Colour Vision Testing
We have three methods of assessing colour vision at our Practice. The HRR test (Hardy/Rand/ Rittler,) the City University Test and the Farnsworth D15 test.
To help interpret and understand the results of these tests, it is useful to briefly explain the accuracy of each test.
This is best understood by looking at the Sensitivity and Specificity of each test. These terms are used in medicine to describe how good a test is at identifying a certain condition.
Sensitivity is the ability of the test to correctly identify a positive result. In this case correctly identifying everyone who has a colour deficiency.
Specificity is the ability of the test to correctly identity a negative result. In this case correctly identifying someone with normal colour vision as ‘normal.’
Studies that look at colour vision tests often comment that there is always a trade-off between sensitivity and specificity.
To explain that in plain English – If you make a test very difficult then some people with normal colour vision will fail the test, but you will capture everyone who has a colour vision deficiency.
Conversely, if you make the test very easy, then all people with ‘normal’ colour vision will complete the test easily, but also some people with a mild colour vision problem may pass the test too.
This trade-off between sensitivity and specificity is sometimes why using multiple tests can be useful because it builds the level of confidence in the testing process and the result found. Each individual test can have its own limitations.
HRR Test
The HRR is a pseudoisochromatic plate test similar to the more familiar Ishihara test.
We prefer using the HRR over the Ishihara because it is capable of grading the extent of any colour vision deficiency, and it is also capable of identifying blue/yellow deficiency. (The Ishihara test can do neither of these things.)
The HRR starts by using 6 screening plates. If 2 incorrect screening plates are used as the criteria for a fail, then the test has a Sensitivity of 100%, and Specificity of 97.5%. Meaning that it will correctly identify all people with a colour vision problem if they get 2 or more plates wrong, but a very small percentage of people with normal colour vision will also get 2 plates wrong.
If 3 incorrect screening plates are used as the criteria for a fail then the test has a Sensitivity of 98% and a Specificity of 100%.
Effectively this means that we can be confident that anyone scoring more than 5/6 on the screening plates have ‘normal’ colour vision perception with a high degree of certainty.
If someone fails the screening plates, the subsequent diagnostic plates can be used to grade the type and the extent of a colour deficiency.
Farnsworth D15 Test
The City University Test
The Farnsworth D15 Test is a colour matching test derived from the more intensive 100 Hue test. It is meant for an easier classification of colour vision deficiencies and the 15 colour shades are picked because they sit along the lines of colour confusion. This makes it easier to visualise which colours someone may confuse.
The physical Farnsworth D15 Test has Specificity of 100% and a Sensitivity of approximately 55%. This means that it is a very good test at identifying people with normal colour vision, but it is less good at diagnosing the type of colour vision deficiency.
The D15 test puts 15 colours into order. It is still considered a ‘pass’ if 1 mistake is made close to consecutive colours.
The City University Test was derived from the Farnsworth D15 test.
We use this as a screening test when there is no time to complete a full D15 test. It can also be used when there is a contradiction in other colour vision tests to help with a diagnosis.
The test consists of screening plates that are marked out of 10, and diagnostic plates (if needed) that are marked out of 6.
Lighting
The other important part of testing colour vision is the illumination of the colour testing. All colour vision tests are designed to be completed in natural daylight, or as close as possible with artificial light.
It is recommended that any artificial lighting used has a Colour Rendering Index (CRI) of over 90%, the colour temperature of bulb is close to 6000K, and that the output of the bulb is over 300 lumens.
We use natural daylight when we can, and have lighting sources that conform to the standards above. We have a specialist ‘daylight’ light source that has a CRI of 95% and a colour temperature of 6000K.