Scores on a new hazard prediction test are associated with both driver experience and crash involvement

Hazard perception skill has been found to be associated with drivers’ crash risk. This skill has typically been measured using computer-based response-time hazard perception tests, in which drivers indicate the earliest point that they detect a potential hazard in video clips of traffic filmed from a driver’s perspective. In recent years, researchers have suggested an alternative type of measure, known as a “hazard prediction test”, in which each test item is a traffic clip that stops (typically cutting to black) just before a hazardous incident unfolds. Drivers taking the test have to predict what happens next in the clip. Measures of this kind have been found to distinguish between high risk (novice) and lower risk (experienced) driver groups, and have been argued to offer several advantages over traditional response-time hazard perception tests. However, a key strength of the response-time hazard perception test is that assessments using this format have been found to predict crash involvement. The same has not yet been demonstrated for hazard prediction tests, raising questions about their validity. In the present research, we created a new hazard prediction test using Australian traffic scenes. In our version of the test, drivers’ scores were based on how many plausible predictions they were able to generate for each traffic clip. We established validity evidence for the test scores in two studies with separate samples, using two different versions of the test (long vs. short) with different response modes (verbal vs. written). As well as distinguishing between novice and experienced driver groups, test scores were also associated with self-reported crash involvement in both studies. We also found a significant correlation between hazard prediction test scores and scores in an established response-time hazard perception test. These findings support the proposal that scores on the hazard prediction test are a valid measure of hazard perception skill.     

Young drivers’ perception of hazards: Variation with experience and day versus night

Young novice drivers have a relatively high crash risk for several years following initial licensing, and while all drivers are at greater risk at night, the night-time increase is greater for inexperienced drivers. Poor hazard perception has been identified as an important contributor to inexperienced drivers’ risk, but research on day-night differences in hazard perception for drivers varying in experience is lacking. This exploratory study investigated the nature of hazards reported by young inexperienced drivers versus more experienced and slightly older drivers. Hazards were not pre-identified by researchers; participants were simply provided with a general definition of ‘hazard’. Analysis focused on how experience level affected qualitative differences in the kinds of hazards reported, with particular focus on day-night differences.The 53 participants ranged in driving experience from learners through to 5+years post licensing, and in age from 16 to 30 years. They viewed 14 day- and night-time video clips of a diverse range of driving situations, pausing the video whenever they identified a hazard and then explaining why they had paused it at that point. Their responses were recorded. Content analysis of responses showed that more experienced drivers reported visibility-related hazards significantly more often than inexperienced ones, and significantly more so at night. They also commented significantly more on hazards related to tight bends in the road and significantly less on hazards concerning compliance with rules. Comments tended to be fewer with higher vehicle speeds, particularly for the least experienced drivers.Results are discussed in terms of how experience-related differences in drivers’ cognitive schemata and mental models are likely to affect hazard perception and crash risk, particularly at night. Some implications for driver training and license testing are suggested.     

The differences in hazard response time and driving styles of violation-involved and violation-free taxi drivers

The present study aimed to investigate the relationships between taxi drivers’ traffic violations in past driving and two domains: driving skill (hazard perception skill) and driving style. Five hundred and fifty taxi drivers aged 25 – 59 were recruited to finish a video-based hazard perception test and the Chinese version of the Multidimensional Driving Style Inventory (MDSI). The relationships between hazard response time, driving style and traffic violations were examined, and the differences in hazard response times and driving styles of violation-involved drivers (n = 220) and violation-free drivers (n = 330) were compared. The results showed that taxi drivers’ traffic violations are closely related to their driving styles and hazard response time. Violation-involved drivers scored significantly higher in hazard response time and maladaptive driving styles (i.e., anxious, risky and angry styles) and lower in careful driving style than violation-free drivers. More importantly, drivers’ hazard response time and driving styles can effectively predict their violation involvement in the last 12 months with an overall classification accuracy of 66.4%. The findings provide evidence for the usefulness of video-based hazard perception tests and the MDSI in taxi driver testing and training.     

Reducing optimism bias in the driver’s seat: Comparing two interventions

Optimism bias combined with sensation-seeking and risky driving have been proposed to be the main contributing factors to young drivers’ involvement in road traffic collisions. The present study aimed to evaluate how two brief interventions, one based on an unambiguous definition of “good” driving and the other on a hazard perception test, might reduce young drivers’ optimism bias. One hundred and twenty-eight university students were randomly allocated to one of three groups: standard definition, hazard perception or control. Measures evaluating optimism bias were completed before and after the intervention, and questions regarding their sensation-seeking and past risk-taking tendencies were asked at follow-up. Both brief interventions reduced optimism bias levels, but hazard perception had the strongest effect. The effectiveness of the two interventions also differed across individuals depending on their sensation-seeking and past risky driving tendencies. The results provide evidence for the effectiveness of brief interventions to reduce optimism bias.     

Behind the wheels with autism and ADHD: Brain networks involved in driving hazard detection

Driving is a cognitively challenging task, and many individuals with autism spectrum disorder (ASD) or with attention-deficit/hyperactivity disorder (ADHD) struggle to drive safely and effectively. Previous evidence suggests that core neuropsychological deficits in executive functioning (EF) and theory of mind (ToM) may impact driving in ASD and ADHD. This functional magnetic resonance imaging (fMRI) study compares the brain mechanisms underlying ToM and EF during a hazard perception driving task. Forty-six licensed drivers (14 ASD, 17 ADHD, 15 typically developing (TD)), ages 16–27 years, viewed a driving scenario in the MRI scanner and were instructed to respond to driving hazards that were either “social” (contained a human component such as a pedestrian) or “nonsocial” (physical objects such as a barrel). All groups of participants recruited regions part of the “social brain” (anterior insula, angular gyrus, right middle occipital gyrus, right cuneus/precuneus, and right inferior frontal gyrus) when processing social hazards, and regions associated with motor planning and object recognition (postcentral gyrus, precentral gyrus, and supplementary motor area) when processing nonsocial hazards. While there were no group differences in brain activation during the driving task, years licensed was predictive of greater prefrontal and temporal activation to social hazards in all participants. Findings of the current study suggest that high-functioning ASD and ADHD licensed drivers may be utilizing similar cognitive resources as TD controls for decisions related to driving-related hazard detection.     

Mechanisms behind hazard perception of warning signs: An EEG study

Warning signs are the most effective way to alert people to hazards in the industries, architecture, and transportation. To better understand how these warning signs prompted hazard, this study investigated the intrinsic mechanism of hazard perception of warning signs, including warning words (low versus high hazard levels) and surrounding shapes (upright triangle versus circle). We recruited 18 participants and recorded electroencephalogram (EEG). We found that a combination of high-hazard-level words and an upright triangle shape significantly elicited more negative N300 and increased theta-band (3–8 Hz) synchronization, but reduced beta-band (15–30 Hz) desynchronization. The N300 component represented the emotional arousal strength, theta oscillations reflected the engagement of retrieval of emotional experiences from episodic memory, and beta oscillation connoted the semantic encoding. Thus, these findings suggested that a high-hazard-level combination had the ability to obtain high priority in cognitive processing. The EEG pattern changes allowed us to assess whether the hazard perception of warning signs was high or low, which could be an accurate indicator to better evaluate the design of warning signs.     

Visual exploration and hazard search strategies in a simulated road crossing task among primary and secondary school students in Tanzania

Pedestrians, cyclists, motorcyclists, and other vulnerable road users represent more than half of all road fatalities globally. In Tanzania, pedestrians account for a significant proportion of the death toll, accounting for 30% of all traffic fatalities (WHO, 2018) and a 2016 study conducted in Dar Es Salaam found that 87% of school-aged children walk to school (Draisin, 2016), highlighting that school-aged children are exposed to a high level of risk.The present work reports the results of a study conducted in primary and secondary schools in the Arusha Region of Tanzania which investigates the students’ road crossing mental representation, as well as their level of hazard perception awareness, through their declared gaze behaviour. The students were asked to identify and tell the areas where hazards could come from within three road crossing scenarios, thus exploring the mental representation of the visual exploration strategies applied by children and teenagers when crossing the road. Results showed the tendency to apply the “compliant gaze behaviour” pattern in a flexible manner and to integrate it with the exploration of other areas in the visual field, particularly by the senior students. Practical implications for planning effective pedestrian road safety training programs are discussed.     

The interactive effect of vehicle signals and sensation-seeking on driver hazard perception

Does sensation seeking—as a personality trait that is closely related to risky driving—affect a driver's hazard perception, and if so, under what circumstances? Using hazard locations and vehicle signals as cues, this study classifies traffic situations into four types: danger ahead with cues, danger ahead without cues, danger behind with cues, and danger behind without cues. Based on these classifications, we examine the characteristics of hazard perception, decision-making criteria, sensitivity, and visual attention patterns of drivers with different sensation-seeking tendencies. Fifty-two experienced drivers were recruited for experiments in Dalian, China. Using computer, participants were shown videos taken from the driver's perspective and were asked to respond to hazards by pressing keys on the left mouse button. Their eye movements were recorded, and a multi-dimensional driving-style scale was used to divide participants into higher and lower sensation-seeking groups. The results showed that the higher sensation-seeking drivers had a lower hazard detection rate, fewer fixation counts, and shorter glances when there were cues, and they responded more slowly when there was danger ahead. For dangers ahead with cues and dangers behind without cues, the more attention the higher sensation-seeking drivers paid to the danger, the slower their response and the lower their hazard detection rate. The results indicated that higher sensation-seeking drivers have poor hazard perception in situations when there are vehicle signals as prompting clues and dangers ahead. In particular, for a danger with clues, the visual agility of the higher sensation-seeking drivers was worse than that of the lower sensation-seeking drivers. The more attention they paid to the danger the worse their hazard perception. The results of this study can provide guidance to teach drivers hazard-perception skills.     

A novel method for assessing rival models of recognition memory

A general comparison is made between the multinomial processing tree (MPT) approach and a strength-based approach for modeling recognition memory measurement. Strength models include the signal-detection model and the dual-process model. Existing MPT models for recognition memory and a new generic MPT model, called the Multistate (MS) model, are contrasted with the strength models. Although the ROC curves for the MS model and strength model are similar, there is a critical difference between existing strength models and MPT models that goes beyond the assessment of the ROC. This difference concerns the question of stochastic mixtures for foil test trials. The hazard function and the reverse hazard function are powerful methods for detecting the presence of a probabilistic mixture. Several new theorems establish a novel method for obtaining information about the hazard function and reverse hazard function for the latent continuous distributions that are assumed in the strength approach to recognition memory. Evidence is provided that foil test trials involve a stochastic mixture. This finding occurred for both short-term memory procedures, such as the Brown–Peterson task, and long-term list-learning procedures, such as the paired-associate task. The effect of mixtures on foil trials is problematic for existing strength models but can be readily handled by MPT models such as the MS model. Other phenomena, such as the mirror effect and the effect of target-foil similarity, are also predicted accurately by the MPT modeling framework.     

Properties of reverse hazard functions

For continuous distributions reverse hazard is defined as the probability density divided by the cumulative probability, F(x); whereas the usual hazard function is the density divided by the survivor function, 1−F(x). Reverse hazard corresponds to the conditional density of an immediate failure or state change, conditioned by the fact that the state change occurred. For example, of all the items that failed, the proportion of those items that immediately failed is reverse hazard. Reverse hazard exhibits many symmetrical properties with hazard. In this paper a set of theorems are developed that explicate the properties of reverse hazard for both continuous and discrete probability distributions. Taken together, hazard and reverse hazard are a powerful set of theoretical constructs that are valuable for understanding stochastic systems.