August 2024
Volume 65, Issue 10
Open Access
Eye Movements, Strabismus, Amblyopia and Neuro-ophthalmology  |   August 2024
Landolt C-Tests With “Fixed” Arcmin Separations Detect Amblyopia But Underestimate Crowding in Moderate-to-Severe Amblyopic Children and Adults
Author Affiliations & Notes
  • Sarah J. Waugh
    Centre for Vision across the Life Span, School of Applied Sciences, University of Huddersfield, United Kingdom
  • Maria Fronius
    Goethe University Hospital, Department of Ophthalmology, Child Vision Research Unit, Frankfurt, Germany
  • Correspondence: Sarah J. Waugh, Department of Optometry and Vision Sciences, University of Huddersfield, Queensgate, Huddersfield, West Yorkshire HD1 3DH, UK; [email protected]
Investigative Ophthalmology & Visual Science August 2024, Vol.65, 33. doi:https://doi.org/10.1167/iovs.65.10.33
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Sarah J. Waugh, Maria Fronius; Landolt C-Tests With “Fixed” Arcmin Separations Detect Amblyopia But Underestimate Crowding in Moderate-to-Severe Amblyopic Children and Adults. Invest. Ophthalmol. Vis. Sci. 2024;65(10):33. https://doi.org/10.1167/iovs.65.10.33.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose: Crowding is exaggerated in central vision of strabismic amblyopia, impacting on reading ability. Crowding magnitude and interocular differences (IODs) in acuity are indicators for detection, assessment, and monitoring of treatment. Lateral masking (including contour interaction) also affects acuity and can mimic or ameliorate crowding. We investigated lateral masking/contour interaction and crowding impact on crowding magnitude and IOD measures in healthy and amblyopic pediatric and juvenile/adult groups using two Landolt C-tests with “fixed” arcmin separations.

Methods: Acuity (logMAR) was measured with Landolt C-tests with specified 2.6’ (“crowded”) and 35’ (“uncrowded”) separations. Crowding magnitudes (crowded - uncrowded acuities) and IODs were calculated. Participants were 69 subjects with strabismic amblyopia (n = 39 pediatric, i.e. children ≤8 years of age), 31 subjects with anisometropic amblyopia (n = 14 pediatric), and 76 healthy controls (n = 36 pediatric). Subjects with amblyopia were subgrouped by acuity as low severity (<0.4 logMAR) or high severity (≥0.4 logMAR) using the 35’ separation C-test.

Results: Crowding magnitudes were greater in strabismic than in anisometropic amblyopia and control/fellow eyes. They were higher in pediatric control/fellow eyes than in juvenile/adult eyes. In high severity strabismic amblyopia, crowding magnitudes progressively and significantly reduced (slope = −0.17 ± 0.07, P < 0.05) with worsening acuity. IODs for this group were higher on the 2.6’ C-test, but lower than expected. In high severity pediatric subjects with anisometropic amblyopia, seven of eight had lower IODs measured with the “crowded” than the “uncrowded” C-tests.

Conclusions: These C-tests detect amblyopia but underestimate crowding in children and adults with high severity strabismic amblyopia. Separate isolated optotype acuity and crowding distance tests may better target specific functions, while minimizing the impact of masking.

Crowding represents an essential bottleneck for object perception with important consequences for peripheral, amblyopic, and impaired macular vision.1 It is a property in which objects, individually resolvable, appear cluttered or pooled together. Crowding occurs in peripheral vision26 as neural resources are dedicated to analyzing the central visual space.7,8 In strabismic amblyopia, excessive crowding occurs in central vision,913 which also negatively impacts on the ability to read.1416 Accurate clinical assessment that includes crowding is therefore diagnostically important. 
Research into the specific role of crowding within visual perception and its relationships to other spatial interaction effects, such as contour interaction, masking, and grouping, is ongoing.9,13,1726 Crowding is thought to result from pooling of information across target and flanker receptive fields or feature detectors1,17,19,2729 leading to target-flanker confusion, rather than decreases in target detectability, specific to masking.17,2426 Crowding distance, or the critical spacing of crowding, is the smallest center-to-center distance (degrees) between resolvable objects free of crowding.17 Crowding distance (degrees) scales with eccentricity (degrees),6,17,30 known as Bouma's law, and is independent of target size.17,24,25 At the normal adult fovea, crowding was thought not to exist,3,17,18,26 however, the crowding distance is too small to measure with traditional optotypes. They abut or overlap before the crowding distance limit is reached.3133 At very close separations, however, lateral masking/contour interaction effects still degrade target recognition.9,19,34 
Lateral masking (also known as overlap,1 simple,2426 or ordinary17 masking) is like superimposition masking, in which a target (sinusoidal grating or Gabor patch,3537 but sometimes letters17,3538) is “masked” or harder to detect (or identify) when superimposed on a similar object (with specific spatial frequency, orientation, and location characteristics).3537 As the target and mask separate from superimposition, masking generally reduces in strength but is still present, hence the term “lateral” masking. Contour interaction, the mostly negative impact of flanking contours on visual acuity (usually with high contrast Cs, or optotypes, surrounded by a box or bars),13,34,3943 has many similarities to lateral masking. Like with lateral masking,17,26 it is thought to occur when target and flankers encroach upon the target's aggregate receptive field or feature detector13,17,19 affecting target visibility9,13,17,19,22 and scales in extent with that receptive field's size.17,26,34,37 This means that the impact of contour interaction or lateral masking on acuity should be constant if the separation between optotype and flankers is fixed in proportion to the optotype size.4449 The magnitude of contour interaction (difference in acuity with and without flankers) at the fovea is close to constant across acuity levels with normal development,20,50 imposed dioptric blur,4 and different target types,41,51 and in some amblyopes.9,13 Remote facilitation (an improvement in detection threshold with versus without a mask), a feature of lateral masking,26,36,52 has not specifically been reported in contour interaction experiments, although it may exist. In both lateral masking36,52 and contour interaction experiments,9,13,43 a reduction of masking has been demonstrated for overlapping, or very closely separated, target and flankers. Perhaps because stimuli and tasks differ in experiments on contour interaction (high contrast edges; gap/letter discrimination) versus lateral masking (Gaussian envelope of filtered edges; detection/perceived contrast), nearby/overlapping effects are more clearly documented for contour interaction. When flanking bars and target abut, or are too close together to be easily resolved, performance improvement (or a reduction in masking) often results.9,13 
When a standard high-contrast acuity optotype is closely surrounded by bars9,39 or by similar optotypes (e.g. Landolt Cs, Tumbling Es, symbols, and letters), measures of foveal acuity in adults are similarly degraded.4,33,34,41 This is true under conditions of dioptric blur4 and across different optotypes (symbols, pictures, and letters)41,53 suggesting that a common mechanism, like lateral masking is engaged. However, it is not true in young children,20,54 the normal periphery,4,55 and in strabismic amblyopia,56 where flanking letters degrade acuity substantially more than bars, like crowding. Thus, lateral masking/contour interaction is distinct from crowding and both are relevant to the clinical testing of visual acuity. 
Experiments by Flom and colleagues,9,13,39 which quantified contour interaction as a subcomponent of crowding, rather than as a separate entity from it, have had considerable impact on vision chart design.4449 Percent correct performance for an isolated acuity Landolt C was remeasured in the presence of four flanking contours across a range of separations. Peak negative spatial interaction occurred at 1 to 2.8’ edge-to-edge separation in non-amblyopic eyes and 0.6 to 6.3’ in amblyopic eyes (approximately 40% of a Landolt C size away) and reduced closer and further away. Improvement in performance when contours neared the target was thought to be due to new contrast, luminance, or blur cues present in the neural image at the gap location.34,42,43 Spatial interactions were absent when contours were greater than 100% to 160% C width, or by 25’, away. 
The focus of this paper is to assess effectiveness of a system of what are known as “crowded” and “uncrowded” acuity tests for detection and management of amblyopia, different from those commonly used in the United Kingdom and the United States41,44,45,48,5759 but with implications for all. The Haase-Hohmann Landolt C-tests60,61 (Oculus, Wetzlar, Germany) allow assessment of acuity and crowding magnitude in amblyopia and in other conditions that may affect reading ability, without influence of literacy, cognitive ability, and knowledge of a country's language.60 These tests are used in several European countries (e.g. Germany, Switzerland, the Netherlands, and Serbia) and are recommended for diagnosis of amblyopia and treatment follow-up. Test design follows German standard DIN 58220, which encourages equal separations between optotypes in arcmin.62 It was suggested that separations proportional-to-optotype size, such as in Early Treatment Diabetic Retinopathy Study (ETDRS),44,48,49 result in unequal crowding, becoming stronger for sizes ≤0.4 logMAR (as arcmin separations become smaller).62 In these C-tests,61 edge-to-edge separations between optotypes are specified as constant in arcmin. Crowding magnitudes are estimated from acuities obtained under “uncrowded” (35’) and “crowded” (2.6’) optotype separation conditions63 and have been used extensively in research.6466 Choice of arcmin separation originates from findings of Flom et al. (1963) in which 2.6’ for normal adults was near the peak of contour interaction and 35’ was beyond any. A 2.6’ separation also mimicked separations found in printed newspapers and book texts (range of 1–4 arcmin).67 C-test acuities ≤0.42 logMAR means that the gap in the target C (35’ C-test) is resolvable, as well as the 2.6’ separations between Cs (2.6’ C-test), so Cs can crowd (become confused with each other) if within that individual's crowding distance. For acuities >0.42 logMAR, 2.6’ separations are not resolvable but are subject to lateral masking/contour interaction effects between Cs, which can harm or help performance (see Fig. 1).9,13,34,39,42,43,68 Although previously for separations >25’, no interactions were found,9,39 this limit might be different in some amblyopes or when target and flankers are similar, like in a row of Landolt Cs. These two factors together could lead to underestimated crowding magnitudes. These predictions were tested on retrospectively analyzed clinical research data obtained in adults and children with amblyopia, for whom amblyopia treatment is most successful.6973 We also assess the impact of using different separation C-tests on interocular differences (IODs) in acuity, a key screening indicator for amblyopia detection and treatment. 
Figure 1.
 
Sections of C-tests (1.0 to 1.4 logMAR) scanned from physical versions (Oculus, Wetzlar, Germany). In (a) and (b) separation on the left is specified as 2.6’ and >35’ on the right. (a) Shows unblurred charts used for testing. (b) Shows a hypothetical visualization of Gaussian blur. When 2.6’ separation Landolt Cs are not resolvable due to blur, ability to localize the gap improves versus 35’ separation (top line, b). Actual separations varied from those specified (see Supplementary Fig. S1).
Figure 1.
 
Sections of C-tests (1.0 to 1.4 logMAR) scanned from physical versions (Oculus, Wetzlar, Germany). In (a) and (b) separation on the left is specified as 2.6’ and >35’ on the right. (a) Shows unblurred charts used for testing. (b) Shows a hypothetical visualization of Gaussian blur. When 2.6’ separation Landolt Cs are not resolvable due to blur, ability to localize the gap improves versus 35’ separation (top line, b). Actual separations varied from those specified (see Supplementary Fig. S1).
Methods
Commercially available Landolt C-tests for near (40 cm) assessment61 (Oculus, Wetzlar, Germany) were used (see Fig. 1). Each C-test comprises 3 cards with lines of C optotypes for a range of acuities (1.4 to −0.1 logMAR) in 0.1 log unit steps, plus lines at 0.15, 0.05, and −0.15 logMAR. Gaps in Cs were in four cardinal directions (up, down, right, and left). Two versions were used: “uncrowded” Cs (specified >35’ separation) and “crowded” Cs (specified 2.6’ separation). In this paper, “separation” is used to describe edge-to-edge distance between Cs (as per clinical literature)9,44 and “spacing,” to describe center-to-center distance between Cs (as per crowding literature).17,18 After testing, actual separations were manually measured with a mm ruler (see Supplementary Material). Actual versus specified dimensions vary but do not affect the research outcomes (reported in Discussion) of this study. 
Procedure
Visual acuity was measured in each eye with both tests (35’ and 2.6’) in the same room and lighting conditions at the Child Vision Research Unit in the Ophthalmology Department of Goethe University Hospital. One examiner conducted the assessment, another documented results and ensured a 40 cm viewing distance. An examiner indicated the target C with a pointer and the participant indicated perceived direction of the gap, verbally or manually. Procedures were explained and demonstrated binocularly before monocular testing: in controls, the right eye (RE) first; in amblyopes, amblyopic eye (AE) first. The 35’ separation “uncrowded” test was conducted first, followed by the 2.6’ separation “crowded” test, for which care was taken to avoid selecting Cs at the end of each line. 
Visual acuity was tested per line, with four of six correct the criterion before proceeding to the next line. Crowding magnitudes and IODs were calculated (see equations below). For 5 subjects with amblyopia, crowding magnitudes were recalculated using isolated Cs (created with a white mask with holes, placed over the 35’ test to expose only one optotype), in case the 35’ (“uncrowded”) separation was not actually large enough to escape crowding. This is because some amblyopic observers reported difficulty in localizing/identifying target optotypes for this test, like they experienced with the 2.6’ (“crowded”) test.  
\begin{equation}\begin{array}{@{}l@{}} Crowding\,magnitude\,\left( {logMAR} \right)\\ = \mathrm{`}\mathrm{`}crowded\text{''}\,VA\,\left( {logMAR} \right)\,on\,2.6^{\prime}\,test \nonumber \\ -\mathrm{`}\mathrm{`}uncrowded\text{''}\,VA\,\left( {logMAR} \right)\,on\,35^{\prime}\,test \end{array}\end{equation}
(1)
 
\begin{equation}\begin{array}{@{}l@{}} Recalculated\,crowding\,magnitude\,\left( {logMAR} \right)\\ = \ \mathrm{`}\mathrm{`}crowded\text{''}\,VA\,\left( {logMAR} \right)\,on\,2.6^{\prime}\,test \nonumber \\ -\mathrm{`}\mathrm{`}isolated\text{''}\,VA\,\left( {logMAR} \right) \end{array}\end{equation}
(2)
 
\begin{equation}\begin{array}{@{}l@{}} IOD\,for\,35^{\prime}\,and\,2.6^{\prime}\,{C\text{-}tests}\\ = VA\,\left( {logMAR} \right)\,in\,amblyopic\,eye\,\left( {AE} \right) \nonumber \\ -VA\,\left( {logMAR} \right)\,in\,fellow\,eye\,\left( {FE} \right)\,or\\ VA\,\left( {logMAR} \right)\,in\,right\,eye\,\left( {RE} \right) \nonumber \\ -VA\,\left( {logMAR} \right)\,for\,left\,eye\,\left( {LE} \right) \end{array}\end{equation}
(3)
 
Participants
Data were collected from 176 participants (see Table 1) aged 4.2 to 60 years (mean ± 1 standard deviation [SD], 12.60 ± 9.13 years, median = 8.95 years). Three main groups were: normal healthy controls (control), subjects with anisometropic amblyopia (aniso), and subjects with strabismic amblyopia (strab). Groups were further subdivided by age (pediatric = ≤8 years and juvenile/adult = >8 years) and if amblyopic, by severity according to visual acuity on 35’ C-test (low = <0.4 logMAR and high = ≥0.4 logMAR). Unlike standard clinical categories of amblyopia that use crowded acuities of mild, moderate, and severe,74,75 severity categories here are based on amblyopic acuity only, that is, resolvability of a 2.6’ C gap, ideally on an isolated C (so using the 35’ C-test). This enables the suitability of a test with 2.6’ separations between Cs (the 2.6’ C-test) as a “crowded” test to be examined. Children ≤8 years were important to recruit to test outcomes in visual systems that may still be developing46,76,77 and to assess whether similar cues might be used by adults and children. Experimental participants (n = 100, 65% male subjects) had amblyopia due to (1) unilateral strabismus with and without anisometropia (strab) and (2) pure anisometropia (aniso). Control participants (n = 76, 50% male subjects) had normal or corrected-to-normal visual acuities and binocular vision. Amblyopic participants had uncrowded IODs >0.1 logMAR.7880 Anisometropic amblyopes had interocular refractive differences >1 DS and/or >1.5 DC, absence of strabismus or microstrabismus on cover test and central fixation on visuoscopy (Cüppers visuscope test).81 
Table 1.
 
Participant Group and Eye Details
Table 1.
 
Participant Group and Eye Details
Participants received comprehensive ophthalmological and orthoptic examinations including evaluation of anterior and posterior segments of eyes, visual acuity, refraction (cycloplegic for patients with amblyopia), and eye alignment (cover test). Suppression and stereopsis were assessed with Bagolini striated glasses, the Stereo Fly Test and Randot Stereotest (both from Stereo Optical Co., Inc.), and the TNO test (Lameris Ootech B.V.). Exclusion criteria were deprivation amblyopia, morphological ocular disorder, impaired acuity due to medication, or brain damage, trauma, or neurological disorder. 
This research adhered to the Code of Ethics of the World Medical Association (Declaration of Helsinki). The Ethics Committee of the University of Frankfurt approved study protocol prior to initiation of the study. After explanation about the study, written informed consent from all parents and adult participants, as well as child assent, was obtained. 
Data Analysis
Statistical analyses were mixed model repeated measures analyses of variance (ANOVAs), followed by planned comparisons, if appropriate, using Statistica (Tibco Software, Inc.). Means ± 1 SE are reported unless otherwise indicated. Post hoc comparisons were conducted with Tukey Honest Significant Difference (HSD) tests. P values < 0.05 were considered statistically significant. Data fitting across participants used linear regression analysis with significance level set at P = 0.05 (IgorPro, Wavemetrics, Inc.). Slopes were tested against zero, expressed as t statistic and compared with a critical t value (2-tailed). Two slopes were compared using multiple regression and Tukey tests for equality. 
Results
Measured Crowding Magnitudes
Results (see Fig. 2) show that crowding magnitude (Equation 1) depends on group (F(2, 170) = 38.81; P < 0.0001) being higher in strabismic amblyopic eyes (strab AE = 0.31 ± 0.14 logMAR), than in anisometropic amblyopic eyes (aniso AE = 0.11 ± 0.021 logMAR), fellow eyes (strab FE = 0.12 ± 0.012 logMAR, and aniso FE = 0.12 ± 0.018 logMAR) and healthy control eyes (RE = 0.093 ± 0.013 logMAR, the left eye (LE) = 0.094 ± 0.011 logMAR). Results were unaffected if data from the RE or LE of controls were used. Crowding magnitude also depended on the eye (F(1, 170) = 53.23; P < 0.0001) and age (F(1, 170) = 426.54; P < 0.0001). Post hoc comparisons revealed that crowding in control eyes was not different from that in fellow eyes of subjects with amblyopia (pediatric: aniso P = 0.976 or strab P = 1.000; juvenile/adult: aniso P = 1.000 or strab P = 0.987). Crowding was significantly higher for fellow and control eyes in the pediatric group, than in the juvenile/adult groups (pediatric: RE 0.16 ± 0.019 logMAR and the LE 0.16 ± 0.016 logMAR versus juvenile/adult: RE 0.028 ± 0.018 logMAR and the LE 0.025 ± 0.015 logMAR, P < 0.0001; aniso FE: pediatric 0.21 ± 0.026 logMAR versus juvenile/adult 0.035 ± 0.024 logMAR, P < 0.0001; strab FE: pediatric 0.17 ± 0.016 versus juvenile/adult 0.057 ± 0.018 logMAR, P = 0.0001), but not in amblyopic eyes (aniso AE: pediatric 0.12 ± 0.031 versus juvenile/adult 0.094 ± 0.028 logMAR, both P = 1.000 and strab AE: pediatric 0.31 ± 0.018 versus juvenile/adult 0.30 ± 0.021 logMAR). Crowding magnitude in strab AEs was higher than in control eyes in both age groups (pediatric: P < 0.0001 and juvenile/adult: P < 0.0001) and higher than in aniso AEs (pediatric: P < 0.0001 and juvenile/adult: P < 0.0001). It was not different between control eyes and aniso AEs (pediatric: P = 0.989 versus juvenile/adults: P = 0.584). Analyses conducted on five subgroups according to visual acuity (controls, low anisos, high anisos, low strabs, and high strabs) confirmed these results with one additional finding: no differences were found between crowding for low aniso versus high aniso groups (pediatric: low aniso AE 0.19 ± 0.047 versus high aniso AE 0.063 ± 0.041 logMAR, P = 0.538; juvenile/adult: low aniso AE 0.090 ± 0.036 versus high aniso AE 0.10 ± 0.043 logMAR, P = 1.000) or for low strab versus high strab groups (pediatric: low strab AE 0.35 ± 0.036 versus high strab AE 0.30 ± 0.021 logMAR, P = 0.977; juvenile/adult: low strab AE 0.28 ± 0.041 versus high strab AE 0.31 ± 0.024 logMAR, P = 1.000). 
Figure 2.
 
Crowding magnitudes (logMAR) from Equation 1 calculated for C-tests (RE and LE of controls and FE and AE of amblyopes) for participants in three groups (control, aniso, and strab) and both age groups (pediatric and juvenile/adult). Stronger colors are data for juvenile/adult participants (>8 years). Lighter colors are data for pediatric participants (≤8 years). Horizontal and vertical bars with * indicate statistically significant differences between groups (Tukey post hoc testing P < 0.0001; all other results were P > 0.05).
Figure 2.
 
Crowding magnitudes (logMAR) from Equation 1 calculated for C-tests (RE and LE of controls and FE and AE of amblyopes) for participants in three groups (control, aniso, and strab) and both age groups (pediatric and juvenile/adult). Stronger colors are data for juvenile/adult participants (>8 years). Lighter colors are data for pediatric participants (≤8 years). Horizontal and vertical bars with * indicate statistically significant differences between groups (Tukey post hoc testing P < 0.0001; all other results were P > 0.05).
Figure 3 (all participants) and Figures 4A and 4B (pediatric and juvenile/adult groups) reveal crowding magnitudes for individuals. From the literature,4,12,82 one would expect crowding magnitude to increase in strabismic amblyopic eyes with worsening acuity, falling between two diagonals (slopes 0.6 and 1).4,18 A constant proportion of acuity is expected in anisometropic amblyopic eyes, like in control or blurred eyes (horizontal dashed line, see Figs. 34).4,18 See Table 2 for slopes of all subgroups (with statistical indicators). 
Figure 3.
 
Crowding magnitudes (logMAR) from Equation 1 plotted against 35’ C-test acuity (logMAR) for all participants including control and amblyopic eyes. Two diagonal comparison lines are provided. One is based on crowding magnitude data from the normal adult periphery4 scaled to match control foveal data from current study (slope = approximately 0.6). In that study, visual acuity for a letter surrounded by four letters versus an isolated letter was calculated. The other comparison line shows crowding magnitude increasing directly with acuity (slope = approximately 1). The horizontal line at 0 indicates equal visual acuity for 2.6’ and 35’ C-tests, so crowding magnitude = 0. The horizontal line with grey shading represents average crowding magnitude (shading = ± 1 SE) for control eyes. Individual symbols are data from five amblyopes who had isolated visual acuities and crowding magnitudes recalculated (Equation 2).
Figure 3.
 
Crowding magnitudes (logMAR) from Equation 1 plotted against 35’ C-test acuity (logMAR) for all participants including control and amblyopic eyes. Two diagonal comparison lines are provided. One is based on crowding magnitude data from the normal adult periphery4 scaled to match control foveal data from current study (slope = approximately 0.6). In that study, visual acuity for a letter surrounded by four letters versus an isolated letter was calculated. The other comparison line shows crowding magnitude increasing directly with acuity (slope = approximately 1). The horizontal line at 0 indicates equal visual acuity for 2.6’ and 35’ C-tests, so crowding magnitude = 0. The horizontal line with grey shading represents average crowding magnitude (shading = ± 1 SE) for control eyes. Individual symbols are data from five amblyopes who had isolated visual acuities and crowding magnitudes recalculated (Equation 2).
Figure 4.
 
(a) (left, pediatric group) and (b) (right, juvenile/adult group). Crowding magnitudes (logMAR) from Equation 1 plotted against 35’ C-test acuity (logMAR) for children ≤8 years (a, left) and juvenile/adults >8 years (b, right) including control and amblyopic eyes. Other details match those for Figure 3 but horizontal lines with grey shading represent average crowding magnitudes (shading = ± 1 SE) for control eyes in each group. Individual symbols are data from amblyopes who had isolated visual acuitities and crowding magnitudes recalculated (Equation 2).
Figure 4.
 
(a) (left, pediatric group) and (b) (right, juvenile/adult group). Crowding magnitudes (logMAR) from Equation 1 plotted against 35’ C-test acuity (logMAR) for children ≤8 years (a, left) and juvenile/adults >8 years (b, right) including control and amblyopic eyes. Other details match those for Figure 3 but horizontal lines with grey shading represent average crowding magnitudes (shading = ± 1 SE) for control eyes in each group. Individual symbols are data from amblyopes who had isolated visual acuitities and crowding magnitudes recalculated (Equation 2).
Table 2.
 
Slopes ± 1 SD from Figures 3 and 4a,b
Table 2.
 
Slopes ± 1 SD from Figures 3 and 4a,b
Crowding magnitudes from low strab AEs look to increase with worsening acuity. As acuity worsens ≥0.4 logMAR (i.e. in high strabs), crowding magnitudes are expected to increase but they decrease. Both for all high strab AEs (see Fig. 3) and only pediatric high strab AEs (see Fig. 4A), descending slopes are significantly different from 0 and from low strab AE slopes. For the smaller number of juvenile/adult strabs (see Fig. 4B), a similar pattern is present, but slopes for high strab AEs do not statistically differ from 0 or from low strab AE slopes, but they differ significantly from those of control eyes. Slopes fit to juvenile/adult strab data were not significantly different from slopes fit to pediatric data. 
Data for all anisometropic amblyopic eyes (aniso AEs) are fit with a slope not significantly different from 0 but different from that for control eyes. For pediatric aniso AEs, crowding magnitudes decrease with increasing amblyopia severity (see Fig. 4A) with a slope of −0.34 ± 0.121, which is significantly different from 0. For juvenile/adult aniso AEs (see Fig. 4B), crowding magnitudes are flat with a slope of 0.012 ± 0.118, not significantly different from 0 or from slopes fit to pediatric data. Crowding magnitudes appear higher for aniso AEs than control eyes (mean ± 1 SE of aniso AE 0.094 ± 0.028 versus control eyes 0.026 ± 0.008; Tukey P = 0.5836). For control eyes, crowding magnitudes are likely underestimated due to a floor effect (–0.15 logMAR is the best acuity possible) affecting 35’ more than 2.6’ acuity measures. True crowding magnitudes would be closer to 0.093 ± 0.012 logMAR, the average across the whole control group and similar to that found in aniso AEs in Figure 4B. 
For 5 subjects with amblyopia (see Fig. 3), recalculated crowding magnitudes using isolated C acuities, rather than 35’ separation acuities, (Equation 2) increased for 4 high strab AEs (by 0.35 ± 0.083 logMAR) but decreased by 0.1 logMAR for 1 high aniso AE of a 12-year-old, within repeatability of acuity estimates in children (± 0.15 to ± 0.18 logMAR)83,84 and adults (± 0.10 to ± 0.14 logMAR).85 
Measured IODs
When IOD data from three groups (the control, strab, and aniso group) and two age groups (pediatric and juvenile/adult groups) were analyzed, there was a significant effect of group on IODs (F(2, 170) = 23.97; P < 0.0001), all significantly different from each other (P < 0.0001). Figures 3 and 4 showed that acuity subgroups are important. IOD results for five subgroups (controls, low anisos, high anisos, low strabs, and high strabs), two age groups (pediatric and juvenile/adult groups), and two separation conditions (35’ and 2.6’) are shown in Figure 5. There is a significant effect of group on IODs (F(4, 166) = 280.01; P < 0.0001) and a significant age effect (F(1, 166) = 5.48; P = 0.0203), but no interaction between them (F(4, 166) = 0.91; P = 0.459). IODs for high strabs were different from low strabs (0.97 ± 0.023 vs. 0.38 ± 0.039 logMAR, P < 0.0001); and IODs for high anisos were different from low anisos (0.67 ± 0.042 vs. 0.30 ± 0.042 logMAR, P < 0.0001). IODs for controls (0.0031 ± 0.028 logMAR, P < 0.0001) were different from all other groups. 
Figure 5.
 
Interocular differences (IODs) in VA obtained for C-tests (35’ and 2.6’ separations) for all five subgroups (control, low aniso, high aniso, low strab, and high strab) and both age groups (pediatric and juvenile/adult). Stronger colors show data for juvenile/adult participants and lighter colors are data for pediatric participants. Horizontal bars with * indicate statistically significant differences between groups (Tukey post hoc testing P < 0.0001; all other results were P > 0.05). The horizontal bar with P < 0.01 is the result from a planned comparison. Data from control eyes (control 35’ and control 2.6’) were statistically different from all other subgroups but bars are omitted for clarity.
Figure 5.
 
Interocular differences (IODs) in VA obtained for C-tests (35’ and 2.6’ separations) for all five subgroups (control, low aniso, high aniso, low strab, and high strab) and both age groups (pediatric and juvenile/adult). Stronger colors show data for juvenile/adult participants and lighter colors are data for pediatric participants. Horizontal bars with * indicate statistically significant differences between groups (Tukey post hoc testing P < 0.0001; all other results were P > 0.05). The horizontal bar with P < 0.01 is the result from a planned comparison. Data from control eyes (control 35’ and control 2.6’) were statistically different from all other subgroups but bars are omitted for clarity.
Higher IODs were measured in the juvenile/adult group than the pediatric group (F(1, 166) = 5.48; P < 0.020) at 0.50 ± 0.021 vs. 0.43 ± 0.022 logMAR, except in controls (0.00 ± 0.026 vs. 0.01 ± 0.027 logMAR). A reduction in crowding magnitude for FEs with increasing age, but not AEs (see Fig. 2), leads to larger IODs. IODs are different for 2.6’ and 35’ separations (F(1, 166) = 35.42; P < 0.0001) but this depends on the age and group (F(4, 166) = 3.30; P = 0.0123). Collapsing across age to remove the development effect, the 2.6’ C-test produced higher IODs than the 35’ C-test in strabismic amblyopia (high strab: 1.06 ± 0.027 vs. 0.88 ± 0.022 logMAR and low strab: 0.49 ± 0.045 vs. 0.27 ± 0.038 logMAR, P < 0.0001). However, differences in IODs for 2.6’ vs. 35’ separation C-tests were similar for high strab (0.18 ± 0.019 logMAR) and low strab (0.22 ± 0.032 logMAR) groups. Overall IODs were not significantly different between 2.6’ vs. 35’ C-tests in controls (0.00 ± 0.018 vs. 0.00 ± 0.021 logMAR, P = 1.00), or in anisometropic amblyopes (low aniso: 0.30 ± 0.049 vs. 0.29 ± 0.041 logMAR; high aniso: 0.66 ± 0.050 vs. 0.69 ± 0.041 logMAR). However, IODs decrease for the 2.6’ C-test in pediatric high anisometropic amblyopes (0.54 ± 0.068 logMAR) compared to the 35’ C-test (0.69 ± 0.056 logMAR) due to higher crowding magnitudes (Equation 1) measured in FEs (by 0.05 to 0.45 logMAR) than in AEs for 7 of 8 children. A planned comparison found this decrease to be statistically significant (F(1, 166) = 9.066, P = 0.00301). 
Discussion
Spatial interactions, including contour interaction13 and crowding,13 have been incorporated into clinical vision tests44,47,49,61,67 to better detect, diagnose, and monitor treatment of conditions, including amblyopia. Considerable variations in test design with regard to optotype separation and arrangements within clinical tests4,41,61,67,86 makes comparison of baseline characteristics of patient groups, treatment effects, as well as calculation of dose-response relationships and treatment efficiency across research laboratories, difficult.8789 This study highlights difficulties in obtaining accurate estimates of acuity and crowding from crowded acuity tests. 
Cues Leading to Reduced Crowding Magnitude in C-Tests
When Cs or letters are placed closer together than the resolution limit, they are not seen as separate so they cannot crowd, but blur together and can mask each other. Under these circumstances, cues such as local luminance cues/irregularities,9 reduced light scatter,68 neural blur,32,39,43 or increased energy at newly created spatial frequencies34 can arise that enhance visibility of the gap in a C (see Fig. 1) and can help with C gap9 and target letter identification.43 Improvement in recognition performance, leading to an upturn in the contour interaction function, has previously been demonstrated for normal participants as well as three of four tested amblyopic eyes (and 4 of 4 fellow eyes),9 suggesting that cues such as those listed above are also used by subjects with amblyopia. 
Crowding magnitudes measured with 2.6’ and 35’ C-tests (see Figs. 3, 4) can be explained because (1) lateral masking/contour interaction extends out to separations approximately 100% to 160% C-size9,13 but in strabismic amblyopia (strab), crowding can extend out even further,12 (2) negative effects of masking reduce as separation decreases to approximately 40% C-size (peak intensity),9,13 and (3) for separations <40% C-size, cues introduced can enhance ability to overcome spatial interactions.9,13,34,39,43,68 Crowding distances for flanked single target optotypes in strab AEs (adult range = 15–1064’)12 are much larger than in normal adult vision (mean ± 1 SE of 2.9 ± 0.24’90 and 3.2 ± 0.15’91) or in normal child vision (8.4 ± 1.1’ for 4-year-olds).91 They increase with severity of acuity loss12,29 and are associated with increased positional uncertainty,7,9294 spatial distortion,95,96 and central suppression scotomas.65,97 Negative effects of masking are thought to have minimal impact on crowding,12 however, luminance/blur cues at very close separations, could ameliorate crowding. 
In Figure 3, in normal, healthy controls (mean acuity ± 1 SE = −0.14 ± 0.003 logMAR; C-size of 3.6’), a 2.6’ separation (spacing of 6.2’) falls outside the crowding distance of most but is subject to negative masking (48% C-size separation). As acuity worsens up to 0.0 logMAR, a 2.6’ separation engages stronger masking effects resulting in a significantly positive slope (see Table 2). In anisometropic amblyopia (low aniso), acuity is worse than normal so masking increases until the 2.6’ separation brings flanking Cs into the performance enhancement region (<40% C-size separation) leading to a slope not different from zero. In the high aniso group, enhancement effects become stronger. For pediatric aniso AEs, with larger crowding zones to contend with than healthy adult eyes, enhancement effects significantly reduce crowding magnitudes as acuities worsen (see Fig. 4A). In strab AEs, as acuity worsens (low strab), so does crowding. As acuity worsens more (high strab), crowding increases but masking does not, as enhancement cues become available (26%–7% C-size separations) partially contributing to a significant negative slope; a similar story occurs for pediatric subjects with high-severity strabismic amblyopia. For subjects with amblyopia too, 35’ C-tests (with actual center-to-center spacings of 58’–192’; see Supplementary Fig. S1) are likely to engage crowding (46–1064’), leading to reduced calculated crowding magnitudes (Equation 1). For four strab AEs retested with an isolated C (see Methods), better “uncrowded” acuities and higher crowding magnitudes were recorded (Equation 2). An alternative possibility that could produce a negative slope of crowding magnitude with worsening acuity in amblyopic eyes, is that acuity and crowding develop differently, with worse acuities and reduced crowding in younger eyes. This is not true in normal vision,20,46,76,77,86,91 but developments of acuity and crowding in anisometropic and strabismic amblyopia, are as yet unknown. 
Impact of C-Tests on Estimating IOD
Acuity measures depend on spatial interactions, therefore IODs depend on C-test separation (2.6’ or 35’). Clinicians rely on IODs for detecting >0.1 logMAR IOD7880 and monitoring treatment in amblyopia. In support of using the 2.6’ C-test, higher IODs were found in low strab (by 0.22 ± 0.033 logMAR) and high strab (by 0.17 ± 0.020 logMAR) groups, than for the 35’ C-test (see Fig. 5). However, as crowding increases with increasing severity of strabismic amblyopia, differences between crowded and uncrowded IODs should be greater in high strab, than low strab groups. In anisometropic amblyopia, IODs for 2.6’ and 35’ C-tests are not significantly different, except in the pediatric high aniso group. In this group, 7 out of 8 (87.5%) participants had lower IODs for 2.6’ than 35’ C-tests (by −0.14 ± 0.048 logMAR) due to measured crowding magnitude being greater in the fellow eye. 
Analysis of data obtained retrospectively, with visual acuity gathered according to clinical research protocol (“uncrowded” AE, “crowded” AE, “uncrowded” FE, and “crowded” FE) means that visual acuity on the FE for the “crowded” 2.6’ separation was always tested last (fourth). Could order effects (eye order or test order) have influenced our results? Fatigue could have negatively affected visual acuity measurement using the 2.6’ C-test in the FE, resulting in worse acuity, higher crowding magnitude, and a lower 2.6’ C-test IOD, than might be present, especially in children. Practice effects also affect visual acuity measures and could lead to the opposite outcome. For example, visual acuity measures using isolated Cs (8 possible orientations) on 14 adults, conducted 14 times within a single session, improved from the first to fourth measurement by −0.03 ± 0.042 logMAR.98 In our control pediatric and juvenile/adult groups, order of testing for the LE was the same as for the FE of amblyopes (and RE same as AE), but measured crowding magnitudes did not differ (pediatric = RE 0.16 ± 0.019, LE 0.16 ± 0.016 logMAR; and juvenile/adult = RE 0.028 ± 0.018, LE 0.025 ± 0.015 logMAR), nor did IODs (35’ 0.008 ± 0.042 and 2.6’ 0.004 ± 0.043 logMAR). In a study of visual acuity repeatability in amblyopic (mean age = 6.44 ± 1.94 years) and control (mean age = 6.69 ± 1.12 years) children using the crowded Lea Symbol chart,84 there was no difference (P = 0.39) in the degree of amblyopia measured if the amblyopic eye was tested first (n = 24; IOD = 0.40 ± 0.37 logMAR), or second (n = 8; IOD = 0.29 ± 0.16 logMAR). Finally, a control experiment was conducted using the C-tests on four amblyopes. Visual acuity measures were taken in reverse order, then the original order, within a single session, exaggerating any fatigue/practice effects on the 2.6’ FE visual acuity measure. These data are provided in Supplementary Figure S2 with details of analysis. Overall, there was no significant difference found in any test measure (all P > 0.05) for the two orders, and outcomes did not depend on the eye (FE versus AE), or separation condition (2.6’ and 35’). One of two juvenile anisos showed a greater IOD for the 35’ than the 2.6’ C-test. A reduction in IOD measures for 16 of 21 anisometropic amblyopic children (≤8 years: visual acuities = 0.05 to 0.45 logMAR) was also found for a test using HOTV letters and a 1 stroke-width (20% optotype) separation compared to that obtained for a test using 1 optotype-width separation (Haine LA, et al. IOVS, 2023; 64: ARVO E-Abstract 1457). These results were obtained using covariate adaptive randomization and minimization99 to determine test order that minimized eye bias, accounting for amblyopia severity (mild, moderate, or marked), strabismus presentation (esotropia versus exotropia), amblyopia treatment (patching or atropine), gender, and refractive error (hypermetropia versus myopia). In that study, crowded and uncrowded trials were interleaved to avoid fatigue and practice effects. 
In summary, to find accurate IODs in terms of acuity, isolated acuities would be best to compare. Additional inter-ocular ratios (IORs) of crowding distance (degrees), obtained through separate spacing measures (with targets that avoid or minimize masking), would be recommended. However, using only IODs based on crowded acuities, can result in inaccuracies. 
Impact of C-Tests on Detecting Amblyopia and Monitoring Amblyopia Treatment
Monitoring crowding magnitude (Equation 1) is not suitable to assess treatment progress in patients with high-severity strabismic amblyopia (>0.4 logMAR on 35’ C-test) as with improvement in acuity, crowding magnitudes could appear to worsen (see Figs. 34). Higher IODs were measured with the 2.6’ than the 35’ C-test in strabismic amblyopes. If treatment is monitored only with 35’ C-test IODs, they can appear to reduce to <0.2 logMAR, suggesting cessation of treatment according to protocol,100 when if measured with the 2.6’ C-test they may be higher, suggesting that treatment should continue. In patients with eccentric fixation, IODs of 0.8 logMAR (35’ C-test) at initiation of treatment, reduced to 0.2 logMAR after 12 months of treatment. They reduced from 1.0 logMAR to 0.7 logMAR with the 2.6’ C-test.88 Caution in only using the 2.6’ C-test for IODs has also been suggested: despite unchanged 2.6’ IODs during several months of treatment, 35’ IODs may reduce, signaling residual functional plasticity and warranting continuation of treatment.101 Others have suggested that IODs should be minimized with both 2.6’ and 35’ C-tests, for acuity gain during treatment to be retained.100 
Optimal Clinical Measurement of Acuity and Crowding
Scaling separation between optotypes in proportion to optotype size on an acuity test means lateral masking/contour interaction effects on acuity will be approximately constant. Spacing in arcmin thereby increases with larger letters, but not enough to benefit detection of strabismic amblyopia, as their crowding increases faster than acuity.12,24,102 Recent studies4,12,103 have suggested that a 1 optotype separation is too large to optimally detect amblyopia and have suggested reducing this to 0.25,4 0.40,90 or 0.1312 separations, (i.e. 1.25, 1.4, or 1.13 spacings). The current study shows that Cs placed too closely together in attempts to optimize crowding, may introduce luminance/blur cues that lead to underestimates of crowding. An 0.4 optotype separation (1.4 spacing)90,91 may prove optimal to reduce masking but enhance crowding magnitude, but the same results may not apply to Landolt Cs, pictures, or symbols. Even if optimal, use of standard optotypes means that mature foveal crowding distance cannot be accurately measured because they are too wide.90,104 A better strategy might be to measure isolated letter/Landolt C acuity to find the acuity limit, and crowding distance with skinny optotypes to find the crowding limit.90,91 A Vernier target with flanking contours,105,106 which would also allow for closer distances to be assessed, is less attractive due to the difficulty of the task for patients and the number of trials needed to attain a suitable threshold. 
Conclusions
C-acuity tests with 2.6’ and 35’ separations may provide information about susceptibility to deterioration when viewing close arrangements of optotypes, for example, reading text. However, calculated crowding magnitudes are considerably underestimated in moderate-to-severe strabismic amblyopia. The 2.6’ C-test for amblyopes may introduce luminance/blur cues that enhance C acuity and the 35’ C-test is not uncrowded for all. Higher IODs are found for 2.6’ versus 35’ C-tests in strabismic amblyopia, but these IODs are not good indicators of severity. Sole use of the 2.6’ C-test can also underestimate IODs in anisometropic amblyopia. This study highlights difficulties in obtaining accurate estimates of acuity and crowding from crowded acuity tests. A better strategy might be to measure isolated acuity (a size limit) and crowding distance (a spacing limit), independently. 
Acknowledgments
The authors thank Iris Bachert, Licia Cirina, Juliane Tittes, Peggy Feige, and Ines Tscherner, for help with data collection. Preliminary results were presented at the 44th European Conference on Visual Perception (ECVP) 2022 in Nijmegen, The Netherlands (Waugh & Fronius Perception 51(1S): 190) and at the Child Vision Research Society meeting in London, United Kingdom, in 2023. We also thank two anonymous reviewers for their valuable contributions leading to an improved manuscript. 
Supported by URN020-01 from University of Huddersfield and Goethe University (Department of Ophthalmology) for travel for this collaboration. Other financial support provided by ERA-NET NEURON grant (JTC2015) to M.F., the German Ministry for Education and Research (BMBF, grant number 01EW1603B) and by the association “Augenstern-e.V.” (a non-profit association supporting research in pediatric ophthalmology) from Frankfurt, Germany. 
Disclosure: S.J. Waugh, None; M. Fronius, None 
References
Levi DM. Crowding—An essential bottleneck for object recognition: a mini-review. Vision Res. 2008; 48: 635–654. [CrossRef] [PubMed]
Korte W. Über die Gestaltauffassung im indirekten Sehen [in German]. Zeitschrift für Psychologie. 1923; 93: 17–82.
Bouma H. Visual interference in the parafoveal recognition of initial and final letters of words. Vision Res. 1973; 13: 767–782. [CrossRef] [PubMed]
Formankiewicz MA, Waugh SJ. The effects of blur and eccentric viewing on adult acuity for pediatric tests: implications for amblyopia detection. Invest Ophthalmol Vis Sci. 2013; 54: 6934–6943. [CrossRef] [PubMed]
Toet A, Levi DM. The two-dimensional shape of spatial interaction zones in the parafovea. Vision Res. 1992; 32: 1349–1357. [CrossRef] [PubMed]
Bouma H. Interaction effects in parafoveal letter recognition. Nature. 1970; 226: 177–178. [CrossRef] [PubMed]
Levi DM, Klein SA. Vernier acuity, crowding and amblyopia. Vision Res. 1985; 25: 979–991. [CrossRef] [PubMed]
Dow BM, Snyder AZ, Vautin RG, Bauer R. Magnification factor and receptive field size in foveal striate cortex of the monkey. Exp Brain Res. 1981; 44: 213–228. [CrossRef] [PubMed]
Flom MC, Weymouth FW, Kahneman D. Visual resolution and contour interaction. J Opt Soc Am. 1963; 53: 1026–1032. [CrossRef] [PubMed]
Müller P. Über das Sehen der Amblyopen [in German]. Ophthalmologica. 1951; 121: 143–149. [CrossRef] [PubMed]
Stuart JA, Burian HM. A study of separation difficulty. Its relationship to visual acuity in normal and amblyopic eyes. Am J Ophthalmol. 1962; 53: 471–477. [CrossRef] [PubMed]
Song S, Levi DM, Pelli DG. A double dissociation of the acuity and crowding limits to letter identification, and the promise of improved visual screening. J Vis. 2014; 14: 3. [CrossRef] [PubMed]
Flom MC. Contour interaction and the crowding effect. Problems in Optometry. 1991; 3: 237–257.
Levi DM, Song S, Pelli DG. Amblyopic reading is crowded. J Vis. 2007; 7: 21. [CrossRef]
Stifter E, Burggasser G, Hirmann E, Thaler A, Radner W. Monocular and binocular reading performance in children with microstrabismic amblyopia. Br J Ophthalmol. 2005; 89: 1324–1329. [CrossRef] [PubMed]
Stifter E, Burggasser G, Hirmann E, Thaler A, Radner W. Evaluating reading acuity and speed in children with microstrabismic amblyopia using a standardized reading chart system. Graefes Arch Clin Exp Ophthalmol. 2005; 243: 1228–1235. [CrossRef] [PubMed]
Pelli DG, Palomares M, Majaj NJ. Crowding is unlike ordinary masking: distinguishing feature integration from detection. J Vis. 2004; 4: 12. [CrossRef]
Strasburger H, Harvey LO, Jr., Rentschler I. Contrast thresholds for identification of numeric characters in direct and eccentric view. Percept Psychophys. 1991; 49: 495–508. [CrossRef] [PubMed]
Wolford G, Chambers L. Lateral masking as a function of spacing. Percept Psychophys. 1983; 33: 129–138. [CrossRef] [PubMed]
Waugh SJ, Formankiewicz MA. Grouping effects on foveal spatial interactions in children. Invest Ophthalmol Vis Sci. 2020; 61: 23–23. [CrossRef] [PubMed]
Herzog MH, Sayim B, Chicherov V, Manassi M. Crowding, grouping, and object recognition: a matter of appearance. J Vis. 2015; 15: 5. [CrossRef] [PubMed]
Levi DM, Carney T. The effect of flankers on three tasks in central, peripheral, and amblyopic vision. J Vis. 2011; 11: 10. [CrossRef] [PubMed]
Reuther J, Chakravarthi R, Martinovic J. Masking, crowding, and grouping: connecting low and mid-level vision. J Vis. 2022; 22: 7. [CrossRef] [PubMed]
Levi DM, Hariharan S, Klein SA. Suppressive and facilitatory spatial interactions in amblyopic vision. Vision Res. 2002; 42: 1379–1394. [CrossRef] [PubMed]
Levi DM, Hariharan S, Klein SA. Suppressive and facilitatory spatial interactions in peripheral vision: peripheral crowding is neither size invariant nor simple contrast masking. J Vis. 2002; 2: 167–177. [PubMed]
Levi DM, Klein SA, Hariharan S. Suppressive and facilitatory spatial interactions in foveal vision: foveal crowding is simple contrast masking. J Vis. 2002; 2: 140–166. [PubMed]
Ester EF, Zilber E, Serences JT. Substitution and pooling in visual crowding induced by similar and dissimilar distractors. J Vis. 2015; 15: 4. [CrossRef] [PubMed]
Rosenholtz R, Yu D, Keshvari S. Challenges to pooling models of crowding: implications for visual mechanisms. J Vis. 2019; 19: 15. [CrossRef] [PubMed]
Kalpadakis-Smith AV, Tailor VK, Dahlmann-Noor AH, Greenwood JA. Crowding changes appearance systematically in peripheral, amblyopic, and developing vision. J Vis. 2022; 22: 3. [CrossRef] [PubMed]
Whitney D, Levi DM. Visual crowding: a fundamental limit on conscious perception and object recognition. Trends Cogn Sci. 2011; 15: 160–168. [CrossRef] [PubMed]
Coates DR, Levi DM. Contour interaction in foveal vision: a response to Siderov, Waugh, and Bedell (2013). Vision Res. 2014; 96: 140–144. [CrossRef] [PubMed]
Coates DR, Levi DM, Touch P, Sabesan R. Foveal crowding resolved. Sci Rep. 2018; 8: 9177. [CrossRef] [PubMed]
Siderov J, Waugh SJ, Bedell HE. Foveal contour interaction for low contrast acuity targets. Vision Res. 2013; 77: 10–13. [CrossRef] [PubMed]
Danilova MV, Bondarko VM. Foveal contour interactions and crowding effects at the resolution limit of the visual system. J Vis. 2007; 7: 25. [CrossRef]
Legge GE, Foley JM. Contrast masking in human vision. J Opt Soc Am. 1980; 70: 1458–1470. [CrossRef] [PubMed]
Foley JM. Lateral effects in pattern vision. J Vis. 2019; 19: 8. [CrossRef] [PubMed]
Chung ST, Levi DM, Legge GE. Spatial-frequency and contrast properties of crowding. Vision Res. 2001; 41: 1833–1850. [CrossRef] [PubMed]
Majaj NJ, Pelli DG, Kurshan P, Palomares M. The role of spatial frequency channels in letter identification. Vision Res. 2002; 42: 1165–1184. [CrossRef] [PubMed]
Flom MC, Heath GG, Takahashi E. Contour interaction and visual resolution: contralateral effects. Science. 1963; 142: 979–980. [CrossRef] [PubMed]
Hairol MI, Formankiewicz MA, Waugh SJ. Foveal visual acuity is worse and shows stronger contour interaction effects for contrast-modulated than luminance-modulated Cs. Vis Neurosci. 2013; 30: 105–120. [CrossRef] [PubMed]
Lalor SJ, Formankiewicz MA, Waugh SJ. Crowding and visual acuity measured in adults using paediatric test letters, pictures and symbols. Vision Res. 2016; 121: 31–38. [CrossRef] [PubMed]
Liu L. Dual oriented mechanisms in contour interaction. Vision Res. 2001; 41: 853–858. [CrossRef] [PubMed]
Siderov J, Pluháček F, Bedell HE. Upturn of the contour-interaction function at small flanking bar-to-target separations. Vision Res. 2020; 167: 1–7. [CrossRef] [PubMed]
Bailey IL, Lovie-Kitchin JE. Visual acuity testing. From the laboratory to the clinic. Vision Res. 2013; 90: 2–9. [CrossRef] [PubMed]
McGraw PV, Winn B, Gray LS, Elliott DB. Improving the reliability of visual acuity measures in young children. Ophthalmic Physiol Opt. 2000; 20: 173–184. [CrossRef] [PubMed]
Atkinson J, Anker S, Evans C, Hall R, Pimm-Smith E. Visual acuity testing of young children with the Cambridge Crowding Cards at 3 and 6 m. Acta Ophthalmol (Copenh). 1988; 66: 505–508. [CrossRef] [PubMed]
Bailey IL, Lovie JE. New design principles for visual acuity letter charts. Am J Optom Physiol Opt. 1976; 53: 740–745. [CrossRef] [PubMed]
Kaiser PK. Prospective evaluation of visual acuity assessment: a comparison of Snellen versus ETDRS charts in clinical practice (an AOS thesis). Trans Am Ophthalmol Soc. 2009; 107: 311–324. [PubMed]
Ferris FL, Bailey I. Standardizing the measurement of visual acuity for clinical research studies: guidelines from the Eye Care Technology Forum. Ophthalmology. 1996; 103: 181–182. [CrossRef] [PubMed]
Manny RE, Fern KD, Loshin DS. Contour interaction function in the preschool child. Am J Optom Physiol Opt. 1987; 64: 686–692. [CrossRef] [PubMed]
Hairol MI, Formankiewicz MA, Waugh SJ. Visual acuity and contour interaction for luminance-modulated and contrast-modulated Cs in normal foveal vision. J Vis. 2010; 10: 1332. [CrossRef]
Polat U, Sagi D. Lateral interactions between spatial channels: suppression and facilitation revealed by lateral masking experiments. Vision Res. 1993; 33: 993–999. [CrossRef] [PubMed]
Waugh S, Formankiewicz M, Pelli D. Cross-optotype metrics for foveal lateral masking. J Vis. 2017; 17: 372. [CrossRef]
Norgett Y, Siderov J. Foveal crowding differs in children and adults. J Vis. 2014; 14: 23. [CrossRef] [PubMed]
Marten-Ellis SM, Bedell HE. A comparison of foveal and peripheral contour interaction and crowding. Optom Vis Sci. 2021; 98: 41–50. [CrossRef] [PubMed]
Norgett Y, Siderov J. Effect of stimulus configuration on crowding in strabismic amblyopia. J Vis. 2017; 17: 5. [CrossRef] [PubMed]
Holmes JM, Beck RW, Repka MX, et al. The amblyopia treatment study visual acuity testing protocol. Arch Ophthalmol. 2001; 119: 1345–1353. [CrossRef] [PubMed]
Hyvärinen L, Näsänen R, Laurinen P. New visual acuity test for pre-school children. Acta Ophthalmol (Copenh). 1980; 58: 507–511. [CrossRef] [PubMed]
Kay H. New method of assessing visual acuity with pictures. Br J Ophthalmol. 1983; 67: 131. [CrossRef] [PubMed]
Hohmann A, Haase W. Development of visual line acuity in humans. Ophthalmic Res. 1982; 14: 107–112. [CrossRef] [PubMed]
Haase W, Hohmann A. [A new test (C-test) for quantitative examination of crowding with test results in amblyopic and ametropic patients (author's transl)]. Klin Monbl Augenheilkd. 1982; 180: 210–215. [CrossRef] [PubMed]
Wesemann W, Heinrich SP, Jägle H, Schiefer U, Bach M. Neue DIN- und ISO-Normen zur Sehschärfebestimmung. Der Ophthalmologe. 2020; 117: 19–26. [CrossRef] [PubMed]
Haase W . Amblyopien. Der Ophthalmologe. 2003; 100: 69–88. [CrossRef] [PubMed]
Huurneman B, Boonstra FN, Cillessen AHN, van Rens G, Cox RFA. Crowding in central vision in normally sighted and visually impared children aged 4 to 8 years: the influence of age and test design. Strabismus. 2012; 20: 55–62. [CrossRef] [PubMed]
Kávássy A, Kaufmann H. Untersuchungen über die hemmende Wirkung der Kontureninteraktion als Teilaspekt der Pathophysiologie der Schielamblyopie [in German]. Spektrum der Augenheilkunde. 1992; 6: 281–284. [CrossRef]
Korinth SP, Gerstenberger K, Fiebach CJ. Wider letter-spacing facilitates word processing but impairs reading rates of fast readers. Front Psychol. 2020; 11: 444. [CrossRef] [PubMed]
Hohmann A, Haase W. Entwicklung der Sehschärfe beim Menschen - ein neuer Test mit standardisierten Einzel- und Reihenoptotypen [in German]. In: Freigang M (ed), Arbeitskreis Schielen. Wiesbaden: BVA Publikationen; 1981: 131–136.
Takahashi ES. Effects of flanking contours on visual resolution at foveal and near-foveal loci, PhD Thesis, School of Optometry: Berkeley: University of California; 1968.
Flynn JT, Schiffman J, Feuer W, Corona A. The therapy of amblyopia: an analysis of the results of amblyopia therapy utilizing the pooled data of published studies. Trans Am Ophthalmol Soc. 1998; 96: 431–450; discussion 450–453. [PubMed]
Flynn JT, Woodruff G, Thompson JR, et al. The therapy of amblyopia: an analysis comparing the results of amblyopia therapy utilizing two pooled data sets. Trans Am Ophthalmol Soc. 1999; 97: 373–390; discussion 390–395. [PubMed]
Hubel DH, Wiesel TN. The period of susceptibility to the physiological effects of unilateral eye closure in kittens. J Physiol. 1970; 206: 419–436. [CrossRef] [PubMed]
von Noorden GK, Crawford ML. The sensitive period. Trans Ophthalmol Soc UK (1962). 1979; 99: 442–446.
Fronius M, Cirina L, Ackermann H, Kohnen T, Diehl CM. Efficiency of electronically monitored amblyopia treatment between 5 and 16 years of age: new insight into declining susceptibility of the visual system. Vision Res. 2014; 103: 11–19. [CrossRef] [PubMed]
Stewart CE, Fielder AR, Stephens DA, Moseley MJ. Treatment of unilateral amblyopia: factors influencing visual outcome. Invest Ophthalmol Vis Sci. 2005; 46: 3152–3160. [CrossRef] [PubMed]
Williams C . Amblyopia. BMJ Clin Evid. 2009; 2009: 0709. [PubMed]
Jeon ST, Hamid J, Maurer D, Lewis TL. Developmental changes during childhood in single-letter acuity and its crowding by surrounding contours. J Exp Child Psychol. 2010; 107: 423–437. [CrossRef] [PubMed]
Semenov L, Chernova N, Bondarko V. Measurement of visual acuity and crowding effect in 3–9-year-old children. Hum Physiol. 2000; 26: 16–20. [CrossRef]
Stewart CE, Fielder AR, Stephens DA, Moseley MJ. Design of the Monitored Occlusion Treatment of Amblyopia Study (MOTAS). Br J Ophthalmol. 2002; 86: 915–919. [CrossRef] [PubMed]
Chen AM, Cotter SA. The amblyopia treatment studies: implications for clinical practice. Adv Ophthalmol Optom. 2016; 1: 287–305. [CrossRef] [PubMed]
Deutsche Ophthalmologische Gesellschaft DOG. Leitlinie 26 a Amblyopie (dog.org). 2010, https://www.dog.org/wp-content/uploads/2009/09/LL-26-Amb-2010-12-29-mit-Inhaltsverz-+Interessenkonfl-Endversion.pdf.
Cüppers C . Moderne Schielbehandlung [Modern treatment of strabismus]. Klin Monbl Augenheilkd Augenarztl Fortbild. 1956; 129: 579–604. [PubMed]
Bonneh YS, Sagi D, Polat U. Local and non-local deficits in amblyopia: acuity and spatial interactions. Vision Res. 2004; 44: 3099–3110. [CrossRef] [PubMed]
Manny RE, Hussein M, Gwiazda J, Marsh-Tootle W. Repeatability of ETDRS visual acuity in children. Invest Ophthalmol Vis Sci. 2003; 44: 3294–3300. [CrossRef] [PubMed]
Chen SI, Chandna A, Norcia AM, Pettet M, Stone D. The repeatability of best corrected acuity in normal and amblyopic children 4 to 12 years of age. Invest Ophthalmol Vis Sci. 2006; 47: 614–619. [CrossRef] [PubMed]
Arditi A, Cagenello R. On the statistical reliability of letter-chart visual acuity measurements. Invest Ophthalmol Vis Sci. 1993; 34: 120–129. [PubMed]
Norgett Y, Siderov J. Crowding in children's visual acuity tests—effect of test design and age. Optom Vis Sci. 2011; 88: 920–927. [CrossRef] [PubMed]
Fronius M, Bachert I, Lüchtenberg M. Electronic monitoring of occlusion treatment for amblyopia in patients aged 7 to 16 years. Graefe Arch Clin Exp Ophthalmol. 2009; 247: 1401–1408. [CrossRef]
Mehmed B, Fronius M, Pohl T, et al. Electronically monitored occlusion therapy in amblyopia with eccentric fixation. Graefes Arch Clin Exp Ophthalmol. 2022; 260: 1741–1753. [CrossRef] [PubMed]
Stewart CE, Stephens DA, Fielder AR, Moseley MJ. Modeling dose-response in amblyopia: toward a child-specific treatment plan. Invest Ophthalmol Vis Sci. 2007; 48: 2589–2594. [CrossRef] [PubMed]
Pelli DG, Waugh SJ, Martelli M, et al. A clinical test for visual crowding. F1000Research. 2016; 5: 81. [CrossRef]
Waugh S, Pelli D, Álvaro L, Formankiewicz M. Crowding distance in healthy children. J Vis. 2018; 18: 855. [CrossRef]
Levi DM, Waugh SJ, Beard BL. Spatial scale shifts in amblyopic vision. Vision Res. 1994; 34: 3315–3333. [CrossRef] [PubMed]
Levi DM, Klein S. Differences in vernier discrimination for grating between strabismic and anisometropic amblyopes. Invest Ophthalmol Vis Sci. 1982; 23: 398–407. [PubMed]
Fronius M, Sireteanu R, Zubcov A. Deficits of spatial localization in children with strabismic amblyopia. Graefes Arch Clin Exp Ophthalmol. 2004; 242: 827–839. [CrossRef] [PubMed]
Bedell HE, Flom MC. Monocular spatial distortion in strabismic amblyopia. Invest Ophthalmol Vis Sci. 1981; 20: 263–267. [PubMed]
Fronius M, Sireteanu R. Monocular geometry is selectively distorted in the central visual field of strabismic amblyopes. Invest Ophthalmol Vis Sci. 1989; 30: 2034–2044. [PubMed]
Cüppers C. Grenzen und Möglichkeiten der pleoptischen Therapie [in German]. Klin Monbl Augenheilkd. 1961; 38: 33–98.
Heinrich SP, Krüger K, Bach M. The dynamics of practice effects in an optotype acuity task. Graefes Arch Clin Exp Ophthalmol. 2011; 249: 1319–1326. [CrossRef] [PubMed]
Hu F, Hu Y, Ma Z, Rosenberger WF. Adaptive randomization for balancing over covariates. WIREs Comput Stat. 2014; 6: 288–303. [CrossRef]
Von Noorden GK, Campos EC. Binocular Vision and Ocular Motility: Theory and Management of Strabismus. 6th Ed. St. Louis, MO: Mosby; 2002.
Kracht J, Bachert I, Diehl C, et al. Electronically recorded occlusion treatment in amblyopes older than 7 years: acuity gain after more than 4 months of treatment? Klin Monbl Augenheilkd. 2010; 227: 774–781. [CrossRef] [PubMed]
Waugh SJ, Haine LA, Formankiewicz MA, Pelli DG. Crowding distance beats acuity and crowded acuity in detecting strabismic amblyopia. J Vis. 2022; 22: 4370. [CrossRef]
Facchin A, Maffioletti S, Martelli M, Daini R. Different trajectories in the development of visual acuity with different levels of crowding: the Milan Eye Chart (MEC). Vision Res. 2019; 156: 10–16. [CrossRef] [PubMed]
Clark A, Intoy J, Yang H, Poletti M. Fixational eye movements and crowding in the foveola. J Vis. 2020; 20: 1514. [CrossRef]
Levi DM, Klein SA, Aitsebaomo AP. Vernier acuity, crowding and cortical magnification. Vision Res. 1985; 25: 963–977. [CrossRef] [PubMed]
Westheimer G, Hauske G. Temporal and spatial interference with vernier acuity. Vision Res. 1975; 15: 1137–1141. [CrossRef] [PubMed]
Figure 1.
 
Sections of C-tests (1.0 to 1.4 logMAR) scanned from physical versions (Oculus, Wetzlar, Germany). In (a) and (b) separation on the left is specified as 2.6’ and >35’ on the right. (a) Shows unblurred charts used for testing. (b) Shows a hypothetical visualization of Gaussian blur. When 2.6’ separation Landolt Cs are not resolvable due to blur, ability to localize the gap improves versus 35’ separation (top line, b). Actual separations varied from those specified (see Supplementary Fig. S1).
Figure 1.
 
Sections of C-tests (1.0 to 1.4 logMAR) scanned from physical versions (Oculus, Wetzlar, Germany). In (a) and (b) separation on the left is specified as 2.6’ and >35’ on the right. (a) Shows unblurred charts used for testing. (b) Shows a hypothetical visualization of Gaussian blur. When 2.6’ separation Landolt Cs are not resolvable due to blur, ability to localize the gap improves versus 35’ separation (top line, b). Actual separations varied from those specified (see Supplementary Fig. S1).
Figure 2.
 
Crowding magnitudes (logMAR) from Equation 1 calculated for C-tests (RE and LE of controls and FE and AE of amblyopes) for participants in three groups (control, aniso, and strab) and both age groups (pediatric and juvenile/adult). Stronger colors are data for juvenile/adult participants (>8 years). Lighter colors are data for pediatric participants (≤8 years). Horizontal and vertical bars with * indicate statistically significant differences between groups (Tukey post hoc testing P < 0.0001; all other results were P > 0.05).
Figure 2.
 
Crowding magnitudes (logMAR) from Equation 1 calculated for C-tests (RE and LE of controls and FE and AE of amblyopes) for participants in three groups (control, aniso, and strab) and both age groups (pediatric and juvenile/adult). Stronger colors are data for juvenile/adult participants (>8 years). Lighter colors are data for pediatric participants (≤8 years). Horizontal and vertical bars with * indicate statistically significant differences between groups (Tukey post hoc testing P < 0.0001; all other results were P > 0.05).
Figure 3.
 
Crowding magnitudes (logMAR) from Equation 1 plotted against 35’ C-test acuity (logMAR) for all participants including control and amblyopic eyes. Two diagonal comparison lines are provided. One is based on crowding magnitude data from the normal adult periphery4 scaled to match control foveal data from current study (slope = approximately 0.6). In that study, visual acuity for a letter surrounded by four letters versus an isolated letter was calculated. The other comparison line shows crowding magnitude increasing directly with acuity (slope = approximately 1). The horizontal line at 0 indicates equal visual acuity for 2.6’ and 35’ C-tests, so crowding magnitude = 0. The horizontal line with grey shading represents average crowding magnitude (shading = ± 1 SE) for control eyes. Individual symbols are data from five amblyopes who had isolated visual acuities and crowding magnitudes recalculated (Equation 2).
Figure 3.
 
Crowding magnitudes (logMAR) from Equation 1 plotted against 35’ C-test acuity (logMAR) for all participants including control and amblyopic eyes. Two diagonal comparison lines are provided. One is based on crowding magnitude data from the normal adult periphery4 scaled to match control foveal data from current study (slope = approximately 0.6). In that study, visual acuity for a letter surrounded by four letters versus an isolated letter was calculated. The other comparison line shows crowding magnitude increasing directly with acuity (slope = approximately 1). The horizontal line at 0 indicates equal visual acuity for 2.6’ and 35’ C-tests, so crowding magnitude = 0. The horizontal line with grey shading represents average crowding magnitude (shading = ± 1 SE) for control eyes. Individual symbols are data from five amblyopes who had isolated visual acuities and crowding magnitudes recalculated (Equation 2).
Figure 4.
 
(a) (left, pediatric group) and (b) (right, juvenile/adult group). Crowding magnitudes (logMAR) from Equation 1 plotted against 35’ C-test acuity (logMAR) for children ≤8 years (a, left) and juvenile/adults >8 years (b, right) including control and amblyopic eyes. Other details match those for Figure 3 but horizontal lines with grey shading represent average crowding magnitudes (shading = ± 1 SE) for control eyes in each group. Individual symbols are data from amblyopes who had isolated visual acuitities and crowding magnitudes recalculated (Equation 2).
Figure 4.
 
(a) (left, pediatric group) and (b) (right, juvenile/adult group). Crowding magnitudes (logMAR) from Equation 1 plotted against 35’ C-test acuity (logMAR) for children ≤8 years (a, left) and juvenile/adults >8 years (b, right) including control and amblyopic eyes. Other details match those for Figure 3 but horizontal lines with grey shading represent average crowding magnitudes (shading = ± 1 SE) for control eyes in each group. Individual symbols are data from amblyopes who had isolated visual acuitities and crowding magnitudes recalculated (Equation 2).
Figure 5.
 
Interocular differences (IODs) in VA obtained for C-tests (35’ and 2.6’ separations) for all five subgroups (control, low aniso, high aniso, low strab, and high strab) and both age groups (pediatric and juvenile/adult). Stronger colors show data for juvenile/adult participants and lighter colors are data for pediatric participants. Horizontal bars with * indicate statistically significant differences between groups (Tukey post hoc testing P < 0.0001; all other results were P > 0.05). The horizontal bar with P < 0.01 is the result from a planned comparison. Data from control eyes (control 35’ and control 2.6’) were statistically different from all other subgroups but bars are omitted for clarity.
Figure 5.
 
Interocular differences (IODs) in VA obtained for C-tests (35’ and 2.6’ separations) for all five subgroups (control, low aniso, high aniso, low strab, and high strab) and both age groups (pediatric and juvenile/adult). Stronger colors show data for juvenile/adult participants and lighter colors are data for pediatric participants. Horizontal bars with * indicate statistically significant differences between groups (Tukey post hoc testing P < 0.0001; all other results were P > 0.05). The horizontal bar with P < 0.01 is the result from a planned comparison. Data from control eyes (control 35’ and control 2.6’) were statistically different from all other subgroups but bars are omitted for clarity.
Table 1.
 
Participant Group and Eye Details
Table 1.
 
Participant Group and Eye Details
Table 2.
 
Slopes ± 1 SD from Figures 3 and 4a,b
Table 2.
 
Slopes ± 1 SD from Figures 3 and 4a,b
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×