Abstract
Introduction: Recognition of ten Sloan letters is commonly used to assess visual acuity in adults. For testing children, who may not be familiar with language based symbols, either letters with a matching card, or more familiar symbols are used. The currently recommended options for children are comprised of four, rather than ten optotypes. Larger sets are available, but these sets have less principled design. Our aim was to generate a set of ten non-language based symbols with consistent stroke width, aspect ratios. We designed the set to elicited similar thresholds and to vanish uniformly into a grey background when pseudo high pass filtering was applied. Methods: Inter-optotype variability was assessed with independent, interleaved Quest staircases on eleven visually normal adults. We ran repeated measures ANOVAs to assess statistical differences within regular and pseudo high pass (vanishing) versions of three optotype sets (newly developed, the EDTRS Sloan letter, and Landolt C's). Maximum threshold differences within each group were determined with post hoc analysis in cases of significant main effects. Results: All four sets which were comprised of unique optotypes had statistically significant inter-optotype variability. Post hoc analysis revealed the variance was smaller for the pseudo high pass than regular versions. The new set had less inter-optotype variability than the Sloan regular and the Sloan vanishing sets. The new vanishing version eliciting remarkably similar internal thresholds, and error analysis further suggested more uniform vanishing in the new set than within the pseudo high pass filtered Sloan set. Conclusion: The newly designing set of ten optotypes had less inter-optotype variability compared to that within the EDTRS Sloan letter set. As this set does not require familiarity with language symbols, application for recognition acuity measurement across age and culture should be explored.
Meeting abstract presented at VSS 2017