Attention-based Binocular Training for Impaired Stereopsis (BRAVE)

Measured using the Randot Stereotest - Circles test (graded disparities from 400 to 20 arcsec, 10 levels) under standardized lighting at 40 cm. The score is recorded as the finest disparity correctly identified (in log arcseconds). Nil stereopsis is assigned 3000 arcsec (3.477 log arcsec). The primary metric is the between-group difference in the change score.

Measured using the Randot Stereotest - Circles test under crossed disparity (log10 arcsec; nil stereopsis assigned 3000 arcsec) to evaluate long-term maintenance of training effects

Measured using the Randot Stereotest - Circles test with the test booklet rotated 180 degrees (log10 arcsec; nil stereopsis assigned 3000 arcsec).

Measured with the Randot Preschool Stereotest (random-dot stereograms, 800 to 40 arcsec, 6 levels) under both crossed and uncrossed disparity conditions (log10 arcsec; nil stereopsis assigned 3000 arcsec).

Psychophysical stereoacuity threshold (75% threshold in log10 arcsec) measured using a 4-alternative forced-choice staircase ring test administered dichoptically.

Monocular (affected/dominant eyes) and binocular distance visual acuity measured using an electronic ETDRS logMAR chart at 4 meters (recorded in logMAR units).

Binocular status classified into categorical states (fusion, suppression, or diplopia) using the Worth 4-dot test at near (40 cm) and distance (3 m).

Effective suppression depth quantified by the interocular contrast ratio at the perceptual balance point measured via a dichoptic letter-polarity task.

Peak latencies (ms) and amplitudes (μV) of the transient pattern-reversal VEP N75-P100-N135 complex across high-contrast and isoluminant red-green conditions, alongside harmonic amplitudes/phases (2F, 4F) from steady-state VEPs, recorded via a 64-channel EEG system.

Mean amplitude (μV) of the posterior contralateral-minus-ipsilateral difference wave (N2pc) extracted across predefined parieto-occipital electrodes (P7, PO7, P8, PO8) during a modified spatial cueing task.

Spatial cueing effects on mean reaction time (ms) and response accuracy (%) derived from the modified spatial cueing task.

Visual search slope (reaction time as a function of set size, in ms/item) measured from a 192-trial zero-disparity dichoptic visual search task.

Visuomotor dexterity and kinematic performance are assessed using the Grooved Pegboard Test separately for dominant and non-dominant hands. Hand movements during the task are video recorded for detailed kinematic tracking. The primary quantitative metric for this outcome is the total completion time (in seconds) required to successfully place all 25 grooved pegs (shorter duration indicates better motor speed and coordination). Secondary kinematic metrics extracted from video analyses include mean peg-insertion duration (seconds), inter-peg temporal variability across the 25 trials (coefficient of variation, CV), and total drop/error counts.

Domain scores assessed via the World Health Organization Quality of Life Instrument-Short Form (WHO QoL-BREF) and the Amblyopia and Strabismus Questionnaire (A&SQ). Higher scores indicate superior functional visual ability and quality of life. For both questionnaires, domain and composite scores are linearly transformed to a standardized 0 to 100 scale, where higher scores represent superior functional visual ability, fewer daily limitations, and better overall quality of life.