index
previous comment
next comment
|
Billie Louise Bentzen, Ph.D., COMS
|
October 24, 2002 |
ACCESSIBLE DESIGN FOR THE BLIND
Billie Louise Bentzen, Ph.D., COMS
Comments on Draft Guidelines for Accessible Public Rights-of-Way (June 17, 2002)
By Billie Louise Bentzen
PREAMBLE
Detectable Warning
The Board asks for response on a question regarding installation of detectable warnings only on curb ramps with a slope of 1:15 or less.
RESPONSE: I support the requirement for detectable warnings on ALL ramps leading to crosswalks, and at sidewalk street transitions, regardless of slope.
Rationale: Two studies confirmed that the absence of a curb was problematic for travelers who are blind. Barlow and Bentzen, found that 39% of blind travelers did not detect the street and stop when they approached a crosswalk on a curb ramp. Repeating the analysis using only the ramps that met ADAAG requirements, (had slope of 1:12 or lower), 48% of travelers stepped into the street. While curb ramps having slopes between 1:12 and 1:15 are probably more detectable than curb ramps having steeper slopes, they are clearly not reliably detected. A detectable consistent surface is needed before the street edge to provide a definitive cue to the location of the street at a curb ramp or blended transition.
I applaud the Board’s strong stance on signalization of roundabout crossings. I expect that there are alternative solutions to provide accessibility for pedestrians with visual impairments, however, roundabout proponents have been slow to respond to concerns of pedestrians with disabilities. Proponents of roundabouts often quote the reduction in crashes as support for the safety of the installations. Crash data do not tell the whole story; there is little or no information on pedestrian avoidance of roundabout locations. Anecdotal information from Europe and Australia, as well as from US installations, indicates that pedestrians who are blind or visually impaired avoid crossing at roundabouts. Residents at one center for blind persons in Fredensborg, Denmark, used to walk independently from the local train station to their residence, using the signalized intersection of two, two-lane streets. Since the installation of a single-lane roundabout, with good deflection at entry and exit lanes, with detectable surface treatment, and a groove in a granite pillar intended to help blind pedestrians position and align for crossing, they are unable to independently travel to and from the train station. Travel on this route is now done by taxi.
1101.3 DEFINED TERMS
ADD: standardized
Detectable Warning. A standardized surface feature built in or applied to walking surfaces or other elements to warn of hazards on a circulation path.
Rationale: Those reading the definition need to understand that the surface of the detectable warning is specified in ADAAG and that various textured surfaces may not meet the requirements of a detectable warning.
CHANGE to: Pushbutton Locator Tone—a repeating sound that informs approaching pedestrians that they are required to push a button to actuate pedestrian timing and that enables pedestrians who have visual disabilities to locate the pushbutton.
Rationale: The above definition is the definition and term used in the MUTCD.
1102. SCOPING
1102.5.2 Protruding objects
1102.5.2 Post-Mounted Objects.
ADD: Where a sign or other obstruction is mounted between posts or pylons and the clear distance between post or pylons is greater than 12 inches (305mm), the lowest edge of such sign or obstruction shall be 27 inches (685mm) maximum or 80 inches (2030mm) minimum above the finish floor or ground. There shall be a bar or similarly detectable element 15 inches (380mm) above the floor or ground connecting the two posts or pylons.
Rationale: I am delighted with the reduction of distance that objects can protrude from posts. This will eliminate many hazards in the sidewalk area.
The sentence regarding posts and pylons seemed to be missing a section regarding the distance between posts and pylons. In addition, I encourage the addition of the requirement for a lower bar on such signs. Signs between poles, and railings with their leading edges higher than 15 inches, are not regularly detectable in cane travel by persons of varying stature and travel technique. There are many independent travelers with visual impairments who are less than adult stature.
The three principal cane techniques are: 1) the touch technique, where the cane is arced from side-to-side and touches the ground at points outside both shoulders; 2) the constant contact technique, where the cane is slid from side-to-side in a path extending just beyond both shoulders; and 3) the diagonal technique, where the cane is held in a stationary position diagonally across the body with the tip just above the ground at a point outside one shoulder and the handle extended to a point outside the other shoulder. When one of these techniques is used and the element is in the detectable range, it gives a person of average adult stature, who uses proficient technique with a long cane, sufficient time to detect the element with the cane before there is body contact. The typical cane techniques do not locate objects extending into the travel path above the hips. Pedestrians who are blind or visually impaired may travel on any part of the public sidewalk and are not limited to the pedestrian access route.
Pilot research (Karnes et al. attached) indicates that numerous body contacts are likely to be made with objects that protrude four inches from walls or twelve inches from a post or pylon, even by adults who travel using a long cane proficiently in one of the three principal cane techniques. For persons of short stature, including children, simple geometry indicates that they will be unlikely to detect objects with a long cane before contacting them with the body when the leading edge is as high as 27 inches above the floor or ground. Many persons do not consistently use proficient techniques with a long cane; they are particularly at risk.
1102.7.1 Bus Route Identification. Exception 2
ADD: If portable receivers are required to access the signs, receivers must be freely distributed to persons with disabilities who cannot read print signs.
Rationale: Allowing such an exception does not provide accessible information unless there is a concomitant requirement to distribute the receivers to those who may wish to access the information.
1102.8 Pedestrian crossings
CHANGE TO: Where a pedestrian crossing is provided, it shall comply with the applicable provisions of 1105. Where pedestrian signals are provided at a pedestrian crossing, where pedestrian signal timing is actuated by pedestrian detectors (pushbuttons) or by passive pedestrian detection, or where leading pedestrian intervals or exclusive pedestrian phasing is used, pedestrian signals shall comply with 1106.
Rationale: The draft language does not require accessible information to be provided at intersections unless pedestrian signals are installed. The Manual on Uniform Traffic Control Devices (MUTCD) only requires (using ‘shall’ language) the installation of pedestrian signals at limited locations, such as school zones, crossings where a signal is installed due to high pedestrian volumes, and where there is exclusive pedestrian phasing. Other locations are discussed in ‘should’ language. I am concerned that this rule may encourage traffic engineers to limit the installation of pedestrian signals, in order to avoid installing accessible pedestrian signals. In the MUTCD, pedestrian signals are not required at some locations because the vehicular signal can be considered adequate, under provisions in the MUTCD, to provide information to pedestrians. Under the current draft guidelines, therefore, there would be no imperative to make the “green ball” information accessible. These locations may not provide adequate information for pedestrians who are blind without installation of accessible pedestrian signals.
The locations suggested above and in the PROWAAC report are ones at which the signal features make it hard to detect the pedestrian crossing phase without provision of accessible information. Locations such as those are known to be problematic for pedestrians who are blind or visually impaired. In addition, there may be other locations where the traffic movement does not provide sufficient information for pedestrians who are blind or visually impaired.
I am unable at this time to suggest language that will cover all possible situations in which accessible information may be needed by an individual pedestrian who is blind. Therefore I recommend that at signalized intersections in new construction where pedestrian facilities such as sidewalks are provided but pedestrian signalization is not, that conduit piping be installed in relation to the curb ramps such that a retrofit with APS if required could be easily accomplished.
Pedestrian actuated traffic signals change with each cycle and usually require the pedestrian to push a button to get enough time in the cycle to cross the street. As well, the vehicular patterns and pedestrian timing may vary, depending on the signal timing plan of the intersection. The pedestrian timing may be concurrent with the traffic moving parallel to the pedestrian’s path, or at a totally different time in the cycle. Pedestrians who misjudge the starting traffic pattern may find themselves in the street when cars are moving perpendicular to their path with a ‘green arrow’. Most newly installed intersections will be traffic actuated and will have complex traffic patterns.
1102.10 Stairs
I agree with this addition that will make stairs in the public rights-of-way more visible to all pedestrians. I suggest a slight revision in the language.
CHANGE TO READ: 1102.10 Stairs. Where provided, stairs
shall comply with 504. Stair treads shall have a 2 inch (51mm) wide strip of
color , contrasting with the tread and riser, dark on light or light on
dark, the full width of the front edge of each tread.
Rationale: Light/dark contrast is the important feature, not hue.
1104.2.1 Perpendicular curb ramps.
ADD: Where possible, the slope of the curb ramp shall be aligned with the sidewalk and crosswalk direction.
Rationale: The orientation of curb ramps toward the intersection can be disorienting for travelers who are blind or visually impaired. In addition, they require an extra turn for wheelchair users. I believe that the guidelines need to encourage orientation of the ramp in the direction of travel on the crosswalk.
PROWAAC debated at great length on the issue of curb ramp orientation. While travelers who are blind or visually impaired do not use the slope of the ramp to determine their crossing alignment, it is difficult to prevent the slope from influencing travel direction. Advocates for pedestrians with visual impairments recognize the safety issues for wheelchair users of warping at the gutter/ramp intersection, however, whenever possible, the slope of the ramp should be aligned with the crosswalk and the grade break should be aligned perpendicular to the crosswalk alignment. The language of the guidelines needs to state that and encourage two ramps more strongly.
1105.4.2 Medians and Pedestrian Refuge Islands – Detectable Warnings
DELETE: Exception
Rationale: Detectable warnings inform the pedestrian who is blind or visually impaired about the presence of a cut-through island or median. They should be required at all medians and islands. Although the pedestrian may not need to stop at that location when the signal timing is adequate for a full crossing, slower pedestrians may prefer to stop and wait, if they know the refuge exists. In the absence of an APS, blind pedestrians frequently begin crossing during the clearance interval because of the difficulty of determining the exact onset of the walk interval, and the resulting inability to “claim” the crosswalk before vehicles turning across the crosswalk. Hence, they may have insufficient time to cross the street. Denying them the information that they have a safe refuge constitutes discrimination and endangers the life safety of pedestrians who are blind in such situations. Even in the presence of APS, because they are unable to make eye contact with drivers, pedestrians with visual impairments have difficulty claiming the crosswalk during the walk interval, and may be delayed in starting crossings relative to sighted pedestrians.
In addition, contacting the side edge unexpectedly when traveling within the cut-through section of the median can be disorienting and confusing if pedestrians do not realize they are within a median area. The detectable warning provides the pedestrian with information about the location of the cut-through refuge area.
1105.6 Roundabouts
1105.6.1 Separation
CHANGE TO: Continuous barriers landscaping
separation or railings shall be provided along the street side of the
sidewalk where pedestrian crossing is prohibited. When railings are used, they
shall have a bottom rail 15 inches (380mm) maximum above the pedestrian access
route.
Rationale: Use of the word barriers is confusing. Landscaping can guide individuals to the crosswalk location as well as a barrier or guardrail.
1105.6.2 Signals
I urge the Board to keep this language and requirement for signals at each crossing of roundabouts. See previous comments in answer to the question in the Preamble.
1105.7 Turn lanes at Intersections
I also applaud the inclusion of pedestrian activated signals at these locations, which have been problematic for pedestrians who are blind for years.
1106. ACCESSIBLE PEDESTRIAN SIGNAL SYSTEMS
1106. 1 General.
I recommend the use of the term “Accessible Pedestrian Signal (APS)”, rather than Signal Systems or Pedestrian Signal Devices in this text to more closely conform to the MUTCD language and typical current terminology.
Rationale: The Board has introduced new language in these guidelines that does not match with language typically used in either the traffic engineering or the orientation and mobility professions. ‘Signal system’ is a defined term in the MUTCD (“Signal System - two or more traffic control signals operating in signal coordination”). ‘Signal system’ is not used with that meaning in these guidelines. It is important that engineers and those who are familiar with the MUTCD understand these guidelines properly.
1106.2 Pedestrian Signals.
CHANGE: Each crosswalk with pedestrian signal indication,
or where pedestrian signal timing is actuated by pedestrian detectors
(pushbuttons) or by passive pedestrian detection, or where a leading pedestrian
interval or exclusive pedestrian phasing is used, shall have a signal
device an Accessible Pedestrian Signal which includes audible and
vibrotactile indications of the WALK interval. Where a pedestrian pushbutton is
provided, it shall be integrated into the signal device Accessible
Pedestrian Signal and shall comply with 1106.3.
Rationale: As stated earlier in our discussion of scoping, 1102.8, pedestrian signals are not required at some locations because the vehicular signal can be considered adequate, under provisions in the MUTCD, to provide information to pedestrians. These locations may not provide adequate information for pedestrians who are blind without installation of accessible pedestrian signals.
The term ‘signal device’ has a different meaning in the traffic engineering profession and may not be understood to mean a device that is commonly known as an Accessible Pedestrian Signal. There’s no need to introduce confusion.
1106.3 Location.
CHANGE TO READ AS FOLLOWS: Pedestrian signal devices
Accessible Pedestrian Signals shall be located 60 inches (1525mm)
maximum from the crosswalk line extended, 120 inches (3050mm) maximum and 30
inchs (760mm) minmum from the crub line, and 120 inches (3050mm) minimum from
other pedestrian signal devices at a crossing. in line with,
or as close
as possible to, the
crosswalk line that is farthest from the intersection, and
no more than
60 inches (1525mm) maximum from
that line.
The APS
shall be as close
as possible to, and no more than 120
inches (3050 mm) maximum, from the curb line of the street it serves.
APS shall be 120 inches (3050mm) minimum from another APS on the same corner.
The control face of the signal device accessible pedestrian
signal shall be installed to face the intersection and be parallel to the
direction of crosswalk it serves.
Rationale: In new construction, the location for the APS can and should be restrictive and consistent. The ‘box’ in the MUTCD was developed in consideration of retrofit situations and local practice; it is a large area 10 feet square. The recommended language attempts to make clear that the closer the APS is to the departure point while remaining accessible from a level surface the better it is. In discussion over the past year I have often recommended that the APS be placed at the point of the flare immediately adjacent to the top of the ramp. For the vibrotactile information to be of value, it should be located as close to the crossing departure location as possible. An exception to this might be at diagonal ramps where the APS might be located further away on either side of the ramp in order to provide clearer information, directionality, and separation between the APS.
The requirement of location more than 30 inches from the curb is a traffic engineering standard which has no relation to accessibility and may result in the placement of devices in less accessible locations. It should not be included here.
1106.2.3 Audible Walk Indication
CHANGE TO READ AS FOLLOWS: The audible indication of the
WALK interval shall be by voice speech message or tone.
Rationale: The use of the term ‘speech message’ is more accurate. I am concerned that use of ‘voice’ will be considered to require the recording of human voice messages. Some methods of digital speech may provide more consistent messaging than individual recordings. Currently, AT&T provides text to speech messaging capabilities on the web and in commercially available software. (http://www.naturalvoices.att.com/demos/index.html)
ADD THE FOLLOWING SECTION
1106.2._ Speech WALK Messages.
1106.2. WALK signals may be in the form of speech messages.
1106.2._._ Speech WALK messages shall contain the words “WALK SIGN.”
1106.2._._ Speech WALK messages shall repeat throughout the WALK interval or be combined with a repeating WALK tone.
Exception: Speech messages may be limited to a maximum of seven seconds in duration where pedestrian signals rest in WALK.
1106.2._._ At intersections having concurrent pedestrian phasing, speech messages shall follow the model: “Howard. Walk sign is on to cross Howard.” Designation of “street, “avenue,” etc. shall be used whenever its omission could lead to ambiguity.
Rationale: In the US, speech WALK messages are commonly used in newer accessible pedestrian signal installations. All of the pushbutton-integrated devices on the market in the US are capable of providing speech messages. Speech WALK messages are not necessary to providing unambiguous information regarding which crosswalk has the WALK interval, provided that pushbuttons are installed in the locations specified, and they will be unintelligible to some users in some ambient noise conditions. However, speech walk signals are perceived as being more user-friendly than tonal WALK signals. Good installation has a number of requirements.
Speech WALK messages should continue to repeat throughout the WALK interval, or be combined with a WALK tone for the balance of the WALK interval so that pedestrians with visual impairments will know when the WALK interval ends, they will be aware of its full duration, and be able to initiate street crossings in knowledge of the status of the pedestrian signal. Combination of a speech WALK signal with a tone signal may have some of the advantages of both. It should be clear that this is permitted.
At many intersections of an arterial with a minor street, the pedestrian signal on the minor street “rests in WALK” during the vehicular green of the arterial, until a pedestrian or vehicle actuates the signal to enter or cross the arterial. WALK signals that sound continuously during what is sometimes a walk interval some minutes long will be particularly objectionable in neighborhoods. Most of the pushbutton-integrated APS have means to limit the WALK signal to a certain number of seconds. Pedestrians can actuate the audible WALK signal multiple times during the same (rest-in-)WALK interval.
The language of speech walk messages must be consistent and these guidelines should repeat at least the same specifications as in the MUCTD. The MUTCD specifies that speech WALK messages should say ‘Walk sign’, in order to provide information about the status of the walk indication without providing a command, such as ‘walk now’. Accessible Design for the Blind completed a survey of pedestrians who are blind, traffic engineers and orientation and mobility specialists last year and developed recommendations for speech WALK messages, as well as pushbutton informational messages (see attached). The recommended wording for WALK messages was “Howard. Walk sign is on to cross Howard.”
1106.2.3.2 Volume
ADD: Exception: When special activation is used to provide audible beaconing, the volume may exceed 5dB above ambient noise level.
Rationale: Special activation of louder signals may be useful in some situations to provide beaconing. If it is provided by special activation, such as a long press of the pushbutton, as suggested by PROWAAC and provided by several US manufacturers of APS, the louder signal will only sound when someone requests such a feature. Allowing a volume increase in response to special activation will provide some flexibility as these features develop in response to needs and as research continues on these issues.
Additional speakers mounted at pedhead might be found to provide directional information to pedestrians who are blind or visually impaired; I do not want these guidelines to prevent further evaluation of that option.
1106.2.3.3
ADD: WALK tones shall repeat throughout the WALK interval.
Exception: WALK tones may be limited to a maximum of seven seconds in duration where pedestrian signals rest in WALK.
Rationale: At many intersections of an arterial with a minor street, the pedestrian signal on the minor street “rests in WALK” during the vehicular green of the arterial, until a pedestrian or vehicle actuates the signal to enter or cross the arterial. WALK signals that sound continuously during what is sometimes a walk interval some minutes long will be particularly objectionable in neighborhoods. Most of the pushbutton-integrated APS have means to limit the WALK signal to a certain number of seconds. Pedestrians can actuate the audible WALK signal multiple times during the same (rest-in-)WALK interval.
1106.3.2 Locator Tone.
ADD: Exception: When special activation.is used to provide audible beaconing, the volume of the locator tone for the succeeding clearance interval may exceed 5dB above ambient noise level
Rationale: When special actuation of louder signals is used to provide beaconing, having the loudness of the locator tone increased for the remainder of the pedestrian phase is essential to enabling those who need this assistance to home in on the opposite corner while they are still in the middle of the crossing.
1106.3.4 Optional Features.
CHANGE TO READ:
An extended button press shall be permitted to activate additional
features. Buttons that provide additional features shall be marked with three
Braille dots forming an equilateral triangle
in
within .5 inches of the center of the pushbutton.
Rationale: Many pushbuttons have a depressed spot in the very center of the button for use with head stick or mouth stick. Placing the symbol somewhere near the center is better than requiring the symbol to be in the center of the pushbutton, which is incompatible with a feature that makes pushbuttons accessible to persons with limited use of hands and arms.
1107.4 Arrow
ADD THE FOLLOWING: 1106.4.Arrow. Signs shall include a tactile arrow aligned parallel to the crosswalk direction. The arrow shall be raised 1/32 (0.8 mm) minimum and shall be 1-1/2 inches (38mm) minimum in length. The arrowhead shall be open at 45 degrees to the shaft and shall be 33 percent of the length of the shaft. Stroke width shall be 10 percent minimum and 15 maximum of arrow length. The arrow shall contrast with the background, light on dark or dark on light.
Rationale:
Contrast in hue may not be perceptible but light/dark will be perceptible for most who use visual information.
1106.4.2 Street name
CHANGE TO READ:
Signs
Accessible pedestrian signals shall include street name information
aligned parallel to the crosswalk direction and complying with 703.32,
or shall provide street name information in audible format.
Rationale: Street name information provided in audible format will be usable by more individuals than street name information provided in tactile format.
Providing street name information in tactile characters will result in signs that are very large. Most persons who read by touch read Braille. If there is a requirement for tactile street name information, it is more reasonable to require Braille information than tactile characters. It will serve those who will use it better, and require smaller, less expensive signs.
1108.1.4 Size
SUGGESTED
CHANGE: Detectable warning surfaces shall extend 24 inches (610 mm)
minimum in the direction of travel and the full width of the curb
ramp, landing, or blended transition.
Rationale: I assume the Board did not establish a maximum depth of detectable warning because there is some research evidence that deeper warnings are more detectable. However, while research by Bentzen and others consistently found that blind persons more reliably detected detectable warnings at 30 inches or 36 inches than at 24 inches, 24 inch deep detectable warnings were repeatedly detected on 85-90% of trials, and currently function well at transit platform edges. Approximately 24 inch deep detectable warnings are commonly used in other countries. Detectable warnings 24 inches deep were recommended in the PROWAAC report and have been recommended previously by B. Bentzen since the original ADAAG public comment period. Consistency in depth of detectable warnings promotes consistent interpretation. Blind pedestrians in any setting of the US should know that when they encounter a detectable warning, they need to stop immediately and check it out, as there is a vehicular or falling hazard immediately ahead. There is very minimal stopping distance.
The deeper the detectable warning, the greater the possible adverse effect on persons with mobility impairments, and the greater the expense of installation and maintenance. Therefore the minimum depth of 24 inches detected on 85-90% of trials in multiple experiments has been recommended. Installing large areas of detectable warning surfaces as entire curb ramps ends up providing less precise information to blind pedestrians than installing a smaller amount in a very predictable location, as well as increased expense and greater possible adverse impact on persons with mobility impairments.
1108.2.1 Curb Ramps and Blended Transitions
Suggested change: Add exception.
Exception: At cut-through islands and medians the detectable warning may be located so that the edge nearest the crosswalk is at the curb line.
Rationale: The PROWAAC recommendation that detectable warnings be 6-8 inches from the curb line was based primarily on construction practices at curb ramps, and the need to be able to place detectable warnings behind a curb which is at street level, rather than alter that curbing. In addition, it may help pedestrians using wheelchairs because they can begin to ascend the ramp before encountering the truncated dome surface. It may also help persons with visual impairments because it places the leading edge of the detectable warning, for persons approaching the street, 6-8 inches farther away from the vehicular travel lane.
However, requiring that detectable warnings be 6-8 inches from the curb line at cut-through islands and medians has the effect of requiring that islands and medians be a total of 12-16 inches wider than may otherwise be necessary or advisable. Paragraph 1105.4.1 requires that these elements be 72 inches minimum in the direction of travel, and paragraph 1105.4.2 requires detectable warnings at each side complying with 1108, separated by a 24 inch minimum length of walkway without detectable warnings. Construction of detectable warnings as currently specified in 1108.2.1, without the proposed exception, results in the requirement that cut-through islands and medians actually be a minimum of 84-88 inches in the direction of travel.
It seems unnecessary to require this additional width if a construction system is used that does not require that detectable warnings be placed behind curbs at street level. There is seldom curbing at the street level at cut-through islands and medians, so the primary reason for the spacing back from the curb line is not relevant here. In addition, there is no benefit for persons using wheelchairs because they are not going up a slope. In alterations there are often limitations on available right-of-way.
Requiring that cut-through islands and medians be 12-16 inches wider than 72 inches could result in the failure to provide such places of refuge. Provision of detectable refuges is more important for the safe travel of pedestrians with visual impairments than having extra-wide refuges.
1108.2.2 Rail Crossings
Suggested Change: Add exception.
Exception: Where automatic gates across pedestrian ways bar pedestrian access to the crossing when rail vehicles are approaching or at a crossing, the detectable warning surface shall be located to the side of the automatic gate farthest from the crossing.
Rationale: Automatic gates can cause serious head injury to blind pedestrians approaching crossings. There is currently no reliable way for pedestrians who are blind to be notified of the location of automatic gates. Gate support is typically outside the normal pedestrian route and not likely to be encountered by blind pedestrians. Where there is a gate, a blind pedestrian may become trapped between the gate and the crossing, with the gate barring the way for the blind pedestrian to move farther away from the crossing.
Determining Recommended Language
for Speech Messages used by
Accessible Pedestrian Signals:
Final Report
Accessible Design for the Blind
Billie Louise Bentzen, Ph.D.,
Janet Barlow, M.Ed., and
Lukas Franck, M.Ed.
Research supported in part by:
The Seeing Eye
American Council of the Blind
California Council of the Blind
Center for the Visually Impaired, Atlanta
U.S. Access Board
January 7, 2002
Acknowledgement
The authors gratefully acknowledge the contribution of Daniel Ashmead, Ph.D.,
Department of Hearing and Speech Sciences, Vanderbilt University Medical Center,
to sections on speech intelligibility in noise.
Introduction
Accessible Pedestrian Signals (APS), to provide information to pedestrians who
are blind, have been installed in some locations in the U.S. for over 30 years.
Until recently, they have all been audible signals, with loudspeakers installed
at the pedestrian signal head (ped head) that broadcast tones or messages across
the intersection during the steady walk interval of the pedestrian crossing
phase. The most commonly used signal, produced by several manufacturers, is one
in which a repeating sound like “cuckoo” is used for crossings that run in a
roughly north-south direction, and a repeating sound like “chirp” is used for
crossings that run in a roughly east-west direction. A survey of orientation and
mobility specialists conducted in 1998 by the Orientation and Mobility Division
of the Association for Education and Rehabilitation of the Blind and Visually
Impaired (Bentzen et al., 2000) and a parallel survey of pedestrians who are
blind or visually impaired by the American Council of the Blind (Carroll &
Bentzen, 1999) both indicated that this type of signal often provides ambiguous
information.
Recently, APS that provide audible information from the pushbutton location have
been introduced in the US. This type of signal has been used extensively in
Europe and Australia for many years. It typically uses a rapidly repeating tone
to indicate the walk interval, and a slower repeating tone during the “flashing
don’t walk” and steady “don’t walk” intervals to indicate that the pedestrian
should not initiate a crossing. This slowly repeating tone, or locator tone,
alerts users to the need to press a button to request a walk signal, and helps
users locate the pushbutton. Vibrotactile information is also provided in most
of these signals. In the U.S., in some of these pushbutton integrated signals,
speech messages are being used to indicate the walk interval information, as
well as additional information when the pedestrian pushbutton is pushed during
the “don’t walk” and “flashing don’t walk” intervals. A third type of APS
provides a transmitted message, usually a speech message, from the pedestrian
signal head to a receiver held by the traveler who is blind or visually
impaired. This research focused on APS that provide directly audible speech
messages, not transmitted messages.
At least five currently available accessible pedestrian signals in the U.S.
offer the option of directly audible speech messages. APS manufactured by Polara
Engineering, Inc., NOVAX Industries Corp., Prisma Teknik AB, Dick Campbell Co.,
and Bob Panich Consultancy Pty. Ltd. are all capable of having real time speech
messages for the walk interval. All of these except the Panich signals are also
capable of providing messages with additional information when the pedestrian
pushbutton is pushed.
These speech signals are intended to be audible from 6-12 feet from the
pushbutton pole, or to the building line, whichever is less, as stipulated in
Guidance in the Manual on Uniform Traffic Control Devices (MUTCD 2000 4E.08).
The speaker for these messages is located in the pushbutton unit, with the
exception of the NOVAX signal, which also has a ped head mounted speaker. The
signals are not intended to be heard from across the street. They are normally
used in association with a quiet locator tone, repeating once a second, that
informs approaching pedestrians that they need to push a button to request a
pedestrian phase. This quiet tone also enables users to locate the pushbutton
easily. In addition, a tactile arrow on the pushbutton unit is aligned in the
direction of travel of the crosswalk to provide tactile indication of the
crosswalk that is controlled by the pushbutton.
Signals having speech messages are seen by many blind pedestrians as being very
user friendly, and able to provide unambiguous information about which street
has the walk sign. They are capable of providing both real-time and fixed
descriptive information. Real-time walk signal information can be provided about
the status of the pedestrian signal. Pushbutton messages can potentially include
the name of the intersection and the street actuated by the pushbutton, the
geometry and/or signalization of the intersection, and nearby landmarks such as
transit stops or public buildings. The pushbutton message on the Polara and
Campbell are typically activated by pressing the button for three seconds or
more; the Prisma message is activated immediately when the button is pushed. The
NOVAX signal has the capability of providing pushbutton messages, but it has not
yet been used in this fashion.
During the deliberations of the Public Rights-of-Way Access Advisory Committee (PROWAAC)
(Access Board, 2001), representatives of the National Federation of the Blind,
the American Council of the Blind, the Council of Citizens with Low Vision, the
Association for Education and Rehabilitation of the Blind and Visually Impaired,
and The Seeing Eye all supported the use of APS having speech messages to reduce
possible ambiguity about which crosswalk has the walk signal at some
intersections. PROWAAC was established by the U.S. Access Board in 1999 to
provide recommendations for minimum guidelines and requirements for implementing
the Americans with Disabilities Act in public-rights-of-way.
There currently is minimal standardization of either walk messages or pushbutton
messages for APSs, however, therefore message content, length, and structure
vary from one APS installation to another. The only existing standard for the
wording of APS speech messages is contained in The Millennium Edition of the
Manual on Uniform Traffic Control Devices (MUTCD 2000 4E.06). This standard
requires that where verbal messages are provided during the walk interval, “the
verbal message . . . shall be the term ‘walk sign,’ which may be followed by the
name of the street to be crossed,” and the “verbal message . . . when the walk
interval is not timing . . . shall be the term ‘wait.’” As an option “Accessible
pedestrian signals that provide verbal messages may provide similar messages in
languages other than English, if needed, except for the terms “walk sign” and
“wait.” The specification of the terms “walk sign” and “wait” was based on
research conducted for The Smith-Kettlewell Eye Research Institute (Myers,
Bentzen & Crandall, 1996).
This research on speech messages for APS was undertaken to develop
recommendations based on data from stakeholders to form the basis for the
establishment of recommended practice for the structure and content of walk
messages and pushbutton messages for directly audible accessible pedestrian
signals.
Other concerns regarding speech messages
Other concerns regarding speech messages for APS, in addition to standardization
of message content and structure, include the following.
• It is important that messages are understandable by all users.
• It is not likely to be possible to make speech signals understandable in all
ambient noise situations, such as loud traffic.
• It is easy for listeners to miss or mistake parts of messages in loud ambient
noise situations.
• Non-English speakers and non-native English speakers are likely to have
difficulty understanding speech messages.
• To be understood, speech messages must be carefully recorded, in a clear
voice, with excellent diction, and moderate pacing.
• Replacement of signals having speech messages necessitates custom recording
rather than off-the-shelf substitution of components.
Research on intelligibility of speech in noise is also relevant to the general
consideration of the use of speech messages in APS. A number of speech-in-noise
tests have been used by audiologists and hearing scientists. In most of these
tests the information is sound from loudspeakers. Typically the background noise
on these tests consists of pre-recorded conversational babble from multiple
speakers (like “cocktail party” noise). The listener’s task is to listen to
speech samples and report the word or sentence which was spoken. The measure is
percentage of utterances reported correctly. The test is usually done at several
different signal-to-noise ratios.
A recent report by Killion (1999) provides an example of the findings. Listeners
with normal hearing required a signal-to-noise ratio of +2 dB to get 50% of
sentences correct, and +5 dB to get 90% correct. Listeners with hearing
impairments required signal-to-noise ratios of +8 dB for 50% correct and +15 dB
for 90% correct. These findings suggest that a signal-to-noise ratio of +5 dB or
more is required for good speech intelligibility by persons with normal hearing,
in listening situations where the listener is not able to see the speaker.
However a large majority of persons with visual impairments are over the age of
60 and thus can be expected to have age-related hearing loss, particularly for
higher frequencies. Thus, speech messages in APS may need to be equal to or
greater than 15 dBA above ambient traffic sound to be intelligible to many
persons with visual impairments.
Speech-in-noise tests of the sort represented by the Killion (1999) study have
some limitations in terms of generalization to audible pedestrian signals. One
is that the background noise was conversational babble, which has the same
spectral composition as the speech signals. However, traffic sound has most of
the energy concentrated in the lower frequencies (Wiener et al., 1997). Persons
with visual impairments who have normal hearing may be able to understand APS
speech messages better in the context of ambient traffic sound than
conversational babble because of the difference in spectral composition.
However, the majority of persons with visual impairments who are likely to have
age-related hearing loss can be expected to have more difficulty hearing the
higher frequencies in speech than the predominantly lower frequencies in traffic
noise. Therefore the results of tests such as that of Killion (1999) may
underestimate the amount by which APS speech messages need to exceed the
intensity of traffic noise in order to be intelligible to older blind
pedestrians.
It may not be possible to provide APS with intelligible speech messages that are
not so loud as to have the potential for causing hearing loss. The MUTCD says
that the maximum volume of APS should be 89 dB, and APS should be no more than 5
dB louder than ambient sound (4E.06). The 89 dB upper absolute limit was set by
the Federal Highway Administration to comply with OSHA standards. The 5 dB upper
limit relative to ambient sound was based on research that found that an 880 Hz
square wave tone used as an APS locator tone was more readily perceived and
localized by blind persons from distances of up to twelve feet when it was 2-5
dB above ambient vehicular sound along an eight lane artery than when it was 10
dB above that ambient sound Bentzen et al, 2000).
Another limitation in generalizing to APS the results of Killion (1999) and
related research is that the background noise levels tend not to be as high as
in traffic situations. Mean traffic noise for traffic accelerating at the onset
of 20 walk intervals on a four-lane street was measured at 94 dBA from the
position at which a pedestrian who is visually impaired would be standing while
waiting to cross the street (Wiener et al., 1997). The microphone of an APS
which senses ambient sound level may be located up to 12 feet, (or sometimes
more), from the position of pedestrians waiting to cross, and hence measure a
lower sound level. It may then control a signal output which is, for example, 5
dB louder than the ambient sound at the APS, but this sound, measured from the
position of the person waiting to cross the street, may not even be as loud as
the vehicular noise heard from that position. In order to make the speech loud
enough to be intelligible by a pedestrian waiting to cross, it might have to be
objectionably loud when heard by a person standing close to the APS.
This research does not address the above issues, which should be considered in
any installation of APS using speech messages. This research only provides
recommendations regarding the structure and content of walk messages and
pushbutton messages, when speech messages are used on APS.
Method
An Expert Panel met by teleconference to develop a survey to be given to
stakeholders, including pedestrians who are visually impaired, orientation and
mobility specialists, transportation engineers, and accessible pedestrian signal
manufacturers. The Expert Panel was comprised of the three co-principal
investigators (authors, all Certified Orientation and Mobility Specialists), and
one representative familiar with speech equipped APSs from each of the following
groups: transportation engineers, APS manufacturers, Council of Citizens with
Low Vision, and the American Council of the Blind. The National Federation of
the Blind declined the invitation to participate in this research.
The Expert Panel developed sample walk interval and pushbutton messages that
were applicable to different intersection geometries and signalization patterns,
and that varied in message content, length, and structure. Each message approved
by the Expert Panel was required to have consensus that it could be appropriate
in content and structure. Appropriate messages were those that were judged by
most Expert Panel members to be unambiguous, and to clearly convey the minimum
necessary information.
The Expert Panel discussed several issues of message content and wording
extensively and made the following recommendations:
• The words “walk sign” and “crossing” should not be used in the pushbutton
message; “to cross” was considered acceptable in the middle of a phrase.
• Landmark information, such as location of public buildings and transit stops,
should be prohibited from pushbutton and walk signal speech messages.
• Construction information should be prohibited from pushbutton and walk signal
speech messages.
Therefore these wordings and types of information were not included in the
sample messages developed by the Expert Panel. (For explanation of the rationale
for these recommendations, see Results and Discussion.)
The Expert Panel unanimously recommended the following order of information for
pushbutton messages
• Intersection identification
• Intersection signalization (if remarkable)
• Intersection geometry (if remarkable)
In this order, the information necessary to the greatest number of users comes
first.
Messages selected by the Expert Panel were developed into a survey instrument
for obtaining data on preference for message types by blind pedestrians,
orientation and mobility specialists, transportation engineers, and APS
manufacturers. The survey also contained items to evaluate the understanding of
various message types, and preferences for use of the pushbutton delay to
actuate pushbutton messages.
The survey was mailed to160 people who are actively involved in decision-making
about pedestrian signals or in teaching pedestrians with visual impairments to
recognize and interpret the geometric and signal information at intersections;
these people represented the following categories of stake-holders: orientation
and mobility specialists, APS manufacturers and distributors, and transportation
engineers. The same survey was administered in Braille, large print, or orally
to 170 pedestrians with visual impairments who were attending the convention of
the American Council of the Blind during the first week of July 2001.
Of the 170 respondents with visual impairments, 84.2% were usually unable to
read street signs, see traffic lights and see Walk/Don’t Walk signs, and 15.8%
were usually able to read street signs, see traffic lights and see Walk/Don’t
Walk signs. Respondents were asked to indicate the frequency with which they
independently crossed streets with traffic lights; 72.5% reported that they
independently crossed streets with traffic lights at least once a week, while
27.5% reported that they independently crossed streets with traffic lights less
often or never. Ages of respondents were: <40=23.5%; 40—54=50.0%; 55—69=21.8%;
>69=8.8%. Respondents were not systematically asked to indicate what travel aid,
(such as a long cane, dog guide, optical aid, or no aid) they typically use,
however, 89 respondents were observed to be using a long cane and 62 were
observed to be using a dog guide in traveling to the survey area.
The panel reviewed the survey results via a conference call and made
recommendations based on the results and comments.
Results and Discussion
Expert Panel Decisions Prior to Conducting the Survey
Expert Panel members had extensive discussion of safety concerns related to the
wording of the walk and pushbutton messages. It was considered essential that
the walk message be worded as “Walk sign” or “Walk sign is on.” This is
consistent with the MUTCD. The wording indicates the status of the signal; it
does not give a command or imply that it is safe to cross.
There was concern that some messages that have been used on pushbutton messages,
such as “Crossing Charles at State,” could be misinterpreted as a walk message
by a pedestrian unfamiliar with such messages. This discussion led to the
decision that the words, “crossing” and “walk sign” should not be included in
the pushbutton messages. There was also discussion of the words “to cross.” The
panel decided that “to cross” could be used as part of a phrase in the
pushbutton message, but not at the beginning of the message. If used at the
beginning of a message, if the word “to” is not heard or understood, pedestrians
could mistakenly understand that “cross” means they should cross now.
The Expert Panel also decided that the word “wait” should be included in every
pushbutton message, because pushbutton messages always come on during either the
flashing or steady don’t walk intervals. This is consistent with the
requirements of the MUTCD.
The Expert Panel discussed the types of information to be provided by the
messages at length. While there was recognition that APS could provide landmark
and construction information, the panel decided that APS should only be used to
provide crossing and intersection information. That is, the APS should be a
pedestrian signal only. The separation of traffic signal information from other
types of information was considered a safety issue. The panel felt that other
types of information should be provided by other means. Providing landmark or
construction information was considered analogous to adding signs to the
pedestrian signal head. In addition, the realities of maintenance, typical
departmental structure of public works departments, and the necessity of
deciding on a case by case basis what messages were included and how they should
be worded were determined to be problematic. Therefore, the sample messages on
the survey did not include this type of information.
The panel also discussed the fact that different messages might be understood
differently by different stakeholders. In particular, the traffic engineering
representative pointed out that a message such as “Howard Street, Walk Sign”
would typically be understood in the traffic engineering community to indicate
that the walk signal was on to make a crossing parallel to Howard Street, while
that message is currently being used on APS to indicate that the walk signal is
on to cross Howard Street. This led to the inclusion of questions to determine
the understanding of messages by the various groups.
Introduction to Survey Results
Surveys were completed by 170 pedestrians who were legally blind. The response
rate for the 160 surveys mailed to the other stakeholders was 37.5%, comprised
of 26 Orientation and Mobility Specialists, 29 Transportation Engineers, and 5
APS manufacturers or distributors.
In eight items designed to determine respondents’ subjective evaluation to walk
interval and pushbutton messages, respondents were asked to rate sample messages
on a three point scale in terms of whether they were “good” messages, that is,
messages they thought had the necessary information and that would be understood
correctly. (1=poor; 3=good) The number of sample messages for each of these
eight items, representing three situations in which walk messages were used and
five situations in which pushbutton messages were used, ranged from two to
eight. For these questions in which the response was a rating, means were
computed for each sample message by groups, blind (totally blind and low
vision), orientation and mobility (O & M) specialists, and transportation
engineers, for the three large groups. Selected planned comparisons were then
performed for the three items having six or eight sample messages, to determine
whether observed differences in mean ratings between the highest rated sample
message in each group and other messages in that group were significant; that
is, whether there was less than .05 probability that the observed differences
were attributable to chance. (Seven planned comparisons were performed for each
of two items and five were performed for one item.)
In five items, respondents were asked to mark a choice that indicated their
understanding of walk interval or pushbutton messages. There were three choices,
a choice that the authors are calling the correct choice (that is, the way
speech messages have typically been configured on APS), an incorrect choice, and
“I’m not sure.” Frequencies were computed, by major respondent group, for each
of these questions.
Respondents were also asked whether they thought it was best to have the
pushbutton message come on only after the button is pushed for a minimum number
of seconds. Responses were “Yes,” “No,” or “No opinion.” Frequencies were
computed by major group, and Chi Square tests performed.
Respondents were given the opportunity to indicate particular wordings they
thought were good or did not think were good, and to suggest alternative
wordings.
Following statistical analysis, the Expert Panel convened by teleconference to
interpret the results, combining the results for preference with those for
accurate understanding of the message, respondents’ additional suggestions, and
their own expertise.
Walk Signal Messages
In each of the first two questions, respondents were asked to rate each of eight
sample messages intended to convey that the walk sign was on to cross the street
in front of them (the perpendicular street). In the first question, respondents
were to rate sample messages for their use at a crossing where it would be
obvious from the location of the pushbutton, or the nature of the intersection,
such as a mid-block crossing, which crosswalk would have the walk signal when a
walk message was heard. In the second question, respondents were asked to rate
the same eight sample messages for use at a corner where it might not be obvious
which of two crosswalks had the walk signal when a walk signal came on, such as
when there are two pushbuttons mounted on the same pole, serving two different
crosswalks. Mean responses of those participants who rated all eight messages in
each group can be seen in Table 1.
Table 1. Mean ratings of sample walk signal messages for locations where it
would be obvious which crosswalk had the walk interval and for locations where
it would not be obvious which crosswalk had the walk interval The highest mean
rating for each situation, for each respondent group, is indicated by an
asterisk. Means in bold type did not differ significantly from the highest mean
for that situation and group.
|
Sample Walk Messages |
Blind
|
O&M Specialists |
Engineers
|
|
|
Mean |
Mean |
Mean |
|
At a signal where it is obvious which crosswalk has the walk signal. |
N=135 |
N=26 |
N=25 |
|
Walk sign. |
1.70 |
1.65 |
1.44 |
|
Walk sign is on. |
1.99 |
1.73 |
1.68 |
|
Howard Street, Walk sign. |
2.33* |
2.35 |
1.64 |
|
Walk sign is on to cross Howard Street. |
2.24 |
2.12 |
2.52* |
|
Howard Street, Walk sign is on. |
2.29 |
2.39* |
2.08 |
|
Walk sign, Howard Street. |
2.13 |
2.19 |
1.96 |
|
Walk sign on, Howard Street. |
2.14 |
2.15 |
1.92 |
|
Howard Street, Walk sign on. |
2.25 |
2.35 |
1.96 |
|
|
|
|
|
|
At a signal where it is not obvious which crosswalk has the walk signal. |
N=131 |
N=24 |
N=24 |
|
Walk sign. |
1.18 |
1.13 |
1.21 |
|
Walk sign is on. |
1.35 |
1.08 |
1.25 |
|
Howard Street, Walk sign. |
2.25 |
2.42* |
1.71 |
|
Walk sign is on to cross Howard Street. |
2.44 |
2.13 |
2.58* |
|
Howard Street, Walk sign is on. |
2.47* |
2.29 |
1.96 |
|
Walk sign, Howard Street. |
2.12 |
2.00 |
1.83 |
|
Walk sign on, Howard Street. |
2.22 |
2.13 |
1.83 |
|
Howard Street, Walk sign on. |
2.44 |
2.38 |
1.88 |
For both types of intersections, respondents in all groups rated walk messages
that included the name of the street to be crossed higher than messages that did
not include the street name. There was a tendency across blind respondents and
orientation and mobility specialists to rate messages beginning with the street
name higher than messages beginning with “Walk sign,” although most differences
were not statistically significant. Traffic engineers rated the message that was
the most complete grammatical sentence significantly higher than all other
messages.
Regardless of preferences for message types, it is essential that walk signal
messages correctly convey to users which crosswalk has the walk signal. APS are
required by the MUTCD (4E.06) and the recommendations of PROWAAC (X02.5.2.2A) to
provide unambiguous information about which crosswalk has the walk interval.
Respondents were therefore asked two questions to determine the relative
importance to understanding the walk signal message, of complete phrasing versus
more concise phrasing. Frequency of responses to each question can be seen in
Table 2.
Table 2. Understanding of sample walk interval messages. Number of respondents
in each group who provided each response. The meaning the APS message would be
intended to convey to pedestrians (the correct response) is indicated in bold
print.
|
Sample Messages, and Responses |
Blind |
O & M Specialist |
Engineer |
|
|
|
|
|
|
Mess: “Walk sign is on to cross Howard St.” |
|
|
|
|
Resp. A. Howard St. is the street parallel to the crossing. |
5 |
0 |
0 |
|
Resp. B. Howard St. is the street to be crossed. |
155 |
26 |
29 |
|
Resp. C. I'm not sure. |
7 |
0 |
0 |
|
|
|
|
|
|
Mess: “Howard St. Walk sign.” |
|
|
|
|
Resp. A. Howard St. is the street parallel to the crossing. |
9 |
0 |
2 |
|
Resp. B. Howard St. is the street to be crossed. |
135 |
23 |
17 |
|
Resp. C. I'm not sure. |
23 |
3 |
10 |
For the walk message “Walk sign is on to cross Howard St., 92.8% of blind
respondents correctly understood that Howard St. was the street to be crossed.
For the walk message “Howard St. Walk sign,” only 80.8% of blind respondents
correctly understood the message to mean that Howard St. was the street to be
crossed, that is, the street for which the walk sign now says WALK.
Twelve respondents spread across all groups stated that more concise messages
are better, however one blind respondent stated that the best messages were
“more phrase like.” This expressed preference for more concise messages does not
correspond with the preference ratings or the data on understanding messages,
however.
In considering the walk interval messages, the Expert Panel recognized that
despite comments to the contrary, participants not only rated the more
completely phrased messages more highly than the more concise messages, but were
also more likely to understand the more completely phrased messages correctly.
However, Expert Panel members were concerned that 7.2% of blind respondents
interpreted the most accurately interpreted message incorrectly or found it
ambiguous.
Another reason Expert Panel members preferred the more fully phrased walk signal
message was that it was consistently interpreted by traffic engineers in the
same way as it was intended to be interpreted by pedestrians who are visually
impaired. Having this agreement in interpretation will facilitate correct
installation of APS having speech messages.
Additionally, Panel members were strongly of the opinion that the street name
should come first in the message. This is because users are primed to hear “Walk
sign,” and may incorrectly assume that the street they desire to cross is the
one that has the walk sign, and begin their crossing without waiting to be sure
which street is being signaled. Users who are familiar with speech signals at a
particular intersection will not be delayed in their crossing by having the name
of the street first, and users who are unfamiliar with speech signals at a
particular intersection, or who are careless at a particular time, will have the
information first that is critical to life safety. This is consistent with the
high ratings given by blind respondents and orientation and mobility specialists
to sample messages in which the street name comes first.
Therefore Panel members concluded that the recommended wording for walk messages
should follow the model “Howard. Walk sign is on to cross Howard.” They
recommended that the designations “Street,” “Avenue,” etc. should not be used
unless necessary to distinguish between roadway names that are similar in a
particular geographic area.
Pushbutton messages
Respondents were asked to rate six sample pushbutton messages which could be
available from a speaker at the pushbutton during the don’t walk and flashing
don’t walk intervals. The messages were intended to identify the intersection
and to identify the particular street whose walk signal would be actuated by the
pushbutton. PROWAAC recommends requiring (X02.5.1.4 A) that unambiguous
information be provided at pushbuttons indicating which walk signal is requested
by that pushbutton. The MUTCD says that unambiguous information should be
provided (MUTCD 4E.08). PROWAAC (X02.5.1.4 C) also recommends requiring
intersection identification information at each APS. Mean ratings of those
participants who rated each of the six sample messages in this item are
presented in Table 3.
Table 3. Mean ratings of sample pushbutton messages providing intersection
identification. The highest mean rating for each group is indicated by an
asterisk. Means in bold type were not significantly different from the highest
mean for that group.
|
Sample Pushbutton messages Providing Intersection Identification
|
Blind
N=118 |
O&M Specialists N=26 |
Engineers
N=25 |
|
Mean |
Mean |
Mean |
|
|
Pushbutton for Howard St. at Grand Ave. Wait. |
1.98 |
2.08 |
1.72 |
|
Requesting Howard St. at Grand Ave. Wait. |
1.92 |
1.62 |
1.40 |
|
Request for Howard St. at Grand Ave. Wait. |
1.97 |
1.81 |
1.48 |
|
Request to cross Howard St. at Grand Ave. Wait. |
2.44 |
2.23 |
2.20 |
|
Wait to cross Howard St. at Grand Ave. |
2.54* |
2.54* |
2.28 |
|
Requesting to cross Howard St. at Grand Ave. Wait. |
2.14 |
2.00 |
2.36* |
Both blind respondents and orientation and mobility specialists rated the
message “Wait to cross Howard St. at Grand Ave.” the highest, but not
significantly higher than “Request to cross Howard St. at Grand Ave. Wait.” Most
messages that began with “Request” or “Requesting,” or that did not contain the
words “to cross,” were rated lower. Traffic engineers again rated most highly
the message that was the most grammatically complete, but this was not
significantly different from the ratings of the two messages most highly rated
by blind respondents and orientation and mobility specialists.
Respondents were also asked to respond to three items to determine their
understanding of the intersection identification messages. Regardless of
preference for a particular wording, if many people misunderstand the wording or
are unclear about its meaning, it is not a good message. Frequency of response
to each item is shown in Table 4.
For the message “Wait to cross Joy St. at Central Ave.” 82.8% (135) of blind
respondents correctly understood the message to mean they had pushed the button
to cross Joy St. For the message “Requesting Central Ave. at Joy St. Wait.”
70.6% (120) of blind respondents correctly understood the message to mean that
they had pushed the button (put in a request to the signal controller), to cross
Central Ave. For the street identification message “Pushbutton for Joy St. at
Central Ave. Wait.” Only 70.4% (119) of blind respondents correctly understood
the message to mean that they had pushed the button to cross Joy St.
Table 4. Understanding of sample pushbutton messages providing intersection
identification. Number of respondents in each group who provided each response.
The response designated as correct is indicated in bold print.
|
Sample messages, and responses |
Blind |
O & M Specialist |
Engineer |
|
|
|
|
|
|
Mess: “Wait to cross Joy St. at Central Ave.” |
|
|
|
|
Resp. A. You have pushed the button to cross Central Ave. |
14 |
0 |
1 |
|
Resp. B. You have pushed the button to cross Joy St. |
135 |
26 |
27 |
|
Resp. C. You don't know. |
14 |
0 |
1 |
|
|
|
|
|
|
Mess: “Requesting Central Ave. at Joy St. Wait.” |
|
|
|
|
Resp. A. You have pushed the button to cross Central Ave. |
120 |
24 |
15 |
|
Resp. B. You have pushed the button to cross Joy St. |
19 |
0 |
2 |
|
Resp. C. You don't know. |
31 |
2 |
12 |
|
|
|
|
|
|
Mess: “Pushbutton for Joy St.at Central Ave. Wait.” |
|
|
|
|
Resp. A. You have pushed the button to cross Central Ave. |
16 |
0 |
0 |
|
Resp. B. You have pushed the button to cross Joy St. |
119 |
22 |
18 |
|
Resp. C. You don't know. |
34 |
3 |
11 |
“Wait to cross Joy St. at Central Ave.” was the only sample message that was not
considered ambiguous by a large percentage of traffic engineers. The traffic
engineer member of the Expert Panel pointed out, prior to administration of the
survey, that to traffic engineers, phrases such as “Requesting Central Ave.” and
“Pushbutton for Central Ave.” would commonly be interpreted as requesting to
cross parallel to the named street.
Three respondents, (1 O & M specialist, 2 traffic engineers) mentioned that
beginning the statement with the word “pushbutton” might be interpreted as
meaning “Push the button,” resulting in repeated pushing of the button.
In considering the intersection identification portion of the pushbutton
message, the Expert Panel recognized that the sample message “Wait to cross Joy
St. at Central Ave.” had a higher mean rating by blind respondents and O & M
specialists than other samples, and its rating by engineers was not
significantly different than the rating of the most highly rated sample.
Additionally, it was correctly understood by more than 10% more respondents than
the other two samples for which this information was obtained. Finally, in the
opinion of the Expert Panel, it is essential that the first word of the message
heard during either the don’t walk or flashing don’t walk intervals should be
“Wait,” so that if no other part of the message is listened to or understood,
users will know it is not time to initiate a crossing. Expert Panel members
were, however, concerned that more than 15% of blind respondents either
incorrectly understood the best understood message or were uncertain of its
correct interpretation. They concluded that standard pushbutton intersection
identification messages should follow the model “Wait to cross Howard at Grand,”
eliminating the street, road, etc. designation where this could not cause
confusion.
Special Applications
Exclusive pedestrian phases. Respondents were asked to rate three sets of sample
messages to be used at intersections having an exclusive pedestrian phase, that
is, a signal timing in which all vehicular approaches have a red light during
the walk and flashing don’t walk intervals, sometimes called a Barnes dance or
scramble crossing. One set of sample messages was for the walk signal, and two
sets of messages were for pushbutton messages.
For three sample walk signal messages, respondents in all groups rated the
message “Walk sign on for all crossings” highest. (See Table 5.) One blind
respondent suggested the alternative message, “Walk sign, all crosswalks.”
Table 5. Ratings of sample walk messages for intersections having exclusive
pedestrian phases. The highest mean rating for each group is indicated by bold
print.
|
Walk Message for Exclusive Pedestrian Phases
|
Blind N=170 |
O&M Specialists N=26 |
Engineers N=29 |
|
|
Mean |
Mean |
Mean |
|
All crosswalks, Walk sign. |
1.92 |
1.78 |
1.98 |
|
All crossings, Walk sign on. |
2.29 |
2.23 |
2.36 |
|
Walk sign on for all crossings. |
2.67 |
2.74 |
2.67 |
The Expert Panel concluded that walk messages for exclusive pedestrian phases
should follow the model “Walk sign is on for all crossings.” This alternative
was not offered in the survey, but differs from the highest rated alternative
only in the addition of the word “is,” which makes the message a full sentence.
The standard walk message expressed as a complete sentence was better understood
than the more condensed message. Therefore the Expert Panel concluded that the
walk message where there is an exclusive pedestrian phase should also be a full
sentence.
Respondents were asked to rate two sets of two sample pushbutton messages for
intersections having exclusive pedestrian phases. The first set of messages was
to be used at intersections at which right turns on red were prohibited. Mean
ratings can be seen in Table 6. The ratings were based only on the second part
of the message, that is, the part that provided information about the signal.
All participant groups gave highest ratings to the message “Request for Howard
St. and Grand Ave. Wait. Red light for all vehicles.” However, four engineers,
four orientation and mobility specialists, and one blind respondent expressed
the concern that listeners could understand the message to mean that the red
light for all vehicles was now on. Alternatives suggested were: “All vehicles
will have red at the same time,” “Wait for red light for all vehicles,” “Walk
all ways on red,” “All cars stopped,” “All crossings permitted,” “Walk sign all
crosswalks,” “Full stop in all directions,” and “All vehicles have red light at
the same time.”
Table 6. Mean ratings of sample pushbutton messages for an intersection where
all cars have a red light during the pedestrian crossing time (exclusive
pedestrian phase), with right turn on red prohibited. The highest mean rating in
each set, for each group, is indicated by bold print.
|
Sample Pushbutton messages
|
Blind N=170 |
O&M Specialists N=26 |
Engineers N=29 |
|
Mean |
Mean |
Mean |
|
|
Request for Howard St. and Grand Ave. Wait. Red light for all vehicles. |
2.50 |
2.46 |
1.96 |
|
Request for Howard St. and Grand Ave. Wait. All vehicles have red at the same time. |
2.12 |
1.96 |
1.70 |
The Expert Panel concluded that the intersection identification information
should follow the same model as the standard pushbutton message, followed by
“Wait for red light for all vehicles.” This is similar to the highest rated
“Wait. Red light for all vehicles,” but should be unlikely to be misunderstood
as meaning that the red light is now on for all vehicles.
The second set of messages was to be used at intersections with a so-called
exclusive pedestrian phase, but at which right turns on red are nonetheless
permitted. Mean ratings for all groups were highest for the message “Request for
Howard St. at Grand Ave. Wait. Red light for all vehicles; right turn on red
permitted,” with the rating based on the part of the message providing
information about the nature of the signal. (See Table 7.) There was also
concern that some users could interpret this message to mean that the red light
for all vehicles was on now. One respondent (engineer) suggested that there was
no need to say “right turn on red permitted,” since this is usually the case.
Another (blind) said that “right turn on red permitted” should precede “Red
light for all vehicles,” as understanding that they needed to watch for turning
vehicles was more important to the safety of blind pedestrians than knowing that
there was a red light for all vehicles at the same time. One orientation and
mobility specialist expressed that the amount of information in both messages
could be confusing to blind pedestrians who were either children or elderly, and
that both might choose to ignore the information.
Table 7. Mean ratings of sample pushbutton messages for an intersection where
all cars have a red light during the pedestrian crossing time (exclusive
pedestrian phase), with right turn on red permitted. The highest mean rating for
each group is indicated by bold print.
|
Sample Pushbutton messages for Exclusive Pedestrian Phase with Right-on-Red Permitted
|
Blind N=170 |
O&M Specialists N=26 |
Engineers N =29 |
|
Mean |
Mean |
Mean |
|
|
Request for Howard St. and Grand Ave. Wait Red light for all vehicles; Watch for turning vehicles. |
1.95 |
1.81 |
1.54 |
|
Request for Howard St. at Grand Ave. Wait. Red light for all vehicles; right turn on red permitted. |
2.69 |
2.69 |
2.33 |
For intersections having exclusive pedestrian phase timing where right turns on
red are permitted, the Expert Panel concluded that the intersection
identification message should follow the same model as the standard pushbutton
message, followed by the message “Wait for red light for all vehicles,” and a
third component, “Right turn on red permitted,” based on the survey ratings.
Angled crosswalks. Respondents were asked to rate two sample pushbutton messages
to be used at an intersection having an angled crosswalk on the street to be
crossed. The beginning of the message identified the intersection and crosswalk.
The rating was based solely on the second part of the message, that part
describing the angled crosswalk. For all groups, the sample message receiving
the highest rating was the same: “Request for Howard St. at Grand Ave. Wait.”
(See Table 8.) However, there were numerous comments from blind respondents that
they did not care for either message. Six blind pedestrians suggested using a
clock analogy for expressing the angle of the crosswalk, for example “Crosswalk
angles to 1:00 o’clock,” and five suggested simply “Crosswalk angles right.”
Table 8. Mean ratings of sample pushbutton messages for an intersection where
there is an angled crosswalk on the street to be crossed. The highest mean
rating for each group is indicated by bold print.
|
Sample Pushbutton messages |
Blind N=170 |
O&M Specialists N=26 |
Engineers N=29 |
|
Mean |
Mean |
Mean |
|
|
Request for Howard St. at Grand Ave. Wait. Crosswalk angles 20 degrees right. |
2.46 |
2.81 |
2.52 |
|
Request for Howard St. at Grand Ave. Wait. Crosswalk 20 degrees right. |
1.81 |
1.38 |
1.79 |
Expert Panel discussion included a number of questions regarding the measurement
of the angle of the crosswalk. Is the angle measured from the curb line at the
crosswalk, which may be curved, or is it measured in relation to the approaching
sidewalk? How does one get contractors to make accurate measurements? If the
measurements are not accurate, may jurisdictions incur liability? Additionally,
if angles are precisely measured and stated, for example, 3 degrees right or
1:45, may this precision be more likely to confuse pedestrians with visual
impairments than to be of assistance? The Expert Panel concluded that the model
for intersections where there is an angled crosswalk should be “Wait to cross
Howard at Grand. Crosswalk angles right.”
Crosswalks to Medians. Participants were asked to rate one set of three sample
messages for use at crosswalks where there is a median with a pushbutton that
must be pushed because pedestrian timing is too short for most pedestrians to
cross the whole street at one time. Highest mean ratings for all groups, based
on the portion of the message providing information about the intersection
geometry and need to push a second pushbutton, were given to the message
“Request for Howard St. at Grand Ave. Wait. Short walk phase; raised [or
cut-through] median with second pushbutton.” (See Table 9.) This was the
longest, most grammatically complete message. An alternative sample message was
“Second pushbutton in median.”
Table 9. Mean ratings of sample pushbutton messages for a crossing having a
median with a pushbutton that must be pushed because the pedestrian phase is too
short for most pedestrians to cross the whole street at one time. The highest
mean rating for each group is indicated by bold print.
|
Sample Pushbutton messages |
Blind N=170 |
O&M Specialists N=26 |
Engineers N=29 |
|
Mean |
Mean |
Mean |
|
|
Request for Howard St. at Grand Ave. Wait. Raised [or cut-through] median with button push required. |
2.23 |
2.12 |
1.82 |
|
Requesting to cross to Howard St. median only. Wait. |
1.66 |
1.31 |
1.67 |
|
Request for Howard St. at Grand Ave. Wait. Short walk phase; raised [or cut-through] median w/ 2nd pushbutton. |
2.25 |
2.46 |
2.04 |
The Expert Panel concluded on the basis of survey ratings and the previously
determined standard model for the intersection identification information, that
the model pushbutton message for crossings having a short walk interval and a
median with pushbutton should be “Wait to cross Howard at Grand. Short walk
phase. Raised [or cut-through] median with second pushbutton.” The Expert Panel
is aware that “walk phase” is incorrect terminology, however, it is terminology
that is used colloquially, while “walk interval” is a technical term that is
unfamiliar to most pedestrians.
Crosswalks to splitter islands. Respondents were also asked to rate one set of
three sample messages for use at a crosswalk having a pushbutton to actuate a
pedestrian signal for crossing to a right turn splitter island. Although in most
cases, right turn lanes are unsignalized, they can be signalized. This question
was intended to determine what pushbutton message would be most appropriate at
such a crosswalk. Blind respondents and engineers gave the highest mean rating
to the message “Requesting to cross right turn lane to island for Howard St. and
Grand Ave. crosswalks,” the longest and most explanatory message. Ratings by
orientation and mobility specialists were essentially equal for all three
messages. (See Table 10.) Although there were no suggestions for alternate
wording to be used for a crosswalk to a splitter island, numerous respondents
who were blind indicated that this was a difficult concept to convey clearly,
and that they would probably find all the sample messages confusing unless they
were familiar with the intersection.
Table 10. Mean ratings of sample pushbutton messages for a crosswalk having a
pushbutton to cross to a right turn splitter island. The highest mean rating in
each set, for each group, is indicated by bold print.
|
Sample Pushbutton messages At Crosswalk to Right Turn Splitter Island |
Blind N=170 |
O&M Specialists N=26 |
Engineers N=29 |
|
Mean |
Mean |
Mean |
|
|
Requesting crossing to island for Joy St. and Grand Ave. |
1.94 |
2.15 |
1.56 |
|
Requesting to cross right turn lane to island for Howard St. and Grand Ave. crosswalks. |
2.27 |
2.23 |
2.22 |
|
Requesting crossing to island for Joy St. and Grand Ave. crosswalks. |
2.20 |
2.31 |
2.04 |
The Expert Panel adapted the message rated most highly by respondents who were
blind and by traffic engineers, to make it more closely resemble the standard
pushbutton message. The model recommended for crosswalks to right turn splitter
islands is “Wait to cross right turn lane to island for Howard and Grand
crosswalks.”
Crosswalks at “T” intersections. Respondents were asked to rate two sample
messages to be used at crosswalks crossing from the top of a “T” intersection to
a corner on one side of the stem. The highest mean rating for both blind
respondents and orientation and mobility specialists, based only on the part of
the message describing the geometry of the intersection, was for the message
“Request for Howard St. at Grand Ave. Wait. Crossing from top of T to stem.” For
engineers, the highest mean rating was for the message “Request for Howard St.
at Grand Ave. Wait. T intersection of Grand Ave.” (See Table 11.)
Table 11. Mean ratings of sample pushbutton messages for crossing from the top
of a “T” intersection to the stem. The highest mean rating for each group is
indicated by bold print.
|
Sample Pushbutton messages for “T” Intersections
|
Blind N=170 |
O&M Specialists N=26 |
Engineers N=29 |
|
Mean |
Mean |
Mean |
|
|
Request for Howard St. at Grand Ave. Wait. Crossing from top of T to stem. |
2.19 |
2.19 |
1.58 |
|
Request for Howard St. at Grand Ave. Wait. T intersection of Grand Ave. |
2.05 |
1.92 |
1.77 |
Two blind respondents and one traffic engineer said that it was not important to
provide the information that the intersection was T-shaped. One blind respondent
suggested the alternate wording, “Crossing Howard St. to east side of Grand.
Parallel traffic on left.” One engineer suggested the alternate wording,
“Requesting to cross Howard at Grand. Grand ends here and all traffic turns onto
Howard.”
The Expert Panel concluded that the only pushbutton message at T intersections
should be intersection information, as additional information seemed to be more
confusing than helpful to respondents who were visually impaired. Thus, the
model “Wait to cross Howard at Grand” is recommended.
Pushbutton Delay
The pushbutton message from the pushbutton can come on as soon as the button is
pressed, or with special actuation in which the button is depressed for a
minimum number of seconds. Participants were asked to indicate whether they
thought the pushbutton message should come on immediately, whether it should
come on after a long button press, or they had no opinion on the matter. The
number of subjects providing each response is shown in Table 12. While more
respondents in each group indicated that the message should come on after a long
button press than immediately, Chi Square tests revealed no significant
differences within groups or in the total population of respondents.
Table 12. Respondent opinion, by group, regarding whether the pushbutton message
from the pushbutton should come on only after the button is depressed for a
minimum number of seconds, or the pushbutton message should come on immediately,
each time the button is pushed. Number of responses in each group.
|
Message Timing |
Blind |
O&M Specialists |
Engineers |
Manufacturers |
Total |
|
Message should come on immediately |
72 |
10 |
5 |
0 |
87 |
|
Message should come on after long button press |
84 |
13 |
6 |
3 |
106 |
|
No opinion |
11 |
3 |
5 |
2 |
21 |
Many participants provided reasons for their opinions on this issue. Reasons in
favor of having the pushbutton message come on immediately were the following.
• “Because you need all the time to align for the crossing.” blind respondent.
• “I like it to come on immediately and then be repeated, especially if there is
a lot of traffic.” blind respondent.
• “If it doesn’t come on immediately, people may miss valuable information.” O &
M specialist.
• “If a long button push is required, blind pedestrians will need special
training.” O & M specialist.
• “It should come on immediately, but the amount of information given should be
limited to the identification of the intersection and crosswalk; more detailed
information should be available after the long button press.” O & M specialist.
• “If it doesn’t come on immediately, how would the blind pedestrian know to
keep the button pressed?” Engineer.
• “Why make blind pedestrians do something more than sighted pedestrians to get
the same information?”
Reasons in favor of having the pushbutton message come on following a long
button press were as follows.
• “It’s good to have the delay, but the delay should be shorter—maybe two
seconds.” Two blind respondents.
• “All pedestrians have the option of hearing additional details.” Two O & M and
one blind respondent.
• “It’s good to have the delay so that the signal is not irritating on every
cycle.” Blind respondent.
• “Reduces neighborhood impact.” Engineer.
• “The long button press is good, but it requires universal education for blind
pedestrians.” Engineer.
• “The long button press is good, but only if the button is large and easy to
push.” Engineer.
• “The long button press is good, but children may always hold the button down
to get the message.” Engineer.
The Expert Panel recognized that the lack of statistical significance for this
question provided no basis for making a recommendation. However, they agreed
unanimously that consistency in function throughout the United States is
essential so that persons who would like to or need to use pushbutton
information will be able to access it readily. They recommended that pushbutton
messages come on following a long button press, (the amount of time to be
standardized following research), preferring, since there was not statistically
greater support for messages that come on immediately, to recommend a standard
that results in less neighborhood impact.
Summary and Conclusions
An Expert Panel of stakeholders developed a survey instrument for obtaining data
on preference for APS speech message structure and wording, by blind
pedestrians, orientation and mobility specialists, transportation engineers, and
APS manufacturers. The survey also contained items to evaluate the understanding
of various message types by respondent groups, and preferences for use of a
pushbutton delay feature to actuate pushbutton messages. The Expert Panel drew
conclusions about the recommended structure and content of walk messages and
pushbutton messages based on their own expertise and on the basis of the survey
results from 170 pedestrians with visual impairments, 26 orientation and
mobility specialists, 25 transportation engineers, and 5 APS manufacturers.
The following model messages are recommended for particular situations. Messages
for actual installations should be developed on the basis of these models. The
structure and content of actual messages should follow the models as closely as
possible. Word order should not be changed. Where complete sentences are used in
the models, they should be used in actual messages for the same situations.
Where sentence fragments are used in models for other situations, they should be
used in actual messages for those situations. In the model messages, such words
as street, avenue and road are not used. In some locations they may be needed to
avoid ambiguity.
SITUATION MODEL MESSAGE
Walk Interval Messages
Model message for the walk interval, applicable to most intersections. “Howard.
Walk sign is on to cross Howard.”
Model walk message for intersections having an exclusive pedestrian phase. “Walk
sign is on for all crossings.”
Pushbutton Messages
Model message for pushbutton intersection identification information “Wait to
cross Howard at Grand.”
Model pushbutton message for intersections having an exclusive pedestrian phase
with right turns-on-red prohibited. “Wait to cross Howard at Grand. Wait for red
light for all vehicles.”
Model pushbutton message for intersections having an exclusive pedestrian phase
with right turns-on-red permitted. “Wait to cross Howard at Grand. Wait for red
light for all vehicles. Right turn on red permitted.”
Model pushbutton message for angled crosswalks. “Wait to cross Howard at Grand.
Crosswalk angles right.”
Model pushbutton message for crosswalks to medians where a second button push is
required. “Wait to cross Howard at Grand. Short walk phase. Raised [or
cut-through] median with second pushbutton.”
Model pushbutton message for signalized crosswalks to splitter islands. “Wait to
cross right turn lane to island for Howard and Grand crosswalks.”
Model pushbutton message for crosswalks at “T” intersections. “Wait to cross
Howard at Grand.” [Not different from standard intersection identification
message.]
Landmark information such as locations of public buildings and transit stops
should not be included in either walk signal or pushbutton messages.
Construction information should also not be included in either walk signal or
pushbutton messages.
The order of information for pushbutton messages should be:
• Intersection identification;
• Intersection signalization (if remarkable); and
• Intersection geometry (if remarkable.
The Expert Panel concluded that pushbutton messages should come on after a long
button press. Currently the length of this extended button press is about three
seconds; this will be standardized in the future based on research.
While APS having speech messages may seem very user-friendly, especially when
evaluated in an indoor setting, away from loud traffic sounds, and by people who
are native English speakers, they will not solve all ambiguity problems that can
occur with APS. It will not be possible to make speech messages from APS loud
enough to be intelligible in all ambient traffic conditions by most people
unless they are also loud enough to potentially cause hearing loss. In even
moderate traffic conditions people who have age-related or other hearing losses,
people who are not native English speakers, and people with cognitive
disabilities are likely to miss hearing or to misunderstand some words, possibly
resulting in misunderstanding entire messages.
APS speech messages are a fairly good way to provide pushbutton intersection
identification, and signalization and geometry information. This information
does not have to be understood by all users. This is additional, optional,
information; it can be provided at a moderate sound level that may preclude its
intelligibility by some potential users. However, tones or other methods may be
a better way to provide real time information about the signal status, in
situations where the signals can be installed to provide unambiguous
information. The combination of the tactile arrow and careful location of the
pushbutton may be more useful and certain in reducing ambiguity than a speech
message during the walk interval.
If the decision is made to use APS with speech messages at a particular
intersection or crosswalk, the messages should be worded carefully, according to
the models above, or the nearest approximation to the special situations in
which they will be used. The messages then must be recorded very carefully, in a
clear, moderately pitched voice, with excellent diction and moderate pacing. For
persons with unimpaired hearing, a female voice will be understood somewhat
better than a male voice because the frequency spectrum of the male voice is
closer to that of traffic. However, for the large number of people who are
visually impaired who also have age-related or other hearing loss, a female
voice may not be as easy to understand as a male voice.
References
Access Board, 2001. Building a true community: Final report, Public
Rights-of-Way Access Advisory Committee, Washington, D.C. U.S. Architectural and
Transportation Barriers Compliance Board.
Bentzen, B.L., Barlow, J.M. and Franck, L. (2000). Addressing barriers to blind
pedestrians at signalized intersections. ITE Journal. 70-9, 32-35.
Bentzen, B.L., Barlow, J.M, and Gubbé, D., 2000. Locator tones for pedestrian
signals. Transportation Research Record 1705, pp. 40-42.
Carroll, J. & Bentzen, B.L. (1999). American Council of the Blind survey of
intersection accessibility. The Braille Forum 38, 11-15.
Killion, M.C. (1999). Guilt-free quick SIN [speech in noise]: When to give up on
4000 Hz. International Hearing Aid Conference V, University of Iowa.
Manual on uniform traffic control devices: Millennium edition, 2001. Washington,
D.C. U.S. Department of Transportation, Federal Highway Administration.
Myers, L., Bentzen, B.L., and Crandall, W.F., 1996. The Talking Signs®system at
crosswalks: A pilot study. Unpublished manuscript. Mobility and Wayfinding
Consultants, Mill Valley, CA.
Wiener, W. R., Lawson, G., Naghshineh, K., Brown, J., Bischoff, A., & Toth, A.
(1997). The use of traffic sounds to make street crossings by persons who are
visually impaired. Journal of Visual Impairment & Blindness, 91, 435-445.
Hazards of Protruding Objects
Complying with the
Americans with Disabilities Act
Accessibility Guidelines
Thomas Karnes, B.A.*
Jackie Schiedeck, M.Ed., Orientation and Mobility
Theresa Postello, M.Ed., Orientation and Mobility
Summary of Results
In a pilot study involving nine persons who are legally blind and who travel
using standard long cane techniques, bodily contact was made with a pole-mounted
object protruding 12 inches at shoulder height, on 67.6% of the approaches on
which the pole was contacted by either the cane or body. Bodily contact was made
with a wall-mounted object having its lower edge approximately 26 inches above a
sidewalk, and which protruded approximately 24 inches into the travel path, on
42.9% of the approaches on which the object was contacted by either the long
cane or the body. (See Table 1.) Both the pole-mounted object and the
wall-mounted object protruded no farther than permitted by the Americans with
Disabilities Act Accessibility Guidelines (ADAAG) 4.4.
The requirements of ADAAG 4.4 seem inadequate to assure safety of competent
blind travelers of average adult stature who use standard techniques with their
long canes. Persons of short stature or who are not using standard cane
techniques would presumably be at greater risk.
Table 1. Obstacle contacts by persons who are legally blind
|
|
Bodily Contact |
Cane contact |
No Contact |
Total approaches |
|
Pole-mounted obstacle |
25 |
12 |
53 |
90 |
|
Wall-mounted obstacle |
9 |
12 |
67 |
88 |