View metadata, citation and similar papers at core.ac.

uk brought to you by CORE

provided by CommonKnowledge

Pacific University
CommonKnowledge
College of Optometry Theses, Dissertations and Capstone Projects

4-23-2015

Comparing Color Vision Testing Using the

Farnsworth-Munsell 100-Hue, Ishihara
Compatible, and Digital TCV Software
Rachel A. Murphy
Pacific University, rachel.murphy@pacificu.edu

Recommended Citation
Murphy, Rachel A., "Comparing Color Vision Testing Using the Farnsworth-Munsell 100-Hue, Ishihara Compatible, and Digital TCV
Software" (2015). College of Optometry. Paper 9.
http://commons.pacificu.edu/opt/9

This Thesis is brought to you for free and open access by the Theses, Dissertations and Capstone Projects at CommonKnowledge. It has been accepted
for inclusion in College of Optometry by an authorized administrator of CommonKnowledge. For more information, please contact
CommonKnowledge@pacificu.edu.
Comparing Color Vision Testing Using the Farnsworth-Munsell 100-Hue,
Ishihara Compatible, and Digital TCV Software
Abstract
It is crucial that eye care professionals be able to provide quick, accurate, and complete testing of color vision,
both to enhance the lives of patients and to satisfy the requirements laid out by industry standards. With the
growing popularity of the use of digital equipment in offices, there is a natural progression to digital color
vision screening tests, which have the advantage of being fast, inexpensive, and readily portable with
automated scoring for greater consistency. Few studies have sought to validate specific digital tests. The aim of
this study is to compare two traditionally accepted manual tests for detecting congenital color vision
deficiency (CCVD) with analogous digital versions. Thirty-five individuals (11 male, 24 female, mean age
25.1 years) with normal color vision and six individuals (all male, mean age 40.0 years) with congenital red-
green deficiency were asked to perform the following four tests for detecting CCVD: Ishihara Compatible
Pseudoisochromatic Plate (Ishihara); Waggoner Computerized Color Vision Test by Konan Medical USA
(TCV); Farnsworth-Munsell 100-Hue Color Cap Rearrangement Test (100-Hue); and an online version of
the Farnworth-Munsell 100-Hue (Online 100-Hue) available for free at color-blindness.com. The
administration time for each test was recorded along with test scores. The Ishihara and TCV had sensitivities
of 83.3% and 100% and specificities of 100% and 94.3%, respectively. The manual 100-Hue and the online
100-Hue had sensitivities of 66.7% and 83.3% and specificities of 88.6% and 85.7%, respectively. The average
test time was 2.3 minutes for the Ishihara and 3.4 minutes for the TCV. The geometric mean completion time
for the manual 100-Hue was 15 minutes; for the Online 100-Hue it was 7.5 minutes, thus reducing the test
time by 50%. A Bland-Altmann analysis shows that the Online 100-Hue tends to give higher scores than the
manual 100-Hue; however, there are several outliers that lead to a wide range and wide variability. Each of the
tests included in this study has specific strengths and weaknesses. An understanding of these can aid the
clinician in selecting the ideal test for a given situation as well as guide research and development of future
digital color vision tests. There are still concerns about consistency and accuracy of digital color tests due to
the variations in screens, but so far, results are promising.

Degree Type
Thesis

Rights
Terms of use for work posted in CommonKnowledge.

This thesis is available at CommonKnowledge: http://commons.pacificu.edu/opt/9

 Copyright and terms of use
If you have downloaded this document directly from the web or from CommonKnowledge, see the
“Rights” section on the previous page for the terms of use.
If you have received this document through an interlibrary loan/document delivery service, the
following terms of use apply:
Copyright in this work is held by the author(s). You may download or print any portion of this document
for personal use only, or for any use that is allowed by fair use (Title 17, §107 U.S.C.). Except for personal
or fair use, you or your borrowing library may not reproduce, remix, republish, post, transmit, or
distribute this document, or any portion thereof, without the permission of the copyright owner. [Note:
If this document is licensed under a Creative Commons license (see “Rights” on the previous page)
which allows broader usage rights, your use is governed by the terms of that license.]
Inquiries regarding further use of these materials should be addressed to: CommonKnowledge Rights,
Pacific University Library, 2043 College Way, Forest Grove, OR 97116, (503) 352-7209. Email inquiries
may be directed to:. copyright@pacificu.edu

This thesis is available at CommonKnowledge: http://commons.pacificu.edu/opt/9

Comparing Color Vision Testing Using the Farnsworth-Munsell 100-Hue,
Ishihara Compatible, and Digital TCV Software

by

Rachel Murphy

A Thesis
Submitted to the Graduate Faculty of the Vision Science Program
in Partial Fulfillment of the Requirements for the Degree of

Master of Science
In
Vision Science
Pacific University Oregon
College Of Optometry

May, 2015

-i-
Copyright © 2015 by Rachel Murphy
All Rights Reserved

- ii -
- iii -
 ABSTRACT

It is crucial that eye care professionals be able to provide quick, accurate, and complete
testing of color vision, both to enhance the lives of patients and to satisfy the requirements laid
out by industry standards. With the growing popularity of the use of digital equipment in offices,
there is a natural progression to digital color vision screening tests, which have the advantage of
being fast, inexpensive, and readily portable with automated scoring for greater consistency.
Few studies have sought to validate specific digital tests. The aim of this study is to compare two
traditionally accepted manual tests for detecting congenital color vision deficiency (CCVD) with
analogous digital versions. Thirty-five individuals (11 male, 24 female, mean age 25.1 years) with
normal color vision and six individuals (all male, mean age 40.0 years) with congenital red-green
deficiency were asked to perform the following four tests for detecting CCVD: Ishihara
Compatible Pseudoisochromatic Plate (Ishihara); Waggoner Computerized Color Vision Test by
Konan Medical USA (TCV); Farnsworth-Munsell 100-Hue Color Cap Rearrangement Test (100-
Hue); and an online version of the Farnworth-Munsell 100-Hue (Online 100-Hue) available for
free at color-blindness.com. The administration time for each test was recorded along with test
scores. The Ishihara and TCV had sensitivities of 83.3% and 100% and specificities of 100% and
94.3%, respectively. The manual 100-Hue and the online 100-Hue had sensitivities of 66.7% and
83.3% and specificities of 88.6% and 85.7%, respectively. The average test time was 2.3 minutes
for the Ishihara and 3.4 minutes for the TCV. The geometric mean completion time for the
manual 100-Hue was 15 minutes; for the Online 100-Hue it was 7.5 minutes, thus reducing the
test time by 50%. A Bland-Altmann analysis shows that the Online 100-Hue tends to give higher
scores than the manual 100-Hue; however, there are several outliers that lead to a wide range
and wide variability. Each of the tests included in this study has specific strengths and
weaknesses. An understanding of these can aid the clinician in selecting the ideal test for a given
situation as well as guide research and development of future digital color vision tests. There are
still concerns about consistency and accuracy of digital color tests due to the variations in
screens, but so far, results are promising.

- iv -
 ACKNOWLEDGEMENT

I would like to express my gratitude to my advisor, Dr. James Kundart, for his continuous
support of my study. This project would not have been possible without him introducing me to
this topic and guiding me through the process of designing, approving, implementing, and
interpreting a research project. His patience, motivation, enthusiasm, and immense knowledge
were incredibly influential along this journey. His door is always open, and he is always
accessible. This helped immensely with my project as well as my personal growth and
enjoyment of my time here at Pacific.

I would also like to thank the rest of my thesis committee: Dr. John R. Hayes and Dr. Karl
Citek. I consider myself privileged to have Dr. Hayes’ guidance in learning and implementing
statistical analysis. I appreciate his enduring patience. I would like to thank Dr. Citek for his
encouragement and insightful comments. His insatiable curiosity has been an inspiration.

I would like to thank Dr. Yu-Chi Tai for her encouragement and help along the way. She
continually pushed me to excellence and I am grateful.

Finally, I would like to give an honorable mention to Dr. James Sheedy. Even though he
wasn’t directly involved in my thesis, my time here at Pacific was greatly enhanced by getting to
know him. His insight has helped me to expand my thinking beyond the box, and I have been
able to see vision in a broader light. He showed me a passion and zeal for optometry and
introduced the philosophy of eye care rather than just the practice.

Rachel Murphy

-v-
 TABLE OF CONTENTS

SECTION PAGE

ABSTRACT …………………………………………………………………………………………………………………… IV
ACKNOWLEDGMENTS ………………………………………………………………………………………………… V
LIST OF FIGURES …………………………………………………………………………………………………………. VII
LIST OF TABLES …...……………………………………………………………………………………………………… VIII
LIST OF IMAGES ………………………………………………………………………………………………………….. IX
INTRODUCTION ….………………………………………………………………………………………………………. 1
METHODS …...……………………………………………………………………………………………………………… 5
RESULTS ………………………………………………………………………………………………………………………. 8
DISCUSSION ………………………………………………………………………………………………………………… 8
CONCLUSION ………………………………………………………………………………………………………………. 11
FIGURES ………………………………………………………………………………………………………………………. 12
TABLES ………………………………………………………………………………………………………………………… 14
REFERENCES ………………………………………………………………………………………………………………… 16
APPENDIX A: Examiner Protocol …………………………………………………………………………………. 19
APPENDIX B: Subject Recording Form ………………………………………………………………………… 21
APPENDIX C: Exclusionary Criteria ……………………………………………………………………………… 22
CURRICULUM VITAE …………………………………………………………………………………………………… 23

- vi -
 LIST OF FIGURES

Figure 1: Mean time taken to complete each task for color normal subjects …………………… 12

Figure 2: Bland-Altmann comparison of the Online Farnsworth-Munsell 100-Hue total

12
error scores with that of the manual Farnsworth-Munsell 100-Hue for all subjects …………

Figure 3: Bland-Altmann comparison of the Online Farnsworth-Munsell 100-Hue total

13
error scores with that of the manual 100-Hue for the color normal subjects ……………………

- vii -
 LIST OF TABLES

Table 1: Raw data for the six male subjects with a previous diagnosis of color vision
14
deficiency …………………………………………………………………………………………………………………….

Table 2: Summary of the results of the 4 color tests when compared to a previous
14
diagnosis of color deficiency ………………………………………………………………………………………..

Table 3: Raw data for the Testing Color Vision program of subjects who failed the
14
initial screening program …………………………………………………………………………………………….

Table 4: Frequency Distribution of TCV and Ishihara …………………………………………………… 15

Table 5: Frequency Distribution of TCV and manual 100-Hue ……………………………………… 15

Table 6: Frequency Distribution of Online 100-Hue and Ishihara ………………………………… 15

Table 7: Frequency Distribution of Online 100-Hue and manual 100-Hue …………………… 15

- viii -
 LIST OF IMAGES

Image 1: Color spectrums for normal and color deficient individuals ………………………… 1

Image 2: Notch filters in the EnChroma lens ……………………………………………………………… 2

Image 3: Trichromatic Confusion Lines ……………………………………………………………………… 3

Image 4: Target of the Nagel Anomaloscope …………………………………………………………….. 4

Image 5: Ishihara Test Plate Number 1 ………………………………………………………………………. 6

Image 6: Waggoner Computerized Color Vision Test …………………………………………………. 6

Image 7: Farnsworth-Munsell 100-Hue ……………………………………………………………………… 7

Image 8: Online Farnsworth-Munsell 100-Hue …………………………………………………………… 7

- ix -
 INTRODUCTION

Congenital color vision deficiency (CCVD) is typically caused by abnormalities of the

photopigments within the retina and has the effect of reducing the visibility of part of the
spectrum of visible hues. Often colors which appear mismatched to a color-normal individual
will look the same to a color deficient individual, causing abnormal color matching and color
confusion.2 Normal color vision is trichromatism, meaning there are three types of
photopigments in the retina. The three types of congenital
color vision deficiency (CCVD) indicate how many
photoreceptors are abnormally affected:
monochromatism, dichromatism, and anomalous
trichromatism. CCVD can further be classified by the
specific photopigment that is either missing
(dichromatism) or deficient (anomalous trichromatism).
These classifications are protanopia (L-cone pigment
missing), protanomaly (L-cone pigment deficient),
deuteranopia (M-cone pigment missing), and
deuteranomaly (M-cone pigment deficient). These
congenital deficiencies have X-linked inheritance with
prevalence as high as 8% of Caucasian males and 0.4% of
Caucasian females.2 A third cone photopigment in the S
cone can be deficient or missing, resulting in tritanomaly or
1
Image 1. Color spectrums for normal tritanopia, but this is extremely rare as a congenital
and color deficient individuals. condition and more likely the result of age or pathology,
like diabetes.

While CCVD varies in objective severity according to the number of receptors affected,
many factors influence subjective severity. This results in overlapping diagnoses, depending on
the type and classification, as well as individual differences in gene expression and variations in
compensation strategies. Regardless of the true severity, most cope very well, and indeed up to
25% are unaware of their condition until discovered by a routine screening well into adulthood.3
However, color is becoming an increasingly important aspect to our society with the advent of
technology, widespread color offset printing, and the ability to manufacture any color dye very
inexpensively. Apart from aesthetics, color is used to convey critical information and provide
useful guidance. A few key areas affected by color vision include early learning and education,
career options, and workplace safety and performance.4

While there does not seem to be an association between presence of abnormal color
vision and level of education achieved,4 5 color is often used in early childhood education. A
color deficient child sees their world of color differently than the normal child and this mismatch
can lead to frustration and confusion on the part of the color deficient child. In addition, color is
one of the earliest concepts taught in school and is often used as an aid in teaching more
complex lessons. If this key means of communication is missing for the child, it could appear to
be a learning deficiency if the presence of a color deficiency is unknown.2 6 7

Career options can also be limited due to CCVD. There are several industries with
positions that have stringent requirements for normal color vision such as aviation, maritime,
railway, armed services, textile and paint manufacturing, printing and fine art, and electrical

-1-
engineering. In these instances, good color vision is critical for safe and efficient operation.
Other industries choose screening tests that allow mild color deficient individuals to pass but
exclude more severe cases; these include firefighters, police officers, and certain positions in
electrical engineering, the armed services, hospital laboratory technicians, and merchant
seamen. Other vocations lead to disadvantages but are not officially regulated such as
horticulture, cartography, histopathology, and pharmacy.2 In one study, 30% of people with
abnormal color vision said their career choice was affected by their color vision.3 Domestic life
can also be affected by color deficiency in areas such as selecting clothes, picking ripe fruit,
choosing decorating materials, and recognizing sunburn.

Road safety has been a large area of debate, both for private and commercial drivers.
Ramachandran performed a rather extensive literature review4 and concluded that CCVD does
not appear to be associated with an increased risk of road traffic crashes in a clear majority of
studies, although experimental tasks show reduced performance.8 Reported difficulties with
driving are also common. It is speculated that the lack of correspondence between reduced
performance in experimental tasks and the incidence of traffic crashes are associated with
compensatory behavior.4

Unfortunately, despite efforts for well over a century, there remains no universally-
accepted treatment for color deficiency. Countless options have been tried, but only a few have
shown any promise. In the early 1970’s Dr. Zelter10 11 designed the X-Chrom lens, a red lens
placed over one eye that decreases the amount of light from short and medium wavelengths of
the spectrum. This darkens objects that are green and blue for this covered eye to allow for
comparison between eyes. A more recent application of filters is the EnChroma lens. These are
blue or gray tinted lenses that have a type of “notch filter” to absorb the wavelengths most
confusing to color-deficient individuals, allowing
for greater separation and discrimination between
colors. These filters, whether the X-Chrom or the
EnChroma lenses, do allow the patient to pass
many of the color screening tests, but they do not
correct the underlying cause. There is also a mixed
view of how much they actually help with color
discrimination as the lenses do not improve the
performance on color sorting tests such as the
Farnsworth-Munsell 100-Hue.12 When worn
monocularly as instructed, the X-Chrom also
decreases depth perception and induces a Pulfrich
effect which can reduce safety in many of the Image 2.9 Notch filters in the EnChroma lens.
applications for which they would be best suited,
such as driving.13

The most recent and promising treatment for color deficiency utilizes gene therapy to
replace the missing or abnormal photopigment. In 2009 Drs. Jay and Maureen, scientists at the
University of Washington, surgically injected a viral vector and therapeutic transgene into the
eyes of color-blind squirrel monkeys. After about 5 months, the monkeys were able to reliably
pass a color test.14 Testing has not yet progressed to humans, but work is underway to develop a
vitreal injection that would be safe for use on humans. Human trials are expected in the next
couple years according to a 2015 news article on NPR’s Shots.15

-2-
 While there are currently no treatments for CCVD, effects can often be minimized with
knowledge of the condition.16 For example, teachers can lessen the reliance on color as a
teaching tool. Redundancy can lead to greater safety, and individuals can consciously pay more
attention to other non-color cues. It is crucial that eye care professionals be able to provide
quick, accurate, and complete testing of color vision, both to enhance the lives of patients and
to satisfy the requirements laid out by industry standards.

Traditionally-Accepted Methods for Testing Color Vision

There are three main categories of color vision tests. Pseudoisochromatic plates utilize
pass/fail test plates to detect and quantify a color defect. Color cap arrangement tests look at an
individual’s ability to accurately put a series of color caps in order and analyze the types and
quantities of errors that are made. Color matching tests create mixtures of light that will look
the same to color deficient individuals but different to color normal individuals. Several tests
within each type are currently in use, but this paper specifically addresses the Ishihara
pseudoisochromatic plate test and the Farnsworth-Munsell 100-Hue test. The Nagel
Anomaloscope, a color matching test, will be discussed briefly.

The Ishihara pseudoisochromatic (PIC) plates is widely considered to be the most

effective test for detecting CCVD and is the most commonly used.2 17 Pseudoisochromatic plate
tests are designed using Stilling's principle, which breaks the target and field into small patches
and varies the reflectance of the patches so that the luminance difference between the target
and field is masked. Thus, color is the only cue available for distinguishing the target from the
background. The target and background colors lie on the dichromatic confusion lines, which are
the series of colors that a color deficient subject will have difficulty distinguishing.2 This test
requires the use of standardized lighting, typically the C.I.E. standard illuminant C (about 6800 K)
in order to maintain these principles.2 18 This test is best used to detect abnormal color vision, as
it has high sensitivity and specificity for this purpose, but it is less reliable for determining type
and cannot identify severity. Also, the booklet of plates can wear out over time and the light
adds additional expense, as they utilize office space and require maintenance and replacement.

The Farnsworth-Munsell 100-Hue (100-Hue)

is a color rearrangement test consisting of 85 color
caps representing a complete hue circle divided into
four 21- or 22-cap sets. The subject is instructed to
start with the marked cap and arrange the rest in
order according to their color for each of the boxes.
The test is scored by calculating and graphing an
error score using the numbers on the back of the
caps. The error score can divide subjects into groups
of superior, average, and poor hue discrimination
ability. Training can improve the error score as much
as 30%.20 Errors for normal subjects will be randomly
distributed around the hue circle, but for color
19
Image 3. Trichromatic Confusion Lines. deficient subjects, the errors will cluster on either
side of the hue circle creating an axis of confusion
where distortions of the chromaticity plane are present. This axis can then be used to classify
the type of color deficiency. This test is also capable of detecting acquired color vision deficiency

-3-
and is often used to track disease progression. This test measures color discrimination ability
rather than presence or absence of a defect, and there can be a wide range of abilities even
among color normal individuals. In color deficient individuals, it indicates ability to compensate
for a defect. As such, there is no direct correlation between severity of a color defect and the
error score on the 100-Hue.20 21 The main uses of the 100-Hue are to demonstrate excellent
color vision for certain vocations and to track the progression of diseases that lead to acquired
color vision loss. Performance is influenced by the illumination used and test administration
technique, as well as the skill, comprehension, and motivation of the subject.21 The largest draw-
back to the 100-Hue is the time it takes to score the classic analog version of the test and the
difficulty in determining an axis of confusion on some of the more ambiguous results.

A third type of color test involves Rayleigh color matching, most commonly in the form
of the Nagel or similar anomaloscope. This is rarely used in a clinical setting, but is considered
the gold standard for research purposes.2 The instrument consists of a circle divided into an
upper and a lower half. The top half of the circle consists of a mixture of 670 nm red and 549 nm
green light with a knob that changes the mixture from 1 to 75 scale units. A reading of 1 is pure
green and a reading of 75 is pure red. The lower half of the circle consists of a pure 590 nm
yellow light that can be adjusted in brightness with another knob. Using the anomaloscope is a
two-step process. First, subjects are asked to make several free matches by turning the knobs
for both the upper and the lower halves. Then the full range of possible matches is determined
using the free matches as a guide. The examiner changes the upper red-green mixture and the
subject is asked if a match can be made by changing the brightness of the yellow. Normal
subjects can create a match with the upper field set at 44 ± 2.5 scale units. Matches outside this
are considered abnormal. Protans can create a match with a greater amount of red present in
the upper field, and deutans can create a match with a greater amount of green.23 This is the
only test discussed in this paper that is capable of quantifiably separating dichromats from
anomalous trichromats. The Nagel
Anomaloscope requires careful
calibration and maintenance and
each machine must be calibrated
and used under identical
conditions, including control of
ambient temperature.24 25 Later
models, like the Oculus
anomaloscope, may require less
maintenance, and allow for testing
of blue-yellow (tritan) defects using
the Mooreland criterion. Image 4.22 Target of the Nagel Anomaloscope.

The Ideal Test for Color Vision

A need exists for an efficient diagnostic method for testing color vision to integrate
seamlessly with current equipment commonly found in a clinical setting. With the growing
popularity of the use of digital equipment in offices, there is a natural progression to the
convenience of having a color vision screening test in digital form, as indeed a few companies
are already offering.26 27 Digital color vision tests have the advantage of being fast, inexpensive,
and readily portable with automated scoring. Currently, a battery of tests must be used in order
to gain a full understanding of an individual's CCVD to include diagnosis and severity.17 The ideal

-4-
color vision test would allow for quick detection, followed by accurate diagnosis, and finally give
an indication of severity. It also needs to have universal criterion to reduce variation among
clinicians. Digital color tests are perfectly positioned to meet this need.

Electronic methods as a whole have been distrusted due to the variation in brightness
and contrast between screens in favor of the well-validated manual tests, but few studies have
sought to validate specific tests to either confirm or disprove this general feeling. The aim of this
study is to compare two traditionally accepted manual tests for detecting congenital color vision
deficiency (CCVD) with analogous digital versions.

METHODS

Subjects

This study was carried out at the Vision Performance Institute at Pacific University.
Thirty-five individuals (11 male and 24 female, mean age 25.1 years) with normal color vision
and six individuals (all male, mean age 40.0 years) with congenital red-green deficiency were
recruited from Pacific University and the surrounding community. Participants had to be over
the age of 12. Excluded from this study were individuals with operable cataracts or acquired
ocular diseases that may affect color vision (such as diabetic retinopathy or optic neuritis,
among other, more rare conditions). These eye diseases can create blue-yellow color vision
defects, in the case of diabetic retinopathy or optic neuritis, or otherwise alter color perception.
Likewise, participants could not be taking any medications that affect color vision, such as
thiazide diuretics for high blood pressure. Appendix C contains a non-exhaustive list of diseases
and medications that were excluded. Pregnant women were not excluded, as testing color vision
is non-invasive, though no pregnant women participated. None of the subjects had a history of
photosensitive epilepsy, and no tinted lenses were permitted. Institutional Review Board
approval and informed consent was obtained for all subjects.

Procedure

All the examinations were made by the same examiner using the same procedure and
the same viewing conditions. A brief history was collected to include self-reported vision of
20/40 or better at near, absence of exclusionary criteria, previous diagnosis of congenital color
vision deficiency (CCVD), and family history of CCVD. Participants were encouraged to wear their
habitual near correction. Subjects then performed four tests for detecting CCVD in randomized
order in a dark room. All tests were performed binocularly.

A Latin Square was used to determine the randomized order prior to testing. A Samsung
Series 5 14" Notebook laptop running Windows 8 was used to administer the two digital tests.
Prior to testing, the Display Color Calibration program provided by the computer was used to
calibrate the brightness and contrast of the screen. The administration time for each test was
recorded along with test scores. Appendix A contains a printout of the exact procedure followed
including instructions sets for each test. Appendix B contains the subject recording for used to
collect and record all data.

-5-
Tests and Scoring

Ishihara Compatible Pseudoisochromatic Plate (24-plate edition)

Plates 1 through 17 were presented sequentially under

Richmond Products' Daylight Illuminator (near illuminant C, 6280
K). This test must be administered under specialty lighting to
maintain the validity of the test.2 18 The subject was instructed to
call out the number seen on the plate or to state that they do not
see a number; only answers given within three seconds were Image 5.28 Ishihara Test Plate
recorded. Number 1.

A score of 13 or more correct on plates 1-15 was regarded as normal color vision.2 For
those scoring below 13 out of 15 correct, plates 16 and 17 were used to classify the failed
subjects as protan or deutan deficiency as well as to determine moderate (both digits visible on
both plates) or strong (only one digit visible on at least one plate). Misreadings, which are
common among normal subjects but are not the expected response of a color-deficient subject,
were not regarded as mistakes.29 30 31 Based on this criterion, previous studies have found the
sensitivity of the Ishihara test to be 99% and the specificity 94%.29 32

Waggoner Computerized Color Vision Test software by Konan Medical USA

The Waggoner Computerized Color Vision Test software (abbreviated TCV for clarity in
this paper) adult diagnostic testing program from www.TestingColorVision.com33 was designed
by Terrence L. Waggoner, OD. It is currently available commercially through Konan Medical USA.
The TCV is to be used on a color-calibrated monitor illuminated with a 6500 K lighting
environment ("Terms-Use"). According to the manufacturer, conditions other than this are most
likely to result in false negatives, especially for low-severity color deficiencies; false positives are
unlikely. A current color calibration is recommended for optimized results. No studies were
found validating this test.

Test plates are programmed to flash for 2 seconds, then replaced by a list of 9 answers
from which to select (8 number choices and one "nothing"). There are 30 diagnostic plates
presented. If this section is passed, the software ends the testing and presents a simplified
scoring certificate with a diagnosis of normal along with the total number of plates correct out
of 30 total plates. If this section is not passed, the software immediately begins further tests to
quantify the defect. This section consists of three sets: 32 plates for Protan, 32 plates for
Deutan, and 12 plates for Tritan. At the end of the test, a score sheet is presented with the
number of plates correct out of total plates for each of the four sections (diagnostic, protan,
deutan, and tritan) with labels of pass, mild,
moderate, or severe. The exact algorithm
used to determine pass/fail and severity is
not known to the authors; however, the
cutoff for pass/fail on the diagnostic portion
appears to be 25/30 (the lowest pass in our
study was 26/30 and the highest fail was
25/30). On several occasions, more than one
Image 6.34 Waggoner Computerized Color Vision Test. diagnosis was present; for example, the

-6-
software might label a subject as a moderate protan and a severe deutan. In these instances,
the most severe diagnosis was used for comparison with the other tests.

Farnsworth-Munsell 100-Hue Color Cap Rearrangement Test

The Farnsworth-Munsell 100-Hue Color Cap Rearrangement Test (100-Hue) was also
administered under the Richmond Products' Daylight Illuminator.36 37 Subjects were instructed
to rearrange the color caps in the four boxes in order according to their color. The boxes were
always presented in order beginning with box 1 and ending with box 4. No time limit was set.
Cap order was recorded and total error score (TES) was
calculated using the excel program "Farnsworth 100 Hue
MS-Excel Template for Scoring" published by Richmond
Products. Pass/fail was based on a TES outside of the
95th percentile of the age-matched norms found by
Kinnear and Sahraie.38 Failing scores were further
classified as protanomaly or deuteranomaly using the
axis of the largest cap error.39 40 Axes falling outside the
Image 7.35 Farnsworth-Munsell 100-Hue.
expected criteria were classified as a tritanomaly.

Online Farnworth-Munsell 100-Hue

An online version of the Farnworth-

Munsell 100-Hue (Online 100-Hue), available
for free at www.color-blindness.com,42 was
selected as an example of an online version of
the 100-Hue. Color-blindness.com is a free
informational website hosted by a color
deficient individual who wanted to understand
his condition better and help others to do the
same. The author has created several online Image 8.41 Online Farnsworth-Munsell 100-Hue.
versions of color tests. No studies were found
validating this test.

The test consists of four rows of colored squares. Subjects were instructed to drag and
drop the color squares to arrange them in order according to their color. Squares can only be
moved back and forth along their own row, thus mimicking the cap sets of the 100-Hue. When
finished, the program gives a total error score and graph of the subject's performance.
Categorization was based on the same criteria as the manual 100-Hue rather than relying on the
algorithm of the program, as this algorithm was unknown to the authors.

Data Analysis

Completion times for each test were analyzed using Analysis of Variance on SPSS
software (Version 19, IBM Inc.). The results of the two digital versions were compared to the
results of the manual versions. Sensitivity and specificity for each test were calculated using
previous diagnosis of a color deficiency as the standard for comparison. Due to the small sample
size, test analysis was based on predictability of the presence of any type of color vision rather
than broken down into the specific type and severity.

-7-
 RESULTS

Each of the 41 subjects performed all four of the color vision tests. Raw data for the six
subjects with a previous diagnosis of a color vision deficiency can be seen in Table 1. The
Ishihara had no false positives with a specificity of 100%, but it did not fail one color deficient
subject resulting in one false negative. The Testing Color Vision program had a sensitivity of
100%, but had two false positives. The Farnsworth-Munsell 100-Hue (100-Hue) had four false
positives and two false negatives. The Online 100-Hue had five false positives and one false
negative. Table 2 shows the sensitivities and specificities for each of the four tests.

The average test time was 2.3 minutes for the Ishihara and 3.4 minutes for the TCV. The
geometric mean completion time for the manual 100-Hue was 15 minutes; for the Online 100-
Hue it was 7.5 minutes. Thus, the Online 100-Hue reduced the time of the test by 50%. All times
were significantly different from each other with non-overlapping 84% confidence intervals as
shown in Figure 1.

A Bland-Altmann analysis was conducted to compare the total error scores of the
manual 100-Hue to that of the Online 100-Hue (Figure 2). The chart shows that the Online 100-
Hue tends to give higher scores than the manual 100-Hue as scores are clustered between zero
and 50. However, there are several outliers that lead to a wide range and wide variability.

To further break down the results of the Testing Color Vision program, Table 3 shows
the results of the three diagnostic programs that are triggered once the subject fails the 29-plate
screening test. Eight subjects failed the screening test and went on to perform the more in-
depth tests.

Based on the performance of the Ishihara, all color deficient subjects were
deuteranomalous, which is the most common color deficiency. Of these six subjects, the TCV
incorrectly diagnosed three of the known color deficient subjects as protanomalous; the 100-
Hue incorrectly diagnosed two subjects as protanomalous; and the Online 100-Hue incorrectly
diagnosed one subject as tritanomalous. Tables 4 and 5 compare the diagnostic outcome of the
TCV program to the Ishihara and Farnsworth-Munsell 100-Hue respectively. Tables 6 and 7 do
the same for the online version of the Farnsworth-Munsell 100-Hue.

DISCUSSION

This study is limited by its use of self-reported data. The anomaloscope is considered the
gold standard for color testing in research settings, although this instrument was not included in
this study as the authors set out originally to compare clinical methods for color vision testing.
The Nagel and similar anomaloscopes are rarely found in use outside of the laboratory. Another
test that might have been included is the American Optical Company Hardy, Rand, and Rittler
(HRR) pseudoisochromatic plate test. This test includes a severity scale and is capable of testing
for tritan defects in addition to protan and deutan defects. In this sense, it is quite similar to the
TCV program used in this study.

Another limitation to this study is the small number of color-deficient subjects included
in this study. Subjects were recruited from Pacific University College of Optometry and the

-8-
surrounding community. The percentage of color-deficient subjects is representative of the
prevalence of color deficiency among the Caucasian population, and the authors struggled to
recruit larger numbers of color-deficient subjects.

The main object of this study was to determine whether digital versions of testing for
color vision are potential options for use in clinic, specifically looking at the TCV program
available from www.TestingColorVision.com43 and the Online 100-Hue available at www.color-
blindness.com. Each of the tests included in this study has specific strengths and weaknesses. An
understanding of these can aid the clinician in selecting the ideal test for a given situation as
well as guide research and development of future digital color vision tests.

The Farnsworth-Munsell 100-Hue is an example of a color cap rearrangement test. The

main strength of the 100-Hue is in determining color discrimination ability, whether in a color-
normal or a color-deficient individual. Scores can be divided into superior, average, low, and
severe. Dichromats tend to be more severe than anomalous trichromats, but performance,
especially among anomalous trichromats, can be quite varied. Ultimately, the error score cannot
reliably be used to distinguish between dichromats and anomalous trichromats, and up to half
of anomalous trichromats may show an error score of less than 100. It is also common to lack an
axis of confusion making diagnosis difficult.21 Vingrys found that the ability to correctly identify
deuteranomaly was 26% with 38% being misclassified as protanomalous.40 This was improved to
71% if the failure criterion was set at a score of 100 with overall misclassifications at 13%. Due
to these variables, the 100-hue is not considered to be an accurate screening test for CCVD.20 40
Additional drawbacks to this test include the time it takes to administer and score and the
reliance on attention and motivation of the patient. Currently, the 100-Hue is most often used
to prove excellent color vision on a yearly basis for industries that require superior color
sensitivity, such as textile, gemology, and paints. Training or a vocation that utilizes continuous
color matching can improve the error score by as much as 30%.20

The online version of the 100-Hue in this study addresses several of the drawbacks to
the manual version. The administration time was cut in half and the scoring instantaneous.
While most online versions automatically calculate the axis of confusion, the results of the built-
in algorithm for the one used in this study were not relied upon for data analysis. The author’s
general sense was that doing so would have increased the number of false positives, but exact
data were not determined. In addition, many scores were classified as undefined for the built-in
algorithm, so these would still have to be manually assessed. It is entirely likely that an accurate
algorithm could be created that out-performs even the traditional 100-Hue and improves the
accuracy and ease of making a diagnosis. Another advantage is that it takes the subjective
interpretation out, which can allow for greater consistency. Unfortunately, the Bland-Altmann
comparison of the Online 100-Hue to the manual 100-Hue shows a range as wide as the
variability indicating that it is a very poor comparison to the performance on the traditional 100-
Hue. As an example, one subject had a perfect score on the manual test and a total error score
of 144 on the online version. Another subject had a perfect score on the online version but a
score of 120 on the manual version. Neither subject was found to be color deficient by the
Ishihara or TCV and neither had a positive family history. This variability is concerning. On the
other hand, the variability was reduced considerably when the color deficient subjects were
removed from the data (Figure 3). If this test is used to prove excellent color vision rather than
to diagnose color deficient patients as the 100-Hue is typically used, an online version might be
an acceptable alternative. Under these circumstances, patients are likely to be highly motivated

-9-
which may reduce some of the variability. Ultimately, the algorithms may be improved upon,
but many of the variables that hinder the manual 100-Hue apply to the online as well, including
a large reliance on attention and motivation, a significantly longer administration time than the
pseudoisochromatic plates, poor results for use as a screening test, and difficulty determining
the best way to score atypical results.

Pseudoisochromatic plates are superior to the 100-Hue as a screening test. The

sensitivities and specificities are very high for the Ishihara. The difficulty with scoring and
classification inherent in the Farnsworth-Munsell 100-Hue, both manual and digital, are not
present, and scoring is not time-consuming. Certain plates are susceptible to misreading, and
this must be taken into account when scoring the test,29 30 31 but otherwise scoring is very
straight-forward. The main drawback of the Ishihara is that it does not give a clear indication of
severity. Accurately diagnosing protans versus deutans can be unreliable as well, although if
patients are asked to compare brightness on the diagnostic plates, accuracy may be as high as
94%.2 Also, very mild anomalous protans and deutans may pass the Ishihara.17

Digital pseudoisochromatic plate tests, such as the TCV software program included in
this study, are quite promising. Though the TCV had a statistically significant longer test
administration time (p<0.001), the effect size was 1.4 minutes. The greatest advantage of a
computerized test is that it is self-run and self-scored once instructions have been given. The
Ishihara requires the presence of an administrator throughout the testing period to turn the
pages and record the answers, and then the test must be hand scored. Thus, the actual time
required of a doctor or technician is quite comparable, if not less for the TCV program. In
addition, the Ishihara requires the use of special lighting which adds extra expense, upkeep, and
desk space. However, the TCV program has flaws as well. The TCV program gives the number of
plates missed and a diagnosis for each category, and most color-deficient individuals missed
plates in both the protanomaly and the deuteranomaly categories. Four of the six known color
deficient subjects scored were classified as having the same severity rating for both
protanomaly and deuteranomaly. The ultimate diagnosis was based on the category that had
the most missed plates, but in all color deficient subjects, these numbers were fairly close
(average difference of 3 plates between the two categories), making the final diagnosis weak. In
the one case that had a difference of 10 plates, the TCV program classified the subjects as a
protan when the Ishihara indicated the subject to be a deutan. The TCV software does include a
Farnsworth D-15 test, which may aid in the separation of severe protans from deutans, but this
was not included in the study. This indicates that the TCV program currently is acceptable as a
screening test for color deficiency in general, but needs to be further analyzed to create a larger
and more accurate gap between the performances of a protan versus a deutan in order to
accurately diagnose the type of deficiency. Further investigation into the reliability of the
severity diagnosis is also needed, but was beyond the scope of this study.

Ultimately, digital color vision testing using pseudoisochromatic plates appears to be a

potentially valid method for screening as our study showed that those who had a color
deficiency did fail the TCV test. It is not recommended for unregistered versions of color vision
tests to be used in clinical applications, and any published product needs to be individually
tested and validated. There is a commercially available 100-Hue studied by Ghose and
associates in 201444 that was unknown to the authors at the time of this study that would have
been a better option due to the greater likelihood of being consistent and available long-term.
Random, free tests are not acceptable for clinical use beyond outside-clinic screening purposes,

- 10 -
but published programs such as the TCV may have a significant role in the coming years with
proper validation. As with any diagnostic test, each version must be individually validated and
algorithms should be designed and enhanced using performance on each plate to find the best
set of plates to present and to further separate the diagnosis of protanomaly versus
deuteranomaly. An equivalent of the study by Birch32 determining the efficiencies of each
individual plate in the Ishihara would give greater guidance in altering the test plate sequence
for greatest accuracy in the TCV test.

As we move in this direction, a few other concerns for digital color testing must be
addressed. What role does the brightness and contrast of screens play in digital color testing?
Studies addressing this were performed in the early 1990s,45 46 but technology has changed
quite a bit in 20 years. Does the calibration of a computer screen change the validity of a test? Is
a native color calibration program adequate for setting up the parameters of a screen for use
with color testing? In addition to these questions concerning the screen, the screens can create
after-images that could potentially interfere with subsequent test plates. These are valid issues
and concerns, but ultimately trials must be conducted to determine if they play a role significant
enough to discontinue the use of digital color tests or to relegate it to use as an initial screening
test only.

CONCLUSION

Digital color vision tests are shorter and easier to administer, allow for automated
scoring, and create consistency across testing environments. They are also cheaper to
manufacture and maintain, and do not take up as much desk or storage space in the office
compared with “hardware” tests and special lighting. With the health industry adapting
electronic health records and equipment becoming increasingly digital, it is an economical and
logical transition to introduce digital color vision tests. There are still concerns about consistency
and accuracy due to the variations in screens, but so far, results are promising. The greatest
challenge will be in developing the ideal set of test plates and algorithms to allow for the most
ideal test set. This is likely to include more than one type of color test depending upon the
application.

Ultimately, the usefulness of a color vision test depends upon the reason for testing,
such as the testing of color vision with regards to qualifying for a particular vocation. In these
instances, the color vision test must be selected based on a pre-determined industry standard
built upon the color vision needs of the particular application. Each of the tests currently used in
clinical practice have specific strengths and weaknesses. Understanding these allows the
clinician to select a certain test for a specific application. It will also allow for the design of better
tests and more accurate algorithms as digital versions become more widespread.

Disclosures

Four free copies of the Waggoner Computerized Color Vision Test software developed
by Terrence Waggoner, OD and distributed by Konan Medical USA were donated for use in this
study. The authors gain no financial benefit or otherwise concerning the outcome of this study.

- 11 -
 FIGURES

Time to completion for normal

color vision subjects
18

16

14
Minutes to completion

12

10

0
Ishihara TCV 100-Hue Online 100-Hue

Figure 1. Mean time taken to complete each task for color normal
subjects. Error bars show the 84% confidence intervals.

Bland-Altmann of Online vs Manual

100-Hue for All Subjects
150

100
Online - Manual

50

0
0 50 100 150 200
-50

-100
Mean

Figure 2. Bland-Altmann comparison of the Online Farnsworth-Munsell 100-Hue total error

scores with that of the manual Farnsworth-Munsell 100-Hue for all subjects. Diamonds indicate
color normal subjects and squares indicate color deficient subjects.

- 12 -
 Bland-Altmann of Online vs Manual
100-Hue for Color Normal Subjects
150

100
Online - Manual

50

0
0 50 100 150 200
-50

-100
Mean

Figure 3. Bland-Altmann comparison of the Online Farnsworth-Munsell 100-Hue total error

scores with that of the manual Farnsworth-Munsell 100-Hue for the color normal subjects.

- 13 -
 TABLES

Table 1: Raw data for the six male subjects with a previous diagnosis of color vision deficiency.
Subject Age Ishihara # Correct TCV # Correct 100-Hue Online 100-Hue
(15) (29) Error Score Error Score
s09 67 2 2 128 272
s11 49 1 1 92 72
s28 51 1 0 156 180
s31 26 5 2 160 92
s38 16 1 0 48 92
s41 31 13 12 164 100

Table 2: Summary of the results of the 4 color tests when

compared to a previous diagnosis of color deficiency.
Test Sensitivity Specificity
Ishihara 83.3% 100%
TCV 100% 94.3%
100-Hue 66.7% 88.6%
Online 100-Hue 83.3% 85.7%

Table 3: Raw data for the Testing Color Vision program of subjects who failed the initial
screening program. Results show the number of plates correct and severity assigned by the
software for each diagnostic set. The total number of plates are shown in parentheses.
Screening Protan Protan Deutan Deutan Tritan Tritan
Subject (29) (32) Diagnosis (32) Diagnosis (12) Diagnosis
s09 2 0 Severe 3 Severe 7 Mild
s11 1 0 Severe 10 Moderate 11 Pass
s16 19 27 Mild 28 Pass 11 Pass
s20 25 27 Mild 26 Mild 12 Pass
s28 0 0 Severe 0 Severe 11 Pass
s31 2 9 Moderate 5 Moderate 12 Pass
s38 0 2 Severe 5 Moderate 12 Pass
s41 12 20 Mild 18 Mild 12 Pass

- 14 -
 Table 4: Frequency Distribution of Table 6: Frequency Distribution of
TCV and Ishihara. Online 100-Hue and Ishihara.
Ishihara TCV Results Ishihara Online 100-Hue Results
Results Normal Protan Deutan Tritan Results Normal Protan Deutan Tritan
Normal 33 3 2 0 Normal 25 0 4 0
Protan 0 0 0 0 Protan 0 0 0 0
Deutan 0 3 2 0 Deutan 1 0 4 0
Tritan 0 0 0 0 Tritan 0 0 0 0

Table 5: Frequency Distribution of Table 7: Frequency Distribution of

TCV and manual 100-Hue. Online 100-Hue and manual 100-Hue.
100-Hue TCV Results 100-Hue Online 100-Hue Results
Results Normal Protan Deutan Tritan Results Normal Protan Deutan Tritan
Normal 28 2 3 0 Normal 24 0 5 3
Protan 0 2 1 0 Protan 1 0 1 1
Deutan 2 0 2 0 Deutan 0 0 3 1
Tritan 3 0 0 0 Tritan 1 0 1 1

- 15 -
 REFERENCES

1. Image Source: http://psych.ucalgary.ca/PACE/VA-Lab/colourperceptionweb/congenital.htm

2. Birch J. Diagnosis of defective colour vision. 2nd ed. England: Oxford; Boston: Butterworth-Heinemann;
2001.
3. Steward JM, Cole BL. What do color vision defectives say about everyday tasks? Optometry & Vision
Science. 1989;66(5):288-95.
4. Ramachandran N, Wilson GA, Wilson N. Is screening for congenital colour vision deficiency in school
students worthwhile? A review. Clin Exp Optom. 2014 Nov;97(6):499-506.
5. Cumberland P, Rahi JS, Peckham CS. Impact of congenital colour vision deficiency on education and
unintentional injuries: findings from the 1958 British birth cohort. BMJ. 2004 Nov 6;329(7474):1074-5.
6. Bacon L. Colour vision defect-an educational handicap. Med Officer. 1971;125:199-209.
7. Thuline HC. Color blindness in children. The importance and feasibility of early recognition. Clin Pediatr
(Phila). 1972 May;11(5):295-9.
8. Pepple G, Adio A. Visual function of drivers and its relationship to road traffic accidents in Urban Africa.
Springerplus. 2014 Jan 24;3:47,1801-3-47. eCollection 2014.
9. Image Source: http://enchroma.com/technology/
10. Zeltzer H. The X-chrom lens. J Am Optom Assoc. 1971;42(9):933-9.
11. Siegel IM. The X-Chrom lens. On seeing red. Surv Ophthalmol. 1981 Mar-Apr;25(5):312-24.
12. Kassar BS, Dresner SC, May JG, Marx MS, Safir A. Evaluation of the X-chrom lens and color deficiency.
Eye & Contact Lens. 1984;10(1):100-4.
13. Hartenbaum NP, Stack CM. Color vision deficiency and the X-Chrom lens. Occup Health Saf. 1997
Sep;66(9):36,40, 42.
14. Mancuso K, Hauswirth WW, Li Q, Connor TB, Kuchenbecker JA, Mauck MC, et al. Gene therapy for
red–green colour blindness in adult primates. Nature. 2009;461(7265):784-7.
15. Hamilton J. University And Biotech Firm Team Up On Colorblindness Therapy [Internet]. National
Public Radio: Shots Health News from NPR; ©2015 [updated 2015 March 25; cited 2015 April 15].
Available from: http://www.npr.org/blogs/health/2015/03/25/395303785/university-and-biotech-firm-
team-up-on-colorblindness-therapy.
16. Cole BL. The handicap of abnormal colour vision. Clin Exp Optom. 2004;87(4-5):258-75.
17. Dain SJ. Clinical colour vision tests. Clinical and Experimental Optometry. 2004;87(4-5):276-93.
18. Dain SJ, Honson V, Curtis C. Suitability of fluorescent tube light sources for the Ishihara test as
determined by colorimetric methods. Colour Vision Deficiencies XI. Springer; 1993. p. 327-33.
19. Image Source: http://www.guldenophthalmics.com/wp-content/uploads/2011/11/Richmond-Color-
Tutorial-Final.pdf
20. Farnsworth D, Color M. The Farnsworth-Munsell 100-Hue Test for the examination of color
discrimination. Munsell Color Company; 1957.
21. Birch J. Use of the Farnsworth—Munsell 100‐Hue test in the examination of congenital colour vision
defects. Ophthalmic and Physiological Optics. 1989;9(2):156-62.

- 16 -
22. Image Source: http://extras.springer.com/2007/978-3-540-32706-6/html/fig/fullres/fig_06_05.jpg
Schiefer, Wilhelm, Hart (Eds.). Clinical Neuro-ophthalmology - A Practical Guide. Springer, Berlin/
Heidelberg/New York, 2007.
23. Birch J. Diagnosis of defective colour vision using the Nagel anomaloscope. Docum Ophthal Proc
Series. 1982;33:231-5.
24. Jordan G, Mollon JD. The Nagel anomaloscope and seasonal variation of colour vision. Nature. 1993
Jun 10;363(6429):546-9.
25. Jägle H, Pirzer M, Sharpe LT. The Nagel anomaloscope: its calibration and recommendations for
diagnosis and research. Graefe's Archive for Clinical and Experimental Ophthalmology. 2005;243(1):26-32.
26. Ghose S, Parmar T, Dada T, Vanathi M, Sharma S. A new computer-based Farnsworth Munsell 100-hue
test for evaluation of color vision. Int Ophthalmol. 2014 Aug;34(4):747-51.
27. Seshadri J, Christensen JO, Lakshminarayanan VF, Bassi CJ. Evaluation of the New Web-Based "Colour
Assessment and Diagnosis" Test. Optometry & Vision Science. 2005 October;82(10):882-5.
28. Image Source: http://www.color-blindness.com/ishihara_cvd_test/img/plates/Ishihara%20Plate%
2002-38.png from http://www.color-blindness.com/ishihara-38-plates-cvd-test/
29. Birch J, McKeever LM. Survey of the accuracy of new pseudoisochromatic plates. Ophthalmic and
Physiological Optics. 1993;13(1):35-40.
30. Miyahara E. Errors reading the Ishihara pseudoisochromatic plates made by observers with normal
colour vision. Clinical and Experimental Optometry. 2008;91(2):161-5.
31. Cosstick M, Robaei D, Rose K, Rochtchina E, Mitchell P. Numerical confusion errors in ishihara testing:
findings from a population-based study. Am J Ophthalmol. 2005;140(1):154-6.
32. Birch J. Efficiency of the Ishihara test for identifying red‐green colour deficiency. Ophthalmic and
Physiological Optics. 1997;17(5):403-8.
33. Waggoner TL. Testing Color Vision. 2013.
34. Image Source: www.TestingColorVision.com
35. Image Source: https://www.good-lite.com/Details.cfm?ProdID=546&category=0&Secondary=0
36. Hovis JK, Neumann P. Colorimetric analyses of various light sources for the D-15 color vision test.
Optometry & Vision Science. 1995;72(9):667-78.
37. Dain SJ, Honson VJ. Selection of an optimal light source for the FM 100-Hue Test. Colour Vision
Deficiencies IX. Springer; 1989. p. 425-32.
38. Kinnear PR, Sahraie A. New Farnsworth-Munsell 100 hue test norms of normal observers for each year
of age 5-22 and for age decades 30-70. Br J Ophthalmol. 2002 Dec;86(12):1408-11.
39. Dain SJ, Birch J. An averaging method for the interpretation of the Farnsworth-Munsell 100-Hue Test--
I. Congenital colour vision defects. Ophthalmic Physiol Opt. 1987;7(3):267-80.
40. Vingrys AJ, Atchison DA, Bowman KJ. The use of colour difference vectors in diagnosing congenital
colour vision deficiencies with the Farnsworth-Munsell 100-hue test. Ophthalmic and Physiological Optics.
1992;12(1):38-45.
41. Image Source: http://www.color-blindness.com/farnsworth-munsell-100-hue-color-vision-test/
42. color-blindness.com [homepage on the Internet]. Colblindor; ©2006-2014. [updated unknown; cited
2015 April 15]. Available from: http://www.color-blindness.com/farnsworth-munsell-100-hue-color-
vision-test.

- 17 -
43. TestingColorVision.com [homepage on the Internet]. Testing Color Vision; ©2013. [updated unknown;
cited 2015 April 15]. Available from: http://www.testingcolorvision.com/index.php.
44. Tucows, Inc. [Software]. Shareware; ©2004. [updated 2004 November 08; cited 2015 April 15].
Available from: http://archive.org/details/tucows_343051_Visual_Acuity_and_Color_Vision_Testing.
45. Birch J, Barbur J, Harlow A. New method based on random luminance masking for measuring
isochromatic zones using high resolution colour displays. Ophthalmic and Physiological Optics.
1992;12(2):133-6.
46. Regan B, Reffin J, Mollon J. Luminance noise and the rapid determination of discrimination ellipses in
colour deficiency. Vision Res. 1994;34(10):1279-99.

- 18 -
 APPENDIX A

Examiner Protocol
1. Greet the patient. Fill out paperwork:
a. Participant Contact Info
b. Informed Consent
2. Explain the experiment and ask if there are any questions.
3. Seat the subject comfortably and fill out the Subject Recording Form.
a. Have subject wear habitual correction
4. Determine the patient's testing order and perform each task according to the guidelines
below.

Ishihara Compatible Pseudoisochromatic Plate Color Vision Test, 24 plate edition.

1. Seat the subject comfortably in front of a desk.
2. Set the Richmond Products’ Daylight Illuminator in front of the subject and turn it on.
a. The stage should be 30 inches from the subject and tilted so that the plane of the
paper is at a right angle to the line of sight.
3. This test should be performed in a dark room for consistency.
4. Explain the test to the subject.
"This task will take about 5-10 minutes. It will be administered under the recommended full-
spectrum illumination using this light. The test consists of a book of 17 plates in which you will
be asked to identify a number presented on the plate. Some plates may not have a number,
and if you do not see a number, please state, “I do not see a number.” You must respond within
3 seconds in order to have your answer counted. Do you have any questions?"
4. Note the time.
5. Set the book on the stage and open the front cover to the first example plate.
6. Ask the subject, "Please call out the number you see."
a. The subject must respond within 3 seconds in order to get credit for the plate.
7. Record the number that the subject said on the "Subject Recording Sheet"
a. If the subject cannot see a number, or does not respond within 3 seconds, record
an "X."
8. Turn the page to the next plate and repeat for all 17 plates.
9. Record the elapsed time for the test on the recording sheet.

Testing Color Vision software version of the Ishihara Pseudoisochromatic Plate Test
1. Seat the subject comfortably in front of a computer at a desk in a dark room.
a. Ensure a mouse is connected to the computer.
b. The screen should be 30 inches from the subject and viewed at a 90 degree angle.
2. Click on the icon for TCV (ensure the CD is in the CD drive).
3. Click on Adult Diagnostic Testing
4. Click on Next
5. Enter the Subject Number as the User Name.
6. Explain the test to the subject.
"This task will take about 5-10 minutes. You will see a test plate made up of dots with a number
on it. The test plate number will appear for TWO seconds. Next you will see a list of NINE
answers to select from. Please click on the number that you saw. If you did not see a number,
click "NOTHING." After selecting your answer, click on next to continue to the next test plate.
Do you have any questions?"

- 19 -
 8. Note the time.
9. Have the subject click the start button when they are ready to begin.
10. Record the subject's score on the Subject Recording Sheet
11. Record the elapsed time on the Subject Recording Sheet.
12. Click Save Certificate and Exit
13. Save results under subject's ID in the desktop folder "TCV"

Farnsworth-Munsell 100-Hue Color Cap Arrangement Test

1. Seat the subject comfortably in front of a desk.
2. Set the Richmond Products’ Daylight Illuminator in front of the subject and turn it on.
a. The stage should be 30 inches from the subject.
b. Lighting should be from 90 degrees above and viewing angle should be 60 degrees.
3. This test should be performed in a dark room for consistency.
4. Explain the test to the subject.
"This task will take about 15 minutes. The object of the test is to arrange the caps in order
according to color. Begin with the cap with a white dot on it. Please use the magnetic wand to
drag the caps into place so they form a regular color series beginning with the first cap. It
should take you about 2 minutes per tray, but accuracy is more important than speed. There is
no time limit. Try to do your best. Do you have any questions?"
5. Note the time.
5. Ensure the first tray is randomized and set it on the stage.
6. Have the patient arrange the caps in color order.
7. Flip the tray over and record the numbers in order on the recording sheet.
8. Repeat for the next 3 trays, in order.
9. Record the elapsed time for the test on the recording sheet.

Online Farnsworth-Munsell 100-Hue Color Cap Arrangement Test

1. Seat the subject comfortably in front of a computer at a desk in a dark room.
a. Ensure a mouse is connected to the computer.
2. Click on the desktop link to access the Online Farnsworth-Munsell 100-Hue Test.
http://www.color-blindness.com/fm100hue/FM100Hue.swf
4. Explain the test to the subject.
"This task will take about 15 minutes. The object of the test is to arrange the color tiles in order
according to color. The first and last tiles on each row are anchored in place, the rest are mixed
randomly within their rows. Please use your mouse to click on a tile, drag, and drop it into place
so that they form a regular color series beginning with the first tile. You cannot move a plate
from one row to another, just back and forth along its own row. It should take you about 2
minutes per row, but accuracy is more important than speed. There is no time limit, but do not
dawdle. Try to do your best. Do you have any questions?"
5. Note the time.
6. Have the subject click start when they are ready to begin.
7. Record the subject's Total Error Score.
8. Take a screen shot of the graph by pressing "Prt Sc." The screen shot will automatically be
saved into the "Screenshots" folder on the Desktop. Open the folder and change the file
name to the subject's ID.
9. Record the elapsed time for the test on the recording sheet.

- 20 -
APPENDIX B

- 21 -
 APPENDIX C

This is a non-exhausted list of diseases and medications that may affected color vision and were
thus excluded from this study.

Exclusionary Diseases
Alzheimer's disease
Cataracts
Diabetes mellitus
Glaucoma
Leukemia
Multiple sclerosis
Macular degeneration
Parkinson's disease
Sickle cell disease
Thyroid eye disease

Exclusionary Medications
Chloroquine
Digoxin
Ethambutol
Hydroxychloroquine
Phenytoin
Sildenafil
Tamoxifen

- 22 -
 Rachel Murphy Curriculum Vitae
4725 SW 139th Ave • Beaverton, OR 97005
706-889-3252 (cell) • rachel.murphy@pacificu.edu

Education:
Candidate for Doctor of Optometry, Pacific University College of Optometry, Forest Grove, OR
To be conferred May 2016

Candidate for Master of Science in Vision Science, Pacific University College of Optometry, Forest Grove, OR
To be conferred May 2016

Bachelor of Science in Biology, Berry College, Mount Berry, GA • May 2010

Minors: Chemistry & Business Administration

Honors and Awards:

Magna cum Laude, Berry College, Mount Berry, GA • May 2010
Dean’s List, each semester of undergraduate studies, Berry College, Mount Berry, GA

Professional Licenses and Certifications:

Basic Life Support for Health Care Providers • SCI8afbb09e3218 • April 28, 2014
HIPAA Certification • MURP241318ESL1960 • September 15, 2012
American Board of Opticianry Certification • #188545 • 2011 – 2014
National Contact Lens Examiners Certification • #188545 • 2011 – 2014

Work Experience:
Vision Performance Institute, Forest Grove, OR May 2013 – July2013
Research Assistant
 Programmed computer activities for research participants using Experiment Builder
 Edited macros for collecting data into a single excel sheet

Jackson Eye Care, Fort Collins, CO February, 2011 – June, 2012

Optometric Technician and Receptionist
 Attended to each patient’s experience including check-in, pretesting, and payment
 Performed contact lens education and handled phone calls and scheduling

Wal-Mart Vision Center, Fort Collins, CO February, 2011 – June, 2012

Dual Certified Optician (ABO & NCLEC)
 Attended to customer’s needs to include sales and eyewear adjustments
 Worked as a team performing daily tasks to run the vision center

Shades of Jackson Hole, Jackson, WY May – August, 2009 & July – November, 2010
Sales Associate/Assistant Manager
 Sold and repaired sunglasses, ordered and managed inventory, and used QuickBooks
 Communicated professionally with customers and interacted effectively with peers and
superiors

Undergraduate Business & Economics Research Journal, Berry College, GA September, 2009 – May, 2010
Copy Editor
 Corrected APA formatting and language before publishing
 Developed excellent written communication skill

- 23 -
 Rachel Murphy Curriculum Vitae

Clinical Experience:
Eye Care Associates of Nevada • Dr. Douglas Devries, Sparks, NV February 22 – May 6, 2016
Koenig & Robertson Optometry • Dr. Gregory Koenig, Fallon, NV November 16 – February 12, 2016
Veterans Affairs Medical Center • Dr. Kirk Halvorson, Salem WA August 24 – November 6, 2015
Oak Harbor Naval Hospital • Dr. Kyle Dohm, Oak Harbor, WA May 18 – August 8, 2015
Pacific EyeClinic • Primary Care • Dr. Susan Littlefield, Cornelius, OR January 5 – April 24, 2015
Pacific EyeClinic • Vision Therapy • Dr. Hannu Laukkanen, Forest Grove, OR January 5 – April 24, 2015
Pacific EyeClinic • Primary Care • Dr. Scott Pike, Forest Grove, OR August 25 – December 16, 2014
Pacific EyeClinic • Ocular Disease • Dr. Ryan Bulson, Portland, OR August 25 – December 16, 2014
Pacific EyeClinic • Primary Care • Dr. Kirk Halvorson, Hillsboro, OR May 5 – August 8, 2014
Pacific EyeClinic • Contact Lens • Dr. Scott Overton, Portland, OR May 5 – August 8, 2014

Research Experience:
Research Assistant, Vision Performance Institute, Forest Grove, OR May, 2013 – August, 2013
 Conducted human subject research
 Designed and programmed computer activities for testing subjects using Experiment Builder

Biology Research Assistant, Berry College, Mount Berry, GA January, 2007 – May, 2008
 Conducted online research, collected field data, and processed images using Image J
 Planned, strategized, and implemented research questions; utilized problem solving skills

Research Presentations:
Master’s Degree Thesis Defense, Pacific University, Forest Grove, OR April 23, 2015
 “A New Generation of Color Testing”

Student Research Symposium, Berry College, Mount Berry, GA April 11, 2008
 “Forest Health, Dogwoods, and the Calcium Connection”

Teaching Experience:
Master’s Seminar Tutor, Pacific University, Forest Grove, OR August, 2013 – May, 2015
 Tutored Master of Science in Vision Science candidates

Chemistry Tutor, Periscope Tutoring, Forest Grove, OR May, 2013 – August, 2013
 Tutored a high school student preparing for final exams

Biology Tutor, Berry College Biology Department, Mount Berry, GA August, 2007 – May, 2010
 Helped fellow students develop personal study skills and improved my own teaching skills

Horseback Riding Instructor, Self Employed, Rome, GA September, 2008 – May, 2010
 Learned to clearly and effectively share my knowledge while maintaining a leadership role

- 24 -
 Rachel Murphy Curriculum Vitae
Conferences and Workshops:
Optometry’s Meeting, Seattle, WA June 24-28, 2015
American Academy of Optometry, Denver, CO November 12-15, 2014
Multifocal Contact Lens Workshop, Forest Grove, OR November 5, 2014
Valley Contax Workshop, Springfield, OR July 12, 2014
Toric Soft Lens Workshop, Forest Grove, OR May 21, 2014
American Academy of Optometry, Seattle, WA October 23-26, 2013
Great Western Council of Optometry, Portland, OR September 21, 2013
Northern Rockies Optometric Conference, Jackson, WY July 25-27, 2013
Optometry’s Meeting, San Diego, CA June 26-29, 2013
Great Western Council of Optometry, Portland, OR September 29, 2012

Community Vision Screenings:

Harvey Clark Elementary School, Cornelius, OR February 2, 2015
Project Homeless Connect, Hillsboro, OR January 30, 2015
Migrant Head Start: Linden, Cornelius, OR October 31, 2014
Pacific University Baseball Team, Forest Grove, OR October 15, 2014
Migrant Head Start: Linden, Cornelius, OR June 18, 2014
Cornelius Elementary School, Cornelius, OR November 22, 2013
Atlas, Yamhill, OR September 5, 2013
VSP Mobile Clinic, Virginia Garcia Vision Clinic, Cornelius, OR January 14, 2013
Tigard Compassion, Tigard, OR October 13, 2012

- 25 -