|
Research Review Of Potential Safety Effects Of Electronic Billboards On Driver Attention And Distraction
The synthesis of the review indicated several areas in which
more information would contribute to a better understanding of the safety
implication of EBBs and tri-vision signs.
These areas are expanded into research recommendations in this
section. Each of these recommendations
or research questions presented in this section reflect a concern for driving
safety.
To obtain information regarding the influence of EBBs or
tri-vision signs on driving performance, it is necessary to examine these
displays in a variety of situations.
For instance, it is possible that driving performance would be affected
on curves but not on tangent road segments.
The depiction of motion on a screen may be disruptive to driving whereas
a static screen may not. The primary
objective of the research to be proposed is to determine whether there are
conditions under which EBB or tri-vision signs constitute a driving concern as
indicated by crashes or other form of degraded driver performance. If such conditions are identified, then
additional research may be required to gain a detailed understanding of the
issue. These research goals will be
identified as Research Questions.
Information that was available in an area that may support
EBB safety was identified as Research Findings. This information primarily focuses on text
issues, particularly with the legibility of letters and words; information
regarding symbols or graphics was not identified. Since the identified research was not performed in an EBB
context, it may be worthwhile to replicate selected research findings in
conditions based on EBB requirements. Thus,
some of the findings will be followed by research questions.
The issues described below have been researched to varying
degrees. Some questions, such as the
effect of a motion-based display on driving safety, have not been fully
researched yet. Other issues, such as
minimum exposure time for a display, have been addressed by states with little
or no research basis. Questions on
legibility have been addressed, but in a context for providing traffic
information to drivers. The research
questions fall into three general areas of highway safety: the roadway, EBBs
and tri-vision signs, and the driver.
The section on EBBs and tri-vision signs is presented in two parts: the
first addresses relatively global aspects of the billboard such as motion
phenomena and exposure time. The second
part addresses the more detailed issues of individual letters and words.
Different roadway characteristics exert varying demands upon
driver attention and skill. Particular
roadway configurations and their characteristics may be more or less suitable
for EBBs, and are important to consider when evaluating the safety effects of
EBBs. The roadway characteristics
listed below have special considerations relative to the issue of distraction
and safety.
-
Horizontal and Vertical Curves. Compared to tangents, curved roadway
segments require more adept handling from drivers. Thus, any distraction presented by an EBB may be larger in
conditions when the driver is experiencing greater demands.
Research
Questions - Curves: Is driver
performance affected adversely by the presence of EBB or tri-vision signs on
vertical curves, horizontal curves or even at a tangent segment? What is the role of vehicle speed and posted
speed in this question? What effect
does the radius of the curve, or other features of the horizontal curve, have
in this situation? Does vehicle speed
change on a tangent segment in the presence of an EBB or tri-vision sign?
-
Interchanges and Intersections. Demands made upon the driver increase at
interchanges and intersections. For
example, at highway interchanges, drivers execute multiple tasks, including
controlling lateral position in their lane, maintaining appropriate headway and
speed, and navigating through traffic as they merge or turn. At intersections, drivers are required to
attend to multiple sources of potential conflicts, such as pedestrian
crossings, changing traffic control devices, and turning maneuvers. Potentially distracting stimuli, including
EBBs, may be particularly disruptive to drivers under the increased demands
associated with interchanges and intersections. Current federal and state statutes prohibit the placement of
roadside signs and displays near interchanges and intersections. EBBs offer a potential for driver
distraction similar to or greater than conventional billboards, and should be
considered to be subject to the same restrictions on placement near
interchanges or intersections. Current
state statues and regulations employ different values ranging from 152 to 305 m
(500 to 1000 ft). The effects of EBBs
and tri-vision signs on driver performance should be examined to determine if
performance deficits are observed.
Research
Questions - Intersections: Will an
EBB or a tri-vision sign located near an intersection contribute to degraded
driver performance? If driving
performance is found to be degraded, then what should be the distance between
this type of display and the intersection?
Research
Questions - Interchanges: Will an
EBB or a tri-vision sign located near an interchange contribute to degraded
driver performance? If driving
performance is found to be degraded, then what should be the distance between
this type of display and the interchange?
-
Work Zones.
Drivers traveling in work zones need to be alert for changing traffic
patterns, sudden stops, workers, pedestrians, and work equipment. Although drivers should reduce their speed
in these zones, they often do not reduce speed sufficiently. This has resulted in special efforts by
states to encourage safer driver behavior in such zones. Whatever the reason for current selection of
improper speed in a work zone, it is possible that the presence of an EBB or
tri-vision sign would aggravate the problem.
Research about the effects of EBBs in work zones on safety should be
performed, since the presence of additional visual distractions may elevate the
risk of driver distraction and unsafe driving.
Research
Questions - Work Zone: Will an EBB or a tri-vision sign that is located
near a work zone promote unsafe driving?
What constitutes close proximity to a work zone and how should the work
zone dimensions be defined relative to the EBB?
-
Distance between EBBs/Tri-vision Signs. Drivers may not direct sufficient attention
to the driving task if EBB or tri-vision signs are too close to each
other. Federal and state regulations
address conventional billboards on this factor. Due to the greater conspicuity of EBBs relative to conventional
billboards, it may be useful to reexamine the minimum distance between EBBs and
tri-vision signs.
Research Question - Distance Between EBB/Tri-vision Signs: Define
an appropriate distance between EBBs and/or tri-vision signs that ensures
driving safety.
Research on EBBs should be directed toward the
characteristics of EBBs, including placement, visual movement, and
luminance. Research should include the
characteristics of EBBs listed below.
-
Message Content and Comprehensibility. If a driver attempts to read a display
(either text or graphic) that is difficult to read due to factors such as
excessive information, unfamiliar terms, and/or poor choice of word order, then
the driver may not devote sufficient time to the driving task, thus raising a
safety concern.
Research
Finding-Amount of Information: An
analysis of the amount of information on static guide signs recommended that a
sign panel should present no more than six bits of information. Examples of a "bit" of information provided
by McNess and Messer(19) include "I-395, "Washington, D.C." and
"South."
Research
Question-Amount of Information:
This finding should be replicated in an EBB and/or tri-vision sign
context. The terms and symbols used in
the context of a roadway environment may present somewhat different results
due, perhaps, to familiarity of terms.
Another feature that is different is the use of pixels on EBB displays
versus the painted signs in the roadway study.
Research
Questions - Quantity of Text: The Manual of Uniform Traffic Control
Devices (MUTCD),(39) Section 2E-21 (applicable to CMSs for use on
freeway and expressway mainlines), states that a CMS should display no more
than three lines of text. Further,
there should be a maximum of 20 characters per line. An analysis of these numbers for the simpler EBB displays, in
conjunction with analyses of exposure times, may be useful. The acceptable amount of information in the
pictorial and/or dynamic display, both EBB and tri-vision sign, could be
determined at the same time. In these
situations, other display components such as graphics, motion and/or exposure
time should be examined in conjunction with quantity of information.
-
Exposure Times.
The time intervals associated with the display of an image and the
transition between images may be important to safe driving and effective driver
understanding of the display. The
design of an EBB or tri-vision sign message will be more effective when the
time constraints for the driving task and the required time for effective message
transmission are considered together. A
related factor is the length of the transition interval between messages or
images. Transition time can be
important during the rotation of elements on a tri-vision sign due to the
visual effects of the rotation.
Transition in an EBB can be almost instantaneous. The minimum exposure time for a display and
maximum time for display transition have been provided by some states.
Research Questions - Exposure
Time: What factors determine the
interval length for the safe display of a message? How can specific display times be determined? (For CMSs, the MUTCD 2000,(39)
Section 2E-21 states that "The entire message cycle should be readable at least
twice by drivers traveling at the posted speed, the off-peak 85th-percentile
speed, or the operating speed.")
Research Questions - Transition
Time: What is a safe transition
time between messages? During the
transition, should the display be completely dark? Should a transition interval between motion displays be a different
time than a transition interval for static displays?
-
Motion and Other Image Components. Some stimuli are particularly effective in
attracting attention. It is not always
clear which feature(s) of a complex stimulus is the most compelling. Motion is a highly likely candidate for an
effective attention grabber. Motion can be provided in a billboard display in
at least two ways. An EBB display
provides motion when its lighted elements or pixels present an image in
apparent movement across the display.
An example is a high-resolution picture similar to a television image or
video. Another example would be a text
message sliding in from the side, top, or bottom. A tri-vision sign provides mechanical motion when its triangular
elements rotate to present an alternate image.
In this case, the motion is provided during a transition between
displays. The motion component of an
EBB or tri-vision sign may be more of a distraction than message content or
pictorial arrangement. This possibility
exists because a motion requires time and if a driver wants to see the entire
movement for some inherent interest, then the driver will focus more on the
motion display and less on the driving task.
Note that the MUTCD 2000,(39) Section 2E-21 states that,
"Techniques of message display such as fading, exploding, dissolving, or moving
messages shall not be used" for a CMS.
Research
Questions - Motion Effects: To what
extent will motion in an EBB or tri-vision sign contribute to degraded driver
performance? What are the relevant
dimensions of motion (e.g., presence vs. absence; low, medium or high speed;
constant speed vs. variable speed; constant direction of moving item vs. change
in direction)? How might motion
interact with the content in a display to affect driver performance? Is the length of the interval during which
motion occurs an important factor?
Contexts to be considered for this research may include interchanges,
intersections, curves, different vehicle speeds, and various TCDs including
traffic signals.
Research
Questions - Other Image Components:
How should the components of the image, or their combination, (e.g.,
motion, color, message content, amount of information) be examined to determine
which are most likely to be particularly distracting? Or is the distractibility of these components conditional to the
specific situations?
- Maintenance.
Poor maintenance of an EBB or tri-vision sign can affect driving
safety. If the letters and words become
difficult to perceive due to scratched protective covers or because elements
are not functioning due to wear, drivers interested in the expected message may
devote too much attention to interpreting the characters. One phenomena requiring maintenance that may
be unique to programmed light emitting elements is the condition in which some
of the lighted pixels in a display fail to illuminate, thereby removing part of
the word. In the same display, other
pixels that should be off are lighted instead.
In some cases, the display presents a sort of abstract design that may
be engaging for its own sake. Depending
on the particular instance, the message becomes difficult if not impossible to
read, thus lengthening the time that a driver attends to the display.
Research
Questions - Maintenance: Define the
features of a sign that, if not maintained properly, contribute to driver
safety issues. Develop criteria for
achieving a minimal level of sign maintenance including a schedule and process
for monitoring and repairing the sign.
This section focuses on Research Findings that appear
relevant to the use of text and graphical information on EBBs. These findings resulted from research on CMS
displays that are used to transmit roadway information to drivers. The degree to which these findings are
relevant may depend on the visual similarity between the EBB and the CMS. The CMS has a similar appearance to the
simpler version of the EBB that was described in Section 2.1. Recall that the simpler version was
characterized as employing a small character matrix for light emitted letters
that are located on a dark background panel.
It typically presents a limited number of words. The resemblance between CMS and EEB is
likely to become less, however, with an opportunity for employing motion, a
variety of colors, and higher resolution matrices. Such differences emphasize the need for research into EBB
legibility. The present Research
Findings are provided here as a starting point for that research.
Illegible letters, words and images promote longer reading
times. This can result in a longer time
for driver attention to be diverted away from the driving task. Because EBB letters are presented in a
different medium than those in the more familiar road guide sign, they require
special attention. The letters and
numerals on a CMS or EBB are often formed from a group of lighted pixels or
elements that are arranged in a rectangular matrix. The font or design of the letter is constrained by the number and
arrangement of elements.
The resulting font is different than the familiar fonts seen
on paper and painted surfaces in several respects. First, the latter family of fonts can take on a large variety of
styles that take advantage of the flexibility of the brush or the dexterity of
the type designer. Many of the simpler
EBB displays, on the other hand, currently use a basic font matrix composed of
few elements (e.g., 4x6, 5x7). This
small number of elements severely limits the variety of font styles available
and results in a relatively crude font style.
Second, note that the discrete elements in the small matrix present a
discontinuous letter "stroke" that is not present in the traditional medium,
that is, the eye can perceive a separation between the light elements if the
sign is not too distant. Third, the EBB
matrix is represented by elements that emit light compared to the black print
that is read under light. Fourth, the
EBB matrix is viewed in both night and day conditions whereas the traditional
printed page is viewed and read under only lighted conditions. Due to factors such as these, it is
important to have research that directly addresses the legibility needs of the
lighted pixel element display.
A recent summary of legibility data is available for CMSs on
the roadway and was compiled by Garvey and Mace.(15) This information was compiled from a field
survey of in-use CMSs, laboratory experiments and static field studies, and a
partially controlled dynamic field study.
From these results, the authors developed recommendations for CMSs. Information selected from these recommendations
is presented in this section as research findings that may be useful for EBB
design. Note, however, that this
information was developed for CMSs and may not be directly transferable to an
EBB due to differences in how they are used.
EBB signs may have different requirements than the CMS and this may lead
to different findings when researched.
Thus, information on CMS legibility should be considered only as
approximate reference data for EBB design.
This section addresses only the legibility of letters and
numerals that are composed of matrix elements for the simpler EBB. The legibility of pictures and other
non-text graphics is not addressed.
Legibility features that are addressed below include luminance,
contrast, matrix size, letter size and proportion, and letter and letter group
spacing.
Table 2. Recommended Minimum Luminance Values (cd/m2)
for CMS Visibility.
|
|
Sun Behind
|
Sun On Sign
|
Sun Overhead
|
Overcast/Rain
|
Nighttime
|
|
Young
Driver (16-40)
|
1000
|
1000
|
850
|
350
|
30
|
|
Older
Driver (65+)
|
10002
|
10002
|
1000
|
600
|
30
|
Source: Garvey and Mace.(15)
- 85th percentile driver accommodated at 198 m (650 ft).
- Will accommodate less than 50 percent of
drivers at 198 m (650 ft) at any luminance level with extreme sun angles.
Luminance levels for night were
recommended to be between 30 and 150 cd/m2.
Research Question - Luminance In
Inclement Weather: An important
consideration for night viewing of an EBB is whether or not heavy moisture
conditions such as fog, snow, or rain are present. These conditions may have a diffusing effect on luminance such
that a sign with normal luminance may have altered contrast in fog conditions. It would be useful to define maximum
luminance in fog conditions and various glare thresholds for young and older
drivers.
-
Contrast and Contrast Orientation. Clear discrimination of letters from their
background is a basic requirement for seeing the letter and reading the
text.
Research Finding - Contrast Luminance: A ratio of an effective range of contrast luminance for CMS
letters against their background is between 5 and 50 (15). Five represents a relatively low contrast
between letter and background. Fifty
indicates a high contrast.
Research Finding - Contrast Orientation: Light text letters on a darker background are preferred. A black background or colored background is
acceptable.(15)
-
Matrix Size.
The design of individual letters and numerals for the simpler EBB is
based on a matrix of rows and columns in which each matrix element contains one
"pixel." Each pixel can be selected to
emit light or not emit light. The
number of rows and columns in a matrix is established to support the design of
a character (letter or numeral). The
size (number of rows and columns) of the character matrix is often the same for
the design of all letters in the alphabet unless a continuous line matrix or
full matrix is available. A given
letter or numeral is defined by which pixels in the matrix are
illuminated. In the 5 x 7 font
discussed by Garvey and Mace,(15) an "I" is defined with 7 vertical
pixels and an "M" is defined with a combination of 18 vertical and diagonal
pixels. One issue for displays with
larger pixels is to select a matrix with sufficient rows and columns to provide
a legible character. If there are too
few rows and/or columns, the design of the character font will result in a
character that is difficult to recognize and read. If the size becomes overly large, space could be wasted for no
noticeable increase in legibility.
Maintenance may also increase due to an increasing number of lighted
pixels.
Research
Finding - Matrix Definition: A 5 x
7 matrix that used a single stroke font was found to be one of the most
effective matrix sizes for characters used on a CMS.(15) The researchers recommended against using a
double stroke font. One example of a
double stroke font is defined by using two columns of pixels to form a vertical
element in a letter such as "F" or "H."
The researchers report that double fonts yield legibility distances
about 25 percent shorter than regular fonts.
Research
Question - Matrix Definition: As
indicated by the researchers, various legibility factors are difficult to
examine apart from each other. While
the single stroke 5 x 7 matrix appears to be a generally robust minimum matrix
size, it would be useful to replicate this finding for the simpler EBB context
when analyzing other legibility factors.
- Letter
Size and Proportion. The
overall proportion of a letter and its components, as well as its height
is an important determinant of the distance at which a letter can be
recognized. The selection of these
dimensions and proportions should consider a number of factors including
posted speed and driver age.
Research Finding - Letter Height: Letter heights have been researched. For instance, a letter height of 457 mm (18
in) was recommended for a CMS on roadways with an 89-km/h (55-mi/h) speed
limit. This value is intended to
accommodate drivers over 60 years of age.(15)
Research Finding - Letter
Width-to-Height Ratio:
Width-to-height ratio should be at least 0.7 for a CMS. Legibility distance increases as much as 10
to 15 percent as the width-to-height ratio is increased from 0.7 to 1.0. Note, however, that this results in a
corresponding increase in the width of a letter on the CMS(15)
Research Finding -
Stroke-Width-to-Height Ratio: A
stroke-width-to-height ratio of no more than 0.13 is recommended for the CMS
letter. Factors that enter this ratio
are driver age, time of day, luminous intensity of the pixel element and amount
of moisture in the air. For example at
night, halation or irradiation can occur with light emitting elements at night
that results in blurring of elements and letters. The distance at which a letter is legible will decrease by as
much as 10 percent as stroke width ratio approaches 0.2.(15)
Research Question - Letter
Height: The distance at which characters should be legible for roadway
information functions may differ from the commercial requirements of the
simpler EBB. The EBB requirements for
letter height and proportion should be defined and examined with other relevant
factors that determine letter height in an EBB context.
- Spacing
Between Letters and Letter Groups.
The degree of separation between letters influences their
legibility. If they are too close,
the letters appear to run together and require longer recognition
time. This is particularly true at
night for letters composed of light emitting elements. Because of the high contrast, luminous
letters of the CMS tend to blur together more readily than the letters on
standard signs. As a result, the
separation between light emitted letters should be greater.
Research Finding - Inter-Letter Spacing: For CMS signs
that do not allow proportional spacing between letters, research indicates an
inter-letter spacing of one-half the letter height.(15)
Research Finding -
Inter-Word Spacing: For CMS signs
that have an inter-letter spacing of one-half the letter height or have
proportional spacing, research indicates an inter-word spacing of the height of
the letter. For inter-letter spacing
3/7 the letter height or less, inter-word spacing can be equal to 5/7 the
letter height.(15)
Research Finding - Inter-Line Spacing: An inter-line
spacing of 70 percent of letter height is suggested for CMS signs with more
than two lines of text. Twenty percent
of letter height should be adequate for the separation between two lines of
text. The 70 percent separation is
important for providing legibility for the middle line when there are three
lines of text.(15)
Research
Question - Spacing: Commercial
signs may employ different features than used in the research referred to in
this section. For instance, if a higher
resolution matrix were used to present a commercial logo symbol or lower case
letters, then the requirement(s) for letter separation may change. Such features may benefit from further
research.
Research on distraction effects of EBBs should include the
variables of driving skill and vulnerability to distraction. A primary driver characteristic that can be
studied in this regard is driver age, especially since driver-related cognitive
and physiological variables including reaction time, visual acuity, and
attention-sharing capacity correlate with age.
Using young and older drivers as a "design" driver may yield lower
and/or upper limits of parameters in EBB and tri-vision sign variables. For example, since older drivers have
greater sensitivity to glare than do younger drivers, including older drivers
in research on sign glare is important when determining limits to sign
luminance.
-
Young Drivers.
The literature review conducted in Task 2 showed that young drivers are
slower to detect traffic hazards, especially if the hazard is distant or
emerging, and that young driver "situation awareness" of the roadway is less
integrated and less sensitive to context.
Of particular importance to the present report are indications that young
drivers may have difficulty handling distracting events that compete for
attention, whether the source is in-vehicle or external to the vehicle.
Research
Question - Young Drivers: Research
should determine whether young drivers are more likely to demonstrate degraded
driving performance in the presence of EBBs and tri-vision signs and if so, to
what extent. This factor could be
present in most, if not all, of the studies described above.
- Older
Drivers. Cognitive and visual
deficits increase the difficulty of the driving task for older individuals. For example, the speed of visual
processing and the capacity to allocate attention in divided attention
tasks decline with age. Yet, each
of these attributes underlies driving performance. Research on the older driver should
focus on their driving performance in comparison to other age groups when
they are exposed to EBBs and tri-vision signs in roadway configurations of
higher driving task difficulty.
Research
Question - Older Drivers: Research
should determine whether older drivers are more likely to demonstrate degraded
driving performance in the presence of EBBs and tri-vision signs and if so, to
what extent. This factor could be
present in most, if not all, of the studies described above.
In the literature review, research was presented on other
types of stimuli that may distract a driver.
These are the cellular telephone and the in-vehicle information display.
- Although research in cellular telephone and in-vehicle
information displays does not currently appear to have direct application to
EBBs or tri-vision signs, these analyses are ongoing and may still provide
insights into the nature of driver distraction, the definition of distraction
thresholds and approaches to minimizing distraction for safe driving.
Research
Question - Monitoring of Other Driver Distraction Research: Continue to follow the results of
research on driver distraction, particularly in the area of cellular telephones
and in-vehicle information displays to determine what may be learned, either in
general principles or specific information, that is applicable to addressing
potential distractibility from EBBs and/or tri-vision signs. Will the peripheral distraction task be a
useful tool in assessing the distractibility of stimuli external to the
vehicle? What modifications need to be
made to make it applicable for both research and perhaps, even, evaluation of
proposed billboards?
A wide range of research questions for EBB and tri-vision signs
has been addressed above. The general
approach to these issues has been to establish under which conditions, if any,
EBBs and tri-vision signs are most likely to affect driving safety. The area for which the most potentially
applicable research is currently available is legibility. Other areas such as length of exposure time
have received attention through state regulation. Some research areas such as whether motion in a display has any
potential safety effect on driving seem to have received practically no
attention. Since there are several
potential research areas, approaches to efficiency in the analyses are
useful. When planning future research
on EBB and tri-vision sign safety questions, criteria may be developed that
prioritize the questions. Such criteria
might include a determination of which questions are the most urgent, the
availability of existing field data for analysis, and the possibility of
performing laboratory research. A
second approach examines the possibility of combining research questions in one
study. Thus, after prioritization, it
could be useful to determine which issues should be investigated independently
versus in an integrated manner.
This section briefly describes several research methods that
may be applicable for the candidate research topics.
- Documentation
Analysis. Two variations of
this method are described. In the
Before and After method, data (e.g., crashes) from a site without an EBB
or tri-vision sign are compared to the data from the site after sign
installation. If the latter number
of crashes was substantially greater than the former, and there are no
other significant changes to the site (i.e., change in volume, average
speed, new construction, etc.) during the comparison times, then one can
argue that the sign was associated with increased crashes. A variation of this approach compares
the site of interest to a control site during the same period. This method compares data from an EBB
or tri-vision sign site to data at a comparable, control site without a
sign.
-
Field Study.
Field studies normally have observers collect data on driver and vehicle
behavior at the site(s) instead of performing an analysis of
documentation. If there were a substantial
difference in quantity of behaviors, there would again be reason to argue that
the EBB was associated with changed driver behavior. When this type of study is compared to the study of
documentation, this study usually has more controlled conditions of observation
as well as the advantage of examining more behaviors (e.g., swerving, braking)
than just crashes. The analysis of
documentation, however, is more likely to provide a greater number of crashes
since the data collection period is likely to be longer.
-
Test Track.
Another method shares some of the advantages of a field as well as a
laboratory study. A simulation of
infrastructure elements such as pavement and signs is avoided. The dynamics of the vehicle is not a
problem. In addition experimenter can
control instructions to the driver, provide signs and objects of any design,
repeat trials, and control presentation of stimuli. The use of a test vehicle in this situation could avoid crashes
with other cars while allowing observation and documentation of lane keeping,
speed, and other behaviors.
-
Simulation.
A laboratory study allows the researcher to examine specific conditions
by simulating scenarios with a combination of hardware and software. The fidelity of the simulation can vary
depending on the kind of research question being asked. The fidelity will also depend on the type of
apparatus available to study the research question. Lane deviation, change in speed, and crashes can be measured in
the laboratory. An advantage of the
laboratory is the capability to provide a realistic context for the study that
does not expose the participant to actual danger.
|
