Inclusion and Exclusion criteria too stringent, no patient coluld be recruited

Amblyopia, affects 4% of the population. In the presence of a normal retina and optic nerve, the visual cortex does not develop normally secondary to the amblyopic eye being at a disadvantage in the sensitive period such as in strabismus. Amblyopia is associated with visuospatial disorders, and poor to absent stereopsis. Amblyopia is treated by occlusion of the non- amblyopic eye in childhood. The present study aims to evaluate the effectiveness of perceptual learning (PL), that is a training of visual drawing tasks (12 sessions of 30 Minutes duration) with crowding (c) for children with amblyopia compared to an amblyopic control group with a placebo drawing task.

Arm 1: Crowded (PLc)- Training with small spacing Arm 2: Uncrowded (PLu)- Training with large spacing Prior to this, a small group will help optimize parameters such as contrast and distance of visual objects presented

Outcome(s):

Primary endpoints: improvement of amblyopia,crowding ratio (CR) and visual acuity before and after training, at 7 weeks, at 6 months Secondary outcome: hand-eye coordination, evaluated with the Beery VMI test before and after training, at 7 weeks, at 6 months Tertiary outcome over 6 years of age: macular structure measured with the OCT before and after training, at 7 weeks, at 6 months

Study design:

Cohort study, randomised controlled trial, single blind, active control, parallel groups

Inclusion / Exclusion criteria:

• 20 Children included to optimize test criteria (high vs low contrast and spacing).
• 40 Children for 2-armed study.

Measurements and procedures:

12 sessions (2x per week for 6 weeks, 30 Minutes each) where drawing tasks are performed during their occlusion time.

Arm 1: Crowded - task with small spacing Arm 2: Uncrowded task with large spacing. Prior to this, a small group will help optimize parameters such as contrast and distance of visual objects presented The following parameters are evaluated at onset, at week 7 and at 6 months: visual acuity with and without crowding, the Berry Test and in children over 6 years of age, if possible: a reading task and OCT.

Statistical Considerations:

Patients will be randomized by letting a computer program perform the assignment of children in a certain training group. However, some restrictions will be inserted on the amount of randomization: equal number of children in our training groups, age- and acuity matched groups, equal patching times.

The analysis is aimed at answering the primary question: Can the training reduce crowding and improve visual acuity in children with amblyopia? In order to answer this question, a repeated measures ANOVA is conducted in which the crowding ratio (CR) before and after training will be entered as a within-subjects variable (2 levels: CR pre, and CR post/VA pre, VA post). Training group will be entered as a between subjects variable (2 levels: control and experimental). If there is an interaction effect, post hoc ANOVA's will be conducted to disentangle the effect of training on the CR within training groups.

A secondary question is: Can the training improve hand-eye coordination in children with amblyopia. To this regard the results of the Beery VMI test are being evaluated with a repeated measures ANOVA.

Sample size Calculation. As the crowding ratio (CR) is to be expected more important than VA, only this parameter will be powered.

Main parameter for the power calculation will be the difference (CRd) of the CR at the end of the intervention period compared to the CR at screening.

A CR decrease of 0.3 between "control" and "learning" group is assumed to be of clinical relevance.

The population standard deviation (SD) of CR is assumed to be 0.33. Assuming a moderate correlation coefficient of 0.5 between timepoints, the within subject SD is calculated as sqrt(0.33*0.33*0.5)=0.23. Therefore the SD of CRd is sqrt(2)*0.23=0.33. [Details are described in Sample Sizes for Clinical Trials,Steven A . Julious,Chapman and Hall/CRC 2009].

Based on a sample size of 20 subjects for each study group, there is a power of 88% to detect a CRd decrease of 0.3. This calculation is based on a one-sided T-test (alpha=5%).

GCP Statement: This study will be conducted in compliance with the protocol, the current version of the Declaration of Helsinki, the ICH-GCP or ISO EN 14155 (as far as applicable) as well as all national legal and regulatory requirements.

Inclusion Criteria: children aged 4-8 years with

• strabismic amblyopia
• cycloplegic refraction corrected at least 3 months prior to enrolment
• Reliable test results for two previous examinations
• absence of ocular conditions other than strabismus, amblyopia and corrected refractive error

Exclusion Criteria:

• inability to perform reliably when examined
• inability to draw
• unstable visual results on 2 previous examinations