The importance of the visual system as the input channel for sensory information necessary when driving is often stated1,3 The extent of this input channel, in comparison with other sensory input channels has, however, been debated.4 Nevertheless, and regardless of any attempts trying to quantify the extent of it, the visual system is regarded as crucial for driving.5 As an example of its importance, deterioration of the visual system has been used to account for the accident involvement of older drivers.6,7 The visual system has been investigated in relation to accident involvement also for novice drivers and drivers with some experience.8 It was found that there was a moderate relation between the observation task, i.e., the use of the visual scanning system, and accident involvement, and it was concluded that visual search strategies were a major area of concern. Anticipation of potential hazards, i.e., hazard perceptual skills, are developed trough visual search strategies9 as the driver gains experiences and can be assessed by data from the visual search strategies of the driver in real-world traffic.2
Visual search strategies can be detected by measuring eye movements. Analyses of eye movement data take into consideration the anatomy and physiology of the human visual system, as the measurements of the movements relate to what the foveal vision is directed toward. Foveal vision provides the driver with high-resolution information, which supports capabilities such as recognition.10 Peripheral vision supports capabilities such as the driver’s orientation, but without the driver being fully conscious of this process.11 When driving, the driver uses his foveal vision to detect directional cues, whereas the peripheral system is used to maintain lateral control of the vehicle.12 Peripheral vision also provides the driver with a wide range of visual information from which the foveal sampling of features takes place.13
The way to assess the development of these visual search strategies is based on several studies. Mourant and Rockwell14 found that the visual search strategies of novice drivers were unskilled. The novice drivers concentrated their visual search on a smaller area and closer to the vehicle. Miltenburg and Kuiken5 investigated the relation between traffic experience and visual search strategies in a laboratory setting, but the results did not support the findings of the Mourant and Rockwell study. Instead, they found that experienced drivers had a larger number of fixations on irrelevant cues than the novice drivers. This finding could reflect the fact that experienced drivers do not need to allocate as much of their foveal attention to the traffic environment as inexperienced drivers.15 Chapman and Underwood9 found that experienced drivers fixated lower and had less vertical variance in fixation locations than novice drivers. Crundall and Underwood16 found that experienced drivers select visual strategies according to the complexity of the roadway. For example, they found that experienced drivers increased their search in the horizontal meridian relative to the type of road. The strategies of novice drivers were found to be inflexible. The findings concerning horizontal search patterns confirmed an earlier study by Nagata and Masuda.17 Allen et al.18 found differences between experienced and novice drivers regarding amplitude, but not the frequency and length, of eye movements. However, no correlation between fixation durations and the number of fixations was established. In addition to these results, Falkmer and Gregersen2 also found that novice drivers fixate more often on in-vehicle objects than experienced drivers, further supporting the idea that there are differences in visual search strategies between inexperienced and experienced drivers.
Hence, there are several reports on differences in visual search strategies between experienced and inexperienced drivers, as well as in relation to the roadway. However, the results are not unambiguous, and they are not sampled by similar procedures. Based on these previous findings, the aim of the present study was to test all of the following hypotheses in one study, using a fixed test route, by assuming that inexperienced drivers, in comparison to experienced drivers: 1. Fixate closer to the vehicle, as shown by Mourant and Rockwell14; 2. Fixate more often on in-vehicle objects, as shown by Falkmer and Gregersen2; 3. Spread their fixations less along the horizontal meridian, as shown by Chapman and Underwood9; 4. Fixate more often on relevant traffic cues, with respect to number of fixations, as shown by Miltenburg and Kuiken5; and 5. Fixate more often on objects classified as potential hazards, as suggested by Miltenburg and Kuiken.5
The five hypotheses were tested and compared with respect to roadway, i.e., on a city route and on a rural route, based on the findings of Crundall and Underwood.16
Two groups participated in the study and were recruited on a “first-come, first-served” basis. The group “inexperienced drivers” consisted of 20 learner drivers, 18 men and two women, who had almost completed their driver education. Shortly after data collection, all the drivers in this group actually obtained a driving license. The mean age in this group was 19.9 years (standard deviation [SD] 5.5). All participants attended a local driving school.
The “experienced drivers” group consisted of 20 experienced, licensed male drivers. The inclusion criterion was that they had driven more than 100,000 km after obtaining their driving license at the age of 18 to 20 years. The mean age in this group was 34.6 years (SD 6.3). The participants were recruited from a vocational training center with a broad variety of educational programs.
Participation was voluntary and the individual data obtained were kept confidential. The subjects were given an oral presentation of the purpose of the study and were informed that they could leave the experiment whenever they wished. The Regional Ethical Committee of the Faculty of Health Sciences, University of Linköping, Sweden, had approved the outline of the study.
All subjects were given in-car instructions to perform a 30-min drive on a fixed test route in Sweden, in real traffic environments, while wearing an eye tracker, further described in the apparatus section. A driving instructor, sitting in the front passenger seat, gave instructions to the subject on where to drive. The car had dual brakes and an automatic gearbox. The whole test route comprised driving in urban complex traffic environments as well as suburban and less complex rural areas. Two route sections from the fixed route were chosen for analyses of the eye movements.
The first route section included driving in crowded city traffic, with a speed limit of 30 km/h, on a road section with parked cars on both sides of a one-way street. A combined road bump and zebra crossing was situated on this one-way street. It was expected that several pedestrians would be crossing this street and, hence, hazard perception would mostly be related to anticipation of such events. This route is called the “city route.”
The second route section included driving on a rural dual carriageway with a speed limit of 70 km/h, passing over a hill, and crossing a four-way intersection. After driving straight through the intersection, the drivers entered a 90° righthand turn into an avenue. Hazard perception would mostly be related to anticipation and detection of potential hazards both on the other side of the hill and as oncoming traffic in the four-way intersection. This route is called the “rural route.”
The two selected route sections each required approximately 30 to 50 s to drive. They were all driven in daylight and in nonrush-hour conditions. Particular transient safety threats were controlled for in the data analyses, but no test route was excluded as a result of this or any other reason.
Subjects’ eye movements were measured on both eyes simultaneously using a head-mounted eye tracker, NAC EyeMarkRecorder (EMR), model 60019,20, while driving a Chrysler Voyager equipped with an NTSC video recorder for data storage. The video-based data were analyzed using the NAC Data Processing Unit linked to a PC to establish fixation data.
Data Analysis in the Present Study
Fixations are identified from the eye movement data. They are generally defined as mean x,y coordinates of the fixation points, yielding an area of a·a°, lasting a minimum of t ms. The fixation point is assumed to reflect the focus of attention. The fixation identification procedure in the present study is described subsequently. The methodology was adopted from previous work by Falkmer and Gregersen.2 In the present study, both a qualitative and the quantitative approach were used.
The t value in this study was set at 100 ms. Furthermore, an a-value, larger or equal to 1·1°, was used2 based on the fact that foveal vision is restricted to a visual angle of approximately 1° around a fixation point. Hence, the Data Processing Unit was set to recognize fixations in which at least three consecutive data samples, measured at 30 Hz, fell within a minimum of 1·1° of each other using a weighted centroid procedure.20 This gave minimum fixation duration of 100 ms and allowed for pursuit tracking to be classified as a fixation.
The video-based data were analyzed frame by frame for those sets of frames clustered into a fixation. The fixation durations were noted. Each fixation was then labeled in one of three categories to identify the focus of attention of the subject while driving. The first category concerned the object that was fixated, the second category the area or the background of the object that was fixated, and the third category the distance of the object from the driver. The classification of each fixation was based on a classification matrix comprising 81 different categories of objects, nine of them being classified as not traffic relevant; and 72 different categories of areas or backgrounds, 34 of them being classified as not traffic-relevant, divided into nine subsections, i.e., the road, bicycle path, sidewalk, road verge, vegetation, buildings by the road side, the sky, and in-vehicle areas. Additionally, four different distances, i.e., <0.60 m, e.g., in-vehicle objects, >0.60 m, <5.0 m, e.g., area a, and b, in Figure 1, and >5 m, were included. Objects closer than 5 m were further possible to be verified as such using the eye parallax of the subjects relative to the objects. The matrix was further used for classification of potential hazards taking the objects, the areas, and the distances to the vehicle into account for the determination. The principles for scoring and categorizing eye fixations were adopted from the "object–space–time" relations scoring system.21
In a second step, fixations were assigned to one of the two categories: traffic-relevant and not traffic-relevant fixations based on the classification matrix, further verified by manually checking the clustered video frames. A traffic-relevant object would typically be a pedestrian entering the driving lane, whereas a not traffic-relevant object would typically be a commercial board far away from the roadside or a bird in the sky. Furthermore, fixations were also classified as relevant for hazard perception or not by the same method.
All data were compared between the group of inexperienced and experienced drivers. Moreover, all data were also compared within each group and for the whole study population with respect to the two route sections, i.e., the “city route” and the “rural route.”
Statistical Methods and Alpha Level
The alpha level for the χ2 test, handling the nonparametric fixation data, was set at 0.05.
In total, 6686 fixations were analyzed. Of these, 3417 (i.e., 49%) were fixations made by the group of inexperienced drivers and 3269 (i.e., 51%) were fixations made by the group of experienced drivers. A total of 2490 fixations (i.e., 37%) were made while the subjects drove the “city route.” The remaining 4196 fixations (i.e., 63%) were made while the subjects drove the “rural route.”
Fixations Closer to the Vehicle
To determine whether or not inexperienced drivers fixate closer to the vehicle than experienced drivers, the number of fixations on an object further away than 5 m for each group was compared with the number of fixations on objects closer than 5 m. Three fixations were classified as impossible to define with respect to the distance of the fixated object and were thus excluded from the analysis. Table 1 shows that a vast majority of the fixations were on objects further away than 5 m, in total 90% (5981). It was found that experienced drivers had more fixations 378 on objects closer than 5 m than inexperienced drivers (324; χ2 = 7.633, df = 1, p < 0.01). The inexperienced drivers had more fixations further away (2018) on the “rural route” than on the “city route” (1072; χ2 = 7.999, df = 1, p < 0.01). The same holds true for experienced drivers (1833 vs. 1058; χ2 = 65.532, df = 1, p < 0.001). In fact, on the “rural route,” the amount of fixations on objects further away than 5 m was equal between the two groups, i.e., 92%.
To summarize, hypothesis 1 was not verified with respect to these results.
Fixations on In-Vehicle Objects
To determine whether or not inexperienced drivers direct their focus of attention more often than experienced drivers to in-vehicle objects such as the speedometer, mirrors, and radio, the number of fixations on objects closer than 0.60 m was compared with the number of fixations on objects further away than 0.60 m. It was found that inexperienced drivers fixated more often on in-vehicle objects (5%) than experienced drivers (2%; χ2 = 40.510, df = 1, p < 0.001), as shown in Table 1. No significant differences were found for any group between the “city route” and the “rural route” with respect to the number of fixations on in-vehicle objects, as also noted in Table 1.
To further elaborate data concerning fixations on identified objects closer than 0.60 m, the number of fixations related to objects on the dashboard was compared with the number of fixations on other objects within this close distance. The results are presented in Table 2.
As shown in Table 2, inexperienced drivers had more fixations on the dashboard than experienced drivers (χ2 = 38.214, df = 1, p < 0.001). However, experienced drivers were able to adapt their number of fixations on the dashboard according to the route. On the “rural route,” the number of fixations experienced drivers made on the dashboard was higher than on the “city route” (χ2 = 7.410, df = 1, p < 0.01). The inexperienced drivers were not able to do this.
To summarize, hypothesis 2 was verified with respect to these results.
Fixations on Identified Areas
To further elaborate data concerning hypothesis 1, fixation data for the “city route” and the “rural route” were analyzed with respect to two identified areas, named area a (focus of expansion) and b (close to the car in the middle of the driving lane), shown in Figure 1.
The number of fixations on area a and area b for the “city route” and the “rural route” are presented in Table 3.
It was found that inexperienced drivers had a relatively higher number of fixations on area b than on area a compared with experienced drivers (χ2 = 52.026, df = 1, p < 0.001). This may be interpreted as indicating that the focus of the visual “attention” was closer to the vehicle compared with experienced drivers. The focus of the visual “attention” was closer to the vehicle when the inexperienced drove on the “city route” compared with when they drove the “rural route” (χ2 = 6.461, df = 1, p < 0.05). The same was true for experienced drivers (χ2 = 38.900, df = 1, p < 0.001). All these results are presented in Table 3.
To summarize, hypothesis 1 was verified with respect to these results.
Fixations Along the Horizontal Meridian
To determine whether or not inexperienced drivers spread their fixations less in the horizontal meridian than experienced drivers, the number of fixations directed to areas b (close to the car in the middle of the driving lane) and c (right hand road side) in Figure 1 was compared between the two groups. The results are presented in Table 4.
Inexperienced drivers had more fixations on the righthand roadside and the verge, including the combined road bump and zebra crossing, i.e., area c, than experienced drivers (χ2 = 16.202 df = 1, p < 0.001). However, experienced drivers were able to adapt their number of fixations on area b and area c to the route. On the “city route,” the number of fixations experienced drivers made on area c was higher than on the “rural route” (χ2 = 7.111, df = 1, p < 0.01). The inexperienced drivers were not able to do adapt this way.
The number of fixations along the horizontal meridian was analyzed further with respect to the fixations on objects related to the crossing roads in the intersections on the “rural route.” Fixations were categorized according to area a (focus of expansion) and b (close to the car in the middle of the driving lane), shown in Figure 1, but for both ways, i.e., to the left and to the right on the crossing road. In total, 408 fixations were analyzed. It turned out that experienced drivers had 28% fixations on area a and 72% on area b, whereas inexperienced drivers had 5% on area a and 95% on area b (χ2 = 41.977 df = 1, p < 0.001).
To summarize, hypothesis 3 was verified with respect to these results.
Number of Fixations on Relevant Traffic Objects
The fixations classified as relevant to the traffic scene were also analyzed with respect to the number of fixations on relevant traffic cues. As stated previously, the vast majority of fixations were on traffic relevant objects, as shown in Table 5.
A comparison between inexperienced and experienced drivers showed that inexperienced drivers had a relatively higher number of fixations on traffic-relevant objects (χ2 = 45.912, df = 1, p < 0.001). A further analysis revealed that inexperienced drivers showed relatively higher number of fixations on traffic-relevant objects on the “city route” than on the “rural route” (χ2 = 26.988, df = 1, p < 0.001). The same was true for experienced drivers as well (χ2 = 18.479, df = 1, p < 0.001).
To summarize, hypothesis 4, was verified with respect to these results.
Number of Fixations on Objects Classified as Potential Hazards
The fixations on objects classified as potential hazards were analyzed with respect to the number of fixations on these objects. The vast majority, i.e., in total 80%, of the fixations were on objects classified as potential hazards, as shown in Table 6.
A comparison between inexperienced and experienced drivers showed that inexperienced drivers had a relatively higher number of fixations on objects classified as potential hazards than experienced drivers, as shown in Table 6 (χ2 = 28.366, df = 1, p < 0.001). A further analysis revealed that inexperienced drivers showed relatively higher number of fixations on objects classified as potential hazards on the “city route” than on the “rural route” (χ2 = 23.576, df = 1, p < 0.001). The same was true for experienced drivers as well (χ2 = 65.089, df = 1, p < 0.001).
To summarize, hypothesis 5 was verified with respect to these results.
The Type of Routes With Regard to Cognitive Load
In general, the two analyzed routes, i.e., the “city route” and the “rural route” were apparently routes that generated a high demand on the drivers regardless of whether they were experienced or not. The share of fixations that could be attributed to traffic-relevant objects was high on both routes, i.e., approximately 90%. This share is approximately 45% higher compared with previous investigation, using the same methodology but other routes.2 This meant that also the number of fixations on objects classified as potential hazards was high, i.e., 80%. For this reason, it is essential, before discussing the other obtained results, to analyze these facts.
It could be that on the “city route,” the drivers were entering a generally crowded and busy one-way city street, in which it could be expected to find pedestrians crossing the street from both sides. In addition, a combined road bump and zebra crossing halfway through this route made it necessary to brake to ensure that the speed was adjusted to the potential damage that the road bump would cause if passed too fast. At the same time, the driver had to check for pedestrians. Moreover, during the whole “city route,” except for the zebra crossing, the drivers also had to pay attention to possible drivers or passengers stepping out of their parked cars on both sides of the road. This means that hazard perception, as such, does not only refer to objects that constitute a hazard per se, but also to objects that may constitute hazards, e.g., car doors and gaps between parked cars from which a car occupant or a pedestrian crossing the street may emerge.
On the “rural route,” passing an intersection makes hazard perception essential. Furthermore, after passing the intersection, this route included driving straight for 200 m while preparing for a 90° righthand turn into an avenue. This probably meant that there was little time to allocate the “visual attention” to something other than traffic-relevant objects and to objects classified as potential hazards. To summarize, the two selected routes generated high demands on the driver’s “visual attention” with respect to fixations. The high cognitive load of the driver, in turn, probably affected the drivers’ visual search strategies. This fact should be borne in mind when scrutinizing the results.
Inexperienced and Experienced Drivers
The findings of Mourant and Rockwell14 that inexperienced drivers concentrate their visual search to a smaller area, closer to the front of the vehicle, compared with experienced drivers, were not completely confirmed by the present study. By analyzing the results from the calculation of the distances to all fixated objects, and in addition comparing the number of fixations on areas a and b in Figure 1, it may be concluded that inexperienced drivers tended to locate their focus of attention more on in-vehicle objects and on area b, i.e., in a lower vertical position, than experienced drivers. The analysis of fixated objects further away than 5 m did, however, show that experienced drivers had more fixations on objects closer than 5 m than inexperienced drivers. This finding may seem strange in comparison of the findings by Mourant and Rockwell14 and the results from a previous study using the same methodology.2 However, of the fixated objects on distances shorter than 5 m, the experienced drivers had only 2% on objects closer than 0.60 m, i.e., in-vehicle objects, whereas the inexperienced drivers had 5% of their fixated objects on this shorter distance. In addition, the percentage of fixated objects closer than 5 m was actually the same between the two groups of drivers on the “rural route,” indicating that it was on the “city route” that the experienced drivers fixated objects closer than 5 m, but further away than 0.60 m, to a higher degree than inexperienced drivers. Bearing in mind the combined road bump and zebra crossing, the potential hazard of crossing pedestrians between the parked cars and the risk of sudden opening of car doors, this finding may be interpreted that experienced drivers actually adapted their visual search strategy to the type of road and the traffic conditions to a higher degree than inexperienced drivers by fixating the potential hazards on the “city route.” However, this does not explain why there were no differences between the two groups on the “rural route.”
The fixations on in-vehicle objects were further scrutinized and revealed that experienced drivers were able to adapt their number of fixations on the dashboard according to the type of road. In this case, they had a higher number of fixations on the “rural route,” probably as a result of the difficulty in estimating the speed when driving alone on a rural road, which sometimes was the case. This ability to adapt according to the type of road was not evident for inexperienced drivers.
Area a in Figure 1 represents an area in which experienced drivers usually allocate their “visual attention” to a higher degree than inexperienced drivers.14 This area is also traditionally being pointed out as a preferable for visual search strategies in driving schools in comparison to area b.1 As expected, experienced drivers had more fixations in area a compared with the inexperienced drivers. However, the total number of fixations in area a was low. This may be attributable to the fact that only in a limited part of the “city route” was it at all meaningful to fixate objects in area a. On the other hand, on the “rural route,” it was more meaningful to fixate objects in area a, which was also done by the experienced drivers to a higher degree than by inexperienced drivers.
The more flexible search strategies of experienced drivers, found in the previous the study,2 include a wider spread of horizontal search. The present study did not confirm this. On the contrary, inexperienced drivers fixated more often in area c in Figure 1, compared with area b, than experienced drivers. The inexperienced drivers were, however, not able to adapt their horizontal visual search strategies to the type of road and to the traffic conditions. On both routes, they had the same percentage of fixations on objects in area c compared with area b, whereas the experienced drivers had roughly the same percentage on the “city route” as the inexperienced drivers but fewer on the “rural route.” This result may be attributed to a more developed and flexible visual search strategy among the experienced drivers as a result of the fact that on the “rural route,” no immediate hazard could be expected or occurred in area c. Instead, it could be assumed that the higher percentage of fixated objects in area c among the inexperienced drivers could be attributed to a greater need for this group to find visual cues to enhance lane positioning rather than detecting potential hazards. This reasoning does not, however, apply to the intersection part of the “rural route.” Hence, the visual search strategies were examined with respect to intersections, revealing that experienced drivers spread their fixations horizontally much wider compared with inexperienced drivers, which is consistent with previous findings.2,16
Both experienced and inexperienced drivers had a relatively higher number of fixations on traffic-relevant objects on the “city route” than on the “rural route.” Inexperienced drivers were found to have a higher number of fixations on traffic-relevant objects than experienced drivers, which is in accordance with previous findings.2,5
The validity of the NAC 600 eye tracker data may be low with respect to measurement of fixation duration. However, the NAC cornea reflex system has been subjected to a validation study and has been used in several other studies,19,20,22,23 in which no such deficiency in data validity was found. Another explanation may be that by not predefining the a·a° values for fixation definition, to achieve an individual maximum of fixation for each route, data representing saccades may have been wrongly classified as fixation data. This is, however, not plausible as a result of the fact that every fixation was viewed to be able to assign it to an object, an area, and a distance. This means that a misclassification of a saccade into a fixation would have been discovered during this qualitative process.
When, like in this study, an object-based classification system is used, it is important to have a high resolution in data concerning both spatial and temporal aspects. This is of special importance when comparing experienced and inexperienced drivers, because the level of experience in itself may influence the understanding of the traffic environment with respect to when and for which object attention is needed, something that an inexperienced driver is less capable to do. Even if the matrix used in this study has a high resolution with, in total, 81 different categories of objects and 72 different categories of backgrounds, it still does not fully provide unambiguous analyses with respect to spatial and temporal aspects. This means that misclassification with respect to potential hazards may have occurred in the present study. However, as a result of the fact that surprisingly high numbers of fixations on potential hazards were found, this poses no major problems to conclusion drawn from the results. Future research should try to develop the matrix with respect to this issue.
This study provides normative data for the understanding of the development of visual search strategies among drivers. Furthermore, the methodology used in the present study, i.e., to combine a quantitative analysis with a qualitative analysis, proved to be useful to compare visual search strategies among inexperienced and experienced drivers.
This study was part of the TRAINER project GRD1-1999 to 10,024 in the EU Growth program under DgTren coordinated by Dr. Evangelos Bekiaris, Associate Professor at HIT, Greece, to whom the authors owe a great gratitude. Benny Nielsen, formerly at Körkort Handikapp, Lernia, now at SweRoad, and Per Henriksson at VTI both made major contributions to the study.
Torbjörn Falkmer, PhD
Department of Neuroscience and Locomotion
Faculty of Health Sciences
Linköping University, SE-581 85
1. Falkmer T, Gregersen NP. Deliverable D6.1for the TRAINER Project GRD1-1999-10024 as Part of the European Union GrowthProgramme: Pilot Plans. Bruxelles: DG Tren; 2001.
2. Falkmer T, Gregersen NP. Fixation patterns of learner drivers with and without cerebral palsy (CP) when driving in real traffic environments. Transportation Res Part F: Traffic Psychol Behav 2001;4:171–85.
3. Spijkers W. Distribution of eye-fixations during driving. IATSS Res 1992;16:27–34.
4. Sivak M. The information that drivers use: is it indeed 90% visual? Perception 1996;25:1081–9.
5. Miltenburg PGM, Kuiken MJ. The Effect of Driving Experience on Visual Search Strategies. Result of a Laboratory Experiment. Haren: University of Groningen; 1990.
6. Owsley C, Ball K, Sloane ME, Roenker DL, Bruni JR. Visual/cognitive correlates of vehicle accidents in older drivers. Psychol Aging 1991;6:403–15.
7. Maltz M, Shinar D. Eye movements of younger and older drivers. Hum Factors 1999;41:15–25.
8. Groeger JA, Kuiken MJ, Grande G, Miltenburg PGM, Brown ID, Rothengatter JA. Preliminary Design Specifications for Appropriate Feedback Provision to Drivers With Differing Levels of Traffic Experience. Haren: The Netherlands: Traffic Research Centre, University of Groningen; 1990.
9. Chapman PR, Underwood G. Visual search of driving situations: danger and experience. Perception 1998;27:951–64.
10. Samuelsson S, Nilsson L. Om möjligheterna att upptäcka och identifiera perifert presenterad information i bilen [in Swedish]. Linköping: VTI; 1996.
11. Leibowitz HW. Recent advances in our understanding of peripheral vision and some implications. Proceedings of the Human Factors Society 30th Annual Meeting; Santa Monica, CA; Human Factors Society; 1986:605-7.
12. Rockwell RH. Eye movement analysis of visual information acquisition in driving. Paper presented to the Australian Research Board; 1972.
13. Lansdown TC. Visual Demand and the Introduction of Advanced Driver Information Systems Into Road Vehicles [PhD thesis]. Loughborough: Loughborough University; 1996.
14. Mourant RR, Rockwell TH. Strategies of visual search by novice and experimental drivers. Hum Factors 1972;14:325–35.
15. Hughes PK, Cole BL. The effect of additional demand on eye movement behaviour when driving a car. In: Gale AG, ed. Vision in Vehicles II—Proceedings of the Second International Conference on Vision in Vehicles; Nottingham, UK; 14-17 September 1987. Amsterdam: North-Holland; 1988:221–30.
16. Crundall DE, Underwood G. Effects of experience and processing demands on visual information acquisition in drivers. Ergonomics 1998;41:448–58.
17. Nagata M, Masuda K, Kuriyama H. Drivers’ visual behavior as affected by stopping maneuver. SAE Technical Paper 870236. Warrendale, PA: Society of Automotive Engineers; 1987.
18. Allen JA, Schroeder SR, Ball PG. Effects of experience and short-term practice on drivers’ eye movements and errors in simulated dangerous situations. Percept Mot Skills 1978;47:767–76.
19. Kielgast K. The reliability of the NAC-V—a preliminary test. In: O’Reagan JK, Levy-Shoen A, eds. Eye Movements: From Physiology to Cognition: Selected/Edited Proceedings of the Third European Conference on Eye Movements; Dourdan, France; September 1985. Amsterdam: North-Holland; 1987:656-7.
20. Kielgast K. Report on the Validity of the NAC-600. An Analysis of Data and Fixation Generation During Car-Driving. Unpublished report. Copenhagen: Department of Psychology, University of Copenhagen; 1994.
21. Nygaard B. ‘Object–space–time’ relations scoring system. In: Road user Perception 1974-79. Copenhagen: Danish Council for Road Safety Research; 1979.
22. Jacobs AM, O’Regan JK. Spatial and/or temporal adjustments of scanning behavior to visibility changes. Acta Psychol (Amst) 1987;65:133–46.
23. Miura T. Active function of eye movement and useful field of view in a realistic setting. In: Groner R, d’Ydewalle G, Parham R, eds. From Eye to Mind: Information Acquisition in Perception, Search and Reading. Amsterdam: North-Holland; 1990:119–127.
24. Cohen AS. Is the duration of an eye fixation a sufficient criterion referring to information input? Percept Mot Skills 1977;45:766.
25. Miura T. Visual search in intersections—an underlying mechanism. IATSS Research 1992;16:42–9.
This article has been cited 12 time(s).
International Journal of Sports Science & Coaching
Translating Key Methodological Issues into Technological Advancements When Running In-Situ Experiments in Sports: An Example from Sailing
International Journal of Sports Science & Coaching, 8(1):
Accident Analysis and PreventionThe view from the road: The contribution of on-road glance-monitoring technologies to understanding driver behaviorAccident Analysis and Prevention
Accident Analysis and PreventionThe visual control of bicycle steering: The effects of speed and path widthAccident Analysis and Prevention
Accident Analysis and PreventionA comparison of static and dynamic hazard perception testsAccident Analysis and Prevention
Anales De PsicologiaVisual search strategies in expert and novice drivers during the perception of driving scenesAnales De Psicologia
Accident Analysis and PreventionThe visual search patterns and hazard responses of experienced and inexperienced motorcycle ridersAccident Analysis and Prevention
Injury PreventionLegislative advocacy is key to addressing teen driving deathsInjury Prevention
Perceptual and Motor SkillsFixation identification in centroid versus start-point modes using eye-tracking dataPerceptual and Motor Skills
PediatricsNational young-driver survey: Teen perspective and experience with factors that affect driving safetyPediatrics
Transportation Research Part F-Traffic Psychology and BehaviourThe development of hazard perception: Dissociation of visual orientation and hazard processingTransportation Research Part F-Traffic Psychology and Behaviour
Transportation Research Part F-Traffic Psychology and BehaviourThe effect of the driving activity on target detection as a function of the visibility level: Implications for road lightingTransportation Research Part F-Traffic Psychology and Behaviour
Human FactorsLooks are (almost) everything: Where drivers look to get informationHuman Factors
visual search strategies; driver education; eye movements; fixations; driver experience
© 2005 American Academy of Optometry