Specifying colours for colour vision testing using computer graphics.

Numerical Confusion Errors in Ishihara Testing: Findings From a Population-based Study

Daily experience shows that colour of a very distant object cannot be accurately determined. It is assumed that visual acuity (VA) loss is one of the factors at play in this case. The effects of reduced VA as a result of refractive error or optical defocus on colour vision have not been examined. Such study willdictate the need or otherwise for optical correction before assessment of colour vision. The purpose of this study therefore, was to investigate the effects of optical defocus on colour vision in individuals with normal colour vision. Twenty nine young adult subjects (11 male and 18 females) were included in this study. Their ages ranged from 11 to 29 years with a mean of 22.1 ± 3.4 years. All subjects had VA of 6/6 or better and normal colour vision. The colour vision was evaluated with the Farnsworth panel D-15 (desaturated). Each subject was optically defocused to VA of 6/24, 6/60 and 1/60(6/360) at 6 meters respectively and colour vision was assessed at each reduced VA. At VA of 6/24, colour vision was not affected in all subjects. When the VA was reduced to 6/60, however, 15 (51%) of the subjects failed the colour vision test. When the VA was further reduced to 1/60 by optical defocus, 24 (83%) of the subjects failed the colour vision test. It was concluded that optical defocus and reduced VA can adversely affect colour vision test findings. While VA reduction to 6/24 may not affect the colour vision results, VA of 6/60 can. It is therefore, recommended that in clinical colour vision testing and colour vision screening, VA should first be assessed andcompensated, especially if uncorrected VA is 6/24 or worse.

Color vision testing can be accomplished by several means, each of which has its virtues and limitations. The most common test of color perception is the pseudoisochromatic plate. This is a card composed of a field of dots of varying hue and saturation within which persons with normal color vision can see some figure, but persons with a color vision defect cannot. Alternatively, some pseudoisochromatic plates are constructed so that color normals see one figure and color defectives another. These tests rely upon either the reduced hue discrimination or the reduced saturation discrimination of color defectives as a means of identification of these color defectives. Unfortunately, the illuminant under which the test is conducted is critical, and the diagnostic value of this type of test is limited.

Color vision research is not new for the Federal Aviation Administration (FAA); the Civil Aerospace Medical Institute has been conducting color vision research and publishing the results since 1967 ( 3 ). The FAA originally initiated color vision research because of the emerging use of color coding in the airport environment and the FAA has continued a line of color vision research because of the increasing use of color coding resulting from changing technology inside the cockpit, on air traffic control displays, and in the airport environment. Color can be used to convey meaning without supplemental signage such as the ubiquitous traffic signal that alerts drivers to proceed with caution via a yellow flashing light or to stop via a red flashing light. However, that meaning is only conveyed if the driver can distinguish between the yellow and the red colors. Approximately 8 to 10% of the male population ( 5 ) has a congenital color vision deficiency and, depending upon the type and severity of that deficiency, that task of interpreting the meaning of color coding may be difficult or impossible. Consequently, the FAA has long maintained a color vision standard for aero-medical screening to ensure that pilots and air traffic controllers can perform safety-related tasks without adverse consequences. Throughout the past few years, the FAA has explored a variety of color vision tests, searching for a valid screening test that has high sensitivity and specificity, meaning the ability to detect the presence or absence of the deficiency, respectively. Basically, color vision tests can be categorized as diagnostic, screening, or occupational tests. Diagnostic tests are designed to specifically diagnose the type and degree of deficiency, the screening tests focus on differentiating between normal or deficient color vision, and the occupational tests seek to separate those capable versus incapable of certain tasks such as identifying colors of wires or lights (e.g., the Farnsworth Lantern test that was developed to assess the ability of potential Navy signalmen for identifying red, green, and white lights). A few tests have been developed for the purpose of precisely diagnosing and classifying color vision; however, when color vision test scores are compared to performance on occupational tasks such as identifying or discriminating colors used in signal lights, precision approach path indicator (PAPI) lights, colored navigation lights, color coded map reading tasks, color coded air traffic control displays, and cockpit displays, a specific cut-point on those selection tests has not been found that can fully separate those who can from those who cannot accurately perform the color-coded pilot or air traffic control tasks. Some tests, including new computerized instruments, have been designed to differentiate defects involving the long wavelength sensitive cones (protan-type), middle wavelength sensitive cones (deutan-type), and short wavelength sensitive cones (tritan-type). Congenital protan and deutan deficiencies are, collectively, extremely common, affecting 1 in 12 men and 1 in 230 women; however, recent evidence indicates that tritan defects are virtually never present at birth (e.g., congenital) and the inherited forms involve S cone photoreceptor degeneration that develops later in life with the exact onset depending upon the specific mutation ( 1, 4 ). Thus, the exact frequency of inherited tritan defects is uncertain; however, it is probably less than 1 in 500. In part, because the underlying pathophysiology has not been well understood, few tests have been available that are capable of detecting tritan deficiencies. In the past, those tests included the single Farnsworth F2 pseudoisochromatic plate (PIP), the Moreland anomaloscope, the Hardy, Rand, Rittler PIP test, and, most recently, the Oculus anomaloscope. Consequently, the occupational color vision tests used by most agencies only screen for the most common (protan and deutan) types of defects. The newly developed computerized color vision tests, including the Colour Assessment and Diagnostic Test, the Cambridge Colour Test, the Cone Contrast Test, and the Computerized Color Vision Test, are all designed to detect tritan defects. However, tritan weaknesses have been noted in several of the FAA ‘ s recent studies in much higher than the traditionally expected numbers and diagnostic agreement is low among those tests when tritan deficiencies are involved. In the past, the FAA and other regulatory organizations have not, or have rarely, required tritan color vision screening in their occupational screening because of the following three factors: the rarity of the congenital defect, the unknown number of individuals affected by acquired deficiencies, and the lack of effective, reliable, valid, and affordable equipment with which to diagnose the deficiency.

Purpose:To investigate the quality of life (QoL) in a sample of color vision deficit (CVD) patients in India and how color vision deficiency affects them psychologically, economically, and in productivity related to their work and occupation.Methods:A descriptive and case–control study design using a questionnaire was conducted on N = 120 participants, of whom 60 were patients of CVD (52 male and eight female) who visited two eye facilities in Hyderabad between 2020 and 2021 and 60 were age-matched normal color vision participants who served as controls. We validated English–Telugu adapted version of CVD-QoL, developed by Barry et al. in 2017 (CB-QoL). The CVD-QoL consists of 27 Likert-scale items with factors (lifestyle, emotions, and work). Color vision was assessed using the Ishihara and Cambridge Mollen color vision tests. A six-point Likert scale was used, with lower scores indicating poor QoL (from 1 = severe issue to 6 = no problem).Results:The CVD-QoL questionnaire’s reliability and internal consistency were measured, including Cronbach’s α (α =0.70–0.90). There was no significance between the group in age (t = −1.2, P = 0.67) whereas the Ishihara colour vision test, scores showed a significant difference (t = 4.50, P < 0.001). The QoL scores showed a significant difference towards lifestyle, emotions and work (P = 0.001). The CVD group had a poorer QoL score than the normal color vision group odds ratio [OR] =0.31, 95% confidence interval [CI], (P = 0.002, CI = 0.14–0.65, Z = 3.0) . In this analysis, a low CI indicated that the OR was more precise.Conclusion:Color vision deficiency affects Indians’ QoL, according to this study. The mean scores of lifestyle, emotions, and work were lower than the UK sample.Since CVD is underreported and possibly affects developing countries more, advocacy for a new health care plan on CVD is essential. Increasing public understanding and awareness could also help diagnosing the CVD population.

Aims: To investigate whether there is an association between congenital colour vision defects (CVD) and occupational choice and employment history, in order to inform the debate about the value of...

Normal human color vision is usually thought to be based on only three spectrally different cone types. However, two facts suggest the possibility that some color-normal males could have more than three cone pigment types: (1) Most people with normal color vision have more than two photopigment genes on each X-chromosome and (2) there appear to be genetically specified variations in spectral positions of the normal middle-wavelength-sensitive (MWS) and long-wavelength-sensitive (LWS) pigments. For example, a male might have one gene encoding an LWS pigment and two genes encoding slightly different MWS pigments. If all three different X-encoded genes were expressed in different cones, then this person would have four spectrally different cone types. How firm is the assumption that more than two of the X-encoded pigment genes can be expressed? Both analysis of the statistics of photopigment gene number among different color vision phenotypes and analysis of the arrangement of pigment genes on the X-chromosome yield insight into this aspect of photopigment gene expression. These analyses suggest that individuals with multiple pigment genes on the X-chromosome may express more than two of those genes.

The X-linked red- and green-pigment genes are arranged in a head-to-tail tandem array. The colour-vision defect of deuteranomaly (in 5% of males of European descent) is associated with a 5'-green-red-3' visual-pigment hybrid gene, which may also exist in males with normal colour vision. To explain why males with a normal red, a normal green and a green-red hybrid gene may have either normal or deutan colour vision, we hypothesized that only the first two genes are expressed and deuteranomaly results only if the green-red hybrid gene occupies the second position and is expressed preferentially over normal green-pigment genes occupying more distal positions. We used long-range PCR amplification and studied 10 deutan males (8 deuteranomalous and 2 deuteranopic) with 3 visual pigment genes (red, green and green-red hybrid) to investigate whether position of the hybrid gene in the array determined gene expression. The green-red hybrid gene was always at the second position (and the first position was always occupied by the red gene). Conversely, in two men with red, green and green-red hybrid genes and normal colour vision, the hybrid gene occupied the third position. When pigment gene mRNA expression was assessed in post-mortem retinae of three men with the red, green and green-red genotype, the green-red hybrid gene was expressed only when located in the second position. We conclude that the green-red hybrid gene will only cause deutan defects when it occupies the second position of the pigment gene array.

In some countries there is controversy as to whether normal or near normal colour vision is necessary for pilots of aircraft. It has been argued that information displayed by aviation colour codes is always displayed redundantly and that in modern aviation there is less dependence on information acquired visually from outside the aircraft. One aviation signal system that displays information solely by colour coding is the Precision Approach Path Indicator (PAPI) which signals that an approach to land is too high or too low by means of the number of red and white signals. A Farnsworth lantern was modified to simulate PAPI signals viewed from one nautical mile under conditions close to 1.5 km visibility. Normal observers were able to identify the red and white signals and distinguish them from blank signals without difficulty except that some subjects made a few errors confusing a yellowish white (simulating the colour of a white PAPI signal attenuated by the atmosphere) with red. Most colour vision defective observers made errors confusing red and white, and protan observers frequently failed to see the red signal, calling it a blank. Although a few protans could identify the signal colours without error it is concluded that the majority of colour vision defective observers are not able to identify a red/white signal system reliably despite the code being a simple two category code and the fact that brightness differences between the red and white signals were an available cue. The results are surprising since past reports suggest that colour vision defectives are able to recognise red with only a few errors so that a two colour red/white code might have been expected to be within their capacity. The number of red = white and white = red errors made by deutans was also surprising.

Color vision tests use estimative of threshold color discrimination or number of correct responses to evaluate performance in chromatic discrimination tasks. Both approaches have advantages and disadvantages. In the present investigation, we compared the number of errors during color discrimination task in normal trichromats and participants with color vision deficiency (CVD) using pseudoisochromatic stimuli at fixed saturation levels. We recruited 28 normal trichromats and eight participants with CVD. Cambridge Color Test was used to categorize their color vision phenotype, and those with a phenotype suggestive of color vision deficiency had their L- and M-opsin genes genotyped. Pseudoisochromatic stimuli were shown with target chromaticity in 20 vectors radiating from the background chromaticity and saturation of 0.06, 0.04, 0.03, 0.02, 0.01, and 0.005 u’v' units. Each stimulus condition appeared in four trials. The number of errors for each stimulus condition was considered an indicator of the participant's performance. At high chromatic saturation, there were fewer errors from both phenotypes. The errors of the normal trichromats had no systematic variation for high saturated stimuli, but below 0.02 u’v' units, there was a discrete prevalence of tritan errors. For participants with CVD, the errors happened mainly in red-green chromatic vectors. A three-way ANOVA showed that all factors (color vision phenotype, stimulus saturation, and chromatic vector) had statistically significant effects on the number of errors and that stimulus saturation was the most important main effect. ROC analysis indicated that the performance of the fixed saturation levels to identify CVD was better between 0.02 and 0.06 u’v’ units reaching 100%, while saturation of 0.01 and 0.005 u’v’ units decreased the accuracy of the screening of the test. We concluded that the color discrimination task using high saturated stimuli separated normal trichromats and participants with red-green color vision deficiencies with high performance, which can be considered a promising method for new color vision tests based in frequency of errors.

There are two currently available lantern tests that have their design based solely on the practices of the rail industry. These are the CN Lantern (CNLAN) used on the Canadian Railways and the Railway LED Lantern Tests (RLLT). In the same way that the signaling practices differ in the two jurisdictions, the design of the lanterns also differ. The CIE (Commission Internationale de l'Éclairage)-recommended color vision standards for transport predate both lanterns. The study was undertaken to assess the appropriateness of these lanterns in CIE Color Vision Standard 1. CIE Standard 1 is called "Normal color vision" but some very mild anomalous trichromats may pass the specified "lantern test that presents a high level of difficulty." The lantern tests were undertaken by 46 color vision-normal and 37 color vision-deficient subjects. Subjects made more errors on the RLLT, and the pattern of errors is different, partly because there are blank presentations in the RLLT and "no light" is an acceptable response. The two lanterns showed agreement on the pass/fail categorization of 73 of the 83 subjects. The RLLT fails more color vision-normal subjects. Despite the different construction principles, the RLLT and CNLAN have pass/fail levels that are comparable with the Holmes-Wright Type B lantern, which is nominated in CIE Color Vision Standard 1 but is no longer commercially available. The higher failure rate of color vision-normal subjects on the RLLT has been addressed by changing the intensities of the two darkest red lights.

Current color vision (CV) tests used for aviation screening in the U.S. Army only provide pass-fail results, and previous studies have shown variable sensitivity and specificity. The purpose of this study was to evaluate seven CV tests to determine an optimal CV test screener that potentially could be implemented by the U.S. Army. There were 133 subjects [65 Color Vision Deficits (CVD), 68 Color Vision Normal (CVN)] who performed all of the tests in one setting. CVD and CVN determination was initially assessed with the Oculus anomaloscope. Each test was administered monocularly and according to the test protocol. The main outcome measures were test sensitivity, specificity, and administration time (automated tests). Three of the four Pseudoisochromatic Plate (PIP) tests had a sensitivity/specificity > 0.90 OD/OS, whereas the FALANT tests had a sensitivity/specificity > 0.80 OD/OS. The Cone Contrast Test (CCT) demonstrated sensitivity/specificity > 0.90 OD/OS, whereas the Color Assessment and Diagnosis (CAD) test demonstrated sensitivity/specificity > 0.85 OD/OS. Comparison with the anomaloscope ("gold standard") revealed no significant difference of sensitivity and specificity OD/OS with the CCT, Dvorine PIP, and PIPC tests. Finally, the CCT administration time was significantly faster than the CAD test. The current U.S. Army CV screening tests demonstrated good sensitivity and specificity, as did the automated tests. In addition, some current PIP tests (Dvorine, PIPC), and the CCT performed no worse statistically than the anomaloscope with regard to sensitivity/specificity. The CCT letter presentation is randomized and results would not be confounded by potential memorization, or fading, of book plates.

Background: Colour vision assessment correlates with damage of the visual pathway and might be informative of overall brain damage in multiple sclerosis (MS). Objective: The objective of this paper is to investigate the association between impaired colour vision and disease severity. Methods: We performed neurological and ophthalmic examinations, as well as magnetic resonance imaging (MRI) and optical coherence tomography (OCT) analyses, on 108 MS patients, both at baseline and after a follow-up of one year. Colour vision was evaluated by Hardy, Rand and Rittler plates. Dyschromatopsia was defined if colour vision was impaired in either eye, except for participants with optic neuritis (ON), for whom only the unaffected eye was considered. We used general linear models adjusted for sex, age, disease duration and MS treatment for comparing presence of dyschromatopsia and disease severity. Results: Impaired colour vision in non-ON eyes was detected in 21 out of 108 patients at baseline. At baseline, patients with dyschromatopsia had lower Multiple Sclerosis Functional Composite (MSFC) scores and Brief Repeatable Battery-Neuropsychology executive function scores than those participants with normal colour vision. In addition, these patients had thinner retinal nerve fiber layer (RNFL), and smaller macular volume, normalized brain volume and normalized gray matter volume (NGMV) at baseline. Moreover, participants with incident dyschromatopsia after one-year follow-up had a greater disability measured by the Expanded Disability Status Scale and MSFC-20 and a greater decrease in NGMV than participants with normal colour vision. Conclusions: Colour vision impairment is associated with greater MS severity.

Background: The Farnsworth D15 test is designed to categorise colour vision deficiency as severe or moderate. The level of difficulty of the test was set so that those who passed it should be able to recognise surface colour codes, such as those used for electrical wiring. The test is widely used to provide advice to patients with abnormal colour vision and is often used for occupational selection when reliable recognition of surface colour codes is required. However, there has been only one previous study of the correlation between performance at the D15 test and the naming of surface colour codes and there has been no study of whether a person who passes the D15 can reliably name surface colours. Methods: One hundred and two people aged 11 to 65 years with abnormal colour vision were recruited from consecutively presenting optometric patients and were asked to name the colours of fabric, paint and cotton thread samples. There were 10 colours in each class of material and the samples were presented in a large (five to 10 degree angular subtense) and small size (2.5 deg and a single thread). The errors made were compared to those made by an age‐matched control group of equal size with normal colour vision. Results: The correlations between the Farnsworth D15 colour confusion index and colour naming errors were 0.62 for the large stimuli and 0.73 for the small stimuli. Its sensitivity and specificity identifymg those who made more errors than the worst performing colour normal person were 0.80 and 0.69 (large stimuli) and 0.75 and 0.71 (small stimuli). A Nagel anomaloscope range of less than 35 scale units provides essentially the same sensitivity and specificity. Conclusions: About 40 per cent of those with abnormal colour vision can name the main colours correctly under good visibility conditions. The D15 test is an imperfect predictor of those who can name surface colour codes correctly but it does provide useful information for general counselling. It is not suitable as a single test for occupational selection because it will pass 20 per cent who cannot name surface colours correctly and fail 30 per cent who can. In occupations in which recognition of surface colour codes is of critical importance, it may be best not to select people with abnormal colour vision because of the lack of a colour vision test that is a perfect predictor of the ability to recognise surface colours.

Polymorphism in the number of genes encoding long-wavelength-sensitive cone pigments among males with normal color vision