Optotypes are visual symbols used to test recognition acuity. This set of optotypes was designed to be balanced, so that each item would elicit a similar estimate of visual acuity to the other items. We attempted to achieve this working with the following constraints on shape; that symbols should be enclosed shapes, make up of a single continuous stroke of constant width, such that symbols “vanish” in split stroke format (Frisen 1986). Note that despite the recognition acuity for the ten items being similar (Hamm, Yeoman et al. 2018), such balance is imperfect with some observers finding some items (moderately) easier than others. Refinement of balance - as well as determination of the appropriateness of the set for testing across varying age, cultural background and visual ability - would be valuable to help standardise the set for wider use.
Standardisation for the clinic requires one to determine what size of shape should be associated with “normal” vision. The bounding box of Sloan letters on the 0.0logMAR line of an eye chart (the line that should be just legible by someone with “normal” vision) subtends 5 arc min, while the stroke-width is 1 arc min. Our optotypes do not have a 1:5 stroke:bounding-box ratio, but rather have a 1:8.2 ratio. Three strategies for equating performance across tests are: (a) match bounding-box size, (b) match stroke-width or (c) match based on observer-performance. To pursue option (c), matching by comparison to a norm, requires a large–scale study to norm across clinical groups. So far we have been dividing the bounding box height by 7.6 for regular and 12.6 for vanishing to obtain similar results to other tests, but refinement will be needed in a study designed to do so accurately.
With these thoughts in mind, we have opened this set up to the research community to discuss, and modify according to their needs. We are distributing this material under an attribution (please reference Hamm, Yeoman et al. 2018) no-commercialisation licence. Measures of recognition acuity have a material impact on people in many ways, determining if they are able to drive, access financial assistance for a disability, or (for children) whether they will receive help from a teacher’s aide in class. As we move away from relying exclusively on charts to measure acuity, and begin to see wider adoption of digital tests of acuity, new standards will need to be established. Our hope is that the broader vision science community can contribute to the development of such standard in a number of ways - including what symbols we use and how - and that TAO can facilitate this development.
Frisen, L. (1986). "Vanishing optotypes. New type of acuity test letters." Arch Ophthalmol 104(8): 1194-1198.
Hamm, L. M., J. P. Yeoman, N. Anstice and S. C. Dakin (2018). "The Auckland Optotypes: An open-access pictogram set for measuring recognition acuity." Journal of Vision 18(3): 13-13.