ARTICLE IN PRESS

International Journal of Industrial Ergonomics 35 (2005) 157–162

www.elsevier.com/locate/ergon

Short paper

Passenger-side rear-view mirrors: driver behavior and safety

Thomas Ayresa,, Li Lib, Doris Trachtmanc, Douglas Youngc
a
Consultant, Box 60591, Palo Alto, CA 94306, USA
b
NASA Ames Research Center MS 262-2, Moffett Field, CA 94035, USA
c
Exponent, 5401 McConnell Ave., Los Angeles, CA 90066, USA
Received 5 December 2002; received in revised form 4 May 2004; accepted 4 May 2004
Available online 6 October 2004

Abstract

Passenger-side rear-view mirrors (PRMs) have been standard equipment on motor vehicles sold in the US for many
years, although they are not required by the federal motor vehicle safety standards. Numerous studies documented both
the apparent need for PRMs (to overcome visual obstructions) and their apparent value (by increasing visual access to
the passenger-side rear). In addition, surveys of drivers have found a general appreciation of the importance of
sampling visual information from the rear. Very little can be found, however, regarding the actual safety benefit of
PRMs. A review of the research literature and several initial studies (driver observation and accident-data analysis),
suggest that PRMs may not be associated with any substantial accident prevention, perhaps because they are not
consistently used. Implications and research directions will be discussed.
Relevance to Industry

PRMs should have been a success story, having been carefully developed through research to provide important
information for safe driving. The apparent failure of PRMs to reduce accident rates in practice illustrates a potential
problem with designing and deploying safety features or devices without empirical assessment of normal user behavior.
r 2004 Elsevier B.V. All rights reserved.

Keywords: Driver behavior; Mirror; Accident data; Traffic safety

1. Introduction Ayres et al., 1993). Many approaches are used to

address human limitations and errors on the part
Studies of transportation accidents repeatedly of vehicle operators and pedestrians, from changes
have shown that human behavior—rather than in vehicle and roadway design to encouragement
environmental conditions or vehicle function—is and enforcement of safe behavior. Nevertheless, it
the primary contributor to accidents (Treat, 1980; is estimated that more than a million people die
each year in motor vehicle accidents, with injuries
Corresponding author. in the tens of millions (Ross, 1999). The toll is

0169-8141/$ - see front matter r 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.ergon.2004.05.009
 ARTICLE IN PRESS

158 T. Ayres et al. / International Journal of Industrial Ergonomics 35 (2005) 157–162

rising most quickly in developing countries, with such as trucks and buses; GAO Business Services,
higher casualty rates (per vehicle) than in indus- 1978).
trialized countries.
Assessment of the safety value of simple and
practical interventions can play an important role
2. Accident data
in allocating scarce funds for public safety.
Compared to major changes in transportation
Very little can be found, however, regarding the
infrastructure (e.g., new roadways) or large-scale
actual safety benefit of PRMs with respect to
education efforts, vehicle design requirements (for
preventing accidents. Mortimer and VanderMey
new vehicles, or as retrofits for existing vehicles)
(1971) performed a study of accident data for
may offer reasonably cost-effective solutions that
driver-side mirrors, at a time when very few cars
compensate for human limitations. The success of
had PRMs (but all had driver-side mirrors). In a
such changes, however, may depend heavily on
sample of 18 accidents involving passing or lane
how drivers use them.
changes for which faulty rear vision was judged to
As an example, in this paper, we evaluate the
be involved, left- and right-side accidents were
safety benefit provided by passenger-side rear-view
equally likely. This could suggest that the presence
mirrors (PRMs). It is widely recognized, and
of an outside rearview mirror (driver’s side) did
demonstrated in research studies, that driver-side
not reduce accidents.
and interior rearview mirrors often do not provide
drivers with a complete view of the area to the rear
and passenger side of a vehicle. PRMs have been 2.1. Method
standard equipment on motor vehicles sold in the
US for many years, although they are not required As a first step toward assessing the accident
by the federal motor vehicle safety standards. reduction attributable to the introduction of
Numerous studies documented both the apparent PRMs, we performed an analysis of reports of
need for PRMs (to overcome visual obstructions; fatal accidents in the US. Data were obtained from
e.g., Burger, 1974) and their apparent value (by the Fatal Accident Reporting System (maintained
increasing visual access to the passenger-side rear). by the National Highway Traffic Safety Adminis-
In addition, surveys of drivers have found a tration) for all fatal motor vehicle accidents
general appreciation of the importance of sam- reported in the US in 1999 (the most recent year
pling visual information from the rear (Smith with data available at the time of the analysis). We
et al., 1978). examined data for fatal accidents involving side-
On the other hand, it is also clear that there are swipes between vehicles moving in the same
trade-offs associated with the provision of direction (a type of accident included in the coding
PRMs. Flat-surface PRMs do not provide a system); this is a class of accident that would occur
sufficiently wide field of view for the driver during lane changes and for which PRMs could
(Mourant and Donohue, 1979) and convex prove helpful for drivers. If PRMs in actual usage
PRMs are associated with errors in distance tend to help drivers avoid fatal same-direction
estimation, even after adaptation (Flannagan et sideswipes (by preventing collision or reducing
al., 1996, 1997). A limited degree of convex severity), then these accidents should constitute a
curvature appears to be better than a flat mirror, smaller percentage of all fatal accidents involving
according to on-road research (Mortimer, 1971) recent model-year cars (from the last decade or so,
and accident data (Luoma et al., 1995). In when PRMs have become ubiquitous in the US)
addition, there are problems related to glare than for older model-year cars (when PRMs were
from the reflections of headlights of following only an option). The data do not contain
vehicles at night (Flannagan, 1988), and PRMs information about the presence or absence of a
can cause serious or fatal injuries to pedestrians PRM on a vehicle, so model year was used as a
and cyclists (particularly PRMs on larger vehicles substitute.
 ARTICLE IN PRESS

T. Ayres et al. / International Journal of Industrial Ergonomics 35 (2005) 157–162 159

2.2. Results evidence that the addition of PRMs to motor

vehicles have prevented a substantial number of
In 1999, there were approximately 93,000 fatal accidents in the US—either PRMs are not
vehicles involved in a total of approximately potentially helpful for avoiding same-direction
37,000 fatal motor vehicle accidents in the US sideswipes, or they are not used effectively.
(with close to 42,000 fatalities). Sideswipes be-
tween vehicles moving in the same direction
accounted for 1.54% of all vehicles involved in 3. Driver behavior
fatal accidents. Fig. 1 shows the results for the
number of vehicles involved in fatal accidents in Given that PRMs can provide drivers with a
1999 involving cars with model years 1960–2000. view of a portion of the side-rear environment that
Aside from some high variability in the earliest is not available with other mirrors, why might
model years, for which there were few accidents PRMs not contribute more to accident prevention?
overall, there is no indication that fatal same- In most situations, the information that a PRM
direction sideswipes declined across model years; could provide is also available if the driver turns
the trend line actually increases slightly. This her head to look at the passenger side and rear.
pattern becomes even clearer when the analysis is Perhaps drivers are no more likely to seek crucial
restricted to model years beginning with 1970 (Fig. information from a PRM than they are to turn
2); over the period 1970–2000, same-direction and look in the appropriate direction if the vehicle
sideswipes increased significantly as an annual lacks a PRM. Indeed, there is some evidence that
percentage (t ¼ 2:19; po0:05). drivers do not make optimal use of resources
A similar pattern of results (significant regres- available for safe lane changing. Inexperienced
sion with positive slope) was obtained for fatal drivers, particularly, are less likely to use mirrors
accidents in 1994 (not shown). In addition, it was (Mourant and Rockwell, 1972). Drivers in early
found that vehicles in same-direction sideswipes studies were found to be more reluctant to use
(not broken down by model year) increased as a PRMs than driver-side mirrors because of the
percentage of all fatal-accident-involved vehicles greater effort needed for head and eye movements,
across the period from 1994–2002 (from 1.3% in leading to suggestions for mounting PRMs on the
1994 to 1.8% in 2002). Thus, these data offer no front passenger-side fender (where they could be

7%

6%

5%
Percentage

4%

3%

2%

1%

0%
1960 1965 1970 1975 1980 1985 1990 1995 2000
Model Year

Fig. 1. Fatal same-direction sideswipes as percent of all fatal accidents. Data from 1999 for model years 1960–2000.
 ARTICLE IN PRESS

160 T. Ayres et al. / International Journal of Industrial Ergonomics 35 (2005) 157–162

3.5%

3.0%

2.5%
Percentage

2.0%

1.5%

1.0%

0.5%

0.0%
1970 1975 1980 1985 1990 1995 2000
Model Year

Fig. 2. Fatal same-direction sideswipes as percent of all fatal accidents. Data from 1999 for model years 1970–2000.

seen without having to turn so far to the side that the driver looked toward the PRM on only 6
rather than on or just in front of the door, Nagata of 29 occasions (21%) during the last 3 s prior to
and Kuriyama, 1985). Hetrick (1997) found that the lane change, and only 5 (14%) looked toward
8% of drivers did not signal before making lane the passenger-side blind spot, compared with 16
changes, and over half of the remainder did not (55%) who looked toward the interior rearview
signal until after beginning their maneuver, mak- mirror; unfortunately, the report does not specify
ing it difficult for other drivers to avoid them in a overlap, e.g., how many looked either toward the
conflict. PRM or the blind spot or both. Nevertheless,
We have made unobtrusive observations of those findings suggest that drivers often fail to
(unsuspecting) drivers approaching and conduct- gather potentially useful information about adja-
ing rightward lane changes (toward the passenger cent-lane traffic before making rightward lane
side of the vehicle in the US). Many drivers did not changes.
look to either the PRM or the passenger-side rear
before or during a rightward lane change. These 3.1. Method
observations, however, were difficult to make
reliably. In order to gather additional data on visual
In order to document driver looking behavior, information sampling during rightward lane
eye and/or head movements can be studied with changes, we conducted a small-scale roadway
video cameras or other devices mounted inside study. Eight subjects were recruited (6 male,
vehicles. In a recent study, 16 drivers used an 2 female; ages 23, 29, 31, 39, 39, 39, 45, and 50);
instrumented vehicle (an SUV loaned to them for they were told that the study was concerned with
the study) during their normal commuting (Lee driving behavior, specifically posture and seating
et al., 2003), and recorded data for over 8000 lane position. They were asked to drive their own cars
changes. Analysis of a subset of 29 rightward lane (all with PRMs) for the study, with video cameras
changes (out of 500 that were selected to empha- installed to record their behavior. This assured
size those when approaching a slower-moving that they would be comfortably familiar with the
vehicle ahead, and with vehicles fairly close in vehicle they used; all of the subjects had been
adjacent lanes, thus more demanding) indicated driving their vehicles for at least 2 months, and
 ARTICLE IN PRESS

T. Ayres et al. / International Journal of Industrial Ergonomics 35 (2005) 157–162 161

most for at least a year. During a 6.5 mile course in never having used it, despite having looked at it on
daylight on local urban streets, subjects were asked all 10 trials, and the two subjects who reported
by the in-car experimenter to make various lane almost never using it had looked at the PRM for
changes, both leftward and rightward, using their 0 and 7 trials, respectively; another subject said
judgment to make the maneuvers safely. Subjects they had almost always used it, but in fact looked
were not told about the specific aims of the study at the PRM on only 2 of 10 trials. These findings
until afterwards. suggest that self-reported mirror usage can be
Two small video cameras were mounted in each highly unreliable, even immediately after a driving
subject’s vehicle, both aimed at the driver’s head. episode.
One camera was placed next to the center rearview
mirror and the other was next to the PRM; the two
images were captured on a single split-screen video 4. Conclusions
recorder. Eye glance directions were coded from
the videotapes by one viewer, and checked by a It is clear from past research that looking
second viewer. Any differences between the two forward and using an inside rearview mirror
viewers were resolved by watching the correspond- cannot always provide enough information for
ing section again and reaching consensus. safe lane changes. Outside rearview mirrors have
been carefully designed to make blind-spot in-
3.2. Results formation available while balancing trade-offs
such as glare and limited field of view. Accident
The data of interest for this paper involve mirror statistics, however, including the fatal accident
use during rightward lane changes, with each of data analyzed here, have failed to demonstrate a
the 8 subjects making 10 such maneuvers upon safety benefit for PRMs. Perhaps more detailed
request, yielding 80 observations. Overall, subjects analyses, such as review of in-depth accident
looked at the inside rearview mirror on 97.5% of reports, will uncover such evidence. In the mean-
trials prior to rightward lane changes, but they time, the question remains as to why PRMs might
looked at the PRM on only 65% of such trials. not yield substantial safety benefits.
Usage of the PRM varied substantially across One possibility is that, in actual practice, PRMs
subjects. Three subjects used both the inside may not add to the information gathered by
rearview mirror and the PRM on every trial drivers; consequently, drivers may not feel a need
before making a rightward lane change. Another to check PRMs often. In most vehicles, a PRM
three subjects also used the inside mirror every does not provide much information that would not
time, but looked at the PRM on only half of the also be available by turning and looking to the
trials; these subjects presumably felt the inside passenger-side rear. It could even be argued that
mirror often gave them enough information for the PRM, with its narrow field of view, is inferior
their maneuver. Finally, two subjects occasionally to a direct look; some driving manuals recommend
(1 in 10 trials) made a rightward lane change that the driver always make a direct look before
without checking either of the critical mirrors, and initiating a lane change. A PRM could act as a
in general used the PRM infrequently (one never convenience, allowing a driver to perform an
used the PRM, and one used it for only 2 of 10 initial check of conditions before using a direct
trials). look to verify that it is safe to maneuver. In heavy
After the driving, subjects were asked about and fast-moving traffic, however, where lane-
their use of the mirrors during the study. Although changing is made difficult by short following
all subjects looked at the inside rearview mirror on distances (Ayres et al., 2001), drivers may not
9 or 10 of the 10 trials, one reported that he had have the luxury of using both a PRM and a direct
never looked at that mirror when changing lanes, look before initiating a lane change.
and another reported that he used that mirror only In cars without PRMs, drivers were found to
sometimes. For the PRM, one subject reported look back to the right on somewhat more than half
 ARTICLE IN PRESS

162 T. Ayres et al. / International Journal of Industrial Ergonomics 35 (2005) 157–162

of their rightward merges (Robinson et al., 1972). Flannagan, M., 1988. Human performance aspects of
In our study, using cars equipped with PRMs, rearview mirrors: an applied-literature review. University
drivers checked their PRMs on somewhat more of Michigan Transportation Research Institute,
UMTRI-88-20.
than half of the rightward lane changes. Lee et al. Flannagan, M., Sivak, M., Traube, E., 1996. Driver
(2003) found PRM checking on only 21% of trials, perceptual adaptation to nonplanar rearview mirrors. SAE
but their subjects were driving a relatively un- 960791.
familiar vehicle, and their sample emphasized Flannagan, M., Sivak, M., Schumann, J., Kohima, S., Traube,
heavy traffic conditions when drivers may not E., 1997. Distance perception in driver-side and
passenger-side convex rearview mirrors: objects in mirror
have time to check the possibly redundant are more complicated than they appear. University
information in a PRM. of Michigan Transportation Research Institute,
Another possibility is that, despite the some- UMTRI-97-32.
times valuable information available in a PRM, GAO Business Services, 1978. Accident information related to
drivers may not feel a need to look there. There are driver visibility and mirrors. US Department of Transpor-
tation.
numerous documented examples of frequent fail- Hetrick, S., 1997. Examination of driver lane change behavior
ure to use safety equipment (e.g., to use safety and the potential effectiveness of warning onset rules for
belts) or to drive at a safe speed for conditions lane change or ‘‘side’’ crash avoidance systems. Virginia
(e.g., in fog or at night); such behavior might Polytechnic Institute, Masters Thesis.
Lee, S.E., Olsen, E.C.B., Wierwille, W.W., 2003. A compre-
represent under-estimation of risk, or else a
hensive examination of naturalistic lane-changes. Virgina
general tendency to act just adequately rather Tech Transportation Institute, Blacksburg, VA.
than safely (Ayres et al., 1998). Luoma, J., Sivak, M., Flannagan, M., 1995. Effects of driver-
Thus, despite the promise of a simple device that side mirror type on lane-change behavior. Ergonomics 38
appears to fill an informational need, PRMs may (1), 1973–1978.
not substantially reduce accidents if they are not Mortimer, R., 1971. The effects of convex exterior mirrors on
lane-changing and passing performance of drivers. SAE
used, or if they are used instead of direct looking. 710543.
Researchers are now seeking more advanced crash Mortimer, R., VanderMey, T., 1971. Analysis of collisions
avoidance systems as solutions to lane-change involving rear vision. HIT-LAB Reports, University of
safety (e.g., Hetrick, 1997; Lee et al., 2003). Michigan Highway Safety Research Institute.
Mourant, R., Donohue, R., 1979. Driver performance with
right-side convex mirrors. Transportation Research Record
737, 95–100.
References Mourant, R., Rockwell, T., 1972. Strategies of visual search by
novice and experienced drivers. Human Factors 14 (4),
Ayres, T., Gross, M., McCarthy, R., 1993. A retrospective on 325–335.
attempts to reduce vehicular risk through operator training. Nagata, M., Kuriyama, H., 1985. Drivers’ visual behavior with
In: Elia Jr., F.A., Pyatt, D.W. (Eds.), SERA-vol. 1, Safety door and fender mirror systems. SAE 850330.
Engineering and Risk Analysis. ASME Book No. H00894- Robinson, G., Erickson, D., Thurston, G., Clark, R., 1972.
1993. Visual search by automobile drivers. Human Factors 14 (4),
Ayres, T.J., Wood, C.T., Schmidt, R.A., McCarthy, R.L., 1998. 315–323.
Risk perception and behavioral choice. International Ross, A., 1999. Road accidents: a global problem requiring
Journal of Cognitive Ergonomics 2 (1–2), 35–52. urgent action. Roads & Highways Topic Note RH-2.
Ayres, T.J., Li, L., Schleuning, D., Young, D., 2001. Preferred Smith, R., Bardales, M., Burger, W., 1978. Perceived impor-
time-headway of highway drivers. Fourth International tance of zones surrounding a vehicle and learning to use a
IEEE Conference on Intelligent Transportation Systems. convex mirror effectively. US Department of Transporta-
Oakland, CA. tion, DOT-HS-803 713.
Burger, W., 1974. Evaluation of innovative passenger car and Treat, J.R., 1980, A study of precrash factors involved in traffic
truck rear vision system. SAE 740965. accidents. HSRI Research Review 10(6)/11(1), 1–35.