[b] "Low vision" for this table is defined as corrected visual acuity of no better than 20/70 in the better eye
or a maximum diameter of visual field of no more than 30 degrees. The low-vision category does not
include blind individuals.
[c] "Blind" for this table and this report refers to persons with no useful pattern vision, who are referred to
as "functionally blind" in the original reference.
Feature size, viewing distance, and the user's acuity put boundary conditions
on the identification of banknotes. The large corner digits on the portrait
side of $1 and $10 U.S. banknotes would be at the visual limit for someone with 20/400 letter acuity,
if they were viewed at a normal reading distance of 16 inches (40 cm). Suppose
the viewing distance is increased to 40 inches (1 m), roughly the distance from
the eye to a bill on a checkout counter. The corner digits would now be at or
beyond the acuity limit of people with about 20/160 vision. Identification at
the acuity limit is slow and requires optimal viewing conditions, including
good lighting and 100 percent black/white numeral contrast, which is not true
even of crisp, new currency bills. Conservatively, symbols should be at least a
factor of two larger than the acuity limit for practical use. Only people with
acuity better than about 20/80 would meet this more stringent "factor of two"
criterion for viewing the currency digits on a bill on the counter. Following
this informal argument, most people with low vision (acuity less than 20/60)
experience difficulty recognizing the large digits on present banknotes in a
common cash transaction.
Visual-field
size refers to the range of visual directions, centered on the line of sight,
over which a standardized test target can be detected. Field loss results from
damage to portions of the retina, resulting in blind regions in the visual
field. The various scenarios shown in
Figure 2-2
illustrate the effects of different types of visual-field loss and are
discussed in detail below.
Age-related macular degeneration (AMD) afflicts many older people and is the
leading cause of low vision. Because vision is lost in the high-resolution
central portion of the visual field, afflicted individuals must rely on sight
in their low-resolution peripheral vision.
Figure 2-2 shows the same photo
of New York City's Metropolitan Opera House a number of times--once in sharp
focus
(Figure 2-2a) and then as it would be seen with various types of
visual impairment
(Figures 2-2b through 2-2e).
Figure 2-2b shows the picture
with the central field completely blocked, illustrating a severe case of
age-related macular degeneration. Typically, reading is impaired, and
individuals afflicted with this degeneration require large letters (high
magnification) to read.
Some diseases, such as advanced glaucoma or retinitis pigmentosa, can result
in the loss of almost all peripheral vision. Only a small island of central
vision remains (tunnel vision).
Figure 2-2c illustrates vision with this
type of impairment. Individuals with this form of low vision may have fairly
high acuity, but they experience great difficulty in mobility (walking or
driving), and they have problems in visual search tasks (e.g., finding signs,
page numbers, or currency digits). Other diseases, such as diabetic
retinopathy, can result in patchy vision with losses in a number of areas
within the visual field, as illustrated in
Figure 2-2d. Since it is
impossible to accurately portray what vision loss is like for a person with low
vision, the illustrations in these figures are meant only to demonstrate the
types of vision loss that might occur with various diseases and
impairments. However, it should be remembered that the scotomatous (blind)
areas in these simulations would move with the eye, so, for example, a person
with age-related macular degeneration will always have a centrally
located blind spot when viewing different areas of the picture.
In addition to low acuity and reduced field, a third kind of visual deficit,
intensively studied in recent years, is reduced contrast sensitivity (see
illustration in
Figure 2-2e).
Figure 2-3 demonstrates the effects of
this type of visual impairment; imagine reading text printed with contrasts of
60 percent ("he made plans"), 30 percent ("to go camping"), 13 percent ("and
hiking in"), and 6.5 percent ("the mountains"). Many types of eye disease, such
as cataract, can cause a reduction in retinal-image contrast or a loss of
sensitivity to contrast by retinal neurons. The consequence is an effective
reduction in the contrast (difference between light and dark areas) of images.
People with normal vision have a substantial tolerance to contrast reduction
for many visual tasks. In the case of reading, print contrast can be quite low
(down to 10 percent) before there is much effect on reading speed (Legge et
al., 1987). For many people with low vision, however, and for those with normal
aging vision, there is much less tolerance to poor contrast. For such people,
even rather small reductions in print contrast can adversely affect reading
(Rubin and Legge, 1989).
Two aspects of contrast are relevant to currency design. Contrast polarity
specifies whether the letters are light and the background dark, or
vice versa. The contrast polarity of alphanumeric symbols usually has little effect on recognition for people with
normal vision. For some people with low vision (especially those with ocular light scatter due to cataracts), acuity and reading performance are better for the light-on-dark
polarity, which is a property of present U.S. banknotes.
In addition to the 3.7 million visually disabled Americans, there are many
people with mild visual impairment due to peripheral field deficits, glare
sensitivity, or losses in contrast sensitivity. Estimates of the number of
Americans with mild forms of visual impairment, referred to here as visually
impaired, extend to nearly 9 million (Benson and Marano, 1994).
Even in the absence of disease, aging eyes are particularly disadvantaged in
recognition tasks in poor illumination. For example, the reduced pupil size of
an average 80-year-old
compared with that of a 20-year-old
will allow only one-quarter
as much light to enter the eye (Pitts, 1982). This diminished light is then
diffused and scattered by opacities in the optical media (which increase with
age) before it reaches the retina, which is likely to be reduced in its ability
to detect low contrast features. The net result is a severe reduction in visual
performance under adverse lighting conditions for older people.
The committee notes also that special difficulties are encountered by
individuals with multiple disabilities. New currency features relying on sound
would not be accessible to people who are both hearing and visually disabled,
for example, the 10,000 individuals with Usher syndrome in the United States
(RPFFB, 1994). Also, individuals with decreased tactile sensitivity
accompanying visual impairment (e.g., from diabetes) may not benefit from
subtle tactile features (Cholewiak, 1994).
It is important to note that normally sighted Americans would clearly benefit
in ease of use, especially in conditions of adverse visibility, from most
features intended for use by visually disabled people.
Some who are working for the full integration of blind people in American
society have expressed opposition to certain special features (such as obvious
"braille" markings), for fear that these markings would be a ubiquitous signal
that special accommodations are required for blind people (Maurer, 1994). The
committee evaluated prospective features with respect to their ability to
enhance the recognition of the currency by all users.
The currency identification needs of our target population can be conveniently
divided into commercial and daily-living
categories. A commercial need for currency identification arises for those who
are engaged in any kind of retail or other trade in which ability to identify
currency is a necessary part of the job. Lack of suitable banknote
identification features raises obstacles to employment in such jobs. Daily
activities include using currency in many situations, such as purchasing
groceries and using public transportation.
Significant experience and insight comes from visually disabled workers who
operate cafeterias or vending stands, usually in government-owned
buildings, under the Randolph Sheppard Act Business Enterprises Program. In
1992, there were almost 3,500 people employed under this program, doing more
than $395 million in business (Abbott, 1994; RSVFP, 1993). Such workers must
accept bills rapidly and give out accurate change as a routine part of the job.
A blind cashier cannot independently identify bills in a transaction and must
rely on the customer. [5] This arrangement may
result in fraud or costly mistakes.
Visually disabled people face problems in identifying U.S. banknotes in
broader daily living situations. For blind individuals, the overwhelming need
is for denomination without dependence on a sighted person. For people with low
vision, needs vary according to the nature of the visual deficit and the
ambient lighting conditions as discussed earlier in this chapter. An older
person with macular degeneration may have relatively little difficulty in
identifying bills at a well lit checkout counter but may be unable to identify
bills in a dimly lit restaurant or in a taxi at night.
Although denomination of banknotes is the committee's primary focus, it is
frequently useful to know the orientation of a bill. Specific orientation of
bills is required by some money-changing
machines and for stacking purposes by banks. A currency feature that conveys
bill orientation would be useful for such purposes.
In both commercial and daily-living situations, successful currency
transactions require rapid and accurate denomination of banknotes without the
need for lengthy inspection or thought. To achieve this goal, specification of
new or enhanced features should not be aimed at minimal levels of recognition
performance (i.e., threshold levels) but should strive for sufficient
differentiation to permit rapid, effortless performance. For example, research
on reading has shown that letter sizes should be three to five times larger
than acuity values in order to achieve comfortable and rapid reading (Legge,
1991). Similar consideration should be applied to currency identification
features.
This chapter has focused on the difficulties encountered by visually disabled
people in currency identification. The design of new visual, tactile, or other
features for banknote identification should take into consideration perceptual
limitations on discrimination and absolute judgment that limit performance by
everyone. The challenge is to design features that can code six denominations
of bills ($1, $5, $10, $20, $50, and $100) in a way that circumvents these
perceptual limitations.
Discrimination refers to the perceptual ability to compare two physical
magnitudes (e.g., weights, light intensities, tonal frequencies, lengths, etc.)
and tell them apart. The smallest difference that can be reliably discriminated
is called the difference threshold. For perceptual discrimination on many
physical dimensions, the difference threshold is constant (or nearly constant)
in percentage terms across the range of that dimension. This constancy is
termed Weber's Law, and is expressed as:
where D is the difference threshold, D' is the physical
magnitude, and k is a constant (the Weber fraction). Several examples
are shown in Table 2-2. As an example, according to this table, two
weights can be discriminated if they differ by more than 4.3 percent. The
discrimination of length by the sense of touch alone is a very complicated
experimental problem because of the variety of sensory inputs a human being
uses when trying to identify an object by touch. It is not clear that this
simple form of Weber's law is applicable to the task of distinguishing length
by the sense of touch.
Some possible features for coding currency denomination, such as size or
roughness, can be analyzed in terms of their difference thresholds. In
designing a code:
Inherent in discrimination is a comparison between two stimuli. In the case of
banknote identification, the comparison may be between two bills (based on
size, weight, etc.) or between a bill and a measuring gauge.
A second kind of perceptual judgment, called absolute judgment, may be more
important to currency identification. Absolute judgment refers to the ability
to identify any one of N distinct values along a physical dimension. For
example, imagine having N different weights, or N light intensities, or N
loudness levels of a given tone, etc. A human subject has to learn to attach
labels to these stimuli so they can be identified when presented on their own.
Miller (1956) summarized many findings on absolute judgment. He showed that when N is
4 or less,
people tend to be accurate in absolute judgment. When N rises to about 7 (plus
or minus 2), people begin to make errors. There is some variation across
stimulus dimensions. The key point is that if banknotes are coded as six points
along a single physical dimension (such as length, roughness, thickness),
people are likely to make some errors in absolute judgment when a single cue is
present. That is, if presented with a single bill (with no measuring gauge or
other bill for comparison), some errors are likely to be made. Recognizing that
humans are limited to sets of less than seven in their ability to make absolute
judgments is a case for limiting the number of denominations of banknotes in
general circulation to no more than the present six, and even fewer would be
preferable.[6] In the same article, Miller
reviewed evidence that accuracy of absolute judgment can be increased if two or
more independent dimensions are used to code information. In the context of
currency identification, this means that accuracy in identification of six
denominations is likely to be improved if two dimensions are used for coding
(e.g., length and texture, or length and edge profile as in coinage). This
likelihood was confirmed in a demonstration experiment conducted by the
committee, which is briefly described in Chapter;4 in connection with the
size feature.
To summarize the research into discrimination and absolute judgment: (1) Data
on perceptual discrimination can provide lower bounds on the step sizes for
coding denomination along a physical dimension. (2) Coding six items
(denominations) on a single physical dimension (e.g., length) is likely to
result in some errors of absolute judgment. (3) Accuracy can be improved by
using two independent dimensions for coding the six items.
[a] For these stimuli, the Weber fraction is not constant over the range of
stimulus magnitude. The Weber fraction in these cases decreases as the stimulus
magnitude increases, indicating that more subtle changes in visual contrast or
tone loudness are distinguishable at higher absolute stimulus magnitude.
Many visually disabled people must trust others to inform them about the
denomination of bills received. Once identified, these bills must be sorted for
accurate retrieval later. Due to the absence of tactual or other identifying
features in the present bills, many blind individuals employ a system of
folding to assist in this process. Different denominations, once identified by
a trusted sighted person, are folded in different ways (diagonally, crosswise,
etc.) according to denomination. Sometimes, people with low vision forego
visual identification and revert to techniques used by people who are blind.
Under good lighting conditions, persons with low vision identify bills by
holding them close to the eye. An optical magnifier may be necessary for focus.
The ease of identification depends on a host of factors, including the level
and direction of lighting and the nature of visual impairment.
Machines are available that can denominate and to some extent authenticate
banknotes for blind individuals. The smallest of these, costing approximately
$400, is 6x3x1 inches in size, can be carried in a pocket or purse, and
provides a talking output denominating the bill that is inserted. However, the
devices are not always reliable, especially on crumpled and dirty bills, and
are relatively slow. Audio output in the form of synthetic speech is not
usually private, although an earphone can be used if desired. In commercial
situations, blind vendors sometimes use nonportable automatic currency
identifiers retailing for approximately $900. This device is intended to
provide more-reliable authentication abilities than smaller, less expensive
devices and provides a series of auditory "beeps" as output, enhancing privacy
compared with synthetic speech. It is possible to integrate some of these
devices with the cash register. However, reports provided to the committee by a
representative of the approximately 3,500 Randolph Sheppard vendors indicated
that these machines significantly slow down transactions and are usually
bypassed in favor of asking the customer what denomination is handed to the
vendor (Abbott, 1994). The devices, as currently available, do not yet fully
meet the needs of users for fast and confidential banknote identification. In
addition, their bulk and high cost are barriers to widespread use. Development
of these devices by groups not associated with the banknote designers (e.g.,
The American Foundation for the Blind) is limited by their having to determine
denomination-distinctive features and patterns around which to design sensors.
A major cost of developing these devices is in determining these features and
patterns (Schreier, 1994). A discussion of the criteria for an ideal
denominating device is contained in Chapter 3.
Existing approaches in other countries (see Appendix D) that can be used by
visually disabled people to denominate banknotes include the use of bills of
varying sizes and colors. The use of varying sizes reportedly allows a blind
individual to denominate bills accurately using a simple, inexpensive plastic
guide. With practice, discrimination between various bills can be made without
the guide. The use of different sizes also provides a redundant cue for sighted
individuals when sorting money bills, allowing, for example, a $1 note to be
selected from a group of bills carried inside a pocket without withdrawing the
entire group. The ease of using different-sized coins indicates that this
method of differentiating value becomes second nature once a person becomes
familiar with the size/value correlation.
Most other countries also use different predominant colors for each
denomination. The simple detection of color is faster than finding and reading
printed numbers, especially for those with poor letter acuity. It should be
noted, however, that many people with low vision have difficulty in
discriminating subtle shades of color and that color cues are generally less
obvious at low levels of illumination. Consequently, any added color features
should use clearly distinguishable colors.
Some foreign currencies, including the new series of Canadian notes introduced
in 1986, also contain numerals of much larger size and higher contrast than
those found on U.S. banknotes to facilitate identification for people with
normal and low vision.
There are also several countries using tactile denomination symbols, such as
printed bars or shaped grids of small dots, on their banknotes, and visually
disabled people in these countries can use these marks to recognize and
denominate the banknotes.
All the user groups discussed require a means of currency denomination that
can be summarized briefly as follows:
Based on these general user considerations and other considerations of the
technical implementation of such features, the committee developed an
evaluation system for the different possible banknote features under
consideration (see Chapter 3).
It is clear that a major need exists for a better means of banknote
denomination for the 3.7 million Americans with visual disabilities, with the
goal of giving this population the full access to currency handling available
to the rest of society and to visually disabled people in other countries. In
addition, due to the increasing number of older individuals with impaired
vision due to minor eye disease or the normal aging process, such features
would be of great benefit to a far wider population than that represented by
the current statistics on blindness and low vision. Certain new features, such
as color and size, and enhanced existing features, such as larger numerals of
higher contrast, would also benefit those with normal vision by making
denomination more rapid and convenient for all.
2. The level of best-corrected visual acuity at which a person is said to
have "low vision" has several definitions. Measured levels of 20/60 or 20/70
are commonly used and correspond roughly to the more qualitative definition of
inability to read regular newsprint. The definition used by each source of the
data cited in the chapter is given with that data.
3. Because the braille print version of this report will not contain the
graphics, each figure is described in detail in the text.
4. Contrast is defined as 100*(Lmax - Lmin )/(Lmax + Lmin ) where Lmax is the
luminance of the digit, and Lmin is the luminance of the background of the
digit.
5. There are anecdotal stories that some blind people are able to
denominate U.S. currency without any artificial aid. Although some people who
are "legally blind" have sufficient vision to identify bills, the committee
was unable to substantiate through literature search or interaction with
representatives of associations of blind people any such claims for people who
are blind.
6. The replacement of the current $1 banknote with a $1 coin might be used
to reduce the number of banknote denominations, and would be a benefit to
visually disabled people for that reason. However, there are a great many
other issues surrounding the change to a $1 coin, such as the economic impact,
public acceptance, and environmental impact, that the committee was not
qualified to address, so the idea is noted here, but is not fully assessed for
recommendation.
Abbott, G. 1994. Presentation to the Committee on Currency Features
Usable by the Visually Impaired. March 30, 1994.
Benson, V., and M.A. Marano. 1994. Current estimates from the National Health
InterviewSurvey. National Center for Health Statistics, January 1994. Vital and Health
Statistics Series 10(189):95.
Cholewiak, R. 1994. Presentation to the Committee on Currency Features
Usable by theVisually Impaired. March 30, 1994.
Genensky, S. 1978. Data concerning the partially sighted and functionally
blind. Journal ofVisual Impairment and Blindness 72(5):177-180.
Genensky, S. 1994. Personal communication to the Committee on Currency
Features Usableby the Visually Impaired. March 30, 1994.
Jewish Guild for the Blind, The. 1992. Photos taken from visual impairment
poster. NewYork: The Jewish Guild for the Blind.
Laming, D. R. J. 1986. Sensory Analysis. San Diego, California:
Academic Press.
Legge, G.E. 1991. Glenn A. Fry Award Lecture 1990: Three perspectives
on low-visionreading. Optometry and Vision Science 68:763-769.
Legge, G.E. 1993. The role of contrast in reading: Normal and low vision. Pp.
269-287 inContrast Sensitivity, R.M. Shipley and D.M.-K. Lam, eds. Cambridge,
Massachusetts: Massachusetts Institute of Technology Press.
Legge, G.E. 1994. Presentation to the Committee on Currency Features
Usable by theVisually Impaired. February 23, 1994.
Legge, G.E., G.S. Rubin, and A. Luebker. 1987. Psychophysics of reading. V.
The roleof contrast in normal vision. Vision Research 27:1165-1171.
Maurer, M. 1994. Presentation to the Committee on Currency Features
Usable by theVisually Impaired. March 29, 1994.
Miller G. 1956. The magical number seven plus or minus two: Some limits on
our capacityfor processing information. Psychological Review 63(2):81-97.
National Advisory Eye Council. 1993. Vision Research--A National Plan:
1994-1998. NationalInstitutes of Health Publication No. 93-3186.
NRC (National Research Council). 1994. Measurement of Visual Field and Visual
Acuity forDisability Determination. National Research Council Committee on Vision, NRC.
Washington, D.C.: National Academy Press.
Nelson, K. A., and E. Dimitrova. 1993. Severe visual impairment in the
United States andin each state, 1990. Journal Visual Impairment and Blindness 87(3):80-85.
Pitts, D. G. 1982. The effects of aging on selected visual functions: Dark
adaptation, visualacuity, stereopsis, and brightness contrast. Pp. 131-159 in Aging and Human
Visual Function, R.;Sekuler, D. Kline, and K. Dismukes, eds. New York:
Alan R. Liss, Inc.
RPFFB (RP Foundation Fighting Blindness). 1994. Information about Usher
Syndrome.Baltimore, Maryland: National Retinitis Pigmentosa Foundation, Inc., RP
Foundation Fighting Blindness.
RSVFP (Randolph-Sheppard Vending Facility Program). 1993. Annual Report,
Fiscal Year1992. Washington, D.C.: Office of Special Education and Rehabilitative
Services, U.S. Department of Education.
Rubin, G.S., and G.E. Legge. 1989. Psychophysics of reading. VI. The role of
contrast inlow vision. Vision Research 29:79-91.
Schreier, E. 1994. Presentation to the Committee on Currency Features
Usable by theVisually Impaired. March 30, 1994.
Tielsch, J.M., A. Sommer, K. Witt, J. Katz, and R.M. Royall. 1990. Blindness
and visualimpairment in an American urban population: The Baltimore eye survey. Archives
of Ophthalmology 108:286-290.
World Health Organization. 1966. Blindness Information Collected from
Various Sources. Epidemiology and Vital Statistics Report 19:427-511.
NAS Home Page |
NAP Home Page |
Reading Room |
Report Home Page
Acuity is measured by finding the smallest letters a person can read at a
standard distance (traditionally, 20 feet) and expressing the result as the
ratio of this distance to the distance at which a "normal" observer can read
the same letters. For example, a person with 20/60 can just read letters at 20
feet that a person with 20/20 acuity can read at 60 feet. People with low
vision have acuities ranging from less than 20/60 to 20/2000. The denominator
of the fraction in visual acuity notation can also be thought of as the letter
size. This size is defined as the distance at which the observer must stand for
the letter to subtend an angle of 5 minutes of arc. For example, a letter size
indicating 20/20 visual acuity is equivalent to 5 minutes of arc; 20/200 to 50
minutes; etc.
Figure 2-1 Percentage by age group of 1990 U.S. population with visual
disabilities.
It should be noted that clinical definitions of acuity are based on controlled
tests conducted under ideal viewing conditions--bold black letters on a pure
white background (near 100 percent contrast), good illumination, no distracting
symbols, and no time pressure. By comparison, the letters and numbers on U.S.
money bills are of much lower contrast (discussed later in this chapter), are
closely surrounded by distracting abstract designs, and are often viewed under
low illumination with little time for prolonged scrutiny. As a result, direct
extrapolation from acuity testing to feature recognition on bills is likely to
underestimate the number of people who will experience difficulty.
Figure 2-2 Illustrations of various types of visual
impairment using photos of New York's Metropolitan
Opera House: (a) normal vision; (b) central field loss;
(c) peripheral field loss; (d) patchy vision loss;
(e) reduced contrast. Source: The Jewish Guild for the
Blind, 1992.
More important than polarity is the overall contrast level. Photometric
measurements by a member of the committee of the contrast of the large corner digits on the portrait side of $1 notes indicated contrasts of 60 percent (new crisp bill) and 30 percent (worn bill).[4] The reduction from
100 percent contrast undoubtedly has an adverse effect on recognition for many people with low
vision, even under optimal viewing conditions. Under conditions of poor
visibility (especially low lighting and long viewing distance), the suboptimal
contrast will also take its toll on recognition by normally sighted people.
Figure 2-3 Lines of text in decreasing contrast demonstrating visual effects of
loss of contrast sensitivity. Reprinted from Legge, 1993, courtesy of
Massachusetts Institute of Technology Press.
When viewing conditions are bad (often the case for poorly lit indoor settings
or at night), the portion of the population likely to have difficulty
identifying U.S. banknotes is much larger. Under poor lighting conditions, the
visual acuity of persons with "normal" vision is compromised, while the visual
capabilities of those with mild visual impairments due to disease or the
natural cause of aging are further reduced.
Table 2-2 A Summary of Weber Fractions
____________________________________________________________
Stimulus Weber Fraction (%)
____________________________________________________________
Intensity of pure tone 20 [a]
Amplitude of vibration 24.5
Frequency of vibration 10
Visual contrast 10 [a]
Hue - red/yellow 0.6
Hue - green/yellow 2.9
Finger span 2.1
Lifted weight 4.3
Luminance 17
Skin pressure 17
Taste - salt 25
Taste - sucrose 17
____________________________________________________________
Data extracted from Laming, 1986.
ENDNOTES
Previous Section |
HTML Home Page |
Next Section
