The efficacy of hazard perception training and education: A systematic review and meta-analysis.

BACKGROUND: Hazard perception (HP) has been argued to improve with experience, with numerous training programs having been developed in an attempt to fast track the development of this critical safety skill. To date, there has been little synthesis of these methods. OBJECTIVE: The present study sought to synthesise the literature for all road users to capture the breadth of methodologies and intervention types, and quantify their efficacy. DATA SOURCES: A systematic review of both peer reviewed and non-peer-reviewed literature was completed. A total of 57 papers were found to have met inclusion criteria. RESULTS: Research into hazard perception has focused primarily on drivers (with 42 studies), with a limited number of studies focusing on vulnerable road users, including motorcyclists (3 studies), cyclists (7 studies) and pedestrians (5 studies). Training was found to have a large significant effect on improving hazard perception skills for drivers (g = 0.78) and cyclists (g = 0.97), a moderate effect for pedestrians (g = 0.64) and small effect for motorcyclists (g = 0.42). There was considerable heterogeneity in the findings, with the efficacy of training varying as a function of the hazard perception skill being measured, the type of training enacted (active, passive or combined) and the number of sessions of training (single or multiple). Active training and single sessions were found to yield more consistent significant improvements in hazard perception. CONCLUSIONS: This study found that HP training improved HP skill across all road user groups with generally moderate to large effects identified. HP training should employ a training method that actively engages the participants in the training task. Preliminary results suggest that a single session of training may be sufficient to improve HP skill however more research is needed into the delivery of these single sessions and long-term retention. Further research is also required to determine whether improvements in early-stage skills translate to improvements in responses on the road, and the long-term retention of the skills developed through training.

Developing a taxonomy of hazards and hazard mitigation strategies for Motorised mobility scooter users.

The increased use of motorised mobility scooters (MMSs) presents a road safety challenge as using a MMS has risks for the user, pedestrians, and other road users. In relation to enhancing MMS driving safety, much of the training and available literature focuses on training vehicular control. Equally important is the need to investigate higher-order cognitive skills involved in driving MMSs, particularly hazard perception. Through a large questionnaire study with MMS users, we develop a taxonomy of the types of hazard MMS users encounter when crossing roads and strategies that are used to negotiate these hazards. Whilst MMS experience modulated hazard perception and strategy use, a core set of hazards and strategies were identified that have policy and practice implications for training interventions and the built environment. Exploration of the advantages and disadvantages of MMS use indicated its impact on various wellbeing outcomes as well as some potential barriers to use.

Improving attitudes towards vulnerable road users.

Some car drivers pass vulnerable road users (VRUs) at too high a speed and leave too little gap. Research suggests that out-group bias and a lack of empathy may relate to these dangerous incidents. To address this problem, four road safety videos were created with the aim of increasing car drivers' empathy toward Vulnerable Road Users (VRUs). Each video focused on the back-story of a VRU (two horse riders and two cyclists), with a predominantly persuasive approach to inducing empathy and reducing out-group bias. The effectiveness of the videos on car drivers' immediate implicit and explicit attitudes, and on their future intended behaviour around these VRUs, was assessed by comparing intervention participants' responses with those of control participants, both before and after the intervention. Explicit attitudes towards horse riders improved following the intervention for drivers who saw the horse rider videos compared to a control group. The horse rider videos did not however improve explicit attitudes toward cyclists. Similarly, those participants who saw the cyclist videos showed improved attitudes towards cyclists but not horse riders. An Implicit Attitude Test (IAT) found no implicit biases for or against horse riders or cyclists, either before or after the intervention. Intended future passing behaviour was affected by the intervention however, with car drivers reporting greater intended passing distances and lower passing speeds compared to the control group. The results suggest that safety campaigns that adopt this approach can change attitudes and intended behaviours towards VRUs.

Correlations among self-report, static image, and video-based hazard perception assessments: The validity of a new Lithuanian hazard prediction test.

Scholars suggest that improving hazard perception (HP) skills among different road users can lead to safer driving. To this end, a small number of countries have developed a national hazard perception test for their licensing procedures. The purpose of the current research is to develop and validate a new video-based hazard prediction test for the Lithuanian driving context. Eighty-eight drivers participated in the study. Initially, 25 naturalistic clips were included and edited into a hazard prediction test following the "what happens next?" paradigm. The overall score of the test was higher for experienced drivers when compared with novices. On further inspection the overall group difference was driven by 12 clips. The Lithuanian Hazard Perception test HPT-LIT featuring static images (Tuske et al., 2019) and Hazard Perception Questionnaire developed by White et al. (2011) were used to validate the new test. This new version of the test (12 clips, LHP12) showed acceptable psychometric properties. Drivers with less driving experience demonstrated poorer hazard prediction skills than experienced drivers. The results revealed a negative correlation between hazard prediction scores on LHP12 and crash involvement in the novice driver group, as well as a positive correlation between prediction scores and self-reported hazard prediction skills. It was concluded that the new Lithuanian hazard test shows potential for future research.

An online hazard prediction test demonstrates differences in the ability to identify hazardous situations between different driving groups.

Previous research has found that the traditional approach to measuring hazard perception, as used by the UK Government, does not necessarily transfer well to other countries. The speeded response times of a hazard perception test were susceptible to threshold bias, differentiating drivers according to cultural background more than driving experience. An alternative method - the hazard prediction test - appears to mitigate these problems when used across different countries. The current study applies the hazard prediction methodology to a new cultural context (Israel), delivered via an online platform. A further innovation was to include both hazardous and non-hazardous trials allowing measures of criterion and sensitivity parameters of the Signal Detection Theory to be calculated.In total, 74 participants (43 experienced and 31 novices) watched 26 hazard prediction clips that had been filmed in Tel Aviv (Israel), 13 of which contained cues to genuine hazards. Each hazardous clip was occluded prior to the hazardous situation fully materialising, though with sufficient evidence that any participant who was looking in the right place at the right time would be able to correctly identify the unfolding hazard. Participants were first asked to identify whether there was a hazard (a binary decision), and then to predict how the traffic situation was going to develop by choosing one of four options provided after each video.As hypothesised, experienced drivers outperformed novices in the multiple-choice prediction question, provided that they had correctly identified that the clip contained a hazard in the first question. Novices were poor at selecting the correct multiple-choice option, regardless of whether they correctly identified that the clip contained a hazard. Both driving experience and hazard sensitivity significantly predicted multiple-choice accuracy, though criterion did not. The newly developed online test was successful in differentiating between experienced and novice driver groups for accuracy in prediction. These results support the roll-out of the hazard prediction methodology across different countries. Practitioner summary: This study provides new insights into how to develop an online hazard prediction test suitable for international export. The results suggest that this newly developed test is successful in differentiating between experienced and novice drivers. This raises the possibilit y of assessing and training drivers on a global scale.

The Multiple Object Avoidance (MOA) task measures attention for action: Evidence from driving and sport.

Performance in everyday tasks, such as driving and sport, requires allocation of attention to task-relevant information and the ability to inhibit task-irrelevant information. Yet there are individual differences in this attentional function ability. This research investigates a novel task for measuring attention for action, called the Multiple Object Avoidance task (MOA), in its relation to the everyday tasks of driving and sport. The aim in Study 1 was to explore the efficacy of the MOA task to predict simulated driving behaviour and hazard perception. Whilst also investigating its test-retest reliability and how it correlates to self-report driving measures. We found that superior performance in the MOA task predicted simulated driving performance in complex environments and was superior at predicting performance compared to the Useful Field of View task. We found a moderate test-retest reliability and a correlation between the attentional lapses subscale of the Driving Behaviour Questionnaire. Study 2 investigated the discriminative power of the MOA in sport by exploring performance differences in those that do and do not play sports. We also investigated if the MOA shared attentional elements with other measures of visual attention commonly attributed to sporting expertise: Multiple Object Tracking (MOT) and cognitive processing speed. We found that those that played sports exhibited superior MOA performance and found a positive relationship between MOA performance and Multiple Object Tracking performance and cognitive processing speed. Collectively, this research highlights the utility of the MOA when investigating visual attention in everyday contexts.

Assessing Willingness to Engage in Risky Driving Behaviour Using Naturalistic Driving Footage: The Role of Age and Gender.

Young novice drivers are more prone than older drivers to get involved in a risky driving situation. Some young drivers underestimate risk while overestimating their driving abilities, increasing the likelihood of engaging in risky behaviour. Age and inexperience both influence risk estimation, though it is not clear which of these variables is more important. Can drivers' willingness to engage in risky behaviour be assessed in a similar way to hazard perception skill using video-based risky situations? The aim of the current study was to assess whether a video-based tool could measure the willingness to participate in risky driving situations and whether it can distinguish between different types of risky driving scenarios across gender and driver age groups. We also explored the moderating effect of age and gender on drivers' experience in relation to the risky manoeuvres and participants' willingness to engage in risky situations. Participants were presented with naturalistic videos from the perspective of the driver that contained active risky situations (result of driver's own actions) and were asked to make a decision regarding a potential action (to overtake a bus/bicycle or pass through an amber light) and whether they would accelerate at this point. Participants reported that they were more willing to accelerate and overtake cyclists and buses and less willing to pass a light in amber. Young drivers were more willing to both engage in the risky behaviours and accelerate than older drivers, with young males reporting higher scores than the other groups. Gender differences were observed, with males being more prone to overtake and pass through a light in amber than females; however, this difference was not observed for the intention to accelerate. All the above effects remained when we tested the impact of experience on decision making while controlling for age and gender, although driving experience was no longer significant. These results demonstrate that drivers' intention to assume risk can indeed be measured in a similar video-based methodology to that used by hazard perception tests. The findings raise the possibility of assessing and training drivers on a wider range of safety-related behaviours.

A comparison of cybersickness symptoms across 360-degree hazard perception and hazard prediction tests for drivers.

Hazard perception assessment may benefit from VR-presentation by removing field-of-view restrictions imposed by single-screen tests. One concern is whether VR-induced 'cybersickness' will offset any benefits. Self-reported cybersickness ratings were recorded from 77 participants viewing two variants of a 360-degree hazard test: hazard perception and hazard prediction. The latter was hypothesised to be particularly susceptible as clips abruptly cut to a probe question at hazard onset. Such sudden occlusions are thought to increase cybersickness. Overall cybersickness levels were low, with only four participants excluded for above-threshold sickness ratings. The remaining participants showed unexpectedly lower symptoms for the hazard prediction test and rated this test format as more comfortable and engaging. These findings mitigate concerns over the use of 360-degree videos in formative hazard assessments, even when clips involve sudden occlusions. Nonetheless, removal of any participants due to cybersickness raises problems for using VR for formal assessments of hazard perception skill.

Risky Decision-Making and Hazard Prediction are Negatively Related and Could Be Assessed Independently Using Driving Footage.

INTRODUCTION: Traffic collisions are a principal cause of death in Europe, disproportionately affecting young drivers. Driving safety depends not only on our ability to anticipate and respond to dangers on the road but also on the level of risk we are willing to engage within our own driving behaviour. METHODS: Hazard prediction (HPr) and risky decision-making (RDM) tests were given to three groups of young Spaniards (169 participants): 54 non-drivers (M=20), 65 novice (M=21) and, 50 experienced drivers (M=26 years old). Both tests presented participants with video clips of driving recorded from the driver's perspective. The HPr test contained hazardous situations caused by the actions of another road user (eg, a pedestrian crossing the road). Each HPr clip was occluded as a hazard began to unfold and participants were asked to predict "what happens next?" They selected their answer from four on-screen options. The RDM test used clips where any imminent danger would be provoked by the film-car driver's risky behaviour (eg, overtaking illegally). Participants were asked to report the probability of following certain types of risky behaviour (eg, "Would you go forward with the lights on amber?" or "Would you overtake the cyclist/lorry/bus at this point?"). In addition, the effect of the locality of the driving scenarios was manipulated: they could take place in the participant's native country (Spain) or in a different country (UK). RESULTS: Non-drivers and novice drivers were less able to predict upcoming hazards and more likely to make risky decisions. Driving scenarios from another country (UK) provoked riskier decisions than those from the participants' home country (Spain). CONCLUSION: Improvement in HPr skills among novice or new drivers poses a huge challenge as far as driver training is concerned, though it is only part of the solution. Young inexperienced drivers' willingness to engage in risky behaviour also needs to be tackled. Our results suggest that such RDM can be assessed in a similar way to HPr skill, using a naturalistic approach, which raises the possibility of assessing and training drivers on a wider range of safety-related behaviours.

Search for a distressed swimmer in a dynamic, real-world environment.

Visual search is increasingly being explored in dynamic, real-world environments. This includes swimming pools, where lifeguards have shown superior drowning detection in simulated environments. Here, we explored if lifeguard superiority is observed in real-life scenes of a busy swimming pool. Experiment 1 required participants to identify real-life distressed swimmers in clips of busy pool activity via a touchscreen interface. Experiment 2 sought to replicate the first study, with the inclusion of eye-movement measures. Experiment 3 varied the methodology, using an occlusion method where clips were frozen and blurred shortly after target onset. The results demonstrated an experience effect, with lifeguards detecting distressed swimmers more often and faster than nonlifeguards. No clear differences were found in the eye-movements between groups; thus, we cannot conclude that the lifeguards' faster responses are due to better scanning strategies. The different methodological approaches revealed the occlusion method to have the larger effect size, supporting the growing evidence that occlusion may be a better test for dynamic target detection than traditional response-time tests. This research demonstrates that the clear lifeguard experience effect generalizes to real-life pool environments with a large number of swimmers and real incidents. It could be used to inform lifeguard training tools and assessments. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

A novel driving assessment combining hazard perception, hazard prediction and theory questions.

A new hazard test was created using high-fidelity computer animation containing ten hazards. Sixty learner drivers and sixty experienced drivers sat either a hazard-perception version of this test (requiring timed responses to materialized hazards) or a hazard-prediction variant of the test (where the screen is occluded as the hazard begins to appear and drivers are asked 'What happens next?'). Recent studies have demonstrated that the prediction test format outperforms the hazard perception format using naturalistic video, but there has not yet been a study replicating this effect with computer-animated materials similar to the quality of those used in the official UK hazard perception test. The new test also included eleven theory questions designed to probe drivers' knowledge of the rules of the road. The results demonstrated that both test variants differentiated between driver groups with considerable effect sizes. Theory-question scores were comparable across learner and experienced driver groups, reflecting learners' preparation for the test and possible issues with memory decay and overwriting in the experienced group. As an interesting aside, driving-related video game play negatively correlated with hazard perception performance, but not with hazard prediction scores. Some individual hazards better suited the prediction or perception test format, raising the possibility of a future hybrid test that combines the two approaches.

Proof-of-Concept Study: a Mobile Application to Derive Clinical Outcome Measures from Expression and Speech for Mental Health Status Evaluation.

This proof-of-concept study aimed to assess the ability of a mobile application and cloud analytics software solution to extract facial expression information from participant selfie videos. This is one component of a solution aimed at extracting possible health outcome measures based on expression, voice acoustics and speech sentiment from video diary data provided by patients. Forty healthy volunteers viewed 21 validated images from the International Affective Picture System database through a mobile app which simultaneously captured video footage of their face using the selfie camera. Images were intended to be associated with the following emotional responses: anger, disgust, sadness, contempt, fear, surprise and happiness. Both valence and arousal scores estimated from the video footage associated with each image were adequate predictors of the IAPS image scores (p < 0.001 and p = 0.04 respectively). 12.2% of images were categorised as containing a positive expression response in line with the target expression; with happiness and sadness responses providing the greatest frequency of responders: 41.0% and 21.4% respectively. 71.2% of images were associated with no change in expression. This proof-of-concept study provides early encouraging findings that changes in facial expression can be detected when they exist. Combined with voice acoustical measures and speech sentiment analysis, this may lead to novel measures of health status in patients using a video diary in indications including depression, schizophrenia, autism spectrum disorder and PTSD amongst other conditions.

Creating a hazard-based training and assessment tool for emergency response drivers.

Emergency response drivers (ERDs) are often required to engage in high-risk driving manoeuvres on their way to a reported incident. Such risk requires that these drivers receive a high-level of training and continued development. The aim of this paper was to investigate an innovative format for a new potential tool that could support the training and assessment of these drivers: a single-clip Holistic Hazard Test, containing multiple hazards in a single route. In study one, we created a proof-of-concept 15-minute clip containing hazards, multiple-choice questions and probes to collect self-reported safety ratings. ERDs were more accurate on the multiple-choice questions (MCQs) than a control group, though response time scores to hazards did not reach the threshold for significance. In study two, we refined the development process and created a series of new holistic hazard tests across four counties of the East Midlands, UK. Each test contained many hazards and MCQs that assessed situation awareness and decision-making, based on the results of study 1. Participants were recruited across the four counties and were presented with both the test that was specific to their county and one of the unfamiliar-location tests, in order to assess the generalisability of the tests across different locales. The results showed no differences regarding location familiarity, suggesting that tests filmed in one area of the country can be viewed by drivers elsewhere without detriment to performance. ERDs once again responded to MCQs more accurately, and also scored more hazard points on the basis of faster responses to hazards compared to control participants. These results suggest such tests can successfully tap into ERD-specific skills with regard to spotting, predicting and responding to hazards on the road. We recommend refinement of this tool for assessment of emergency response drivers, and further development to extend the materials to create a training tool.

How are distractibility and hazard prediction in driving related? Role of driving experience as moderating factor.

Distraction constitute one of the 'five fatal' behaviours that contribute to road trauma, and some people may be more susceptible to it than others. It is also known that a greater ability to predict danger is related to a lower probability of suffering accidents. It could be hypothesised that drivers with a higher tendency to distraction are worse at predicting traffic hazards, but to what extent might driving experience serve to mitigate this tendency to distraction? The current study collected self-reported attentional errors from drivers by using the Attention-Related Driving Errors Scale (ARDES-Spain) in order to examine whether novice drivers suffered from inattention more than experienced drivers. The results demonstrated that novice drivers scored more highly on ARDES than experienced drivers. ARDES scores were then related to performance in a Hazard Prediction test, where participants had to report what hazard was about to happen in a series of video clips that occlude just as the hazard begins to develop. While experienced drivers were better at the Hazard Prediction test than novice drivers, those participants who reported fewer attention errors were also better able to detect the upcoming hazard following occlusion. In addition, our results demonstrate a relationship between self-reported attentional errors and the ability to predict upcoming hazards on the road, with driving experience having a moderating role. In the case of novice drivers, as their scores in the Manoeuvring Errors ARDES factor increase, their ability in Hazard Prediction diminishes, while for experienced drivers the increase is not significant. Guidance on how to improve training for drivers in order to mitigate the effects of inattention on driving safety can be addressed.

A comparison of hazard perception and hazard prediction tests across China, Spain and the UK.

Hazard perception (HP) is the ability to spot on-road hazards in time to avoid a collision. This skill is traditionally measured by recording response times to hazards in video clips of driving, with safer, experienced drivers often out-performing inexperienced drivers. This study assessed whether HP test performance is culturally specific by comparing Chinese, Spanish and UK drivers who watched clips filmed in all three countries. Two test-variants were created: a traditional HP test (requiring timed hazard responses), and a hazard prediction test, where the film is occluded at hazard-onset and participants predict what happens next. More than 300 participants, across the 3 countries, were divided into experienced and inexperienced-driver groups. The traditional HP test did not discriminate between experienced and inexperienced drivers, though participant nationality influenced the results with UK drivers reporting more hazards than Chinese drivers. The hazard prediction test, however, found experienced drivers to out-perform inexperienced drivers. No differences were found for nationality, with all nationalities being equally skilled at predicting hazards. The results suggest that drivers' criterion level for responding to hazards is culturally sensitive, though their ability to predict hazards is not. We argue that the more robust, culturally-agnostic, hazard prediction test appears better suited for global export.

Prediction and perception of hazards in professional drivers: Does hazard perception skill differ between safe and less-safe fire-appliance drivers?

Can hazard perception testing be useful for the emergency services? Previous research has found emergency response drivers' (ERDs) to perform better than controls, however these studies used clips of normal driving. In contrast, the current study filmed footage from a fire-appliance on blue-light training runs through Nottinghamshire, and endeavoured to discriminate between different groups of EDRs based on experience and collision risk. Thirty clips were selected to create two variants of the hazard perception test: a traditional push-button test requiring speeded-responses to hazards, and a prediction test that occludes at hazard onset and provides four possible outcomes for participants to choose between. Three groups of fire-appliance drivers (novices, low-risk experienced and high-risk experienced), and age-matched controls undertook both tests. The hazard perception test only discriminated between controls and all FA drivers, whereas the hazard prediction test was more sensitive, discriminating between high and low-risk experienced fire appliance drivers. Eye movement analyses suggest that the low-risk drivers were better at prioritising the hazardous precursors, leading to better predictive accuracy. These results pave the way for future assessment and training tools to supplement emergency response driver training, while supporting the growing literature that identifies hazard prediction as a more robust measure of driver safety than traditional hazard perception tests.

Commentary driver training: Effects of commentary exposure, practice and production on hazard perception and eye movements.

Commentary driving typically involves being trained in how to produce a verbal running commentary about what you can see, what you are doing, what might happen and what action you will take to avoid potential hazards, while driving. Although video-based commentary training has been associated with subsequent hazard perception improvements, it can have a negative impact on hazard perception when a live commentary is produced at test (Young, Chapman, & Crundall, 2014). In the current study we use balanced training and testing blocks to isolate the effects of commentary exposure, production of a commentary with and without practice, and learning from earlier self-generation of commentary on behavioural and eye movement measures. Importantly, both commentary exposed and unexposed groups gave hazard perception responses during the commentary video, ensuring that the unexposed control group remained engaged in the procedure throughout. Results show that producing a live commentary is detrimental to concurrent hazard perception, even after practice, and does not enhance any later effect of commentary exposure. Although commentary exposure led to an initial increase in the accuracy of hazard perception responses, this effect was limited to the first occasion of testing, and showed no later benefits relative to engaged hazard exposure.

What happens when drivers face hazards on the road?

The current study aims to obtain knowledge about the nature of the processes involved in Hazard Perception, using measurement techniques to separate and independently quantify these suspected sub-processes: Sensation, Situation Awareness (recognition, location and projection) and decision-making. It applies Signal Detection Theory analysis to Hazard Perception and Prediction Tasks. To enable the calculation of Signal Detection Theory parameters, video-recorded hazardous vs. quasi-hazardous situations were presented to the participants. In the hazardous situations it is necessary to perform an evasive action, for instance, braking or swerving abruptly, while the quasi-hazardous situations do not require the driver to make any evasive manoeuvre, merely to carry on driving at the same speed and following the same trajectory. A first Multiple Choice Hazard Perception and Prediction test was created to measure participants' performance in a What Happens Next? Task. The sample comprised 143 participants, 47 females and 94 males. Groups of non-offender drivers (learner, novice and experienced) and offender drivers (novice and experienced) were recruited. The Multiple Choice Hazard Perception and Prediction test succeeded in finding differences between drivers according to their driving experience. In fact, differences exist with regard to the level of hazard discrimination (d' prime) by drivers with different experience (learner, novice and experienced drivers) and profile (offenders and non-offenders) and these differences emerge from Signal Detection Theory analysis. In addition, it was found that experienced drivers show higher Situation Awareness than learner or novice drivers. On the other hand, although offenders do worse than non-offenders on the hazard identification question, they do just as well when their Situation Awareness is probed (in fact, they are as aware as non-offenders of what the obstacles on the road are, where they are and what will happen next). Nevertheless, when considering the answers participants provided about their degree of cautiousness, experienced drivers were more cautious than novice drivers, and non-offender drivers were more cautious than offender drivers. That is, a greater number of experienced and non-offender drivers chose the answer "I would make an evasive manoeuvre such as braking gradually".

Risk-taking on the road and in the mind: behavioural and neural patterns of decision making between risky and safe drivers.

Drivers' risk-taking is a key issue of road safety. This study explored individual differences in drivers' decision-making, linking external behaviours to internal neural activity, to reveal the cognitive mechanisms of risky driving. Twenty-four male drivers were split into two groups (risky vs. safe drivers) via the Drivier Behaviour Questionnaire-violation. The risky drivers demonstrated higher preference for the risky choices in the paradigms of Iowa Gambling Task and Balloon Analogue Risk Task. More importantly, the risky drivers showed lower amplitudes of feedback-related negativity (FRN) and loss-minus-gain FRN in both paradigms, which indicated their neural processing of error-detection. A significant difference of P300 amplitudes was also reported between groups, which indicated their neural processing of reward-evaluation and were modified by specific paradigm and feedback. These results suggested that the neural basis of risky driving was the decision patterns less revised by losses and more motivated by rewards. PRACTITIONER SUMMARY: Risk-taking on the road is largely determined by inherent cognitive mechanisms, which can be indicated by the behavioural and neural patterns of decision-making. In this regard, it is feasible to quantize drivers' riskiness in the cognitive stage before actual risky driving or accidents, and intervene accordingly.

Hazard prediction discriminates between novice and experienced drivers.

Typical hazard perception tests often confound multiple processes in their responses. The current study tested hazard prediction in isolation to assess whether this component can discriminate between novice and experienced drivers. A variant of the hazard perception test, based on the Situation Awareness Global Assessment Technique, found experienced drivers to outperform novices across three experiments suggesting that the act of predicting an imminent hazard is a crucial part of the hazard-perception process. Furthermore three additional hypotheses were tested in these experiments. First, performance was compared across clips of different length. There was marginal evidence that novice drivers' performance suffered with the longest clips, but experienced drivers' performance did not, suggesting that experienced drivers find hazard prediction less effortful. Secondly, predictive accuracy was found to be dependent on the temporal proximity of visual precursors to the hazard. Thirdly the relationship between the hazard and its precursor was found to be important, with less obvious precursors improving the discrimination between novice and experience drivers. These findings demonstrate that a measure of hazard prediction, which is less confounded by the influence of risk appraisal than simple response time measures, can still discriminate between novice and experienced drivers. Application of this methodology under different conditions can produce insights into the underlying processes that may be at work, whilst also providing an alternative test of driver skill in relation to the detection of hazards.