Table 2.
Characteristics of initial studies included in the systematic review
|
Author/
Year of publication |
The purpose of the study | Country | Type of study | Test method | Sample size | The overall result | Quality appraisal | |
|---|---|---|---|---|---|---|---|---|
| 1 | Feizabadi et al. 1394 (27) |
The relationship between risk perception and driving experiences | Iran | Cross-sectional study | Video | 126 drivers | There was a positive and significant correlation between driving experience and traffic risk perception score and driving experience could predict risk perception. | 40 |
| 2 | Mahmoodi et al. 1394 (28) |
The effect of driving experience on the perception of traffic hazards | Iran | Cross-sectional study | Video | 53 novice drivers, 30 experienced drivers | There was a significant difference between the two groups in terms of risk perception. Novice drivers did not recognize hazardous situations as potentially hazardous situations or reacted to those situations later than the group of experienced drivers. | 38 |
| 3 | Kamran Madadian, Salah Soufi 1397 (29) |
Predicting high-risk driving behaviors based on hazard perception | Iran | Cross-sectional study | Driving risk perception questionnaire Ulleberg, & Rundmo |
303 drivers of public transport | driving hazard perception could predict high-risk driving behavior. | 36 |
| 4 | Horswill, M., Helman, S. 2003 (30) |
Factors affecting the risk of an accident | United Kingdom | Cross-sectional study | Simulator | 48 car drivers and 47 motorcycle drivers | The results showed that car drivers who also ride motorcycles have a faster hazard perception than other people who were just car drivers or only motorcyclists. | 38 |
| 5 | Pradhan et al. 2005 (31) |
hazard perception assessment | US | Cross-sectional study | Simulator with eye tracking | 24 novice drivers, 24 experienced young drivers, 24 experienced older drivers | Novice drivers had weaker hazard detection and perception than other drivers. | 38 |
| 6 | Underwood et al.2005 (32) | Scanning the fixed pathways of the eyes of young and old drivers in assessing hazard perception | United Kingdom | Cross-sectional study | Video | 12 young drivers and 12 old drivers | There was no significant difference in perception of risk and ocular stability between young and old groups. Older drivers received videos of more dangerous scenes than the younger group. | 38 |
| 7 | Sagberg & Bjørnskau 2006 (33) |
Assessment of the driving experience on the hazard perception | Norway | Cross-sectional study | Video | 48 novice drivers and 28 experienced drivers | No significant difference was found in the perception of hazard or response time between the groups. | 40 |
| 8 | Wallis & Horswill, 2007 (34) | Investigation of why experienced drivers are faster at hazard perception than novice drivers | Australia | Cross-sectional study | Video | 25 novice drivers trained, 27 novices untrained, 17 experienced | Trained novice drivers and experienced drivers had significantly better hazard perception than untrained novice drivers. | 40 |
| 9 | Horswill, et al. 2008 (35) |
The ability of hazard perception in older drivers | Australia | Cross-sectional study | Video | 16 novice drivers and 17 experienced drivers | Experienced drivers perceived hazards significantly faster than novice drivers. | 40 |
| 10 | Lee, et al.2008 (36) | Investigating road hazards from the perspective of novice and experienced drivers | US | Cross-sectional study | Driving test | 42 novice teen drivers and 42 experienced drivers | Experienced drivers understood the hazard much more often than novice drivers. | 40 |
| 11 | Horswill, et al. 2009 (37) |
Comparison of drivers’ ability to perceive hazards | Australia | Cross-sectional study | Video | 22 young drivers, 34 older drivers, and 23 older drivers | Older drivers perceived hazards less significantly than older and younger drivers. | 36 |
| 12 | Isler, et al. 2009 (38) |
Improvement of driver hazard perception through video tutorials | New Zealand | Cross-sectional study | Video | 24 inexperienced young drivers, 8 experienced drivers | Experienced young drivers were significantly less aware of the hazards encountered than experienced drivers and required a longer response time. | 38 |
| 13 | Jackson, et al. 2009 (39) |
Predicting the behavior of other drivers | United Kingdom | Cross-sectional study | Video | 41 novice drivers, 39 experienced drivers | Novice drivers were less careful than experienced drivers in the field of hazard perception. | 38 |
| 14 | Liu, et al.2009 (40) | Assessing the ability of hazard perception in experienced and novice drivers | Australia | Cross-sectional study | Simulator | 12 inexperienced motorcyclists, 12 experienced motorcyclists, 12 novice car drivers, and 12 novice drivers with test certificates | The results showed that experienced drivers had a higher hazard perception than other drivers. They were also less likely to have a car accident than other drivers. | 38 |
| 15 | Smith, et al. 2009 (41) |
Assessing hazard perception in novice and experienced drivers | Australia | Cross-sectional study | Video | 32 novice drivers and 30 experienced drivers | Experienced drivers were significantly more aware of the hazards than other drivers and were more careful in answering questions. | 40 |
| 16 | Borowsky, et al. 2010 (42) |
Assessment of drivers hazard perception | Israel | Cross-sectional study | Video with eye tracking | 21 young drivers, 19 experienced drivers, and 16 older drivers | Differences in risk perception depended on the risk situation. Older drivers were the calmest people in response to traffic accidents. Experienced and older drivers had a wider visual scan than younger drivers. | 38 |
| 17 | Huestegge, et al. 2010 (43) |
Investigating the effect of hazard perception training on driving | Germany | Cross-sectional study | A static scene with eye tracking | 20 inexperienced drivers, 20 experienced drivers | Experienced drivers were significantly able to perceive more hazards than inexperienced drivers. | 38 |
| 18 | Shahar, et al. , 2010 (44) |
Assessment of hazard perception | United Kingdom | Cross-sectional study | Video | 20 drivers with one screen and 19 drivers with three screens | The number of hazards perceived by drivers who tested with three screens was significantly better than by drivers who tested with one screen. | 36 |
| 19 | Wetton, et al. 2010 (45) |
Development and validation of two complementary criteria for the driver’s hazard perception ability | Australia | Cross-sectional study | Video and image | 24 novice drivers and 24 experienced drivers | Experienced drivers have hazard perception significantly faster than novice drivers. The two dynamic and static methods were not related to each other. The reliability (Cronbach’s alpha coefficient) of the risk perception test was 0.93. | 36 |
| 20 | Cheng, et al. 2011 (46) |
Comparing the hazard perception ability of motorcycle drivers with accident and without an accident | Hong Kong | Cross-sectional study | Simulator with eye tracking | 63 motorcyclists without accidents and 46 motorcyclists with accidents | Non-crash motorcyclist’s hazard perception was significantly faster than non-crash motorcyclists | 36 |
| 21 | Rosenbloom, et al. 2011 (47) |
Assessing the hazard perception in motorcyclists and car drivers | Israel | Cross-sectional study | Video | 35 motorcyclists and 25 car drivers | Motorcyclists performed significantly better in hazard perception than car drivers. | 40 |
| 22 | Scialfa, et al. 2011 (48) |
Conducting a hazard perception test for novice drivers | Canada | Cross-sectional study | Video | 29 young novice drivers and 146 experienced young drivers | Young novice drivers’ hazard perceptions were significantly slower than experienced drivers. | 38 |
| 23 | Wetton, et al. 2011 (49) |
Hazard perception test for novice drivers | Australia | Cross-sectional study | Video | 175 car drivers | The results of this study showed that the employed test used could distinguish different groups of drivers who had different driving experiences. The ability to detect road hazards and predict accident risk was poor in novice drivers. Novice drivers were slower to respond to hazards than experienced young drivers. A subset of well-reliable scenes successfully distinguished the two groups. | 38 |
| 24 | Lyon, et al. 2011 (50) |
Development of static hazard perception test in North America | US | Cross-sectional study | Static images | 27 experienced young people, 29 novice young people | Novice drivers reacted less quickly to road hazards while their hazard perception was lower than that of young and experienced drivers. It was found that a short hazard perception test using static images could classify individuals with high accuracy. | 40 |
| 25 | Boufous, et al. 2011 (51) |
Hazard perception test for novice drivers | New Zealand | Cohort study | Static images | 20 822 young drivers | Drivers who failed at least twice in the hazard perception test were those who feared being involved in a traffic accident. The risk of accidents was high for those who failed the hazard perception test at least twice, especially among men and among villagers and remote areas. | 40 |
| 26 | Bellet & Banet, 2012 (52) |
Designing a conceptual model of risk awareness | France | Cross-sectional study | Video | 12 people under training, 12 novice drivers, 12 experienced drivers, and 12 police | Experienced police and drivers understood the hazard significantly better than novices and trainees. | 38 |
| 27 | Borowsky, et al. 2012 (53) |
Comparing the hazard perception ability of motorcycle drivers with accident and without an accident | Israel | Cross-sectional study | Video with eye tracking | 10 young novice drivers, 10 active trained drivers, and 21 experienced drivers | There was no significant difference between the studied groups in terms of hazard perception. Experienced drivers had better image-scanning patterns than other people. | 38 |
| 28 | Bromberg, et al. 2012 (54) |
Hazard perception of passers-by from the perspective of experienced people | Israel | Cross-sectional study | Simulator and video | 22 experienced drivers, 20 experienced senior drivers | Older drivers had longer response times than younger drivers, but groups did not differ significantly in pedestrian hazard perception. | 40 |
| 29 | Crundall, et al. 2012 (55) |
Assessing drivers’ perceptions of various hazards | United Kingdom | Cross-sectional study | Simulator with eye tracking | 14 people under training, 17 experienced drivers, and 18 training drivers | Experienced drivers and trainers understood the hazards significantly, more accurately, and quickly than those under training. | 38 |
| 30 | Scialfa, et al. 2012 (56) |
Evaluation of the effect of driving experience on static hazard perception test | Canada | Cross-sectional study | Static images | 25 young novice drivers, 26 experienced young drivers | Novice drivers perceived hazards significantly slower and less accurately than experienced drivers. | 38 |
| 31 | Crundall, et al. 2013 (57) |
Investigating the experience of motorcycling in hazard perception | United Kingdom | Cross-sectional study | Video | 20 novice drivers, 21 experienced drivers, and 20 advanced drivers | Advanced drivers perceived hazards much faster than novice drivers. Advanced drivers, because of their experience, better perceived the hazards. | 38 |
| 32 | Horswill, et al. 2013 (58) |
Assessing the ability of hazard perception in drivers | Australia | Cross-sectional study | Video | 42 experienced drivers, 26 police officers | Police officers perceived the hazards significantly faster and more accurately than experienced drivers. | 36 |
| 33 | Lim, et al. 2013 (59) |
Investigating the effect of intercultural factors on drivers’ hazard perception | Malaysia | Cross-sectional study | Video with eye tracking | 20 novice English drivers, 25 experienced British drivers, 26 novice Malaysian drivers, 27 experienced Malaysian drivers | There was no significant difference between drivers’ hazard perception with different cultures. | 36 |
| 34 | Scialfa, et al. 2013 (60) |
Comparison of dynamic and static hazard perception tests | Canada | Cross-sectional study | Static video and image | 56 experienced drivers | The relationship between dynamic and static tests was low to moderate and both tests had good validity. In the dynamic test, novice drivers perceived the hazard later than experienced drivers, but they responded faster to the static test than experienced drivers. | 40 |
| 35 | Oron-Gilad & Parmet, 2014 (61) | Evaluate the effectiveness of different educational scenarios in hazard perception | Israel | Cross-sectional study | Scenario | 39 young novice drivers and 6 experienced drivers | The use of different educational scenarios is effective in hazard perception and makes a difference between the studied groups. | 40 |
| 36 | Castro, et al. 2014 (62) |
Development and validation of the Spanish hazard Perception Test | Spain | Cross-sectional study | Video | 14 trainees, 16 novice drivers, and 14 experienced drivers | Experienced drivers received higher scores than other groups in response to the hazard perception test. The reliability of the test (Cronbach’s alpha coefficient) was 0.77. | 40 |
| 37 | Meyet, et al. 2014 (63) |
Understanding the hazard of traffic on children | Norway | Cross-sectional study | Video | 540 pedestrians | Children had the slowest time responding to the hazard perception test compared to young individuals and adolescents. | 40 |
| 38 | Vlakveld 2014 (64) |
Comparison of the effect of two methods of video presentation on hazard perception | Netherlands | Cross-sectional study | Video | First method: 30 people under training, 34 professional drivers Second method: 30 people under training, 30 professional drivers |
Professional drivers scored higher on both methods than those trained. The first method was relatively better in distinguishing between the two groups. | 38 |
| 39 | Mackenzie & Harris, 2015 (65) | Evaluation of eye movement and hazard perception in active and inactive driving | United Kingdom | Cross-sectional study | Video, simulator, and eye tracking | 17 drivers and 17 non-drivers | Those who drove identified and perceived the hazards faster than those who did not, but their accuracy did not differ significantly. | 38 |
| 40 | Malone & Brünken, 2015 (66) | Assess hazard perception | Germany | Cross-sectional study | Video | 35 people under training and 31 experienced drivers | Skilled drivers performed better in hazard perception than those trained. Although they were more accurate, they were not statistically significant. | 38 |
| 41 | Meir, et al.2015 (67) | Measuring the ability of young pedestrian’s hazard perception | Israel | Cross-sectional study | Simulator with eye tracking | 27 minors and 20 adults | Younger children were less accurate in diagnosing and perceiving hazards than older children. The younger ones responded less slowly than the older ones. | 40 |
| 42 | Rosenbloom, et al. 2015 (68) |
Hazard perception test for pedestrians | Israel | Cross-sectional study | Video | 158 children, 113 young and 88 elderly | Young people had the highest hazard perception. They were followed by children and then the elderly. Children were least likely to increase vision using the left and right arrow keys. | 40 |
| 43 | Yeung, & Wong 2015(69) |
Investigating the effect of experience on hazard perception | Singapore | Cross-sectional study | Simulator with eye tracking | 14 young novice drivers, 14 young experienced drivers, 12 old experienced drivers | No differences were observed between different groups in hazard perception. The first fixation was significantly slower in specific hazard scenarios for older drivers, with less scanning ability compared to younger drivers. | 40 |
| 44 | Horswill, et al. 2015 (70) |
Video Scene hazard Perception Test | Australia | Cross-sectional study | Simulator with eye tracking | 244 Australian drivers | Drivers who failed the risk perception test were 25% more likely to be involved in an active crash (such as an accident that occurred when the driver’s vehicle was moving). Unsuccessful drivers were also 17% more likely to have been involved in pre-test active accidents while on a temporary license. | 40 |
| 45 | Castro, et al. 2016 (71) |
Investigating the effect of education on hazard perception | Spain | Cross-sectional study | Video | 20 trainees, 62 novice drivers, and 40 experienced drivers | Experienced drivers performed better than other groups. Compared to gradual start-up hazards, there were more differences between groups of drivers regarding sudden hazards. | 36 |
| 46 | Crundall 2016 (72) |
Investigating the difference in hazard perception between experienced and novice drivers | United Kingdom | Cross-sectional study | Video | 15 novice drivers and 15 experienced drivers | Experienced drivers were significantly more accurate in hazard identification and perception. | 36 |
| 47 | Johnston & Scialfa, 2016 (73) | Hazard perception assessment in emergency drivers | Canada | Cross-sectional study | Video | 20 city drivers and 28 emergency drivers | Emergency drivers were significantly faster at hazard perception than urban drivers. | 36 |
| 48 | Malone & Brünken 2016 (74) | The role of ecological validity in hazard perception assessment | Germany | Cross-sectional study | Video | 101 drivers under training, 49 experienced drivers | Experienced drivers performed better than learners in responding to the hazard scenario. There was no significant difference between the groups in terms of potential hazards. The high-validity test showed better performance for differentiating different drivers in terms of experience. | 36 |
| 49 | Vansteenkiste, et al. 2016 (75) |
Hazard perception assessment in cyclists | Belgium | Cross-sectional study | Video with eye tracking | 11 teenagers and 17 adults | There was no difference in accuracy or hazard stabilization between the two groups. Adults were quicker to answer video questions than teens. | 36 |
| 50 | Ventsislavova, et al. 2016 (76) |
Drivers hazard perception | Spain | Cross-sectional study | Video | 34 trainees, 36 crime-free novice drivers, 4 criminal novice drivers, 54 experienced non-criminal drivers, 21 experienced criminal drivers | Hazard detection, situational awareness, and caution were significantly higher for experienced drivers than for novice drivers and learners. Offenders were less cautious and identified fewer dangers than unauthorized individuals. | 38 |
| 51 | Meir, et al. 2016 (77) |
Teaching the hazard perception to young driver | Israel | Cross-sectional study | Video | 40 inexperienced young drivers, 21 inexperienced young drivers | Providing active, school, and combination training improved drivers’ hazard perception. | 38 |
| 52 | Gugliotta, et al. 2017 (78) |
Assessment of hazard perception experience | Spain | Cross-sectional study | Video | 20 trainees, 62 novice drivers, and 40 experienced drivers | Trained drivers were significantly less accurate in identifying hazards than novice and experienced drivers, and answered decision-making questions more accurately than situational awareness questions. | 36 |
| 53 | Zeuwts, et al. 2017 (79) |
Assessing the hazard perception in child and adult cyclists | Belgium | Cross-sectional study | Video with eye tracking | 75 teenagers and 41 adults | Adolescents had a significant delay compared to adults in responding to hazard perception test questions. | 36 |
| 54 | Caparelli-Daque, et al.2017 (80) | Hazard perception test in Brazil | Brazil | Cross-sectional study | Static images | 314 male and female drivers | Drivers’ hazard perception clearly depended on variables such as expertise, age, and sex. | 40 |
| 55 | Feng, et al.2018 (81) | Hazard perception in old drivers | Canada | Cross-sectional study | Static images | 16 young drivers and 21 old drivers | Older drivers were slower and less accurate in answering hazard perception test questions. | 40 |
| 56 | Tuske, et al. 2019 (82) |
Development of Lithuanian Drivers hazard Perception Test | Lithuania | Cross-sectional study | Static images | 34 experienced drivers and 125 drivers with diverse driving experience | The hazard perception test was able to differentiate between inexperienced and experienced drivers, and between experienced drivers who have had less than three or more accidents in their driving history. The reliability (Cronbach’s alpha coefficient) of the test was 0.77. | 38 |
| 57 | Malone & Brunken 2019 (83) | Measuring hazard perception with traditional, verbal, and motor test methods | Germany | Cross-sectional study | Video with eye tracking | 69 drivers | Drivers in the traditional test group (pushing a button) were slower to perceive hazards but were faster at responding to test questions than participants in verbal and motor skills. As a special improvement, the way of verbal testing was differentiated between different types of groups. | 40 |
| 58 | Manley, et al.2020 (84) | Development and validation of hazard perception test for Thai drivers | Thailand | Cross-sectional study | Video | 87 Experienced drivers and 48 novice drivers | On average, experienced drivers perceived the hazard faster than novice drivers. Hazard perception test scores distinguish beginner and experienced Thai drivers. The reliability (Cronbach’s alpha coefficient) of this test was 0.93. | 38 |
| 59 | Castro, et al. 2020 (85) |
Hazard perception and prediction test for walking, cycling, and driving | Spain | Cross-sectional study | Video | 30 pedestrians, 14 cyclists, 13 novice drivers, and 22 experienced drivers | This test was able to differentiate between different studied conditions: A) Between traffic hazards that have been recorded from different perspectives: walking, cycling, and driving; B) Between participants with different user profiles: pedestrian, cyclist, and driver. |
36 |
| 60 | Arslanyilmaz2020 (86) | Improve hazard perception skills | Turkey | Cross-sectional study | Video | 22 high school students | The results showed that computer-based education could increase the level of hazard perception in high school students and improve their skills in this field. | 40 |
| 61 | Wu et al. 2021 (87) |
Development and validation of the Chinese version of the hazard Perception Test | China | Cross-sectional study | Video | 54 novice drivers and 47 experienced drivers | The test had very good internal consistency. Drivers who had driving errors scored lower than drivers who did not. This test measured hazard perception well. The reliability (Cronbach’s alpha coefficient) of this test was 0.86. | 38 |