Transportation Research Part F: Traffic Psychology and Behaviour
The development of hazard perception: Dissociation of visual orientation and hazard processing
Introduction
Eye movements are indicators for the selection of to-be-processed objects and thus an essential prerequisite for visual information processing. Effective ways of eye guidance are crucial for driving a vehicle and allow to detect and process potentially dangerous objects to avoid accidents. Especially young drivers are reported to exhibit a significant risk of crash involvement (Gregersen & Bjurulf, 1996). Furthermore, studies concerning the causes of accidents revealed that insufficient visual orientation is a major factor in the prediction of accidents (Dingus et al., 2006, Horswill and McKenna, 2004, Pelz and Krupat, 1974). For example, Borowsky, Shinar, and Oron-Gilad (2007) found that novice drivers were the least sensitive in responding to unplanned hazards as compared to two other groups of experienced drivers. Taken together, these pieces of evidence suggest that the mechanisms of visual information processing in hazard perception especially in novice drivers require closer attention.
Previous research revealed many details about differences in visual orientation between novice and expert drivers. In their seminal study, Mourant and Rockwell (1972) analyzed visual search strategies during driving on freeway and neighborhood routes. They observed that novices concentrated their search on a smaller area than experts and tended to fixate closer to the vehicle. Similar to these findings, Crundall and Underwood (1998) also reported significant differences between novices and experts while driving on dual carriageways, namely that experienced drivers showed a higher variance of horizontal gaze positions compared to novices. However, in easier traffic situations experts adjusted their scanning patterns accordingly, whereas novices exhibited rather inflexible visual routines, regardless of situational demands (see also Borowsky et al., 2007, Crundall et al., 2003, Crundall et al., 1999). Further studies confirmed this tendency of novices towards stereotypical scanning patterns during driving (e.g., Underwood et al., 2003). A recent study by Falkmer and Gregersen (2005) replicated most of the findings reported above, additionally showing that inexperienced drivers fixate more often on in-vehicle objects and relevant traffic cues compared to experts. Finally, Summala, Nieminen, and Punto (1996) demonstrated that only experienced drivers were able to use ambient visual resources in addition to focal visual processing (see Leibowitz and Post, 1982, Previc, 1998, for this distinction of separate visual processing channels) to maintain vehicle control, although subsequent work revealed that ambient vision did not support hazard detection (Summala, Lamble, & Laakso, 1998). In sum, these studies indicate that experience goes hand in hand with a development of visual strategies that can also adapt to specific demands of the traffic situation.
Apart from such rather general differences in visual orientation between novices and experts, some studies explicitly focused on the role of experience in hazard perception (e.g., Underwood, Crundall, & Chapman, 2002b). For example, inexperienced drivers exhibited longer fixation durations on potentially dangerous objects (Chapman and Underwood, 1998, Falkmer and Gregersen, 2005). Some studies reported direct evidence for slower hazard detection in inexperienced drivers (Grayson and Sexton, 2002, Hull and Christie, 1993), whereas others only reported non-significant trends (Sagberg & Bjornskau, 2006).
However, little is known about what aspects of hazard perception performance are developed from novice to expert drivers. In the studies reported above, the main parameter of interest was the time between the onset of a scene and a response, such as a decision whether the situation is dangerous or not (Hull & Christie, 1993). However, at least two different ways of gaining expertise are imaginable in the context of hazard perception: (1) a faster visual search for potentially dangerous objects as a result of “knowing where to look” (early processing skills related to visual orientation), or (2) a faster decision whether a spotted object is dangerous or not (late processing skills related to hazard processing). The studies reported above did not report data on such distinct sub-processes of hazard perception.
Section snippets
Objectives
The aim of the present study was to separately assess the time until the first fixation of a potentially dangerous object and the subsequent time until the final response to determine which of these distinct measures develops from inexperienced to experienced drivers, which is only possible by combining standard response time and eye movement analyses. To this end, we presented static traffic scenes with either low, medium, or high braking affordance (see Gibson, 1977) to inexperienced and
Participants
Twenty experienced and 20 inexperienced drivers, aged from 16 to 28 years, took part in this study. Experienced drivers were mainly students from RWTH Aachen University. Inexperienced drivers were recruited from local schools. The experienced drivers were outside their German probationary period, resulting in a driving experience from 24 months to 8 years (M = 5 years), whereas the inexperienced drivers were subjects within their probationary period, resulting in a driving experience from 0 months
Results
Pretest: The pretest was implemented to rule out the possibility of different general picture scanning routines that were not related to hazard perception between groups. To do so, we separately analyzed the mean number of fixations, the mean saccade amplitude, and the mean fixation durations for landscape and safe traffic scenes. T-tests revealed that none of these parameters differed as a function of expertise, all t < 1.02. Furthermore, we analyzed fixation distributions by segmenting the area
Discussion
The aim of the present study was to analyze the development of hazard perception performance from inexperienced to experienced drivers. To this end, we had subjects respond to static traffic scenes whenever the scene demanded a braking response or speed reduction. The pictures depicted traffic scenes of either low, medium, or high braking affordance. Most importantly, we additionally registered participants’ eye movements throughout the experiment. This allowed to divide overall RT into two
Acknowledgements
The authors would like to thank D. Crundall for his suggestion to check for any effects of saliency on the data, and G. Underwood for comments on earlier drafts of the manuscript.
References (34)
- et al.
Methods for assessing drivers’ perception of specific hazards on the road
Accident Analysis and Prevention
(1989) - et al.
Estimating hazards in traffic situations
Accident Analysis and Prevention
(1983) - et al.
Young novice drivers: Towards a model of their accident involvement
Accident Analysis and Prevention
(1996) - et al.
A saliency-based search mechanism for overt and covert shifts of visual attention
Vision Research
(2000) - et al.
Caution profile and driving record of undergraduate males
Accident Analysis and Prevention
(1974) - et al.
Hazard perception and driving experience among novice drivers
Accident Analysis and Prevention
(2006) - et al.
Driving experience and perception of the lead car’s braking when looking at in-car targets
Accident Analysis and Prevention
(1998) - et al.
Visual search while driving: Skill and awareness during inspection of the scene
Transportation Research Part F
(2002) Teenage drivers: Patterns of risk
Journal of Safety Research
(2003)- Biermann, A., Skottke, E.-M., Anders, S., Brünken, R., Debus, G., & Leutner, D. (2008). Entwicklung und Überprüfung...
Age, skill
A review of the ‘looked but failed to see’ accident causation factor. In Behavioural Research in Road Safety: Eleventh Seminar Proceedings
Risk perception and decision making during the transition between novice and experienced driver status
Ergonomics
Visual search of dynamic scenes: Event types and the role of experience in viewing driving situations
Eye movements and hazard perception in police pursuit and emergency response driving
Journal of Experimental Psychology
Driving experience and the functional field of view
Perception
Effects of experience and processing demands on visual information acquisition in drivers
Ergonomics
Cited by (88)
Improving hazard perception for drivers based on driving styles with visual alerts
2023, Transportation Research Part F: Traffic Psychology and BehaviourResearch on drivers' hazard perception in plateau environment based on visual characteristics
2022, Accident Analysis and PreventionA novel driving assessment combining hazard perception, hazard prediction and theory questions
2021, Accident Analysis and PreventionCitation Excerpt :There are a number of potential reasons for mixed findings in the literature, not least the fact that there is no agreed method for developing a hazard test. Some tests developed for research purposes use static imagery instead of dynamic clips (Huestegge et al., 2010; Scialfa et al., 2012; Vlakveld, 2014); some use video-recorded hazards (McKenna and Crick, 1994; Pradhan and Crundall, 2016) while others employ computer-generated imagery (e.g. Malone and Brünken, 2016); and some use naturalistic hazards (e.g. Horswill et al., 2008, 2015) while others stage dangerous events (e.g. Jackson et al., 2009; Borowsky et al., 2010). The recorded measures of hazard perception skill differ also.
A cross-cultural comparison of visual search strategies and response times in road hazard perception testing
2020, Accident Analysis and PreventionMethodological considerations for longitudinal investigations of young drivers
2020, Transportation Research Part F: Traffic Psychology and Behaviour