Locator Tones for Pedestrian Signals

Page 1

I Transportation Research Record 1705

Paper No. 00 -1700

(Committee Draft ISO/CD 1159.2, 5.2.1, November, 1997) requires

that for locator tones, "an intermittent sound shall be used with a

maximum repetition rate of 1.2 Hz" (72 cycles per minute). An

international literature search revealed no research on the repe-

tition rate for the locator tone. However, stakeholders in various

countries strongly advocate for various repetition rates.

This research was undertaken to determine whether repetition rate

of locator tones was related to efficiency in location of push buttons.

The research was conducted in conjunction with the mid-year meet-

ing of the American Council of the Blind (Los Angeles, February

1314, 1999). Participants who were blind were tested individually,

on a sidewalk along a busy street, in order that objective and sub-

jective data on efficiency of push-button location could be obtained

for push buttons having three different locator tone repetition rates.

METHOD

Subjects

All participants were attendees at the mid-year meeting of the Amer-

ican Council of the Blind, and their participation was solicited by

meeting organizers. Participants were required to be totally blind

or have insufficient vision to enable them to visually locate poles.

Forty-two persons volunteered to participate. Thirty-five participants

used long canes and seven used dog guides. Participants represented

a range of ages and varying degrees of hearing (see Table 1).

Materials

An 880-Hz square wave tone, having multiple harmonics at onset, an

attack time of 3 ms, a 15-ms sustain, and 82-ms release, was recorded

at the three different repetition rates (0.5 Hz, 1.0 Hz, and 1.5 Hz) on

two accessible pedestrian signal push buttons supplied by NOVAX

Industries Corporation of New Westminster, British Columbia. An

880-Hz tone has been found to be highly localizable by most pedes-

trians, including those with age-related hearing loss (

). Each of

the push buttons was mounted on a lightweight movable pole. So that

each pole could be repositioned as silently as possible, it was given

a rubber-covered bottom.

The intensity of the tones automatically varied in response to ambi-

ent sound. Initially, the intensity was set at 10 dB above ambient

sound. Several of the first twelve participants volunteered that they

thought the locator tone was louder than necessary and perhaps objec-

tionably loud. Researchers then set the intensity of the locator tone at

5 dB above ambient sound for the next nineteen participants. When

participants continued to volunteer that the locator tone was louder

than necessary, researchers reduced it to 2 dB above ambient sound

for the last eleven participants. Maximum locator tone intensity was

always set at 90 dB.

The two primary problems experienced by visually impaired persons at

pedestrian-actuated intersections are determining whether there is a push-

button and locating the push button. Many countries use accessible pedes-

trian signals much more widely than has been done in the United States,

and a number of these-including Australia, Hong Kong, Sweden, Den-

mark, Germany, Belgium, and Austria-routinely require the use of a

locator tone. Typically emanating from the push-button housing, a push-

button locator tone indicates to pedestrians that they are expected to push

a button to request a pedestrian phase. It enables visually impaired pedes-

trians to locate the push button quickly and efficiently. Research was

undertaken to determine the effect of locator tone repetition rate on effi-

ciency of pedestrians' location of the push-button pole. Repetition rates of

1.0 and 1.5 Hz resulted in equal pole location speed, faster than that for

the 0.5 Hz repetition rate, and were preferred over the 0.5 Hz repetition

rate. Locator tones 2 dB above ambient sound resulted in faster pole loca-

tion than did tones 5 dB and 10 dB above ambient sound. Push-button

locator tones should have a standardized repetition rate between 1.0 Hz

and 1.2 Hz so that it may be ensured that visually impaired pedestrians

can efficiently locate push buttons. Locator tones need be no more than 5

dB louder than ambient traffic sound.

A survey of orientation and mobility specialists (conducted by the

Association for Education and Rehabilitation of the Blind and

Visually Impaired) and a survey of pedestrians who are blind (con-

ducted by the American Council of the Blind) both indicate that the

two primary problems experienced by visually impaired persons at

pedestrian-actuated intersections are determining whether there is

a push button and locating the push button. These surveys confirm

the findings of previous research in San Diego and Japan (

14

BACKGROUND

Many countries use accessible pedestrian signals much more widely

than has been done in the United States, and a number of these-

including Australia, Hong Kong, Sweden, Denmark, Germany, Bel-

gium, and Austria-routinely require the use of a locator tone. A

push-button locator tone typically emanates from the push-button

housing, although it sometimes comes from a unit located higher up

on the pole that has the push button. The tone indicates to pedestri-

ans that they are expected to push a button to request a pedestrian

phase. It also enables pedestrians who are visually impaired to locate

the push button quickly and efficiently. The locator tone is most

commonly repeated at a rate of approximately 0.5 to 1.25 Hz (30 to

75 cycles) per minute. The International Standards Organization

(ISO) draft standard on acoustic and tactile signals for traffic lights

B. L. Bentzen, Accessible Design for the Blind, P.O. Box 1212, Berlin, MA

01503. J. M. Barlow, Center for the Visually Impaired, 763 Peachtree Street,

N.E., Atlanta, GA 30308. D. Gubbé, NOVAX Industries Corporation, 658 Der-

went Way, British Columbia V3M508, Canada.

Locator Tones for Pedestrian Signals

Billie Louise Bentzen, Janet M. Barlow, and Douglas Gubbé

Page 2

Bentzen et al.

Paper No. 00 -1700

The experiment was conducted on the sidewalk along the east

and west sides of Century Boulevard, an eight-lane artery in front

of the Airport Westin Hotel in Los Angeles, near the T-shaped

semi-actuated intersection with Concourse Street. A 3.7-m (12-ft)

length of each sidewalk was marked with positions for pole placement

and for participant starting position.

Procedure

Participants were tested individually between 7:30 a.m. and 9:00 p.m.

Twenty-one participants were tested on each side of the street.

Each participant located the push button three times for each of

three locator tone repetition rates, from starting distances and direc-

tions counterbalanced across trials and participants and varying from

1.8 to 3.6 m (6 to 12 ft) from the pole. Participants were randomly

headed approximately 45 degrees to the right or left of the pole. The

pole was positioned from 0.3 to 1.8 m (1 to 6 ft) from the outer edge

of the sidewalk, which was at the curb line.

Participants were instructed that as soon as they heard a locator

tone, they should face the direction of the tone and then move as

quickly as possible to touch the pole on which the push button

(source of the tone) was located. As soon as they located the pole,

one experimenter guided them to the next starting position while

another experimenter moved the pole.

Experimenters used a digital stopwatch to measure the time from

the onset of the first tone in each trial until the participant touched

the pole.

Following this procedure, each signal repetition rate was demon-

strated and participants were asked to rate the use of each of the three

signals as a locator tone, according to a five-point scale in which 1 was

equal to "not good at all" and 5 was equal to "very good."

RESULTS

Repetition Rate

To assess the effect of locator tone repetition rate on efficiency of pole

location as indicated by rate of travel toward the pole, the researchers

performed a two-way mixed analysis of variance (ANOVA) [east

versus west side of street-between; repetition rate (0.5, 1.0, 1.5)-

within]. (The east-versus-west-street-side factor was a control mea-

sure and is not considered in the results presented). Analysis revealed

a significant effect of repetition rate for rate of travel,

(2, 80)

20.5,

.001. Mean rate of travel for the 0.5, 1.0, and 1.5 repetition rates

were 0.43, 0.49, and 0.49 m/s (1.4, 1.6, and 1.6 ft/s), respectively.

Planned contrasts revealed that the 0.5 repetition rate resulted in sig-

nificantly slower rates of travel than did the 1.0 or 1.5 repetition rate

(

.001), the latter two not differing.

Loudness

To assess the effect of loudness (dB) of the locator tones relative

to ambient sound on the rate of travel toward the pole, a two-way

mixed ANOVA (2, 5, 10 dB-between; 0.5, 1.0, 1.5 repetition

rate-within) was used. Analysis revealed a significant effect of loud-

ness

(2, 39)

3.0,

.06, as well as repetition rate,

(2, 78)

17.2,

.001. For the tones with loudness of 2, 5, and 10 dB above ambi-

ent sound, the average rates of travel were 0.52, 0.45, and 0.44 m/s

(1.7, 1.48, and 1.43 ft/s), respectively. Planned contrasts revealed

that the 2 dB signal resulted in faster rates of travel than did either

the 5 or 10 dB tones (

.01), the latter two not differing from one

another. For repetition rates of 0.5, 1.0, and 1.5, the average speeds

were 0.43, 0.49, and 0.49 m/s (1.4, 1.6, and 1.6 ft/s), respectively,

as reported above.

Subjective Ratings

The repetition rate was judged from 1 to 5 for preference as a loca-

tor tone, according to a 5-point scale in which 1 was equal to "not

good at all" and 5 was equal to "very good."

A one-way within-subjects ANOVA (repetition rate-0.5, 1.0,

1.5) revealed significant effects for repetition rate,

(1, 38)

89.4,

0.000. The mean ratings for the 0.5, 1.0, and 1.5 repetition rates

were 1.9, 3.4, and 4.1, respectively. Planned contrasts revealed that

all pair-wise comparisons were significant (

0.00).

CONCLUSIONS

1. The 1.0 and 1.5 repetition rates resulted in equal and faster

pole location than did the 0.5 repetition rate.

2. The 1.5 repetition rate was preferred over the 1.0 rate, and the

1.0 rate was preferred over the 0.5 rate.

3. Tones 2 dB above ambient traffic sound resulted in faster pole

location than did tones 5 and 10 dB above ambient sound.

DISCUSSION OF RESULTS

The fastest locator tone repetition rate permitted by ISO draft stan-

dards on acoustic and tactile signals for traffic lights is 1.2 Hz. There

is no proposed minimum repetition rate. This limitation was set in

light of many years' positive experience with push-button locator

tones in other countries, in the knowledge of strong preferences for

different repetition rates at or below 1.2 Hz in different countries,

and to permit the use of a pedestrian Walk tone whose repetition rate

is reliably perceived as being different than the locator tone, but is

readily achieved by existing accessible pedestrian signal technolo-

gies. The draft ISO standard requires a Walk tone with a repetition

rate of equal to or greater than 2 Hz.

The 1.2 Hz repetition rate limit falls between the two faster repe-

tition rates tested in this research, 1.0 Hz and 1.5 Hz. While only

three repetition rates were tested, and they did not include the 1.2 Hz

ISO (draft) limit, the greatest difference in efficiency of pole loca-

tion observed in this research occurred between 0.5 Hz and 1.0 Hz;

the trajectory flattened somewhat above 1.0 Hz, and there also was

no significant difference in preference between the 1.0 and 1.5 rep-

etition rates. A 1.2 Hz repetition rate seems to be an appropriate

upper limit for locator tones, given the factors-in addition to objec-

tive efficiency of pole location-that informed the ISO draft standard.

TABLE 1 Participant Age and Hearing Status

Page 3

Paper No. 00-1700

Transportation Research Record 1705

However, a lower limit of 1.0 Hz is also necessary to assure good

localizability of push buttons.

That pole location is faster at 2 dB above ambient sound than it is

at 5 or 10 dB above ambient sound is puzzling. One possible expla-

nation is that low sound frequencies heard when the locator tone is

louder may mask the frequencies in the tone that are better for local-

izability. The ear is less sensitive to lower frequencies at low sound

levels, and low frequency components may not have masked the more

localizable frequencies at lower sound levels. The ISO draft standard

says that the signal/noise level shall be

5 dB to the ambient noise

level. This level may be higher than necessary for locator tones.

RECOMMENDATIONS

To ensure that pedestrians who are visually impaired can efficiently

locate push buttons, push-button locator tones should have a stan-

dardized repetition rate of between 1.0 Hz and 1.2 Hz. A signal level

between 2 and 5 dB greater than ambient noise level appears to be

adequate for efficient localization of poles equipped with locator

tones.

Additional research should be conducted on locator tones so that

tonal qualities such as frequency and harmonics, which result in opti-

mal detection and efficient push-button location, may be identified.

Additional research is also needed in order that researchers may clar-

ify the relationship between ambient sound compensation factors and

location of push buttons having locator tones.

ACKNOWLEDGMENTS

This research was supported in part by The Seeing Eye, Inc., the

American Council of the Blind, and the Blinded Veterans Associa-

tion. The authors thank the American Council of the Blind for mak-

ing subjects available, along with Lukas Franck and Linda Myers

for their assistance in data collection, and Randolph Easton, Daniel

Ashmead, and Chantal Laroche, for their assistance in research

design and interpretation of results.

REFERENCES

1. Franck, L., and J. Barlow. What We've Got Here Is A Situation! Envi-

ronmental Access Committee Report.

AER Division Nine Newsletter

1999, pp. 912.

2. Carroll, J., and B. L. Bentzen. American Council of the Blind Survey of

Intersection Accessibility.

The Braille Forum

(in press).

Evaluation of Audible Pedestrian Traffic Signals

. San Diego Association

of Governments, San Diego, Calif., 1988.

4. Murakami, T., M. Ishikawa, M. Ohkura, H. Sawai, J. Takato, and

M. Tauchi. Identification of Difficulties of Independent Blind Travelers

to Cross Intersections With/Without Audible Traffic Signals.

Proc., 9th

International Mobility Conference

, Atlanta, Georgia, 1998.

5. Hulscher, F. Traffic Signal Facilities for Blind Pedestrians.

Australian

Road Research Board Proceedings

, Vol. 8, 1976, pp. 1326.

Publication of this paper sponsored by Committee on Pedestrians.