Visuo-Cognitive Executive Functions in Amblyopia and Strabismus: Associations With Function and Quality of Life

Abstract

Purpose

The purpose of this study was to evaluate higher-order visual processing and executive function in adults with amblyopia and/or strabismus under habitual binocular viewing conditions and to explore their association with functional capabilities and vision-related quality of life (VRQOL).

Methods

This study included 114 adults (23 with amblyopia, 52 with amblyopia and strabismus, 20 with strabismus, and 19 healthy controls). Assessments included visual acuity and binocular function score (BFS), visuo-cognitive tests (Useful Field of View [UFOV], Trail Making Test [TMT], and Stroop Color and Word Test [SCWT]), and functional tests (fine motor skills [FMS] and reading rate). Group differences were evaluated using ANOVA, and regression models assessed the contribution of clinical and cognitive measures to functional outcomes and VRQOL.

Results

Performance significantly varied across groups (F _(3,108) = 753, P < 0.001), for all three visuo-cognitive tests (UFOV F _(3,108) = 2.982, P = 0.034; TMT F _(3,108) = 5.289, P = 0.002; and SCWT F _(3,108) = 4.068, P = 0.009). Adults with amblyopia and/or strabismus had significantly slower completion times on UFOV, TMT, and SCWT compared with healthy controls (P < 0.05), with pronounced deficits in divided attention, visual search, and cognitive flexibility. Performance among the three visuo-cognitive tests was strongly correlated (r’s³ = 0.7, P = 0.01). Regression analyses revealed that SCWT times significantly predicted FMS and reading speed, explaining 42% and 31% of the variance, respectively. Psychosocial VRQOL scores were significantly influenced by the presence of strabismus (P = 0.001), whereas functional impact scores were associated with BFS (P = 0.04) and UFOV times (P = 0.02).

Conclusions

Amblyopia and/or strabismus in adults are associated with executive function deficits, poorer functional performance, and reduced VRQOL. These findings highlight the long-term visuo-cognitive and functional consequences of abnormal visual development and underscore the need for targeted interventions.

Amblyopia and strabismus are common vision disorders characterized by disrupted binocular vision. These conditions are associated with reduced monocular and binocular visual performance, including deficits in visual acuity, contrast sensitivity, and vernier acuity, as well as diminished or absent stereopsis and oculomotor anomalies.¹ Whereas amblyopia arises in childhood, typically from anisometropia, strabismus, or deprivation during early visual development, it remains the most frequent cause of monocular vision loss from childhood through to middle age, highlighting its long-term clinical importance.^2–4 Although amblyopia and strabismus often co-occur, strabismus does not always lead to amblyopia but frequently results in absent or severely degraded stereopsis.⁵ Past studies have demonstrated that amblyopia and strabismus impact functional capabilities, such as fine motor skills and reading speed.^6–9 Recent work conducted by our group confirmed that adults with amblyopia and/or strabismus exhibit poorer manual dexterity and slower reading rates compared with those with normal visual development, with these functional deficits only being partially explained by clinical measures of vision, such as visual acuity and binocular function.¹⁰ The limited explanatory power of clinical vision measures highlights the need to explore further contributors to the observed functional impairments.

Executive functions—including visual attention, processing speed, and cognitive flexibility—are essential for daily tasks, such as navigating crowded environments, driving, reading, and engaging in visually guided activities.¹¹ These higher-order visuo-cognitive processes enable selective focus on relevant stimuli, effective decision making, and task switching while filtering out distractions.^12–15 Previous research has shown that individuals with amblyopia experience deficits in visual attention and processing speed, often manifesting as slower reaction times and reduced accuracy performing complex tasks.^11,16–22 Recent reviews have highlighted persistent attention deficits in individuals with amblyopia, suggesting that early visual disruption may have enduring effects on attentional networks.²³ Deficits in visual search efficiency, spatial attention, and attentional engagement have been reported, although some aspects like covert attention may remain intact—highlighting selective but persistent visuo-cognitive impairments.^24–27

Although growing evidence reports the impact of amblyopia on broader visuo-cognitive function, most studies have assessed using monocular testing paradigms, limiting the extent to which findings can be generalized and applied to real-life settings. However, recent studies have demonstrated that even under binocular viewing conditions, children with amblyopia continue to exhibit deficits in higher-order visual processing skills—such as visual attention, visual search, and visual memory.^16,22 Nonetheless, it remains unclear whether such visuo-cognitive deficits persist into adulthood and how they relate to everyday functional performance. Demonstrating persistent deficits would suggest that early visual disruption has lasting neurocognitive consequences beyond the critical period of visual development.

This study aimed to evaluate performance on standardized cognitive tasks assessing higher-order visuo-cognitive functions in adults with and without amblyopia and/or strabismus, and to examine the relationship between these functions and tasks relevant to everyday performance. A secondary aim was to determine the relative contributions of visual acuity and binocular function to visuo-cognitive and functional outcomes. The current study builds on previous work reporting impairments in fine motor skills and reading proficiency among adults with amblyopia and/or strabismus.¹⁰ Importantly, the inclusion of a non-amblyopic strabismus group allowed us to explore whether executive function deficits are primarily attributable to reduced visual acuity, or whether abnormal binocularity alone may also impact visuo-cognitive and functional performance. This group often receives less clinical attention than those with amblyopia, despite potential challenges in functional and psychosocial domains. Specifically, divided and selective visual attention, visual search and scanning, and cognitive flexibility were assessed under binocular viewing conditions, and their relationship to functional capabilities manual dexterity and reading speed was evaluated. By examining these higher-order cognitive processes, the study sought to provide a more comprehensive understanding of the broader consequences of amblyopia and strabismus in adulthood, including contributors to reduced vision-related quality of life (VRQOL).

Methods

Participants

A total of 114 adults, aged 18 to 40 years, were recruited from the outpatient department of the L. V. Prasad Eye Institute (LVPEI) in Bhubaneswar, India. Twenty-three of the adults had amblyopia without strabismus (etiology: anisometropic, n = 20 and deprivation, n = 3), 52 had both amblyopia and strabismus (etiology: strabismic, n = 20; mixed, n = 20; and deprivation, n = 12), and 20 had strabismus without amblyopia, and 19 served as controls with normal visual development. Participants in the amblyopia groups had a confirmed amblyogenic factor prior to age 8 and reduced visual acuity despite appropriate refractive correction and/or surgical alignment at least 16 weeks prior. Amblyogenic factors included anisometropia (≥1.00 diopters [D] spherical or ≥1.50 D astigmatic difference between eyes), constant unilateral strabismus, or monocular image degradation (e.g., cataract). Inclusion criteria for the amblyopia groups were best-corrected visual acuity (BCVA) of 0.2 logMAR or worse in the amblyopic eye, 0.1 logMAR or better in the fellow eye, and an interocular difference ≥0.2 logMAR. Inclusion criteria for the strabismus groups were a constant deviation of ≥10 prism diopters (PD) horizontally or ≥4 PD vertically, confirmed by alternating cover test. Nineteen visually healthy adults with orthotropic alignment were recruited as controls. All participants wore required refractive correction during testing.¹⁰

Clinical diagnosis categories included anisometropic amblyopia, strabismic amblyopia, mixed anisometropic-strabismic amblyopia, and deprivation amblyopia. Data on cycloplegic refraction (within the preceding 12 months) and treatment history were extracted from the medical records. All participants were free of known neurological, intellectual, or other ocular disorders.

This study was approved by the research ethics committees of LVPEI, Bhubaneswar, India (Approval code: 2020-61-BHR-35), and Queensland University of Technology, Brisbane, Australia (Approval number: 1700000249). All procedures adhered to the Declaration of Helsinki guidelines, and informed consent was obtained from all participants.

All data presented in this paper were collected as part of a single testing session that included assessments of executive function, functional performance skills, and VRQOL. Although the current paper focuses on executive function outcomes, data related to fine motor skills and reading performance from the same participant group have been reported separately.¹⁰

Visual Function Assessment

Visual acuity was measured monocularly and binocularly using the electronic Early Treatment Diabetic Retinopathy Study (ETDRS) standardized logMAR chart. Binocular function scores (BFS) were calculated based on stereoacuity (log arcsec) assessed with the Randot Preschool Stereoacuity Test (Stereo Optical Co., Inc.) and the presence or absence of suppression determined using the Worth 4 Dot App (iPad display, working distance 90 cm).^28,29

Executive Function Assessments

Aspects of executive function were assessed using the Useful Field of View (UFOV), Trail Making Test (TMT), and Stroop Color-Word Test (SCWT). UFOV and TMT tasks used in this study have previously been used in a pediatric cohort with primarily anisometropic amblyopia.¹⁶ Whereas UFOV and TMT primarily assess visual attention, processing speed, visual search, and cognitive flexibility, the SCWT uniquely evaluates inhibitory control and cognitive interference.

Visual processing speed for increasingly complex tasks that require divided and selective attention were assessed with a web-based commercially available version of UFOV (https://www.brainhq.com/partners/brainhq-for-clinicians/ufov/). Subtest 1 (Central Processing) involved identifying a central target, subtest 2 (Divided Attention) required simultaneous identification of a central target and localization of a peripheral target (9 degrees from fixation), and subtest 3 (Selective Attention) included central target identification with localization of a peripheral target embedded among distractors. All tasks were presented on a 35 × 20 cm computer monitor viewed at a 32.5 cm working distance, with white targets displayed on a black background (subtending 100 minutes arc, approximately 6/120). Processing speed was recorded as the minimum presentation time at which participants could accurately perform the task 75% of the time, and practice trials were provided for each subtest. Subtest scores were summed to produce an overall UFOV processing time (total UFOV). Completion time for the 3 subtests was approximately 10 minutes.³⁰

Visual search and scanning, psychomotor speed, and cognitive flexibility were assessed using static and dynamic presentations of the TMT, with the computer-based version requiring participants to use a mouse to select optotypes.^14,31,32 TMT-A involved connecting numbers in ascending order, whereas TMT-B alternated between numbers and letters in sequence (e.g., 1-A-2-B). Each task was conducted in two versions: static (targets remained stationary) and dynamic (targets changed position after each response). Tests were presented on a 35 × 20 cm computer monitor at a 40 cm working distance, with black optotypes displayed on a white background (48 minutes arc, approximately 6/60). Completion times (in seconds) were recorded, and the tasks progressed only when correct responses were provided. Scores for the four subtests (TMT-A static, TMT-A dynamic, TMT-B static, and TMT-B dynamic) were summed for an overall visual search and completion time (total TMT).

Inhibitory control and cognitive flexibility were evaluated using a commercially available app version of the SCWT, EncephalApp Stroop (https://encephalapp.com/).³³ In the OFF state, participants identified the color of hash marks (e.g., ###), whereas in the ON state, they identified the color of incongruent word stimuli (e.g., the word “RED” displayed in blue). The task was presented on an iPad (21 × 15.5 cm) viewed at a 40 cm working distance. After two practice runs, performance was measured for both states, the program recording time to complete five correct ON and five correct OFF runs, total time (OFF + ON), and the difference between ON and OFF times (ON − OFF). The combined OFF + ON time was used to reflect overall task performance, as this approach captures both baseline processing speed and the added cognitive load of interference resolution.

Functional Skills and Health-Related Quality of Life Assessments

Fine motor skills (FMS) were assessed using the manual dexterity (MD) subtest of the Bruininks-Oseretsky Test of Motor Proficiency, Second Edition (BOT-2).³⁴ The MD subtest comprises five timed tasks requiring hand/finger dexterity and speed (dotting, penny placement, peg displacement, shape sorting, and bead stringing). Tasks were performed with the preferred hand, except the bimanual penny task. Correct completions in 15 seconds were summed for a total point score, which was converted to a scale score using normative data. Reading performance was assessed using three sixth-grade level English texts (texts 3, 5, and 7) from the standardized International Reading Speed Texts (IReST) that is designed to reflect everyday reading.³⁵ Participants read aloud; audio recordings were analyzed for: (i) reading time (s), (ii) incorrect words, (iii) correct words, and (iv) reading speed (words per minute [wpm] = correct words ÷ time × 60).^35,36 Self-reported VRQOL was assessed in participants with amblyopia and/or strabismus using the 26-item Amblyopia and Strabismus Questionnaire (A&SQ), that covers 5 domains: (1) fear of losing the better eye, (2) distance estimation, (3) visual disorientation, (4) diplopia, and (5) social contact/appearance.^37–39 Scores ranged from 0 (maximum impact) to 100 (no impact). The total score is the average of all items. Psychosocial and functional impact scores were derived from domains 1 and 5 and domains 2 to 4, respectively. Detailed methods and analyses for these measures for this participant cohort have been previously described.¹⁰

Statistical Analysis

Differences between the control and combined abnormal binocular vision groups were compared using the independent samples t-test. Cohan's d was calculated to determine the effect size of significant differences [Cohen's d = (M₁–M₂)/SD_pooled where SD_pooled = √[(SD₁² + SD₂²)/2]].

A Multivariate Analysis of Variance (MANOVA) was conducted with the four participant etiology groups as the between-subjects factor and the total scores from three executive function tests (total UFOV, total TMT, and combined OFF + ON SCWT) as the dependent variables. A mixed-design ANOVA was performed for each executive function test, with participant group (amblyopia without strabismus, amblyopia with strabismus, strabismus only, and controls) as the between-subjects factor and subtest (e.g. central processing, divided attention, and selective attention for the UFOV task) as the within-subjects factor. Where a significant main effect of group was found, tests of simple effects of group were conducted for each subtest, followed by Bonferroni post hoc pairwise comparisons as appropriate. Effect sizes were calculated using partial eta squared (${\eta }_{\mathrm{p}}^2$) in the ANOVA, with benchmarks of 0.01 for a small effect, 0.06 for a medium effect, and 0.14 for a large effect.

Multiple regression models were used to evaluate the influence of clinical vision measures—including interocular visual acuity difference (IOVA), binocular visual acuity, BFS, and the presence of strabismus—on functional and cognitive outcomes. One model was developed for each executive function task: total UFOV (sum of all 3 UFOV subtests), total TMT (sum of all 4 TMT subtests), and Stroop OFF + ON (combined score of ON and OFF conditions). The dependent variable for each executive function task was selected to reflect overall task performance by summing or combining subtest scores, providing a comprehensive measure of the targeted cognitive domain. Additional models were conducted for fine motor skills, reading speed, and two VRQOL outcomes. Strabismus was entered as a binary variable (0 = absent and 1 = present). Pearson Correlation examined relationships between functional outcomes (fine motor skills and reading speed), executive function tests, and clinical vision measures. Statistical significance was defined as P < 0.05. Multicollinearity was defined as variance inflation factors (VIFs) greater than 5. IBM SPSS version 25 was used for all analyses.

Results

Age, clinical vision measures, manual dexterity scores, calculated reading rate, and VRQOL scores for all participant groups are summarized in Table 1. These data have been previously reported.¹⁰

Table 1.

Age, Vision Characteristics, Functional Skills, and Vision Related Quality of Life (n = 114)

Group	Amblyopia (A) (n = 23)	Amblyopia and Strabismus (AS) (n = 52)	Strabismus (S) (n = 20)	Control (C) (n = 19)	Group Comparison: Factorial ANOVA F (3,110), P Value
Age, y	22 (4)	24 (5)	27 (6)	25 (6)	F = 2.7, P = 0.05
Vision
VA worse eye/non-dominant eye (logMAR)	0.51 (0.25)*,†,‡	0.71 (0.27)*	−0.01 (0.05)†	−0.07 (0.07)	F = 91.6, P < 0.01
VA better eye/dominant eye (logMAR)	−0.01 (0.08)*	−0.01 (0.07)*	−0.04 (0.05)	−0.08 (0.06)	F = 5.4, P < 0.01
Interocular VA difference (IOVA) (logMAR)	0.53 (0.27)*,†,‡	0.72 (0.29)*	0.04 (0.05)†	0.05 (0.04)	F = 61.9, P < 0.01
Binocular VA (logMAR)	−0.03 (0.09)*	−0.03 (0.06)*	−0.06 (0.04)*	−0.14 (0.07)	F = 12.8, P < 0.01
Binocular function score (BFS)	3.6 (1)*,†	4.7 (0.6)*	3.5 (1.4)*,†	1.6 (0.1)	F = 65.3, P < 0.01
Functional skills
Manual dexterity (scale score)	7.6 (2.6)*,†	5.9 (1.3)*	6.9 (1.9)*	11.4 (4)	F = 27.0, P < 0.01
Calculated reading speed (wpm)	122.7 (40)*	118.1 (42.4)*	122 (46)*	164.5 (25.6)	F = 6 .5, P = 0.01
Vision related quality of life
Functional Impact Score	85.2 (12.9)	86.4 (12.3)	85.6 (13.7)	N/A	F = 0.09, P = 0.92
Psychosocial Impact Score	79.9 (16.0)*,†	65.3 (21.9)	54.5 (26.5)	N/A	F = 7.5, P < 0.01

The figures in bold represent significant outcomes.

Data are mean (standard deviation). Post hoc LSD:

^*Significant difference between abnormal binocular vision group and control.

^†Significant difference between A and AS, or S and AS.

^‡Significant difference between A and S.

Executive Function

The mean and standard deviation of UFOV, TMT, and SCWT outcomes of the control participants and those with abnormal binocular vision from amblyopia and/or strabismus are summarized in Table 2. Participants with abnormal binocular vision had poorer executive function test outcomes than controls (P < 0.05, Cohen's d = 0.36 to 1.18).

Table 2.

Useful Field of View, Trail Making Test, and Stroop Color and Word Test Outcomes (n = 114)

	Normal Controls and Combined Abnormal Binocular Vision Participants Mean (SD)			Sub-Group Results Mean (SD)
Group	Control (C) (n = 19)	Abnormal Binocular Vision (BV) Groups (n = 95)	(Abnormal BV Groups versus C) Independent t-test, P Value, Cohen's d	Amblyopia (A) (n = 23)	Amblyopia and Strabismus (AS) (n = 52)	Strabismus (S) (n = 20)
Useful field of view
Subtest 1: Central Processing, ms	17 (0)	19.9 (11.5)	t = 2.5, P = 0.02,	17.3 (1.4)	21.5 (14.5)	18.8 (8.1)
			d = 0.36
Subtest 2: Divided Attention, ms	50.4 (25.2)	84.1 (57.1)	t = 4.1, P < 0.01	70.7 (52.6)	90.8 (60.7)	81.9 (51.7)
			d = 0.76
Subtest 3: Selective Attention, ms	125.3 (66.5)	162.8 (77.3)	t = 2.0, P = 0.05	143.6 (68.5)	167.6 (80.6)	172.5 (77.9)
			d = 0.52
UFOV 1 + 2 + 3, ms	192.7 (87.2)	266.8 (123.7)	t = 2.5, P = 0.01	231.6 (105.8)	279.8 (129.9)	273.3 (124.4)
			d = 0.70
Trail Making Test
TMT-A Static, s	31.4 (6.6)	46.7 (18.8)	t = 6.3, P < 0.01	44.4 (13.6)	50.9 (22.1)	38.6 (10.1)
			d = 1.09
TMT-A Dynamic, s	44.2 (6.4)	58.2 (23.9)	t = 4.9, P < 0.01	57.6 (22.6)	61.0 (26.9)	51.4 (15.0)
			d = 0.8
TMT-B Static, s	40.2 (9.4)	61.3 (25.6)	t = 6.2, P < 0.01	57.4 (20.1)	64.3 (28.7)	57.7 (22.4)
			d = 1.09
TMT-B Dynamic, s	55.0 (14.2)	79.5 (41.4)	t = 4.6, P < 0.01	71.7 (26.5)	87.7 (51.1)	67.1 (15.1)
			d = 0.79
Sum of TMT static, A + B	71.6 (14.0)	108.0 (41.3)	t = 6.9, P < 0.01	101.8 (31.8)	115.2 (47.7)	96.3 (28.7)
			d = 1.18
Sum of TMT dynamic, A + B	99.1 (17.6)	137.6 (62.9)	t = 5.1, P < 0.01	129.2 (47.5)	148.7 (75.7)	118.5 (27.8)
			d = 0.83
Total TMT time, s	170.7 (29.0)	245.6 (99.5)	t = 6.2, P < 0.01	231.0 (77.1)	264.0 (117.7)	214.8 (52.2)
			d = 1.02
Stroop Color and Word Test
OFF time, s	63.9 (7.7)	75.1 (16.4)	t = 4.6, P < 0.01	71.4 (17.1)	76.8 (16.6)	74.7 (15.2)
			d = 0.87
ON time, s	71.6 (11.0)	86.5 (21.3)	t = 4.5, P < 0.01	81.1 (20.7)	89.6 (22.1)	84.6 (19.3)
			d = 0.88
OFF + ON, s	135.5 (17.9)	161.5 (36.3)	t = 4.7, P < 0.01	152.5 (36.1)	166.4 (37.6)	159.4 (32.0)
			d = 0.88
ON + OFF, s	7.7 (6.4)	11.4 (11.4)	t = 2.0, P = 0.05	9.7 (11.5)	12.8 (10.6)	9.9 (13.2)
			d = 0.4

A MANOVA was conducted to examine differences across the three executive function tests (total UFOV, total TMT, and OFF + ON SCWT). The main effect of participant group was significant (F _(3,108) = 2.38, P = 0.013, ${\eta }_{\mathrm{p}}^2$ = 0.61). Significant differences between participant groups were found across all three executive function tasks: total UFOV F _{(3, 108)} = 2.98, P = 0.034, ${\eta }_{\mathrm{p}}^2$ = 0.08; total TMT F _{(3, 108)} = 5.289, P = 0.002, ${\eta }_{\mathrm{p}}^2$ = 0.13; and OFF + ON SCWT F _(3,108) = 4.068, P = 0.009, ${\eta }_{\mathrm{p}}^2$ = 0.10.

Post hoc comparisons showed that the control group had significantly faster total UFOV response times than the strabismus alone group (P = 0.03) and amblyopia and strabismus group (P = 0.007; Fig. 1d). The control group had faster total TMT completion times than the amblyopia alone group (P = 0.03) and the amblyopia with strabismus group (P = 0.001), and the strabismus alone group performed better than the amblyopia with strabismus group (P = 0.04; Fig. 2e). The control group had faster OFF + ON SCWT times than both the strabismus alone (P = 0.03) and the amblyopia with strabismus (P = 0.001) groups (Fig. 3c).

Figure 1.

Individual participant (circles) and group mean (error bars = ± SD) minimal display time to enable completion of the UFOV subtests, that is, visual processing speed for (a) central processing, (b) divided attention, and (c) selective attention. The Y-axis scales have been varied between sub-parts of the figure. UFOV total summed time is also displayed (d). Groups are amblyopia without strabismus (A, red), amblyopia with strabismus (AS, blue), strabismus without amblyopia (S, green), and control (C, purple). The P values report significance of least significant difference (LSD) post hoc tests between subgroups.

Figure 2.

Individual participant (circles) and group mean (error bars = ± SD) times to complete the TMT tests (a) TMT-A static, (b) TMT-B static, (c) TMT-A dynamic, (d) TMT-B dynamic, and (e) TMT total time (sum of all TMT tests). Groups are amblyopia without strabismus (A, red), amblyopia with strabismus (AS, blue), strabismus without amblyopia (S, green), and control (C, purple). The P values report significance of least significant difference (LSD) post hoc tests between subgroups.

Figure 3.

Individual participant (circles) and group mean (error bars = ± SD) time taken to complete the SCWT; (a) OFF time, (b) ON time, (c) OFF time + ON time, and (d) ON time + OFF time. Groups are amblyopia without strabismus (A, red), amblyopia with strabismus (AS, blue), strabismus without amblyopia (S, green), and control (C, purple). The P values report significance of least significant difference (LSD) post hoc tests between subgroups.

Useful Field of View

Visual processing speeds for the three UFOV subtests and the total UFOV score for participants with amblyopia and/or strabismus, as well as controls, are presented in Table 2 and Figure 1. Overall, the control group demonstrated significantly faster processing times across all UFOV subtests and the total UFOV score compared with participants with abnormal binocular vision (all P ≤ 0.05), with a moderate to large effect size (Cohen's d of 0.36 to 0.76).

UFOV subtest outcomes were analyzed across etiology subgroups (amblyopia without strabismus [A], amblyopia with strabismus [AS], strabismus only [S], and controls [C]). A mixed-design ANOVA revealed a significant main effect of group (F _{(3, 330)} = 4.81, P = 0.003, ${\eta }_{\mathrm{p}}^2$ = 0.04) and subtest (F _{(2, 330)} = 151.19, P < 0.0001, ${\eta }_{\mathrm{p}}^2$ = 0.48), but no significant group × subtest interaction (F _{(6, 330)} = 0.99, P = 0.44, ${\eta }_{\mathrm{p}}^2$ = 0.02). Given the significant main effect of the group, tests of simple effects were conducted to examine group differences within each subtest. For the central processing subtest, the effect of group was not significant (F _{(3, 110)} = 0.051, P = 0.985, ${\eta }_{\mathrm{p}}^2$ = 0.001). For the divided attention subtest, there was a significant effect of group (F _{(3, 110)} = 2.829, P = 0.039, ${\eta }_{\mathrm{p}}^2$ = 0.07). Post hoc comparisons indicated that the AS group had significantly longer response times than the controls (P = 0.03). For the selective attention subtest, there was a significant effect of group (F _{(3, 110)} = 3.899, P = 0.009, ${\eta }_{\mathrm{p}}^2$ = 0.10). Post hoc comparisons indicated that both the S and AS groups had significantly longer response times than the controls (P < 0.05). See Figure 1 for a visual summary of these results.

Trail Making Test

Mean completion times for the TMT-A and B tests, for both the static and dynamic mode, in the adults with amblyopia and/or strabismus and controls are summarized in Table 2 and Figure 2. Participants with abnormal vision from amblyopia and/or strabismus took longer than the controls to complete each TMT task (P values < 0.05), with large effect size (all Cohen's d > 0.8).

TMT subtest outcomes were analyzed across etiology subgroups (amblyopia without strabismus [A], amblyopia with strabismus [AS], strabismus only [S], and controls [C]). A mixed-design ANOVA revealed a significant main effect of group (F _{(3, 440)} = 15.76, P < 0.0001, ${\eta }_{\mathrm{p}}^2$ = 0.10) and subtest (F _{(3, 440)} = 19.90, P < 0.0001 ${\eta }_{\mathrm{p}}^2$ = 0.12), but no significant group × subtest interaction (F _{(9, 440)} = 0.58, P = 0.82, ${\eta }_{\mathrm{p}}^2$ = 0.01). Given the significant main effect of the group, tests of simple effects were conducted to examine group differences within each TMT subtest. For TMT-A static, the effect of the group was significant (F _{(3, 440)} = 2.940, P = 0.033, ${\eta }_{\mathrm{p}}^2$ = 0.02); for TMT-A dynamic, the effect of the group was not significant (F _{(3, 440)} = 2.146, P = 0.094, ${\eta }_{\mathrm{p}}^2$ = 0.01); for TMT-B static, the effect of the group was significant (F _{(3, 440)} = 3.934, P = 0.009, ${\eta }_{\mathrm{p}}^2$ = 0.03); for TMT-B dynamic, the effect of the group was significant (F _{(3, 440)} = 8.478, P = 0.000, ${\eta }_{\mathrm{p}}^2$ = 0.05). Where significant group effects were observed, post hoc comparisons were performed to identify specific group differences (TMT-A static AS worse than C, P = 0.035; TMT-B static AS worse than C, P = 0.004; TMT-B dynamic AS worse than C, P < 0.001, and AS worse than S, P = 0.018). See Figure 2 for a visual summary of these results.

Stroop Color and Word Test

Mean completion times for the SCWT OFF and ON tests, and for their sum and difference, in the adults with amblyopia and/or strabismus and controls are summarized in Table 2 and Figure 3. Participants with abnormal vision from amblyopia and/or strabismus were significantly slower than controls on both Stroop conditions (P < 0.05), with large effect size (Cohen's d > 0.8).

SCWT subtest outcomes were analyzed across etiology subgroups (amblyopia without strabismus [A], amblyopia with strabismus [AS], strabismus only [S], and controls [C]). A mixed-design ANOVA revealed a significant main effect of group (F _{(3, 440)} = 11.59, P < 0.0001, ${\eta }_{\mathrm{p}}^2$ = 0.07) and Stroop subtest (F _{(3, 440)} = 706.74, P < 0.0001, ${\eta }_{\mathrm{p}}^2$ = 0.83), but no significant group × subtest interaction (F _{(9, 440)} = 1.22, P = 0.28, ${\eta }_{\mathrm{p}}^2$ = 0.02). Given the significant main effect of the group, tests of simple effects were conducted to examine group differences within each Stroop subtest. For the SCWT ON task, the effect of group was significant (F _{(3, 440)} = 3.367, P = 0.019, ${\eta }_{\mathrm{p}}^2$ = 0.02); For the SCWT OFF task, the effect of group was not significant (F _{(3, 440)} = 1.721, P = 0.162, ${\eta }_{\mathrm{p}}^2$ = 0.01); post hoc comparisons found that the AS group was longer than C for the Stroop ON time (P = 0.013). Figure 3 displays a visual summary.

Clinical Vision Factors Associated With Executive Function

As shown in Table 3, several vision characteristics were significantly associated with executive function performance. Greater IOVA was linked to slower performance on the TMT (r = 0.23). Binocular visual acuity (VA) also showed a moderate correlation with TMT performance (r = 0.23). Notably, poorer binocular function (BFS) was consistently associated with slower performance across all tasks (UFOV, TMT, and SCWT [r = 0.19–0.31, all P < 0.05]), indicating that poorer binocular function was associated with slower performance across a range of visual and cognitive tasks.

Table 3.

Correlation Analysis Among Vision Parameters, UFOV, TMT, and SCWT of all Participants (n = 114)

	Vision Characteristics		Executive Function Assessments
	Binocular VA	BFS	UFOV Total Time	TMT Total Time	SCWT – OFF + ON Time
IOVA	0.12 (P = 0.19)	0.66 ( P < 0.01)*	0.07 (P = 0.46)	0.23 ( P = 0.02)*	0.16 (P = 0.09)
Binocular VA	–	0.33 ( P < 0.01)*	0.15 (P = 0.12)	0.23 ( P = 0.02)*	0.14 (P = 0.15)
BFS		–	0.19 ( P = 0.04)*	0.31 ( P < 0.01)*	0.23 ( P = 0.02)*
UFOV total time			–	0.69 ( P < 0.01)*	0.69 ( P < 0.01)*
TMT total time				–	0.79 ( P < 0.01)*

The figures in bold represent significant outcomes.

Data are outcome of Pearson correlation.

BFS, binocular function score.

^*Correlation is significant at the 0.05 level (two-tailed).

Participants were grouped based on their BFS score: (i) those with BFS 2.3 and less (n = 38) and (ii) those with a score of 2.9 or more (n = 76). Comparison of the two groups performance on total UFOV (233 vs. 264 ms) and OFF + ON SCWT tests (148 vs. 161 seconds) were not significant. A group difference was only observed for total TMT (196 vs. 251 seconds; F = 6.3, P = 0.01). Again, this outcome highlights the relatively small contribution of monocular VA decrement and stereoacuity to performance on these binocular function tasks.

In addition, strong associations were observed among the executive function tasks themselves. UFOV total time was significantly correlated with both TMT total time and SCWT OFF + ON time (r = 0.69 for both, P < 0.01), indicating consistent patterns of slower performance across tasks. TMT and SCWT times were also highly correlated (r = 0.79, P < 0.01).

The results of the multiple regression analysis, reported in Table 4, show the relationship between various clinical measures and cognitive task performance (UFOV, TMT, and SCWT). For UFOV total time, none of the vision measures (inter-ocular VA difference, binocular VA, and BFS) were significantly associated, as indicated by the high P values (all > 0.05). However, the presence of strabismus was significantly associated with slower performance on UFOV (β = −63.98, P = 0.006). Similarly, SCWT OFF + ON performance was only significantly associated with the presence of strabismus, (β = −19.67, P = 0.004); the other vision measures (inter-ocular VA difference, binocular VA, and BFS) did not significantly predict performance on the SCWT (all P values > 0.05). TMT performance was significantly associated with binocular function score (β = 21.67, P = 0.001), with worse BFS associated with slower TMT performance. The VA measures or presence of strabismus did not significantly impact TMT performance.

Table 4.

Determinants of UFOV, TMT, and SCWT Performance (n = 114)

	UFOV: Total					TMT Total					SCWT – OFF + ON Time
Vision Measure	Regression Coefficient (β)	SE	t	P Value	VIF	Regression Coefficient (β)	SE	t	P Value	VIF	Regression Coefficient (β)	SE	t	P Value	VIF
IOVA	−14.26	39.55	−0.36	0.72	1.83	18.98	30.41	0.62	0.53	1.83	8.64	8.74	0.99	0.33	1.09
Binocular VA	125.96	146.25	0.86	0.39	1.08	170.47	114.77	1.49	0.14	1.12	28.29	42.39	0.67	0.51	1.08
BFS	3.51	10.28	0.34	0.73	1.6	21.67	6.27	3.46	<0.01	1.0	0.78	3.86	0.20	0.84	1.09
Presence of Strabismus	−63.98	22.86	−2.80	0.01	1.0	−12.88	21.98	−0.59	0.56	1.54	−19.67	6.61	−2.98	<0.01	1.0
R 2	0.07					0.10					0.07

The figures in bold represent significant contribution of the factor to the model.

The overall R² values were low for all three models (0.07 for UFOV, 0.10 for TMT, and 0.07 for SCWT), indicating that the vision measures explained a small proportion of the variance in performance on these cognitive tasks (between 7% and 10% of variance).

Table 5 reports multiple regression analyses used to evaluate the contributions of visual and executive function measures to FMS manual dexterity scores and calculated reading speed. Significant visual predictors of FMS include IOVA (β = −1.42, P = 0.02), and binocular VA (β = −10.95, P < 0.01) indicating that more severe amblyopia and poorer binocular resolution are associated with poorer manual dexterity scores. The presence of strabismus was positively associated with manual dexterity scores (β = 2.05, P < 0.01). SCWT OFF + ON also emerged as a significant predictor (β = −0.02, P = 0.01), with longer Stroop times associated with lower manual dexterity scores. The model explained 42% of the variance in manual dexterity scores (R² = 0.42).

Table 5.

Determinants of Fine Motor Skills and Reading Speed (n = 114)

	Fine Motor Skills (BOT-2 Manual Dexterity Score)					Calculated Reading Speed (IReST)
	Regression Coefficient (β)	SE	t	P Value	VIF	Regression Coefficient (β)	SE	t	P Value	VIF
IOVA	−1.42	0.60	−2.37	0.02	1.10	1.06	12.31	0.09	0.93	1.86
Binocular VA	−10.95	2.90	−3.78	<0.01	1.09	−45.58	46.34	−0.98	0.33	1.15
BFS	0.02	0.27	0.06	0.95	2.75	−4.25	2.56	−1.67	0.10	1.05
Presence of strabismus	2.05	0.50	4.11	<0.01	1.22	5.08	8.97	0.57	0.57	1.61
UFOV total time	0.001	0.003	0.37	0.71	2.19	−0.01	0.04	−0.31	0.75	2.17
TMT total time	−0.002	0.004	−0.63	0.53	2.79	−0.07	0.06	−1.16	0.25	2.76
SCWT – OFF + ON time	−0.02	0.01	−2.53	0.01	1.09	−0.63	0.10	−6.29	<0.01	1.05
R 2	0.42					0.31

The figures in bold represent significant contribution of the factor to the model.

The strongest predictor of calculated reading speed was SCWT OFF + ON time (β = −0.63, P < 0.01), where longer Stroop times were associated with slower reading speeds. Other variables did not reach statistical significance. The model accounted for 31% of the variance in reading speed (R² = 0.31).

Table 6 presents the results of multiple regression analyses examining the determinants of VRQOL scores derived from the A&SQ in adults with amblyopia and/or strabismus (n = 95). The analyses highlight the relative contributions of clinical, functional, and executive function measures to VRQOL outcomes. Key outcomes were that BFS and UFOV total times significantly influence functional impact scores, whereas strabismus is a major determinant of psychosocial impact scores. For the functional impact score derived from the A&SQ, better binocular function (BFS, β = 2.40, P = 0.04) and shorter UFOV total times (β = −0.02, P = 0.02) were significant predictors. The outcomes on functional skills tests (FMS manual dexterity and reading speed) did not significantly associate with the Functional Impact Score.

Table 6.

Determinants of Functional and Psychosocial Impact Scores (n = 95)

	Functional Impact Score					Psychosocial Impact Score
	Regression					Regression
Measure	Coefficient (β)	SE	t	P Value	VIF	Coefficient (β)	SE	T	P Value	VIF
IOVA	−4.77	3.98	−1.20	0.23	1.39	11.50	6.12	1.88	0.06	1.03
Binocular VA	−14.42	18.51	−0.78	0.44	1.05	56.22	32.65	1.72	0.09	1.03
BFS	2.40	1.17	2.04	0.04	1.0	0.27	2.77	0.10	0.92	1.64
Strabismus	−0.99	3.43	−0.29	0.78	1.37	17.41	5.18	3.36	<0.01	1.0
Manual dexterity	0.38	0.70	0.54	0.59	1.15	1.34	1.25	1.07	0.29	1.19
Reading speed	−0.03	0.03	−0.94	0.35	1.27	0.06	0.06	1.004	0.32	1.27
UFOV total time	−0.02	0.01	−2.29	0.02	1.0	−0.03	0.03	−1.32	0.19	2.01
TMT total time	0.03	0.02	1.45	0.15	1.97	0.02	0.02	0.92	0.36	1.10
SCWT − OFF + ON time	−0.05	0.06	−0.77	0.45	3.22	0.01	0.11	0.10	0.92	3.28
R 2	0.09					0.16

The figures in bold represent significant contribution of the factor to the model.

For psychosocial impact scores, the presence of strabismus was the only significant predictor (β = 17.41, P = 0.001). Although binocular VA and IOVA approached significance for psychosocial impact (P = 0.09 and P = 0.06, respectively), they did not reach the threshold for statistical significance. Despite these significant predictors, the explained variance for both models is modest, with R² values of 0.09 and 0.16 for functional and psychosocial impact scores, indicating that additional factors beyond the measured variables contribute to VRQOL in this population.

Discussion

This study highlights significant deficits in higher-order visual processing and executive function in adults with amblyopia and/or strabismus under habitual binocular viewing conditions, which reflect real-world scenarios. Participants exhibited slower performance compared with controls across key visuo-cognitive tasks, including divided and selective visual attention (UFOV), visual search and scanning (TMT), and cognitive flexibility (SCWT), reflecting challenges in processing complex visual information under time constraints. These findings are consistent with accumulating evidence that amblyopia is associated with persistent attention deficits,²³ which may be compounded in strabismic amblyopia,⁴⁰ and have broader implications for functional outcomes and quality of life.⁴¹

Executive function impairments were evident across all three visuo-cognitive tasks, with participants with abnormal binocular vision consistently requiring longer times to complete these tasks than the controls. The strong intercorrelations among UFOV, TMT, and SCWT test outcomes suggest that these tasks may tap into overlapping cognitive mechanisms, particularly those related to processing speed, attentional control, and cognitive flexibility, which are core components of executive function. Although these associations suggest the possibility of shared underlying cognitive processes or deficits, further study would be required to explore common underlying mechanisms. These findings align with prior studies reporting prolonged visuo-cognitive task times in children with amblyopia.^16,22 Notably, these previous pediatric studies have primarily involved children with anisometropic amblyopia, whereas the present study included observers with strabismic amblyopia who demonstrate worse binocular function. Similarly, the observed deficits in visual search and attention are consistent with previous research on dorsal stream vulnerabilities in amblyopia.^42,43

Relationships Between Vision and Cognitive Measures

Whereas the executive function tests were visually guided, the relationships between clinical visual measures and task performance were relatively weak. Greater IOVA difference, an indicator of amblyopia severity, was modestly associated with slower performance on the TMT (r = 0.23). Similarly, poorer binocular VA showed a moderate correlation with TMT performance (r = 0.23). In contrast, the BFS demonstrated consistent correlations across tasks, with poorer binocular function linked to slower UFOV, TMT, and SCWT performance. These findings align with prior studies reporting prolonged task times in children with amblyopia,^16,22 including Black et al. (2021),¹⁶ which examined similar executive function tasks in a pediatric cohort. However, the stronger association between binocular function and task performance observed in our adult sample may reflect differences in study participant characteristics or developmental context. Notably, Black et al. (2021) included predominantly anisometropic participants,¹⁶ whereas our sample included individuals with strabismus, both with and without amblyopia, which may be associated with more pronounced binocular dysfunction. Further, adults likely have experienced longer durations of visual disruption, which may explain the relationship between binocular dysfunction and task performance observed in our sample.

The regression findings suggest that the level of binocular function and the presence of strabismus are more predictive of executive function task performance than VA alone. The presence of strabismus emerged as a significant predictor for UFOV and SCWT performance, suggesting that ocular misalignment may contribute to attentional deficits and cognitive delays. However, strabismus per se was not significantly associated with TMT performance, which was more strongly linked to the BFS. This nuanced relationship highlights the differing impacts of clinical vision measures on various cognitive tasks, underscoring the complexity of the interplay between visual and executive function deficits.

Real-World Implications

The slower processing speeds for visual attention tasks determined by UFOV performance among adults with amblyopia and/or strabismus, carry significant real-world implications. Poor UFOV has been previously linked to an increased risk of motor vehicle crashes, as it reflects limitations in visual processing speed and divided attention critical for safe driving.⁴⁴ The slower visual processing speeds in this study, particularly in divided attention tasks, suggest that individuals with amblyopia and/or strabismus may face elevated risks in driving situations that require rapid responses to multiple visual stimuli. These findings underscore the importance of assessing and addressing visual and cognitive processing deficits to mitigate real-world safety risks for this population.

The increased completion times on the TMT, indicating deficits in executive functions related to visual search and processing speed, were closely linked to fine motor skills, a relationship not previously explored in amblyopia and strabismus. Poorer binocular visual attention and slower visual search performance further align with prior reports of difficulties in manual dexterity^45,46 and writing.⁴⁷ These persistent deficits into adulthood emphasize the long-term impact of abnormal visual development on real-world tasks. Similarly, SCWT performance strongly predicted reading efficiency, supporting previous findings that reading and Stroop interference share common executive processes, such as inhibition and attention.^48,49 These results align with research in neurodevelopmental and neurodegenerative conditions, where the SCWT is widely used to evaluate executive function.^50,51

Functional and Psychosocial Impacts

Both clinical and functional measures contributed to VRQOL outcomes, although their roles varied by domain. Functional impact scores were significantly influenced by binocular function and visual processing speed, as reflected in UFOV times, whereas functional skills tests, such as manual dexterity and reading speed, did not significantly predict self-reported functional impact. This disconnect suggests that perceived functional limitations may be influenced by subjective experiences and external factors beyond objective performance. In contrast, psychosocial impact scores were strongly influenced by the presence of strabismus, reflecting the emotional and social challenges associated with visible ocular misalignment. The modest explained variance for both functional (R² = 0.09) and psychosocial (R² = 0.16) impact models highlights the need to consider unmeasured psychosocial and environmental influences on VRQOL outcomes.

Study Limitations

This study has several limitations that may affect the interpretation of the findings. The reliance on computer-based tasks, such as UFOV, TMT, and SCWT, may have favored participants who are regular computer users, potentially influencing performance and limiting generalizability. The sample size, while sufficient to detect significant differences, may not fully capture the heterogeneity of amblyopia and strabismus populations, such as variations in severity, etiology, and treatment history. Additionally, unmeasured confounders, including socioeconomic status, cognitive baseline, and educational background, could have influenced task performance and VRQOL outcomes. In our multiple regression analyses examining the factors influencing VRQOL, one predictor variable (SCWT) exhibited VIFs of 3.22 and 3.28, indicating moderate multicollinearity. Although this level of multicollinearity does not significantly compromise the model’s stability or predictive accuracy, it can increase the standard errors of the coefficients, potentially obscuring the significance of some predictors.

Conclusions

This study provides critical evidence of the broader impacts of amblyopia and strabismus on cognitive function and quality of life in adults, emphasizing deficits in visual attention, executive function, and functional performance. The persistence of these deficits beyond the critical period of visual development suggests long-term neurocognitive effects. Future research should explore neural substrates through imaging studies, assess longitudinal changes in these deficits, and evaluate interventions targeting both visual and cognitive impairments to improve outcomes for affected individuals.