The unformatted htm text below is from
|Copyright © 2004 by Leonard Evans|
|The text below is © Leonard Evans. Portions totaling not more than 400 words my be freely used with appropriate citation.|
|For more extensive use, permission in writing must be first obtained by emailing the author.|
8 Driver performance
This html version contains only the text (no figures, tables, equations, or summary and conclusions). To check printed book appearance see pdf version of Chapter 1 or pdf version of Chapter 16.
In this chapter we explore the elements that constitute the driving task and their relationship to safety. We use the term driver performance to refer to the driver's knowledge, skill, and perceptual and cognitive abilities. This is distinct from how the driver actually uses these attributes, for which we use the term driver behavior, the subject of the next chapter.
Components of the driving task
The driving task is a closed-loop compensatory feedback control process, meaning that the driver makes control inputs (to the steering wheel, brakes, and accelerator pedal), receives feedback by monitoring the consequences of the inputs, and in response to these consequences, makes additional inputs. An open-loop process is one in which additional inputs cannot be applied after initiating the process, such as throwing a ball. The ball's trajectory cannot be changed once it has left the hand.
When decomposed into fine detail, the driving task has much complexity, involving as it does the simultaneous control of lateral and longitudinal position through the use of steering wheel, accelerator, and brakes, together with many pattern recognition and other higher level cognitive skills, such as estimating future situations from present information. While the basic skills required to propel a vehicle are usually learned quickly and with ease, some of the higher-level skills that affect safety can be acquired only after many years of experience.
Predominance of visual feedback
The feedback used to monitor driving is overwhelmingly visual. I see no reason to dissent from a 1972 statement that vision provides over 90% of information used to drive. (p 150) Drivers tend to ignore information on signs, or even be unaware of a sign's existence, if the relevant information can be derived directly from the driving environment. The driver's preferred mode of operation is to pursue a visual search, and resort to other information sources only when problems arise, perhaps somewhat like the way most people consult owners' manuals only after preferred methods of trying to solve problems have failed.
The preponderance of visual information over that from all other senses probably increases yet further with increasing skill levels. For example, proprioceptive cues (those from the force and position of hands and arms in supplying control inputs) are of minor importance, and, surprisingly, are even less likely to be noticed by more experienced than less experienced drivers. A skilled driver is relatively unaware of the gain in the steering system (the amount the steering wheel must be turned to alter the vehicle's direction by a given angle). When transferring to vehicles with different steering system gains, experienced drivers do not travel more, or less, sharply around corners, or have difficulty maintaining lane position. Instead, they react to the visual information by making the steering input necessary to achieve the desired visual result without being much aware how much they moved the wheel and in such a manner that there are no observable changes in the trajectory of the vehicle. Similar comments apply to different force characteristics, or, in the extreme, to power versus manual steering. Less experienced drivers are more aware of steering system gain and force-feel characteristics, and their driving can be noticeably influenced by changing them.
The dominance of visual feedback in driving is similar to dominance of aural feedback in the playing a stringed musical instrument. Intonation (playing in tune) is not controlled by the proprioceptive sense of remembering where to place the fingers, but by listening to the sounds produced. A learner trained on an instrument of one size will play one of a different size (on which all the finger placements are different) more out of tune, whereas a skilled player will be less aware that there is even a difference, just as in the steering gain case.
Given the predominance of the visual sense in driving, one might expect that visual performance and crash risks would be intimately related. Innumerable studies over many decades have failed to show any clear relationship between the most basic measure of visual performance, visual acuity, and crash risk. Crash rates decline to a minimum at about age 45, by which time visual acuity and contrast sensitivity have already begun to decline, as have other visual capabilities relevant to driving, such as the ability to withstand glare.
Even so dramatic a visual impairment as the non-use of one eye does not have an overwhelming effect on safety, although it has been shown to have some influence. , However, the magnitude is sufficiently modest, and indeed uncertain, that a strong case is made that monocular drivers should not be excluded as racing drivers. Although US inter-state truck drivers are subject to stringent license requirements, the agency responsible for licensing them approves licensing monocular drivers.
Changes in higher-level visual characteristics, in particular the useful field of view, the area from which useful visual information can be extracted in a single glance, has been shown related to crash involvement risk. Pattern recognition skills are central to driving task. From a loosely-structured, but stimuli-rich, visual environment the driver must select information that is relevant from much that is not.
Judgment of speed
Of the various quantities a driver is called upon to judge, speed is the only one for which instrumented quantitative feedback must, by law, be available. Each time a driver consults a speedometer, perceived speed can be compared to actual speed. Such consultations are additionally motivated by the need to obey speed limits. The repetitive practice, with feedback, of this task might suggest that drivers would become very good at estimating their speed. Many studies have examined the extent to which this is so.
In an experiment on a British test track, drivers of cars with obscured speedometers were instructed to double or halve an initial speed, the magnitude of which was known only to the experimenter. The subjects' attempts to halve or double the initial speeds were biased by large amounts in the direction of the initial speed. For example, the goal of doubling an initial speed of 30 mph produced an average speed of 44 mph, while the goal of halving 60 mph produced 38 mph. In a study in Japan, drivers instructed to travel at their chosen speeds on closed roads drove, on average, 3 km/h faster when the speedometer was concealed.
Subjects in other speed-estimation experiments traveled as passengers in vehicles with speedometers visible only to the drivers who conducted the experiments. This allowed greater task flexibility. Subjects instructed to keep their eyes straight ahead consistently underestimated the speed at which they were traveling.9 The instructions more specifically asked subjects to fixate on the focus of expansion, the geometrical point from which a straight road appears to emerge as one travels forward. Two studies asked subjects to estimate speed without telling them where to look. Speeds were estimated without large average systematic errors; the errors averaged over all subjects tested was typically less than 5 km/h. , When hearing was restricted, both studies found systematic speed underestimation, typically by about 8 km/h.10,11 Further evidence that hearing can play a role in estimating speed is provided by the ability of blindfolded passengers to judge speed without systematic error,11 and by decreased ability of subjects in a driving simulator to maintain set speeds when auditory cues were removed.
While the above experiments indicate that hearing can play a contributory role in estimating speed, it is the changing size and position of objects in the visual field that provide the main cues to speed, and variations in these can generate different sensations of motion. For example, a geometric pattern of bars with decreasing spacing on a roadway produced a sensation of increasing speed, which in turn led drivers to reduce speed. This concept has been applied, for example, to slow traffic in work zones. The main cue for speed comes from peripheral vision. When peripheral vision is eliminated leaving only the central field of view to determine speed, estimates become inaccurate because the vehicle's forward movement produces little change at the focus of expansion.
A sensation familiar to nearly all drivers is that after prolonged driving at high speeds, slower speeds seem even slower than they really are. This phenomenon, referred to as speed adaptation, has been examined in a number of studies. In one, subjects were instructed to drive at 70 mph for specified distances, and then, without guidance from a speedometer, slow down to 40 mph. It was found that the longer the exposure to 70 mph, the higher is the speed later produced to represent 40 mph. After 40 miles driving at 70 mph, the average driver slowed to 53 mph in response to the request to produce 40 mph. A simulator study found that a subject's selection of a target speed is highly influenced by the subject's previous speed. After simulated driving at about 70 mph for three minutes, subjects underestimated a simulated 30 mph by between 5 to 15 mph; the perception that the speed was lower than actual persisted for at least 4 minutes.
Another approach to examining speed adaptation is to observe, in traffic, groups of vehicles that previously have been traveling at different speeds. Speeds of vehicles traveling in opposite directions on a four-lane divided highway were compared. One direction of traffic had been exposed previously to expressway speeds of about 60 mph, while vehicles in the other direction had been exposed to about 40 mph. For each of seven categories of vehicles examined, higher speeds were observed for those exposed to the higher prior speed. The magnitude of the effect is that those previously exposed to 60 mph traveled about 7% faster than those exposed to 40 mph. It is not possible to determine to what extent this difference is due to speeds being perceived differently, or to drivers merely tending to continue driving close to their prior speeds because of behavioral inertia. This distinction was addressed in another study using sites that required drivers to slow down or stop prior to entering the section of roadway on which their speeds were measured. The observed effects were about half of the 7% observed without the slow-down or stop. It is, however, worth noting that the act of slowing down after prolonged freeway driving may itself influence the speed adaptation phenomenon, in that the prior speed becomes not the freeway speed, but the briefly experienced low or zero speed.
The tendency to drive faster on a given road because of prior high speeds on a different road, regardless of the extent to which it is due to perceptual biases in speed estimation or to speed perpetuation, has important safety implications. Through this phenomenon, speed limits, and changes in speed limits, may have spillover effects that influence safety on roads other than the ones directly affected. There are many indications that the 1974 reduction in the speed limit on US Interstate highways from 70 mph to 55 mph led to reductions in speeds on other roads with unaltered speed limits, and that this spillover effect is responsible for some of the reduction in fatalities from 54,052 in 1973 to 45,196 in 1974. The 16% drop is the largest yearly decline ever recorded in peacetime in the US. After 1987, when the US Congress relaxed, and in 1995 removed, the 55 mph limit, increased speed limits on rural sections were associated with higher speeds on urban sections with unchanged limits.
Speed adaptation appears to be largely a perceptual illusion not unlike many optical illusions in which how part of a simple drawing is perceived is greatly influenced by adjacent parts of the drawing. As visual training and experience do not make optical illusions disappear, it seems unlikely that experience or training would make speed adaptation disappear. This underlines the importance of speedometer use, especially when exiting freeways after prolonged travel, or when traveling on streets with low speed limits after traveling at higher speeds. The speedometer provides important information that drivers are unable to obtain using only their unaided senses.
Judgment of relative speed
Much driving is spent following vehicles that are following other vehicles. The field of traffic science originated in elegant mathematical descriptions of vehicle following. Each vehicle (except the lead) in a platoon of vehicles is assumed to react, after a time delay, to a stimulus arising from its relationship with the vehicle it is following. A typical time delay for test track experiments is 1.6 s. The reaction is an acceleration or deceleration. Various forms of the stimulus have been explored, but the one most successful at explaining a great deal of experimental data is the relative speed divided by the spacing.
Drivers' abilities to judge relative speeds have been measured in a number of experiments. In keeping with the results from the vehicle-following experiments, it is found that the ability to judge relative speed is approximately inversely proportional to inter-vehicle spacing. This is consistent with drivers reacting to changes in the perceived area of the followed vehicle rather than to changes in a linear dimension.
The ability to judge the sign of relative motion in a car-following situation was investigated by occluding the vision of subjects who rode in the right-front passenger seat of an instrumented car that followed another instrumented car on a freeway. When the experimenter in the following car judged that the relative speed between the vehicles to be sufficiently close to zero to make judging its sign difficult, the subject was permitted to see the lead car for four seconds. The subject's task was to indicate whether the vehicles moved closer together or further apart. Instructions called for a forced choice - one or other response was required for each stimulus. As is common in forced choice experiments, even for stimuli so small that subjects indicated that they were only guessing, correct responses were in fact well above the chance level.
One surprising result of this experiment was a highly consistent bias in favor of judging that the cars were approaching when they were not. This bias, in the direction of increased safety, is likely induced by peripheral vision cues related to the forward motion of the vehicle in which the subject is traveling. Because of the bias, which increased in magnitude with inter-vehicle spacing, it is not possible to express the results in terms of one threshold value because different values for positive and negative relative speed pertained at each spacing. However, the experiment showed high capabilities at judging the sign of relative motion. For example, if a lead car 60 m away is approaching the following car at 5 km/h, the following driver's probability of correctly judging that the vehicles are closing rather than pulling further apart was 0.99. The results show that it is unlikely that a factor in rear-end crashes is attentive drivers being unable to judge that they are approaching a lead car.
Judgment of spacing
People tend to be able to judge distance reliably over a wide distance range. The short distance cues of accommodation (the focusing of the eye's lens) and binocular disparity (the eyes having to aim more towards each other as viewed objects become nearer) are of little consequence in judging distances of objects outside a vehicle. Most distances that require judgment are in the range 5 m to 500 m. Many factors have been shown to influence spacing judgments. For example, size constancy, the built-in knowledge we have about the size of familiar objects. Vehicles that are larger are judged to be further away. The finding that approaching motorcycles appear further away than trucks provides a likely explanation for why drivers give smaller safety margins to the motorcycles.
Judgment of factors influencing spacing in car following was investigated by projecting static views of the rear of a lead car photographed from the driver's eye position of a following car. Subjects judged whether a particular view represented a greater or lesser inter-vehicle spacing than a standard view. It was found that the same distance was perceived to be greater when viewed from a vehicle with hood geometry that exposed more roadway between the vehicles. This was additionally confirmed by viewing from the same vehicle with its rear raised in order to make more roadway visible. The lead car is actually the same distance from the camera in both photographs in Fig. 8-1.
The finding that the same spacing is perceived to be different from different vehicles has safety implications. Say a driver familiar with a vehicle with a long hood transfers to one with a less obstructed view. If the driver follows at his or her normal perceived spacing, then the vehicle with the less obstructed view will be driven closer to the one followed. Such an effect was observed directly in test-track experiments in which small cars (with short hoods) were observed to follow at closer headways than large cars driven by the same drivers. The perceptual effect would cause drivers of sport-utility vehicles (SUVs) to follow closer than car drivers without knowing they were doing so. This could explain why one hears so many complaints that SUV drivers tailgate.
On a two-lane roadway the task of overtaking a lead vehicle in the face of an oncoming vehicle involves judging the distance of the oncoming vehicle, and the relative speed between the oncoming vehicle and the driven vehicle, which may be in excess of 200 km/h. Drivers' judgments and decisions in overtaking were investigated in extensive experiments conducted on one side of a completed but unopened four-lane section of Interstate freeway. Subjects in one car followed another, while a third car approached in an adjacent lane. It was found that while drivers make reliable estimates of the distance to the oncoming car, they are insensitive to its speed. Basically, at distances required for this task, cues to relative speed (mainly the angle subtended at the driver's eyes by the oncoming car) provide minimal information. When the subjects were informed of the speed of the oncoming car, passing occurred at smaller, and less varying, spacing. These results parallel findings that pedestrians base road-crossing decisions on how far away approaching vehicles are, rather than on their speed.
A follow-up overtaking study found that unsuspecting drivers on two-lane rural roads overtook slower moving cars with greater likelihood the greater the available passing distance, and the lower the speed of the lead car. At night, drivers were more conservative and more variable in the passing distances they were willing to accept than in daytime driving. The inability of drivers to estimate oncoming speed leads them to decline safe passing opportunities when the oncoming car is traveling slower than expected, and to initiate unsafe passing maneuvers when the oncoming car is traveling faster than expected.
Reaction times are influenced by many factors, but, for driving, the two most important are, first, the number of stimuli and possible responses, and second, expectancy. If a subject is instructed to fixate on an unlit lamp, and press a switch as soon as possible after it lights, then simple reaction times on the order of 0.15 s are generally recorded. If the number of stimuli and responses increase (say a number of lights, each with its own switch), then choice reaction times become progressively longer. If the lamp lights every few seconds, reaction times will be far shorter than if the lamp lights every few hours. Expectancy is crucial -- reaction times to expected events are short, to unexpected events much longer.
Reaction times in driving involve identifying a variety of events in a complex environment, so it is not surprising that reaction times bear little resemblance to the minimum possible in laboratory tests. Indeed, it is convenient, conceptually, to divide the time from stimulus to driver response into two phases, decision or perception reaction time (time to decide to brake, for example), and response or movement reaction time (time to place foot on brake pedal), even though they are generally observed as one composite reaction time. While there is fairly extensive literature on reaction times relating to driving, the most difficult factor to investigate, especially as it relates to crashes, is that of expectancy.
The reaction time that produced the best fit to the previously discussed car-following data is 1.6 s. It should be noted that this is for drivers specifically focusing on the car ahead in a test-track experiment. To address expectancy, an experiment was conducted in which young and old drivers of an instrumented vehicle suddenly encountered an object after traveling over a crest-vertical curve (a straight road traveling over a hill). On the first trial, the drivers had been driving for about 10 to 15 minutes, and the object was unexpected. In subsequent trials subjects knew the goal of the experiment, but the location of the object changed. Perception and response times were considerably longer for the trial in which the drivers were not alerted than for the subsequent ones. The older subjects had longer perception and reaction times than the younger, in keeping with much research that shows that reaction times increase with age. For all the subjects combined, the 95th percentile total reaction time for the trials in which drivers were not alerted was 1.6 s. However, the authors point out that while driving an instrumented vehicle with an experimenter present, a driver may be more alert than an average driver. They recommend the use of a reaction time of 2.5 s for surprised drivers, a value that is the common choice in US traffic engineering practice for such purposes as computing sight distances in freeway design.
Reaction times in normal driving were measured by presenting an unexpected stimulus to actual drivers in Finland through the use of a parked instrumented vehicle. When it was safe to do so, the door of this vehicle was opened presenting oncoming motorists with a view of the door close to, but not encroaching upon, the lane on which they were traveling. By means of eight pairs of infrared photocells, the moment at which the oncoming vehicle's trajectory first changed in response to the stimulus of the opened door was measured for 1,326 oncoming drivers. It was found that the average response time was about 2.5 s, with most response times being between 1.5 s and 4.0 s. Thus the 2.5 s value mentioned above finds additional support in this study, and is used in the following example constructed to bring out the importance of reaction time and stopping time.
An example illustrating reaction time and braking
Suppose a car traveling at speed v1 drives over the crest of a crest-vertical curve, and is suddenly confronted by a large obstruction completely blocking the roadway (say, an overturned truck blocking all lanes). Let distances from the crest of the hill be represented by x. The car will travel to d1 = v1T before braking commences, where T is reaction time. Assume that applying maximum braking imparts a constant deceleration, a. The speed, V1(d), is given by
where d is the distance traveled since braking commenced.
Let us proceed by assuming specific values. We take a = 5 ms-2, a reasonable value for good tires on dry level pavement (we ignore the hill which was for expository convenience only). This value is just over half the 9.8 ms-2 acceleration due to gravity. For reaction time we take T = 2.5 s, and for initial speed, v1 = 55 mph (89 km/h, or 24.6 m/s). The driver will begin to brake at x = d1 = 61.5 m, and the car will come to a complete stop (if it does not crash) at x = D1 where
The trajectory of this car is shown as the dashed line in Fig. 8-2. Also shown, as the solid line, is the trajectory of a second car that differs only in that its initial speed is v2 = 70 mph (113 km/h, or 31.3 m/s), the value used to compute d2 = 78.2 and D2 = 176.2.
Figure 8-2. Schematic representation of how the speed of a vehicle varies along a roadway from the location x = 0 at which a large obstruction inviting maximum braking first appears (top) and how the probability of driver fatality depends on the location of the obstruction (bottom).
If the obstruction is located at x > D2 neither car
will crash into it. If it is located between D1 and D2 then
the faster car will crash into it but the slower car will
not. If it is located at x < D1 then both cars will crash
into it, but the faster car at a higher impact speed. If it
is located at x < d1 both cars will crash into it at
their unaltered initial speeds v1 and v2.
If we assume that the obstruction on the highway does not move or crush when impacted by a car, then the striking car will experience a change in speed, or Dv, equal to its traveling speed on impact. Figure 4-7 (p. 72) shows that the probability, P, that an unbelted driver is killed is given approximately by P = (Dv/114)3.54 provided Dv < 114 km/h. This is used to compute the probability of death as a function of where the obstruction is located along the roadway, as shown in the bottom graph in Fig. 8-2.
If the impact occurs prior to any braking (x < d1) the driver of the faster car is (70/55)3.54 = 2.3 times as likely to be killed as is the driver of the slower car. If the impact occurs just as the faster car begins to brake (x = d2), the driver of the faster car is 4.2 times as likely to be killed as the driver of the slower car, which has slowed from 89 km/h to 75 km/h when it reaches d2. As x becomes greater than d2 the ratio of the risk to the faster driver to that to the slower driver becomes larger and larger until the risk to the lower-speed driver becomes zero at x = D1, while the faster driver still has some probability of death for D1 < x < D2.
The values of d1 and d2 are proportional to the reaction time, which was assumed to be 2.5 s. Outcomes are sensitive to this choice. If we chose a reaction time 10% shorter than this (T = 2.25 s) then, if the obstruction was at x = 90 m, the probability that the slower driver is killed decreases from its initial 13% to a lower 9%, and the probability that the faster driver is killed decreases from an initial value of 75% to a lower value of 65%. For x = 100 m the corresponding changes are from 7% to 4%, and from 61% to 51%.
This simple example illustrates three themes of central importance:
1. Small reductions in reaction time can produce large reductions in the probability and severity of crashes.
2. The probability of crashing increases with speed.
3. Given that a crash occurs, fatality risk increases steeply with speed.
In the next chapter relationships are provided suggesting that fatality crash risk is proportional to the fourth power of speed, so that traveling at 70 mph has a fatality risk (70/55)4 = 2.6 times the risk traveling at 55 mph, a ratio that is plausible in terms of the illustrative values presented above.
Rear impact crashes
As rear-impact crashes generally involve vehicles traveling in the same direction, with perhaps one of them stationary, they tend to be of below-average severity, accounting for 5% of US fatal crashes. However, a total of 1.9 million rear-end crashes occurred in 2000, accounting for 30% of all crashes. They also accounted for 30% of the crashes for which injuries were reported, even though some of these may be due more to litigation than impact, as discussed in Chapter 2.
Technology to reduce the risk of rear impact appeared as early as 1916 in the form of a rudimentary stop lamp. Because small reductions in reaction time promise large reductions in crash rates, there has been much research on refining the details of stop lamps. Such factors as light configuration, color, and brightness have been examined, as well as methods of indicating the magnitude of deceleration of the lead car. ,
Center high mounted stop lamps
A major change in alerting following drivers that a lead vehicle was braking occurred with the introduction of the center high mounted stop lamp, a red stop lamp mounted on the centerline of the rear of vehicles. It is generally higher then the other two side-mounted stop lamps, leading to a triangular configuration. Federal Motor Vehicle Safety Standard FMVSS-108 required that the system be installed on all new cars sold in the US after 1 September 1985. The required features of the system were determined based on a number of large-scale experiments in actual traffic. In the first, the experience of a fleet of Washington, DC taxicabs fitted with this type of device or other innovative stop lamps was compared to that of a control group of the same makes, models and driver characteristics, but with the conventional stop lamps of the time. Drivers reported details of all crash involvements. The study analyzed changes in the number of impacts on the rear during braking -- the only type of crash subject to potential influence from changing stop lights. In the field tests, 67% of the taxis struck in the rear were struck while braking. The key finding in the experiment is that the Washington taxicabs with center high mounted stop lamps were struck in the rear while braking 54% less often for the same distance of driving as the taxis in the control group.
A follow-up study used 5,400 telephone company passenger vehicles driven 55 million miles during a 12-month period in locations scattered widely throughout the US. For the same distance of driving, the 2,500 vehicles equipped with center high mounted stop lamps were struck in the rear while braking 53% less than those not so equipped. Another study found a 51% reduction.
The three studies find close agreement that center high mounted stop lamps reduce the risk of being rear-impacted while braking by about 50%. Since about two-thirds of all rear impact crashes involve pre-impact braking by the lead vehicle, these results are equivalent to a 35 percent reduction of rear-impact crashes of all types.
Based on such large risk reductions, the devices were mandated for all cars, and effectiveness in actual use estimated in many studies. All found reductions in rear impacts, but by amounts well short of the 35% reduction suggested so consistently in the experiments. Indeed, after trends became apparent in the first few years of evaluation there was speculation that effectiveness of the device was trending to zero.
In 1998 an evaluation was performed to examine the effectiveness over a long period by estimating the effect on rear impacts for each year in the same manner. This involved using police-reported crash data from eight states to compare the ratio of rear impacts to non-rear impacts for model year 1986-89 cars (all equipped) to the corresponding ratio for 1982-85 cars (mostly not equipped). The same calculation was performed for data for each calendar year from 1986 onwards. The ratios were adjusted for vehicle age because when newer cars are involved in crashes, they are more likely to be struck in the rear than are older cars (possibly because they use higher levels of braking).
Figure 8-3 shows the findings of the study43 together with the 35% reductions reported in the pre-introduction fleet experiments. Although the effectiveness declines in time, it appears to have reached a stable level of about 4%. The benefits from such a risk reduction far exceed the modest cost of the device.
Figure 8-3. Percent reductions in rear-impact crashes associated with center high mounted stop lamps estimated in experiments using large fleets of vehicles equipped with prototypes, and in police-reported rear-impact crash rates in eight states.43
The reason for the lower
effectiveness in use than in the trials as well as for the
subsequent further declines may be related to what might be
called the novelty effect. Anything unusual on the rear of a
vehicle might invite a following driver to fixate on that
vehicle and increase caution, thereby reducing the chances
of crashing into it. The finding of positive effectiveness
in 1995, when the vast majority of vehicles on the roads
were equipped, supports the interpretation that the device
is providing superior cues than the earlier lighting systems
that the vehicle in front is braking. As time goes forward
and there are fewer vehicles without the device, evaluation
becomes more and more difficult, so there does not appear to
be any possibility of an empirical evaluation of the effect
when all vehicles are equipped. Even in the unlikely event
that it did become zero, all the accumulated crashes
prevented in the meantime would pay for decades of future
There are additional approaches to further reducing reaction times. The lights in a traditional or center mounted stop light are incandescent. That is, when a switch completes a circuit, electricity flows through a tungsten filament, heating it to a high enough temperature to glow brightly. This process takes about 200 ms to reach near full intensity. Accordingly, there are proposals to replace incandescent bulbs with other types of light sources with shorter rise times, including light emitting diodes.
A vehicle lighting feature addressing frontal impacts is daytime running lamps. These are reduced-intensity lights on the front of vehicles that automatically illuminate when a vehicle is started, making the vehicle easier to see by other drivers and pedestrians. Although first introduced in Sweden, Norway, and Finland where the greatest benefit from increased conspicuity might be expected, since they have long periods of dusk due to their northerly latitudes, daylight running lamps have also been shown effective in the US, especially at reducing pedestrian risk.
The difficulty, lack of control and reproducibility, and danger of conducting various types of driving research in actual traffic provided the impetus to develop driver simulators, devices which replicate driving with varying degrees of fidelity within the confines of the laboratory. Driving simulators are of two types, fixed base and moving base.
The most rudimentary fixed base simulator consists of little more than a screen presenting pictures to which subjects react, or a mock-up of a vehicle to familiarize students with control devices. Such equipment has proved useful in research and training. It is relatively inexpensive to build, maintain, and use. Valuable research information has come from such simulators, including selecting the road signs that offer superior visibility and earlier detection.
Moving base simulators provide acceleration cues by moving a cab containing a mock-up of a vehicle in all directions within a large interior space. The sensation of accelerating, for example, may be simulated by tilting the cab upwards as well as accelerating it forward. Moving base simulators cost vastly more than fixed base simulators and, because of set-up time, can generally accommodate fewer subjects in a day at vastly greater running cost.
It was the success of sophisticated moving base aircraft simulators that led to the application of similar technology to the driving case. Yet there is little in common between the two situations. The aircraft simulator is a device costing tens of millions of dollars representing an aircraft costing hundreds of millions of dollars. For the automobile case, it seems harder to justify a device costing tens of millions of dollars, when the real article can be purchased for under 20 thousand dollars. High realism simulators appear to offer nothing for training regular drivers. An accompanied learner driver can practice starting and stopping a real car every 15 s or so; a simulator offers little difference in training rate or safety. In contrast, it would be difficult to fit in more than a few real aircraft take-offs and landings in an hour, not to mention the risks and the cost of the aircraft and fuel. The aircraft simulator allows take-offs, followed by take-offs without intervening landings, to be repeated under varying conditions. While the performance skills learned in simulators can be critical in emergencies in the air, car driving emergency situations usually arise because of violations of expectancy which allow little time for corrective actions.
Enthusiasm for driving simulators ignores some of the most basic understanding about the nature of traffic crashes. The discussion above on reaction time showed the primacy of expectancy. Even in experiments using actual instrumented vehicles, reaction times are substantially shorter than in normal driving. Any reliance by traffic engineers on reaction times determined on a simulator, no matter how realistic, could produce unfortunate results. However, the reason that simulators are unlikely to produce knowledge relevant to traffic safety is more fundamental than this. Simulators measure driver performance, what the driver can do. However, safety is determined primarily by driver behavior, what the driver in fact chooses to do. It is exceedingly unlikely that a driver simulator can provide useful information on a driver's tendency to speed, drive while intoxicated, run red lights, pay attention to non-driving distractions, or not fasten a safety belt. Twenty-year-olds perform nearly all tasks on simulators better than the 50-year-olds, but it is the 50-year-olds who have sharply lower crash risks.
Driving simulators are far from new. A 1972 article refers to an earlier 1970 article listing 28 devices then in use, 17 of them in the US. Since the 1960s, driver simulators have incorporated moving bases and multiple movie projectors to provide visual information, including to the rear view mirror. Figure 8-4 is a reproduction of a list of research topics alleged to be suitable for research using driving simulators. The list was published in 1972.47 The research literature provides scant evidence that research agenda was advanced by simulators, neither by those in existence in 1970, nor by the much larger number of far more expensive and sophisticated simulators that have since been built. More than a decade ago I wrote:
Can the lack of progress be traced specifically to
insufficient realism in the simulator, thus justifying a
more sophisticated simulator? Any decisions regarding major
investments in additional driver simulators should identify
what specific problems they can be used to solve, and why
they can solve them when only slightly less sophisticated
simulators could not.
The following thought experiment helps
address such questions. Consider a make-believe simulator
consisting of an actual car, but with the remarkable
property that after it crashes a reset button instantly
cancels all damage to people and equipment. What experiments
could be performed on such make-believe equipment which
would increase our basic knowledge about driving? The
answers provide an upper limit on what might be done using
improved simulators. Defining subject areas, such as alcohol
and driving, should not be confused with defining specific
questions; there are already over 500 technical papers on
how alcohol affects performance. Increased knowledge about
driving is most likely to be discovered using the normal
processes of science. In these, problems are first defined,
and if they can be solved using existing equipment, they
are. If they cannot be solved using existing equipment, new
equipment is developed only if it is considered likely to
contribute to the solution, and not for its own sake. (p
Alas, the remarks fell upon deaf ears. The US supports a National Advanced Driving Simulator with a project cost of $50 million, claiming it to be the most sophisticated simulator available. Although the research literature documents 1,733 papers on alcohol and skill, the first sentence of justification for the $50 million expenditure is The effects of alcohol, drugs, visual impairments and aging on driving will all be safely studied using the new research tool. How like the 1972 list this justification sounds, and I fear that in the decades to come there will be just as little research progress to report.
Acquisition of driving skill
A remarkable feature of vehicle driving is that almost everyone can do it. Not only can most people learn to drive, but they acquire in a matter of weeks the necessary skills to start, stop, and propel a vehicle down a road and around corners. This is achieved without intensive study or extended practice. In 1901 Karl Benz thought that the global market for the automobile was limited because There were going to be no more than one million people capable of being trained as chauffeurs. Given the state of knowledge at the time, his conclusion was not unreasonable.
If automobiles and stringed musical instruments did not exist, but were suddenly invented, even today there are no known general principles of how people learn that would predict which would be easier to master. Given that music is about as old as humanity, it might seem natural to expect people to realize quickly that you just slide your finger up and down the string until you hear the desired note. A common-sense guess might therefore be that within an hour of first encountering a stringed instrument just about everyone could rattle off any tune they knew, but only the gifted few, after years of dedicated training, could reliably keep a 1,500 kg vehicle traveling round a curve at 100 km/h within a 4 m freeway lane surrounded on all four sides by other vehicles.
Although there are no effective models to predict the rate of learning and proficiency at one task compared to another, some patterns have been observed common to the acquisition of complex skills in general. These have been considered to occur in three phases:32
1. Early, or cognitive phase
2. Intermediate, or associative phase
3. Final, or autonomous phase
This categorization fits well the acquisition of driving skill. In the early, or cognitive phase, the learner tries to understand the components. For driving, the location of the controls and what vehicle responses they produce must be learned. In the intermediate phase, different strategies are explored, and the learner is acutely attentive to feedback. The learner-driver devotes full attention to the task, and increases skill by responding to feedback either from observed consequences of inputs, or from directions from an instructor. The skill of knowing what output is required in specific traffic situations develops together with the skill of knowing what input produces the desired output. In the third, or autonomous phase, the task is performed at a high level with minimal effort, in part because behavior becomes rather fixed and inflexible. In this autonomous phase, the task can be performed using a small fraction of the driver's attention. Other tasks, such as navigation, conversation, admiring the scenery, listening to the radio, talking on a cell phone, or thinking about other matters can be performed. Although the mental capacity devoted to driving is small in this autonomous phase, it is still such that, if a threat occurs and is recognized, all attention is quickly switched to the driving task. Most drivers have personally experienced this many times in, say, driving along waiting for specific information from a radio broadcast. An incident occurs in traffic, the driver reacts to the incident, and later realizes that the sought-after radio information, although broadcast, has not been perceived. Of course, if the threatening incident is not recognized because the driver's attention is elsewhere, such as talking on a cell phone, the result can be a crash.
The beginning driver
As people learn to drive, the direction in which they look changes in ways that relate to the three learning stages mentioned above. Experimental studies reveal that in the first hours of driving experience, drivers scan over a wide area, including well above the horizon.49(p 102) After about a month's experience, fixations are more confined in the vertical direction, but still vary horizontally. After three months' experience, fixations are more concentrated at the focus of expansion, with a much greater reliance on peripheral vision for cues to control the vehicle's position in the lane. As drivers gain experience they concentrate their eye fixations in smaller areas. Novice drivers look closer in front of the vehicle and more to the right of the vehicle's direction than experienced drivers, and are more likely to glance at the curb to estimate the vehicle's lane position. Novice drivers sample the rear-view mirrors much less frequently than experienced drivers.
These findings indicate that during the first few times behind the wheel almost all information processing capacity is absorbed in simply maintaining the vehicle's position in the lane. As experience is gained, peripheral vision is used more to locate the vehicle in the lane, with fixations focused further down the road to allow more time to process information that becomes increasingly relevant with increasing vehicle speed. When specifically instructed to pay attention to road signs, novice drivers are more likely to miss them than experienced drivers, another indication that the task of controlling the vehicle is placing more mental workload on novice drivers.
In an experiment in which novice and experienced drivers watched video-recordings taken from a car traveling along a variety of roads, the experienced drivers showed more extensive scanning in attempting to recognize hazards. The authors interpreted the result to mean that the inspection of the roadway by novices is limited not because they have limited mental resources residual from the task of vehicle control, but that they have an impoverished mental model of what is likely to happen in freeway driving. Another study concludes that, compared to experienced drivers, novice drivers detect hazards less quickly and efficiently and perceive them less holistically.
The early stages of learning to drive are generally accompanied by anxiety, tension and fear. Training courses aimed at producing relaxed and confident drivers may reduce fear that in some situations could be protective. Although driving remains one of the riskiest activities, it soon becomes relatively unconnected with fear. Evolution has implanted in us much greater fears of less dangerous activities. Experiments have shown that babies refuse to crawl in the direction of a simulated sharp drop even in response to their mothers' voices. This fear of heights is so ingrained, perhaps even instinctive, that we retain it in the absence of reinforcing experiences to ourselves or acquaintances. We do not lean far out of a window on the fourth floor, from which height a freely falling object would strike the ground at 50 km/h. Yet we travel at much higher vehicle speeds without anxiety. As smooth locomotion through the environment is not part of our evolutionary heritage, we have no instinctive fear of it. Once facility is acquired at basic driving skills, driving becomes relaxed and unassociated with danger. We largely lose that protection described by Shakespeare, "Best safety lies in fear." (Hamlet: Act I, Scene 3).
The material introduced in this chapter shows that, beyond the elementary control skills that are quickly learned, there are many higher level skills involved in driving that cannot be learned quickly. The only way to gain high level performance at these skills, like so many others, is practice, and a learning curve extending over many years is to be expected. However, unlike improving your golf game, practicing to improve driving skill comes with the risk of crashing.
Early stages of driving and crash rates
Many of the fatal-crash relationships in Chapter 7 show sharply higher risks at the earliest ages of driving compared to rates just a year or so later. Younger drivers pose the greatest fatality threats to themselves and to other road users. For involvement in crashes of all types, 16-year-old drivers have crash rates for the same distance of travel about 10 times those of 40- to 50-year-old drivers. Among teenagers, crash rates decline consistently and steeply with each yearly increase in age.
Specific evidence that lack of skill and knowledge is a factor in crashes of beginning drivers is provided by an examination of narrative descriptions of more than 2,000 crashes involving 16- to 19-year-old drivers.57 The results indicated that the great majority of non-fatal crashes resulted from errors in attention, visual search, speed relative to conditions, hazard recognition, and emergency maneuvers.57 High speeds and patently risky behavior accounted for only a small minority of crashes. Differences in the types of errors by first year novices and more experienced youth were relatively few in number and small in magnitude, indicating that the benefits of experience apply rather generally across all aspects of driving. Another study found that crash rates drop most precipitously during the first 6 months of driving. Involvement in certain types of crashes, such as run-off-the-road, single vehicle, night, and weekend crashes had the largest declines. The findings suggest that novices improve their driving in a relatively short period of time.
Lack of skill likely has a large effect on rollover risk. A beginning driver is more likely than an experienced driver to run off the right side of the road because of less skill in maintaining the vehicle's lateral position, and perhaps through increased fear and poorer judgment of oncoming traffic. A beginning driver will have less experience in handling a vehicle that has left its lane, and is more likely to overcompensate, thereby either crashing or, as is more common, receiving a valuable lesson in what not to do. As is common in skilled tasks, the inexperienced make more errors than the experienced.
Another contribution is overall higher levels of risk-taking by drivers less than 30, particularly male drivers. If skill were the sole factor, then the observed lower crash rates for 45-year-old drivers than for 30-year-old drivers would imply major additional skill acquisition even after more than a decade since first learning to drive. While additional experience might reduce crash risk, it is not a plausible explanation of effects of the magnitude observed. It is not possible to separate the roles of skill and youthful risk-taking in a completely satisfactory way. In motorized societies almost all the inexperienced drivers are also young.
However, one can examine data from drivers with little experience who are not young, based on their possession of a learner's permit rather than a full driver license. The plot in Fig. 8-5 uses all 877 fatally injured drivers coded in FARS 1994-2002 as driving with a learner's permit. More than half were teenagers. It is plausible to interpret that driving with a learner permit indicates a comparable lack of driving experience. Given that a driver is killed, the probability that it is in a rollover crash is much higher for younger inexperienced drivers than for older inexperienced drivers, showing that age, as such, is exercising a large influence. For all drivers, Fig. 7-18 (p. 164) shows that, when a driver is killed, the probability that rollover is the most harmful event is higher for male and for younger drivers, thus associating rollover crash fatalities with increasing risk taking. The data in Fig. 8-5 therefore indicate that an important component of the higher risks for younger drivers is due to their youth, and not just to their inexperience.
The relationship between driver skill/performance and
The evidence above shows that lack of skill contributes to higher crash risk. However, it does not follow that higher and higher levels of skill lead to lower and lower crash risk. Once the basics of driving are mastered and the task has become autonomous, its main characteristic is that it becomes what has been called a self-paced or self-controlled task.31 The driver chooses the level of difficulty that feels appropriate and comfortable, so that increased skill may translate into, for example, higher speeds. In Chapter 4 we found that although antilock braking produces superior braking, it was associated with higher fatality risk. Through similar processes, increased skill may translate into increased crash risk.
Peak performance in tests of reaction time relevant to driving and of visual acuity are achieved in the late teens/early twenties. Compared to females, males tend to be more interested and knowledgeable about driving and vehicles. The group with the fastest reaction time, best visual acuity, and most knowledge about vehicles and driving, namely young males, is the group with the highest crash risks.
The clearest indications of performance affecting safety are the increases in crash risks as drivers age. Here deterioration in such performance-related attributes as visual capabilities, reaction times, and information-processing speeds and other cognitive skills leads to increasing crash rates. Eventually such performance degradation produces crash-rate increases even in the presence of likely reductions in risk taking.
One can consider each end of the U-shaped relationships (Chapter 7) to originate from different sources. The elevated rates for the young flow from a combination of lack of skill and higher risk taking, and the higher rates for the old from driver performance limitations.
With increasing experience drivers acquire the impression, reinforced by vast numbers of safe trips, that driving is a safe and effortless task requiring only a small fraction of their total attention. Such an impression encourages drivers to perform a variety of other tasks while driving. A study in which subjects drove vehicles equipped with a video camera provided the following results. During three hours of driving, nearly all subjects manipulated vehicle controls (such as air conditioning or window controls) and reached for objects inside their moving vehicle. Nearly as many were observed manipulating music/audio controls, or had their attention drawn to something outside the vehicle unrelated to driving. Approximately three-fourths ate or drank something while driving or conversed with a passenger. Reading, writing, and grooming activities were also relatively common, but declined to less than half of the participants when observations were restricted to moving vehicles only. About a third of the subjects used a cell phone while driving, and nearly as many were distracted by passengers riding in their vehicle. Taking into account the shorter amount of time that children and especially babies were present in vehicles, children were about four times and infants almost eight times more likely than adults to be a source of distraction to the driver, based on the number of distracting events per hour of driving.
While it is reasonable on intuitive grounds to surmise that, for example, distractions from an infant (especially one in a rear seat) might increase crash risk, it would be extremely difficult to examine this empirically, let alone measure the magnitude of the effect. The effect on safety of one source of distraction in the above list has however been quantified.
The influence on crash risk from the use of a cellular telephone while driving was investigated in a case-crossover analysis, a technique for assessing a temporary change in risk associated with a transient exposure.52 A major strength of the method is that each person serves as his or her own control. This eliminates confounding due to age, gender, personality, and other fixed characteristics. The study used 699 drivers, all of them involved in crashes and all having cell phones available in their vehicles. Comparing the time of their crashes to telephone company records showed that 170 of the drivers were using their telephones just prior to the crash. The telephone records further showed that in a control period, chosen for each driver to be 24 hours before their crashes, only 37 drivers were using cell phones. If telephone use had no influence on crash risk, one would expect similar numbers of drivers to be using the telephone during their crashes and during the control period. Because some drivers were on the telephone during their crashes and also during the control period, the simple relative risk associated with phone use is larger than 170/37 = 4.6. Taking into account the probability that the crash-involved driver may not have been driving 24 hours prior to his or her crash, and other details, led to a conclusion that the relative risk of crashing while using (compared to not using) a cell phone was 4.3 (95 percent confidence interval, 3.0 to 6.5).
This reliable indication of a large effect stimulated many studies to address various issues, such as the extent to which the effect was from the manipulative demands of dialing compared to the cognitive demands of the content of the telephone. The consensus seems to be favoring an interpretation in terms of cognitive demands, so that hands-free telephones may not make all that much difference. Many jurisdictions have passed legislation prohibiting the use of cell phones while driving. While such legislation certainly enhances safety, it still raises benefit-cost questions, because using a cell phone while driving does offer benefits.
Unfamiliarity with vehicle
There is convincing evidence that unfamiliarity with a vehicle increases crash risk. One review examined incidental evidence in a number of previously published studies and concluded that driving an unfamiliar car increased crash risk by about a factor of two. A more specific examination found that 8.9% of all drivers in NASS-reported crashes in 1981 had less than 150 miles driving experience with the crashed vehicle, compared to an approximately estimated 1.5% of all driving in such vehicles. Older drivers with cognitive deterioration experience additional difficulties which pose potential risks in unfamiliar vehicles.
Driver education and training
The evidence above shows that lack of skill can contribute to young drivers' higher crash risks. Education and training have the goal of imparting knowledge and skill. It therefore seems compelling to think that driver education and training must necessarily lead to enhanced safety. Such an intuitively appealing belief has helped spawn a massive worldwide driver education and training industry. There has been a correspondingly vast amount written on the subject, starting particularly in the 1970s.
One review of this literature by Canadian researchers concludes: The international literature provides little support for the hypothesis that formal driver instruction is an effective safety measure. Another review by an Australian researcher similarly concludes: The research evidence suggests that driver training of a traditional and conventional nature contributes little to reductions in accident involvement or risk among drivers of all age and experience groups.
The Cochrane Library in the UK surveyed 19 reference data bases (MEDLINE, TRIS, etc.), the Internet, and other sources. Their search was not restricted by language or publication status. They searched for randomized controlled trials comparing post-license driver education versus no education, or one form of post-license driver education versus another. They concluded: This systematic review provides no evidence that post-license driver education is effective in preventing road traffic injuries or crashes.
Despite these findings, the British Government still included driver education as a key element in an effort to reduce traffic crashes. The intuitive belief that it must be effective was reinforced by a driver education industry sponsored study that did not address crashes, but instead examined knowledge and stated intended behavior before and after a safety presentation. In disagreeing with the British Government's decision to include driver education, the authors of the comprehensive review of the world's literature comment on the need for evidence-based policy.
One effect of driver education is that it enables students to qualify for licensure at earlier ages. Having acquired the licenses, the drivers then experience crash rates typical for their age, and as a consequence end up with more crashes than if they had not received driver education.
Because the evaluations of driver education have been conducted in motorized countries, the results should not be assumed to apply equally to countries in the early stages of motorization. The vastly higher crash rates in less motorized countries (Chapter 3) may have a component resulting from lack of basic skill and knowledge. Children in motorized countries have a large body of information about the rules of the road and how to behave in traffic long before they have driving licenses. They have been riding in, and getting out of the way of, motorized vehicles since infancy. The few weeks of driver education makes but a modest increment to this large pool of knowledge, and therefore renders it unlikely to reduce crash risk. People who start with a lesser pool of knowledge may gain more through driver education, so conclusions for less motorized countries must await specific evaluation studies.
The absence of proven safety benefits from driver education does not prove that training cannot increase safety, but merely that none of the methods so far applied have been demonstrated to do so. There is no theoretical principal stating that some type of education cannot reduce crash risk. The importance of traffic safety justifies supporting research to discover if there are training techniques that do reduce crash risk, but detached objective evaluation is crucial.
Longer term experience
While skill at the components of driving increases rapidly in early learning, the ability to identify and extract relevant information from a complex cluttered traffic environment appears to come more slowly. Perhaps a distinction should be drawn between perceptual-motor skills and total performance that additionally incorporates higher level skills. These additional abilities, which might be described as road sense, or good traffic judgment, develop over many years.
Although such effects may be of the utmost importance, there is limited empirical information. I share the view of most observers that higher-level driving skills continue to increase with driving experience over time frames of the order of decades. The ability to extract and appropriately process relevant information from a complex visual field appears to increase, and there appear to be ongoing increases in driver abilities to project further in time. We saw above that the novice driver is grimly focused on the present location of the vehicle, whereas as skills increase, visual attention focuses more on the vanishing point ahead - where the vehicle will be in the future. As each task becomes more thoroughly learned, the driver acquires more spare mental capacity that, through learning by feedback, focuses further ahead.
Driving seems to abound with examples in which events more and more in the future can beneficially influence present decisions. For example, a driver with a few years experience will likely approach a car stopped at a red light on a straight road in a manner that is independent of how many vehicles are stopped, or when the light turned red; all attention being focused on the rear of the vehicle ahead. A more experienced driver may slow down gently a long way from the light if it has just turned red or if there is a long line of stopped vehicles, but maintain a higher speed if the light has been red for some time and there are only a few vehicles waiting. The more experienced driver is more likely to have learned that in the first case stopping is nearly inevitable, whereas in the second case stopping, or even slowing down, may not be required. Which of these cases applies depends on perceiving, monitoring, processing and projecting into the future much information well beyond judging that the lead vehicle is slowing down or is stationary.
It should be emphasized that some less experienced drivers exhibit more advanced behavior, while some experienced drivers less advanced -- there are large variations among drivers at all stages of experience.
Even drivers with high crash rates complete the vast majority of trips without crashing. A driver with a crash rate ten times the average would still drive approximately 10 months (Table 1-1) between crashes. For such a driver,
even the frequency of near-misses would still be insufficient to teach
which actions are likely to lead to crashes. Drivers learn to negotiate corners skillfully by practicing such maneuvers thousands of times. Each time it is
done badly, corrections can be planned for the next time. Thus driving skills
are learned and polished largely by experimentation and frequent direct feedback. Learning by Shakespeare's recipe, "The injuries that they themselves procure must be their schoolmasters", (King Lear, Act III, Scene 1) is not effective for crash avoidance. Safety must be based on the knowledge of the whole society, as expressed in traffic law, rather than each driver learning from individual experience.
Graduated driver licenses
While young beginning drivers have highly elevated crash risks, seven US states issue learner-driver permits to drivers under age 15. The most common US practice is to issue learner permits at 15 and full driver licenses at 16. In the simplified list in Table 8-1 the full license for younger drivers may still differ from the license for other drivers, for example, by being differently colored to indicate that the holder is under 25. In all US states licenses are issued to drivers at younger ages than in most other countries. Additional details for each state are available in the source providing the information in Table 8-1.
A major contributor to the elevated rates of younger drivers is lack of driving experience, yet the only way to get experience is by driving. But when they drive, they crash. This has been called the young driver paradox. What is needed is a way to gain experience while minimizing risk. This is the goal of graduated licensing.
Graduated licensing is a way to phase in on-road driving by allowing beginners to get their initial experience under conditions that involve lower risk. Three stages are typically involved. The first is a supervised learner period of typically 6 months, then an intermediate licensing phase that permits unsupervised driving in less risky situations, and finally a full license becomes available when conditions of the first two stages have been met.
The concept originated in 1970s research identifying the high crash risks of younger drivers, and was first applied in New Zealand in 1984, with Michigan being the first US state to adopt a graduated licensing program in 1997. Early evaluations were so positive that, by 2003 one or more elements of graduated licensing have been adopted in 58 North American jurisdictions (District of Columbia, 47 US states, 9 Canadian provinces, and 1 Canadian
Table 8-1. The minimum age for a learner's permit and full driver license in all 50 US states and the District of Columbia.
territory). Although most North American programs are too
new for formal evaluation, impressive crash and injury
reductions have been reported in California, Florida,
Kentucky, Michigan, North Carolina, Nova Scotia, Ontario,
Although the basic goals of all graduated licensing programs is the same, specific implementation details differ widely among different jurisdictions.72 In a few cases graduated licenses do not apply to teenagers only, but to all newly licensed drivers. In a typical case an adult (usually a parent) must certify that the beginning driver has driven under supervision for 50 hours. After this first six-month phase is completed the beginning driver may drive alone, but not at night, and not with teenage passengers. At the completion of the second phase, or in some US states, when reaching the legal adult age of 18, a full license permitting unrestricted driving is granted. While such variation makes it impossible to associate a single effectiveness with the concept, reductions in crash rates to the affected populations in excess 10%, and in some cases far in excess, have been observed. Declines as large as 50% have been associated with the first six months.
Graduated licenses constitute an effective approach to providing drivers with the experience that is crucial in acquiring the skills necessary for safe driving while at the same time lowering the risk intrinsic in acquiring these skills. Ongoing refinement and expansions of graduated licensing programs72 will prevent large numbers of crashes by beginning drivers. This will not only reduce injuries and deaths to beginning drivers, but also to passengers and other road users at all levels of experience.
Summary and conclusions (see printed text)
References for Chapter 8 - Numbers in [ ] refer to superscript references in book that do not correctly show in this html version. To see how they appear in book see the pdf version of Chapter 1.
Rockwell TH. Skills, judgment and information acquisition in
driving. In: Forbes TW, editor. Human Factors in Highway
Traffic Safety Research. New York, NY: Wiley-Interscience;
1972, p. 133-164.
 Peli E, Peli D. Driving with
Confidence: A Practical Guide to Driving with Low Vision.
River Edge, NJ: World Scientific; 2002.
 Johnson C, Keltner J. Incidence
of field loss in 20 000 eyes and its relationship to
driving performance. Arch Ophthalmol. 1983; 101: 371-375.
 Wood JM, Troutbeck R. Effect of
visual impairment on driving. Human Factors. 1994; 36:
 Westlake W. Is a one eyed racing
driver safe to compete? Formula one (eye) or two? Br J
Ophthalmol 2001; 85: 619-624.
 Federal Motor Carrier Safety
Administration, US Department of Transportation.
Qualification of Drivers; Exemption Applications; Vision.
Notice of final disposition. Federal Register, Vol. 68, No.
158, 48989-91; August 15, 2003.
 Owsley C, Ball K, McGwin G Jr,
Sloane M, Roenker DL, White MF, Overly ET. Visual processing
impairment and risk of motor vehicle crash among older
adults. JAMA. 1998; 279: 1083-1088.
 Denton GG. A subjective scale of
speed when driving a motor vehicle. Ergonomics. 1966; 9:
 Noguchi K. In search of optimum speed: From the users’ viewpoint. J Intl Assoc Safety Sciences. 1990; 14: 66-75.
 Milosevic S. Perception of
vehicle speed. Revija za psihologijy (Yugoslavia). 1986; 16:
 Evans L. Speed estimation from a
moving automobile. Ergonomics 1970; 13: 219-230.
 McLane RC, Wierwille WW. The
influence of motion and audio cues on driver performance in
an automobile simulator. Human Factors 1975; 17: 488-501.
 Denton GG. The influence of
visual pattern on perceived speed at Newbridge M8
Midlothian. Transport and Road Research Laboratory,
Crowthorne, Berkshire, UK; 1973.
 Meyer E. Evalution of data from
test application of optical speed bars to highway work
zones. Project No K-TRAN KU-00-4, Final Report. Kansas
Department of Transportation; July 2000.
 Smiley A. Driver speed
estimation: What road designers should know. Paper presented
at: Transportation Research Board 78th Annual Meeting,
Workshop on Role of Geometric Design & Human Factors in
Setting Speed; January 1999. http://www.hfn.ca/driver.htm
 Schmidt F, Tiffin J. Distortion of drivers’ estimates of automobile speed as a function of speed adaptation. J of App Psychology. 1969; 53: 536-539.
 Denton GG. The influence of
adaptation on subjective velocity for an observer in
simulated rectilinear motion. Ergonomics. 1976; 19: 409-430.
 Matthews ML. A field study of the effects of drivers’ adaptation to automobile velocity. Human Factors. 1978; 20: 709-716.
 Casey SM, Lund AK. Three field
studies on driver speed adaptation. Human Factors. 1987; 29:
 Insurance Institute for Highway
Safety. Status Report, Special Issue: Speeding. Vol 38 (10);
22 November 2003. http://www.iihs.org/srpdfs/sr3810.pdf
 Herman R, Potts RB. Single-lane
traffic theory and experiment. In: Herman R., editor. Theory
of Traffic Flow. Amsterdam, Netherlands: Elsevier; 1961, p.
 Evans L, Rothery R. Perceptual
thresholds in car-following -- a comparison of recent
measurements with earlier results. Trans Science. 1977; 11:
 Evans L, Rothery R. Detection of
the sign of relative motion when following a vehicle. Human
Factors. 1974; 16: 161-173.
 Sedgwick HA. Space perception.
Chapter 21 in Boff KR, Kaufman L, Thomas JP. editors.
Handbook of Perception and Human Performance, Volume 1,
Sensory Processes And Perception. New York: John Wiley,
 Caird JK, Hancock PA. Perception
of oncoming vehicle time-to-arrival. Proceedings of the
Human Factors Society Annual Meeting, p. 1378-1382; 1992.
 Evans L, Rothery R. The
influence of forward vision and target size on apparent
inter-vehicular spacing. Trans. Science. 1976; 10: 85-101.
 Herman R, Lam T, Rothery R . An
experiment on car size effects in traffic. Traf Eng &
Control. 1973; 15: 90-93,99.
 Farber EI, Silver CA. Knowledge of oncoming car speed as a determiner of driver’s passing behavior. Highway Research Record. 1967; 195: 52-65.
 Simpson G, Johnston L,
Richardson M. An investigation of road crossing in a virtual
environment. Accid Anal Prev. 2003; 35: 787-796.
 Farber EI. Passing behavior on
public highways under daytime and nighttime conditions.
Highway Research Record. 1969; 292; 11-23.
 Näätänen R, Summala H.
Road-User Behavior and Traffic Accidents. Amsterdam,
Netherlands: North Holland; 1976.
 Fitts PM, Posner, MI. Human
Performance. Belmont, CA: Brooks/Cole; 1967.
 Shinar D. Psychology on the Road
-- The Human Factor in Traffic Safety. New York, NY: John
 Olson PL, Sivak M.
Perception-response time to unexpected roadway hazards.
Human Factors. 1986; 28: 91-96.
 Summala H. Driver/vehicle
steering response latencies. Human Factors. 1981; 23:
 National Highway Traffic Safety Administration. Traffic Safety Facts 2000 – A Compilation of Motor Vehicle Crash Data from the Fatality Analysis Reporting System and the General Estimates System. Report No. DOT HS 809 337. Washington DC: US Department of Transportation; December 2001. (Table 32, p. 54).http://www-fars.nhtsa.dot.gov/pubs
 Motor City Dream Cars, Auto
history time line.
 Mortimer RG. A decade of
research in vehicle rear lighting. What have we learned?
American Association for Automotive Medicine, 21st Annual
Proceedings, p. 101-112; September 1977.
 Mortimer RG, Kupec JD. Scaling
of flash rate for a deceleration signal. Human Factors.
1983; 25: 313-318.
 Kohl JS, Baker C. Field test
evaluation of rear lighting systems. Report DOT HS 803 467.
Washington, DC: National Highway Traffic Safety
 Reilly RE, Kurke DS, Buckenmaier
CC. Validation of the reduction of rear-end collisions by a
high-mounted auxiliary stoplamp. SAE paper 810189.
Warrendale, PA: Society of Automotive Engineers; 1981.
 Rausch A, Wong J, Kirkpatrick M.
A field test of two single, center high mounted brake light
systems. Accid Anal Prev. 1982; 14: 287-291.
 Kahane CJ, Hertz E. The
long-term effectiveness of center high mounted stop lamps in
passenger cars and light trucks. Report DOT HS 808 696.
Washington, DC: Department of Transportation; March 1998.
 Evans L, Rothery R. Comments on
effects of vehicle type and age on driver behaviour at
signalized intersections. Ergonomics. 1976; 19: 559-570.
 Sivak M, Flannagan MJ, Sato T,
Traube EC, Aok M. Reaction times to neon, LED, and fast
incandescent brake lamps. Ergonomics. 1994; 37: 989-994.
 National Highway Traffic Safety
Administration. A preliminary assessment of the
crash-reducing effectiveness of passenger car daytime
running lamps (DRLs). Technical Report DOT HS 808 645.
Washington, DC: NHTSA; June 2000.
 Hulbert S, Wojcik C. Driving
task simulation. In: Forbes TW, editor. Human Factors in
Highway Traffic Safety Research. New York, NY:
Wiley-Interscience; 1972, p. 44-73.
 Kuratorium fűr Verkehrssicherheit. Verkerspsychologie IV, Vienna, Austria, p. 149-184; May 1970.
 Evans L. Traffic Safety and the
Driver. New York, NY: Van Nostrand Reinhold; 1991.
 NHTSA. The National Advanced
Driving Simulator (NADS) is the most sophisticated.
 Mackay M. Towards grand
unification of traffic science. J Intl Assoc Safety
Sciences. 1990; 14: 19. Also quoted in slightly different
form by Macrae N. The next ages of man. London, UK: The
Economist, p. 5-20; 24 December 1988.
 Redelmeier DA, Tibshirani RJ.
Association between cellular-telephone calls and motor
vehicle collisions. New Eng J. Med. 1997; 336: 453-458.
 Summala H, Näätänen, R.
Perception of highway traffic signs and motivation. J Safety
Res. 1974; 6: 150-153.
 Underwood G, Chapman P, Bowden
K, Crundall D. Visual search while driving: Skill and
awareness during inspection of the scene. Transp Res Traffic
Psychol Behav. 2002; 5: 87-97.
 Deery HA. Hazard and risk
perception among young novice drivers. J Safety Res. 1999;
 Gibson EJ, Walk RD. The “visual cliff”. Scientific American. 1960; 202: 64-71.
 McKnight AJ, McKnight AS. Young
novice drivers: Careless or clueless? Accid Anal Prev. 2003;
 Mayhew DR, Simpson HM, Pak A.
Changes in collision rates among novice drivers during the
first months of driving. Accid Anal Prev. 2003; 35: 683-691.
 Warshawsky-Livne L, Shinar D.
Effects of uncertainty, transmission type, driver age and
gender on brake reaction and movement time. J Safety Res.
2002; 33: 117-128.
 Stutts J, Feaganes J, Rogman E,
Hamlett C, Meadows T, Reinfurt D, Gish K, Mercadante M,
Staplin L. The role of driver distraction in traffic
crashes: Distractions in everyday driving. AAA Foundation
for Traffic Safety. Washington, DC; July 2003.
 Cohen JT, Graham JD. A revised
economic analysis of restrictions on the use of cell phones
while driving. Risk Analysis. 2003; 23: 5-17.
 Perel M. Vehicle familiarity and
safety. Report DOT HS 806 509. Washington, DC: US Department
of Transportation, National Highway Traffic Safety
Administration; July 1983.
 Hoxie P. Assessment of driver
inexperience with an automobile as a factor which
contributes to highway accidents. Report DOT HS 806 593.
Washington, DC: US Department of Transportation, National
Highway Traffic Safety Administration; December 1984.
 Lundberg C, Hakamies-Blomqvist
L. Driving tests with older patients: Effect of unfamiliar
versus familiar vehicle. Trans Res, Part F. 2003; 6:
 Mayhew DR, Simpson HM. The safety value of driver education and training. Inj Prev. 2002; 8: 3‑8.
 Christie R. The effectiveness of
driver training as a road safety measure: A review of the
literature. Report No. 01/03, Royal Automobile Club of
Victoria, Australia. ISBN 1 875963 26 X; November 2001.
 Ker K, Roberts I, Collier T,
Renton F, Bunn F. Post-licence driver education for the
prevention of road traffic crashes (Cochrane Review). The
Cochrane Library, 2003, Issue 4. Chichester, UK: John Wiley
& Sons, Ltd; 2003.
 Achara S, Adeyemi B, Dosekun E, Kelleher S, Landley M, Male I, Muhialdin N, Reynolds L, Roberts I, Smailbegovic M, van der Spek N. The Cochrane injuries group driver education reviewers:·Evidence-based road safety: The Driving Standards Agency’s schools programme (editorial). Lancet 2001; 358: 230-232.
 State-by-state driving rules for
teenage drivers. http://golocalnet.com/drivingage/
 Warren RA, Simpson HM. The young
driver paradox. Ottawa, Ontario, Canada: Traffic Injury
Research Foundation; 1976.
 Waller PF. The genesis of GDL. J
Safety Res. 2003; 34: 17-23.
 Insurance Institute for Highway
Safety/Traffic Injury Research Foundation. Graduated
Licensing: A blueprint for North America. Arlington
VA/Ottawa Canada; July 2003.
 Branche C, Williams AF, Feldman D. Graduated licensing for teens: Why everybody’s doing it. J Law Med Ethics. 2002; 30: 146-149. Mayhew DR, Simpson HM, Desmond K, Williams AF. Specific and long-term effects of Nova Scotia’s graduated licensing program. Traf Inj Prev. 2003; 4: 91-97.