Original Research

Evaluation of the performance of the Waggoner computerised colour vision test

Ali Almustanyir

Received: 10 Dec. 2024; Accepted: 18 Aug. 2025; Published: 30 Oct. 2025

Copyright: © 2025. The Author. Licensee: AOSIS.
This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license (https://creativecommons.org/licenses/by/4.0/).

Abstract

Background: Individuals with colour vision impairments have a significant probability of misjudging colour. This high risk of colour-related inaccuracies is crucial for evaluating the ability of an individual to distinguish between different hues. The Waggoner computerised colour vision test (W-CCVT) is a novel colour test designed to detect colour vision deficiencies.

Aim: This study aimed to validate the performance of the W-CCVT relative to that of the anomaloscope and Ishihara tests.

Setting: This study evaluated the W-CCVT relative to standard colour vision tests by recruiting 52 participants with colour-normal vision and 135 with colour vision deficiencies from different locations in the capital city of Saudi Arabia.

Methods: The performance of the W-CCVT was compared with that of the Ishihara test and the Heidelberg Multi-Colour Oculus anomaloscope. Participants were directed to identify the numbers on the Ishihara plates, and their responses were documented on a recording sheet.

Results: Concerning pass or fail agreement, good agreement was observed between the W-CCVT and anomaloscope, with a first-order agreement coefficient of 0.97. The sensitivity value was 97.7% and 98% value for the specificity.

Conclusion: The Waggoner computerised colour vision test may serve as a reliable alternative screening tool for detecting colour vision deficiencies.

Contribution: The W-CCVT could be an appropriate screener colour vision test and a suitable alternative to the Ishihara test when implemented on an iPad.

Keywords: colour vision; colour vision deficiency; Ishihara; Waggoner; iPad; colour vision testing.

Introduction

For the past 50 years, clinicians have employed various colour vision assessments, including pseudoisochromatic plate (PIP) tests like the Ishihara and Hardy–Rand–Rittler (HRR) tests, the Farnsworth D15 (D15) test and the Heidelberg multi-colour oculus anomaloscope, to identify colour vision deficiency (CVD). In certain situations, these standard methods are supplemented with additional assessments such as the Farnsworth–Munsell 100 hue test, desaturated D15 versions and the Medmont C100.^1,2 PIP tests are used to screen for CVD, whereas D15 provides diagnostic information, such as identifying protan or deutan, and assessing severity.^1,2 Among these, the anomaloscope is not frequently used in clinical settings because it is expensive and time consuming. However, the anomaloscope is considered the gold standard for determining the type and severity of red–green defects.³

The healthcare industry is adopting several health tests and digitised products in an increasingly digitalised world. These tests can be performed using various digital devices, including personal computer screens, smartphones and tablets.^{4,5,6,7,8,9,10,11} Better healthcare outcomes can be achieved at lower costs using automated diagnostic methods and digitised tests. These digitised tests offer several advantages, including lower costs by eliminating the need for physical availability, time saving for clinicians, provision of a broad variety of colours, reduction of administrator biases, seamless integration with electronic medical records and prevention of colour plate ageing (for colour vision tests). Despite these benefits, digital testing of colour vision presents challenges, as different digital devices exhibit variations in display size, calibrations and screen analysis.^{4,5,6,7,8,9,10,11} Hence, creating a precise computerised colour vision test for commercial and public use is challenging, as the test must consistently display specific colours across various computers and types of display hardware.

Several computerised colour vision tests are available and can be applied to different digital devices. The Ishihara test is the most common colour vision screening test for red–green colour defects and provides information regarding the defect type (protan or deutan).¹ The Ishihara test is available in an electronic version and has been implemented using smartphones¹² and PC displays.¹³ Similarly, the Colour Vision Suit (Waggoner Diagnostics) is a colour vision test designed as a screener test to detect the presence of CVD. Furthermore, this test can also diagnose the type of defect, such as deutan, protan or tritan. The Waggoner computerised colour vision test (W-CCVT) screens for red–green and blue–yellow CVDs, classifies the type of red–green defect, and diagnoses the severity of the defect. The design of the test plates is similar to that used in the Ishihara test. Figure 1 illustrates the W-CCVT test. The W-CCVT test involves two main parts. Initially, a series of 25 stimuli is used to screen for red–green colour vision defects, with the screening phase concluding if the subject makes five errors. Upon completion or termination of this phase, the test moves on to the blue–yellow series. This is followed by a tritan screening and classification phase, which utilises 12 specialised plates. After completing the blue–yellow series, the test proceeds to a more detailed red–green diagnostic assessment. This comprehensive approach allows for a thorough evaluation of various aspects of colour vision, focusing on both red–green and blue–yellow colour perception abilities. If the individual fails the red–green screening plate test, the red–green diagnostic plates are administered. Half of the diagnostic plates display figure colours that are likely missed by individuals with protan vision, while the other half targets deutan vision. Individuals who fail to detect more saturated colours are classified as having more severe defects. Almustanyir and Hovis used a prototype of the W-CCVT on a desktop monitor and determined that the sensitivity and specificity of the test against the anomaloscope were 0.96 and 0.99, respectively.¹⁴

FIGURE 1: Screen shots of the Waggoner computerised colour vision test. The top figure represents an example of the screening (red–green) plates, while the bottom figure shows an example of the tritan plates. Letter ‘C’ on the tablet enables the participant to delete the selected option, and ‘N’ is the response when no number can be detected.

Digital colour vision tests are considered sufficient for the screening and diagnosis of individuals with dichromacy and anomalous trichromacy.^13,15 Rapid changes in relevant programmes and operating systems necessitate continued colour testing to confirm these findings. Consequently, this study aimed to evaluate the performance of W-CCVT on an iPad relative to the anomaloscope and Ishihara colour vision tests. This study can contribute directly to digitalising the clinical setting, and the W-CCVT on an iPad could serve as a potential substitute for conventional colour vision tests used in CVD screening.

Methods

Fifty-two participants with colour-normal vision (CNV) and 135 with red–green CVD were recruited from hospitals, schools, universities and public places in Riyadh, the capital city of Saudi Arabia. Colour vision deficiency was classified based on Rayleigh colour matching using the Heidelberg Multi-Colour (HMC) Oculus anomaloscope (Oculus Optikgeräte GmbH, Wetzlar, Germany). The right eye was examined using the anomaloscope. The research protocol adhered to the ethical guidelines outlined in the Declaration of Helsinki and received approval from the Office of Research Ethics at King Saud University’s College of Applied Medical Science (CAMS-046-3940).

Inclusion/exclusion

The study participants exhibited normal visual systems, demonstrating a binocular distance visual acuity of at least 6/9. A short questionnaire was used to exclude individuals with known acquired CVD. All the participants reported being free from ocular or systemic diseases. Although the participants were permitted to use their regular distance or near-vision correction devices during testing, the use of tinted spectacles or contact lenses was not allowed.

Testing procedure

Participants were tested with the screening plates. These plates aimed to detect the presence of CVD or not (normal colour vision vs CVD). The tests administered included the Ishihara 38 plate edition (Kanehara Trading INC, 2011 [Tokyo, Japan]), the W-CCVT and the HMC Oculus anomaloscope, with the sequence of testing determined by a predetermined random block design. Participants were asked to read the numbers presented on the Ishihara screening plates, and their answers were recorded on a response sheet. All tests were conducted at an approximate viewing distance of 50 cm. For the Ishihara screening plates (numbers 1–17), a failure was defined as making three or more errors, where an error was considered either misreading or failing to identify the correct number. The study utilised a light-emitting diode (LED) lamp (NEW POWER, Model G45-14 E27) with a correlated colour temperature of 6500 K to illuminate the test room. The testing area, consisting of a white table, was illuminated to 800 lux (± 5%) in the horizontal plane. This illumination setup ensured standardised conditions for colour vision assessment.

The W-CCVT was administered using a 10-inch Apple iPad (MQDT2AB/A) running iOS 13.4.1. Each test stimulus was displayed for 2 s, and participants were granted 12 s to input their responses using either a keyboard or touch screen, entering the number they perceived or ‘N’ if no figure was visible. The participant viewed the test binocularly. The testing sequence began with red–green screening plates (detecting if the participant has CVD or not), followed by tritan test plates. In cases where participants failed the red–green screening, the blue–yellow test plate series was followed by the red–green diagnostic series. Red–green screening figures were presented in random order, while blue–yellow screening and diagnosis figures and red–green diagnostic plates started with the most saturated colours, progressively decreasing in saturation. Testing was conducted in a dim room (1 lx), with participants seated approximately 50 cm from the device. The criteria for failure were defined as making three or more errors on the blue–yellow screening plates or five errors on the red–green screening plates. The type and severity of red–green colour vision deficiencies were assessed based on the highest number of errors recorded on the protan and deutan diagnostic plates, while the severity of blue–yellow defects was determined by the total number of errors made during the screening. Twenty-eight participants with CVD completed the test twice, with a 1-week interval between sessions.

The screening is usually for the rapid identification of individuals who may have colour vision deficiencies or not. However, the diagnostic is for in-depth analysis, specifying the type and severity (protan vs deutan) and is used for clinical, occupational or regulatory decisions. Both modes are included in the WCCVT suite, allowing users to select the appropriate test based on their needs.

Analysis

The first-order agreement coefficients (AC1) were calculated between the Ishihara and W-CCVT and between the anomaloscope and the W-CCVT.^16,17 The AC1 agreement index employs a particular approach to adjust for chance agreement compared to the κ coefficient established by the Working Group. In particular, AC1 values align more closely with raw percentage agreement, particularly in datasets with imbalanced marginal distributions. In scenarios where in few subjects pass either test and most fail both tests, both the percentage of agreement and the AC1 coefficient are expected to be high owing to the large number of concordant failures. However, κ can yield a significantly lower value. This discrepancy arises because the κ coefficient is more sensitive to a small number of subjects who pass either test, thereby assigning disproportionate weight to rare instances of disagreement between tests. Consequently, while percentage agreement and AC1 reflect the overall concordance, κ provides a more nuanced view that accounts for the rarity of positive outcomes. The AC1 coefficient, which ranges from −1 to 1, is used to assess agreement – where 1 represents perfect agreement, 0 indicates agreement by chance and –1 signifies complete disagreement.^16,17 Agreement analyses were performed using AgreeStat software (version 2013.2; Advanced Analytics, Gaithersburg, MD, United States). The Ishihara test and the anomaloscope served as the gold-standard references for comparison. Sensitivity was defined as the proportion of participants who failed both the Ishihara and anomaloscope and W-CCVT tests, while specificity referred to the proportion who passed both assessments. This methodology was used to validate the diagnostic accuracy of the W-CCVT in relation to established colour vision testing methods. Furthermore, sample size calculation for the repeatability analysis was performed according to standard formulas for test-retest reliability. Assuming an expected reliability coefficient of 0.85, a minimum acceptable value of 0.60, a significance level of 0.05 and 80% power, a minimum of 10 participants was required. Therefore, including 28 individuals in the repeat sample provides sufficient power to assess test-retest reliability.

Ethical considerations

An application for full ethical approval was made to the King Saud University Office of Research Ethics and ethics consent was received on 12 December 2018. The ethics approval number is CAMS-046-3940.

Results

Red–green defect

Among the participants, 55 (41%) were identified with deuteranomaly, 35 (26%) with deuteranopia, 23 (17%) with protanomaly and 21 (15%) with protanopia. Among the CNV participants, 55% were women and 45% were men. Moreover, among the 135 participants with red–green CVD, 98.5% were men and 1.5% were women. This difference in the proportions of men and women in the two groups is attributed to the X-linked recessive inheritance of red–green CVDs. The average ages of participants in the CVD and CNV groups were 24.9 ± 9.4 years and 24.6 ± 7.1 years, respectively. The age of the CVD group ranged from 12 to 50 years, whereas that of CNV ranged from 15 to 60.

Screening plates

Using the suggested standard of five or more mistakes in the red–green screening series, the majority of participants with colour vision deficiencies did not pass the W-CCVT red–green screening series during the first session. Three participants with deuteranomaly passed their first session. Additionally, three participants with deuteranomaly passed the test in the second session.

Table 1 compares the outcomes of the W-CCVT with those from the anomaloscope and Ishihara tests administered in the initial session. A strong concordance was noted, evidenced by an AC1 coefficient of 0.97 (standard error: ± 0.018). Sensitivity and specificity measures compared to the anomaloscope and Ishihara tests matched exactly, at 97.7% and 98%, respectively.

All participants in the CVN group and only 21% of those in the CVD group returned for the second visit. Table 2 presents the repeatability results for red–green screening plates of the W-CCVT. The test demonstrated high consistency, with an AC1 coefficient of 0.95 (standard error: ± 0.031), indicating strong agreement between sessions. The slight variation observed is attributed to two participants with deuteranomaly who failed the test in the initial session but passed it in the follow-up session.

Diagnostic plates

These plates aimed to detect the type of CVD (protan vs deutan). According to the suggested criteria, the W-CCVT diagnostic plates (protan versus deutan) showed strong classification agreement with the anomaloscope, yielding an AC1 coefficient of 0.84 (standard error: ± 0.031). The results are listed in Table 3. Four participants with deuteranomaly were classified as protan deficient using the W-CCVT. In addition, one participant with protanopia and seven with protanomaly were classified as deutan deficient by the W-CCVT.

Assessing the severity

Figure 2 shows the association between the W-CCVT severity classifications and the corresponding anomaloscope matching ranges. Higher W-CCVT severity grades correlated with wider anomaloscope-matching ranges. The wider anomaloscope-matching ranges mean worse colour discrimination. This means that the matching range value of 70 indicates severe CVD, whereas values below 10 indicate mild CVD. The Spearman rank correlation analysis revealed a moderate but significant relationship (ρ = 0.48, p = 0.0017). Although statistically significant, the correlation results were not perfect, possibly because 8% of participants with dichromacy (3% with deuteranopia and 27% with protanopia) were classified as mild or moderate in the W-CCVT. Moreover, two individuals with deuteranomaly exhibited unexpectedly narrow anomaloscope matching ranges (< 20) but were categorised as severe by the W-CCVT test.

FIGURE 2: Relationship between the severity classifications from the Waggoner computerised colour vision test and the anomaloscope matching ranges among participants with colour vision deficiency. For clarity in displaying participant distribution across severity levels, data points representing dichromacy cases (anomaloscope range of 70) were vertically offset. The closed and open circles represent the participant with mild and moderate deficiency determined by the W-CCVT. The black arrows are the individuals with severe defects determined by the Waggoner computerised colour vision test. The blue ‘X’ marker indicates the mean anomaloscope matching range within each Waggoner computerised colour vision test severity group.

Blue–yellow defects

In the blue–yellow screening series, nine individuals (six with deutan deficiency and three with protan deficiency) did not pass the test when the threshold was set at more than two errors on the tritan plates.

Discussion

The study demonstrated that the W-CCVT was a valid colour test in identifying red–green colour vision deficiencies, showing performance comparable to that of the printed tests evaluated in this study. The W-CCVT exhibited high sensitivity and specificity, which were not significantly different from those of the Ishihara and anomaloscope tests that were used as baseline. Although not statistically significant based on the 95% confidence intervals, the W-CCVT exhibited a lower level of agreement with the anomaloscope findings compared with the findings of previous studies.^14,15,18 Similar to the results reported in other studies,^14,15,18,19 all participants with dichromacy failed the screening plates (red–green) at either visit in this study.

In terms of the diagnosis series, the W-CCVT was relatively good at categorising participants as having either a deutan or protan deficiency (84% were correctly classified compared with that of the anomaloscope). Ng et al.¹⁵ showed that 89% of their sample (CVD = 59, CNV = 361) were correctly classified by W-CCVT. The minor difference between the current study and Ng et al.¹⁵ in terms of classification is likely attributed to the different screening plates used to display the test, as well as the variations in the number of patients with dichromacy between the studies.

The W-CCVT provides a qualitative assessment of defect severity, categorising it as mild, moderate or severe. The findings of this study align with those of Cole et al.², showing a good correlation between the severity ratings from the W-CCVT and the anomaloscope-matching ranges. Only two individuals with dichromacy were identified as having mild or moderate deficiencies by the W-CCVT. Consistent with previous research,^14,15 the W-CCVT sometimes classified participants with dichromacy as having moderate deficiencies.

The W-CCVT can screen for blue–yellow defects, and several participants with deutan deficiency failed these screening plates. Bailey et al.²⁰ also reported that patients with red–green CVD occasionally have an error on the HRR test blue–yellow plates. There are multiple reasons for blue–yellow failures in the W-CCVT. The increased number of errors on the blue–yellow plates could be related to short presentation times. The 2-s presentation time used by the W-CCVT could be short for participants with reduced sensitivity in the red–green zone. Furthermore, the 2-s presentation time resulted in errors on the printed blue–yellow screening plates compared with the red–green screening plates for participants in the CNV group.²¹ It is possible that the effect of shorter presentation times was more significant for participants with a red–green defect, even though the test colours were approximately orthogonal to the red–green axis of confusion. Another possible explanation for failing the blue–yellow screening series is that participants may not expect to see a figure or letter, which can lead to inattention and cause them to miss the figure.

The new W-CCVT in iPad devices can be used to detect red–green CVDs. In particular, the level of agreement between W-CCVT and the anomaloscope test was good in the dataset. Additionally, the W-CCVT can screen for blue–yellow defects. Individuals with deutan deficiency were prone to errors on these plates, particularly in W-CCVT. The blue–yellow failures were likely caused by multiple factors, including age, presentation times, expectation errors and incorrect responses.

The adoption of computerised colour vision tests offers several significant advantages over traditional manual methods that rely on physical colour plates or slides. Unlike conventional tests, which are susceptible to fading over time, damage from fingerprints and variability because of inconsistent lighting conditions, computerised tests maintain consistent colour presentation and are not subject to physical degradation.^15,22 Furthermore, computerised platforms allow for precise control of background illumination, ensuring standardised testing environments and minimising external influences on test results.^15,22 These digital tests can also be easily updated through software enhancements and transferred across devices, such as replacement iPads or computers, thereby extending their usability and relevance without the need for physical replacement of materials.²³ Additionally, computerised systems facilitate automated scoring, integration with electronic health records and efficient data management, further streamlining clinical workflows and improving reliability.²³ Collectively, these features not only enhance the accuracy and reproducibility of colour vision assessments but also reduce long-term costs and logistical challenges associated with manual testing methods.

While the W-CCVT is a recent advancement, there is a lack of published studies evaluating its validity and reliability. Comprehensive validation and independent assessments are still needed. This study provided initial evidence regarding the W-CCVT’s clinical utility, but further research is required to confirm these findings. Furthermore, the W-CCVT showed limitations in classifying defects as either protan or deutan deficient. However, the accuracy of the diagnostic plates was higher than that of the Ishihara test. Furthermore, the W-CCVT can qualitatively diagnose the severity of the defect; however, its agreement with the anomaloscope was not adequate. Some participants with dichromacy were classified as having mild or moderate deficiencies, whereas those with mild defects were classified as moderate or severe. This indicates that this test must not be relied upon solely to assess the severity of defects. For precise assessment, particularly when the severity of the defect is crucial for occupational purposes, the D15 or anomaloscope tests are recommended. Another limitation of this study was the relatively small number of participants who returned for the second testing session. This limited follow-up sample reduces the statistical power and may affect the robustness of the test-retest reliability analysis. Future studies with larger and more representative follow-up cohorts are needed to confirm and extend the repeatability assessment of the W-CCVT. Although the W-CCVT includes 12 plates specifically designed to detect tritan defects, no participants with tritanopia were identified in this study. This condition is a very rare condition, which likely explains its absence in our sample. Consequently, the test’s ability to detect tritan colour vision deficiencies could not be evaluated.

Conclusion

The W-CCVT demonstrates satisfactory performance as a colour vision screening tool, offering significant benefits and providing reasonable accuracy in the identification of CVD. Incorporating this test along with other established colour vision evaluations could enhance the overall diagnostic toolkit available to practitioners. Furthermore, the W-CCVT tended to classify participants with more severe symptoms than the anomaloscope, a trend consistent with other PIP assessment tests.

Acknowledgements

College of Applied Medical Sciences Research Centre and the Deanship of Scientific Research at King Saud University for their funding of this research. The author would also like to thank the OD students’ club for their assistance.

The author declares that they have no financial or personal relationships that may have inappropriately influenced them in writing this article.

A.A. is the sole author of this research article.

This work was supported by the College of Applied Medical Sciences Research Centre and the Deanship of Scientific Research at King Saud University.

The author confirms that the data supporting the findings of this study are available within the article.

The views and opinions expressed in this article are those of the author and are the product of professional research. It does not necessarily reflect the official policy or position of any affiliated institution, funder, agency or that of the publisher. The author is responsible for this article’s results, findings and content.

References