment complexity and age on driving performance
Accident Analysis & Prevention
(2006) - et al.
Effects of distraction and experience on situation awareness and simulated driving
Transportation Research Part F: Traffic Psychology and Behaviour
(2007) - et al.
Combining cognitive and visual distraction: less than the sum of its parts
Accident Analysis & Prevention
(2010) - et al.
Driver performance while text messaging using handheld and in-vehicle systems
Accident Analysis & Prevention
(2011) - et al.
Driver behaviour questionnaire: a follow-up study
Accident Analysis & Prevention
(2006)
Effects of naturalistic cell phone conversations on driving performance
Journal of Safety Research
Driver distraction and driver inattention: definition, relationship and taxonomy
Accident Analysis & Prevention
Managing error on the open road: the contribution of human error models and methods
Safety Science
An analysis of common patterns in aggregated causation charts from intersection crashes
Accident Analysis & Prevention
Human error taxonomies applied to driving: a generic driver error taxonomy and its implications for intelligent transport systems
Safety Science
Factors correlated with traffic accidents as a basis for evaluating Advanced Driver Assistance Systems
Accident Analysis & Prevention
Examining the relationship between driver distraction and driving errors: a discussion of theory, studies and methods
Safety Science
Advances in Industrial Ergonomics and Safety I
Intrusiveness of a visual detection task on secondary and driving task performance
Cited by (61)
Analysis of mobile phone use engagement during naturalistic driving through explainable imbalanced machine learning
2023, Accident Analysis and PreventionHow is the visual working memory load of driver influenced by information density of traffic signs?
2022, Transportation Research Part F: Traffic Psychology and BehaviourThinking aloud on the road: Thematic differences in the experiences of drivers, cyclists, and motorcyclists
2021, Transportation Research Part F: Traffic Psychology and BehaviourUsing naturalistic driving data to examine how drivers share attention when engaging in secondary tasks
2020, Safety ScienceCitation Excerpt :Secondary tasks are of concern for road safety as they can distract drivers from the primary task of driving at times when driving demands undivided attention. Previous research has consistently found that engaging in secondary tasks, particularly with in-vehicle and portable electronic devices, can impair driving performance and increase crash risk (Dingus et al., 2016); (Engstrom et al., 2005); (Owens et al., 2011); (Young et al., 2013). Using in-depth crash data, distracted driving was found to be a major contributing factor in almost 16 percent of serious casualty road crashes resulting in hospital attendance in Australia (Beanland et al., 2013).