Journal of Pediatric Ophthalmology and Strabismus

Original Article 

Short-term Perceptual Learning Game Does Not Improve Patching-Resistant Amblyopia in Older Children

Yoon H. Lee, MD; Marcello Maniglia, PhD; Federico Velez, MD; Joseph L. Demer, MD, PhD; Aaron R. Seitz, PhD; Stacy Pineles, MD

Abstract

Purpose:

To investigate self-administered, at-home use of a perceptual learning–based video game consisting of target detection of stimuli in different sizes, spatial frequency, orientation, and contrast as a potential dichoptic therapy to improve binocular function in amblyopic patients resistant to patching.

Methods:

Children (ages 8 to 18 years) with strabismic and/or anisometropic amblyopia were recruited from a single institution. All participants (n = 25) were prescribed 6 weeks of patching for 2 hours per day, and those whose visual acuity did not improve were randomized to binocular perceptual learning (n = 7), monocular perceptual learning (n = 8), or patching (n = 10) groups for 8 weeks in this prospective cohort study. After an 8-week long period of treatment cessation, during which participants stopped patching or perceptual learning, participants in the patching group were randomized to binocular or monocular perceptual learning training; those in the perceptual learning groups remained the same. Visual function was assessed by visual acuity, low contrast acuity, reading speed, stereoacuity, and binocularity; compliance was evaluated by exercise logs.

Results:

There were no significant improvements in visual function parameters, which did not vary by treatment group. However, some visual outcomes, such as binocular summation and reading speed, correlated positively with compliance to perceptual learning therapy.

Conclusions:

At-home, self-administered use of this perceptual learning–based video game–based visual training does not consistently add therapeutic benefit to those with amblyopia resistant to patching. Future investigation is required to determine whether methods to increase compliance will lead to more reliable outcomes.

[J Pediatr Ophthalmol Strabismus. 2020;57(3):176–184.]

Abstract

Purpose:

To investigate self-administered, at-home use of a perceptual learning–based video game consisting of target detection of stimuli in different sizes, spatial frequency, orientation, and contrast as a potential dichoptic therapy to improve binocular function in amblyopic patients resistant to patching.

Methods:

Children (ages 8 to 18 years) with strabismic and/or anisometropic amblyopia were recruited from a single institution. All participants (n = 25) were prescribed 6 weeks of patching for 2 hours per day, and those whose visual acuity did not improve were randomized to binocular perceptual learning (n = 7), monocular perceptual learning (n = 8), or patching (n = 10) groups for 8 weeks in this prospective cohort study. After an 8-week long period of treatment cessation, during which participants stopped patching or perceptual learning, participants in the patching group were randomized to binocular or monocular perceptual learning training; those in the perceptual learning groups remained the same. Visual function was assessed by visual acuity, low contrast acuity, reading speed, stereoacuity, and binocularity; compliance was evaluated by exercise logs.

Results:

There were no significant improvements in visual function parameters, which did not vary by treatment group. However, some visual outcomes, such as binocular summation and reading speed, correlated positively with compliance to perceptual learning therapy.

Conclusions:

At-home, self-administered use of this perceptual learning–based video game–based visual training does not consistently add therapeutic benefit to those with amblyopia resistant to patching. Future investigation is required to determine whether methods to increase compliance will lead to more reliable outcomes.

[J Pediatr Ophthalmol Strabismus. 2020;57(3):176–184.]

Introduction

Amblyopia affects 2% to 4% of children, making it the most common cause of non-refractive vision loss in children.1 Current amblyopia treatments focus on penalizing or occluding the unaffected eye with the goal of stimulating the amblyopic eye and its associated visual pathways. Whereas patching or atropine drops have proved to be effective in improving visual acuity for most children, there is a small population of patients who are resistant to conventional therapy2 or who suffer recurrence after initial response to treatment.3 Furthermore, even with optimal therapy, children with amblyopia have lower stereoacuity4 and slower reading speeds than unaffected children.5

These findings suggest that there are some aspects of amblyopia not addressed by conventional therapy. One possible element is the concept of suppression: the visual cortex may perceive a discrepancy in the contrast level between the input from the amblyopic eye and the fellow eye, which may induce suppression of the noisier data from the amblyopic eye and loss of complete binocularity.6 Suppression may also decrease plasticity of the visual cortex in response to amblyopia treatment because patients with amblyopia who have higher suppression levels have decreased response to occlusion therapy.7

Several novel amblyopia treatments use binocular approaches, with the goal of reducing suppression and training the eyes to work together.8–10 Instead of completely occluding the non-amblyopic eye, dichoptic approaches present balanced contrast to each eye by providing decreased stimulus to the non-amblyopic eye. Perceptual learning studies, in particular, have successfully used both monocular and dichoptic training to enhance contrast sensitivity and visual acuity in the amblyopic eye.1 An important distinction between perceptual learning and other dichoptic therapies is that specific stimuli and tasks are chosen to purposefully stimulate various areas within the visual cortex.11 Current approaches in the field of perceptual learning focus on understanding the mechanisms that lead to larger and more generalized learning effects and how to implement training paradigms to achieve this outcome. In particular, recent perceptual learning studies have found that binocular training approaches that either put the amblyopic eye and fellow eye into competition12 or help the amblyopic eye to better compete with the fellow eye13 lead to impressive benefits in individuals with amblyopia.

There is evidence that video games and gamified perceptual learning paradigms present advantages in terms of learning and generalization over classic perceptual learning,14–16 especially when targeting children and young adults. We conducted a feasibility study examining the potential benefit of self-administered, at-home use of a tablet-based perceptual learning game, which presents a series of visual stimuli in various orientations and spatial frequencies in a game-play format. This program has been used in previous studies showing benefits in contrast sensitivity in normally seeing individuals17 and those with presbyopia.11 In the current study, the game was tested monocularly (patching the nonamblyopic eye) and binocularly (using polarized glasses that allow only the amblyopic eye to view the images on the screen while allowing both eyes to see the surroundings around the screen).

Patients and Methods

Participants

Inclusion criteria were age 8 to 18 years with strabismic and/or anisometropic amblyopia and visual acuity between 20/40 and 20/200 with at least two Snellen lines worse in the amblyopic eye compared to the fellow eye. Patients were excluded if they had any organic lesions to preclude visual recovery (ie, structural lesions of the eye), developmental delay or inability to perform the perceptual learning task, or a neurologic disorder. Prior patching history did not preclude patients from participating in the study. Twenty-five participants (amblyopia types: strabismic [n = 11], anisometropic [n = 10], mixed [n = 4]) were enrolled in the study. This study was approved by the University of California Riverside and Los Angeles institutional review boards.

Procedure

Participants who met the study inclusion criteria were recruited during routine clinical visits or by phone call by a study investigator at a single institution. All participants were prescribed patching of their dominant eye for 2 hours per day for a minimum of 6 weeks, and those whose visual acuity did not improve were enrolled into the study. Patients were then randomized to one of three groups, with an initial goal of randomizing 10 patients per group: (1) binocular perceptual learning (20 min/day of video game played with polarized lens), (2) monocular perceptual learning (20 min/day of video game played with non-amblyopic eye patched), or (3) patching (2 hours/day) (Figure 1). A duration of 20 minutes was determined to be an appropriate amount of time without causing too much burden on a child's daily activities and was similar to the amount of time that had been successfully used in other populations.11,18 On the other hand, patching was selected for 2 hours because this was found to be an effective amount of patching for moderate amblyopia.19 The following series of assessments were measured before and after 8 weeks of treatment: visual acuity (Early Treatment Diabetic Retinopathy Study test, crowding bars) of both eyes, amblyopic eye, and unaffected eye, low contrast acuity (Sloan charts at 2.5% contrast; both eyes, amblyopic eye, unaffected eye), reading speed (both eyes, amblyopic eye), and stereoacuity (near and distance) were measured before and after 8 weeks of treatment. Then, participants stopped patching or perceptual learning therapy for 8 weeks. Following this, the participants underwent another 8 weeks of therapy. During this period, participants in the patching group were assigned to binocular perceptual learning (n = 6) and monocular perceptual learning (n = 4), whereas those previously in perceptual learning groups remained the same. The same assessments were measured after another 8 weeks of treatment. Additionally, a binocular summation score was calculated as the difference between the binocular low contrast acuity score and the low contrast acuity score of the better eye.

Experimental design. After the first training period, participants in the patching group were randomly assigned to the binocular or monocular perceptual learning (PL) group.

Figure 1.

Experimental design. After the first training period, participants in the patching group were randomly assigned to the binocular or monocular perceptual learning (PL) group.

The training was conducted at home via a self-administered tablet-based video game in which participants had to respond to target stimuli (Gabor patches) of different sizes, spatial frequency, orientation, and contrast while reaction time and contrast thresholds were recorded (Figure 2). The program adapts on contrast thresholds and creates personalized training for each participant.18 A trained researcher reviewed the game data logs (comprising number of exercises per day), daily score, and threshold of target detection, and assigned participants to the following compliance levels: good (more than 80% sessions conforming to the following criteria: [1] at least 10 exercises completed, [2] scores above 1,000, which correlate the number of items responded to correctly, and [3] consistent daily contrast thresholds [within or less than a standard deviation of running average of previous sessions]); mediocre (more than 50% of sessions conforming to the above rules); and poor (less than half of sessions conforming to the above rules and overall highly inconsistent performance across sessions, with many skipped sessions or sessions with poor performance). Patching compliance was not measured, and participants were assumed to be in the same compliance group for patching as perceptual learning.

Example of the video game layout. Participants were asked to tap on the target (oriented Gabor patches, red arrows) while avoiding distractors (blue arrows). Contrast of both targets and distractors was adapted based on performance on the task.

Figure 2.

Example of the video game layout. Participants were asked to tap on the target (oriented Gabor patches, red arrows) while avoiding distractors (blue arrows). Contrast of both targets and distractors was adapted based on performance on the task.

Statistical Analysis

Data analysis was performed using Stata software (version 15; StataCorp, College Station, TX). Two-tailed paired t tests were used to compare the measurements before and after treatment. One-way analysis of variance was used to compare the change in measurements (difference between pretreatment and posttreatment values) among treatment groups (binocular perceptual learning, monocular perceptual learning, and patching) and among compliance groups (good, mediocre, and poor); the compliance group was only a subset of participants due to availability of compliance logs in certain patients. Twoway analysis of variance was used to compare the effect of interaction between treatment group and age, as well as between treatment group and time.

Results

Details of participant characteristics are shown in Table 1. Visual function was measured by visual acuity (both eyes, amblyopic eye, unaffected eye), low contrast acuity (both eyes, amblyopic eye, unaffected eye), binocular summation score, reading speed (both eyes), and stereoacuity (near, distance). Recruitment initially targeted 30 patients, but due to patient drop-out after randomization, 25 patients completed the study. Initial analysis considered each 8-week study period as a discrete sample, so total sample size of the study was 50 (25 participants in two 8-week periods), divided into binocular perceptual learning (n = 20), monocular perceptual learning (n = 20), and patching (n = 10).

Participant Characteristics

Table 1:

Participant Characteristics

Visual Function Parameters

Low contrast acuity (amblyopic eye) increased from 12.9 to 16.6 letters with 8 weeks of perceptual learning or patching (P = .002, |t| = 3.3). There was no improvement in other visual function parameters, as shown in Table 2. Change in visual function parameters did not vary with treatment group, as shown in Table 3.

t Test Before and After Treatment

Table 2:

t Test Before and After Treatment

One-way ANOVA Comparison Among Three Treatment Groups for Difference of Means Before and After Treatment

Table 3:

One-way ANOVA Comparison Among Three Treatment Groups for Difference of Means Before and After Treatment

Effect of Age

There was a significant interaction between age and treatment groups on low contrast acuity (amblyopic eye) (F(2,39) = 7.00, P = .0025). Of note, low contrast acuity was not obtained from every participant and sample size was as follows. For those in the binocular perceptual learning group (n = 18), the mean change in low contrast acuity (amblyopic eye) was 5.2 (95% confidence interval [CI] = −2.3 to 12.7) letters in younger children (ages 8 to 13 years, n = 10) and 1.9 (95% CI = −0.6 to 4.4) letters in older children (ages 14 to 18 years, n = 8) (P = .30). For those in the monocular perceptual learning group (n = 17), the mean change in low contrast acuity (amblyopic eye) was 1.9 (95% CI = −0.6 to 4.5) letters in younger children (n = 16) and 26 letters in older children (n = 1) (P < .01). For those in the patching group (n = 10), the mean change in low contrast acuity (amblyopic eye) was 5.3 (95% CI = −1.1 to 11.6) letters in younger children (n = 8) and −0.5 (95% CI = −6.9 to 5.9) letters in older children (n = 2) (P = .30).

Effect of Compliance Level

Compliance logs, recording of game use downloaded from tablets, were obtained from only 12 patients (good [4], mediocre [3], poor [5]) due to technical errors in recovering log data from some tablets.

Higher compliance level positively correlated with improvement in binocular reading speed for those in the binocular perceptual learning group, but not the monocular perceptual learning group. Specifically, participants with good compliance with binocular perceptual learning had a change of reading speed (both eyes) of +93 wpm (95% CI = −39 to 148) (– = slower reading speed; + = faster reading speed) compared to those with poor compliance who had a change in their reading speed of −21 wpm (95% CI = −76 to 33) (F = 5.4, P = .03) (Figure 3). Similarly, reading speed (amblyopic eye) increased more in those with higher compliance of 39 wpm (95% CI = −28 to 105) for good, 9 wpm (95% CI = −58 to 75) for mediocre, and −9 wpm (95% CI = −75 to 58) for poor compliance, but this was not statistically significant (F = 0.6, P = .60). Compliance level did not have an effect on other measurements, including visual acuity, low contrast acuity, and stereoacuity.

Visual function by compliance group. Change in visual acuity (VA), binocular summation score (BSS), and reading speed (both eyes [OU], amblyopic eye [AE]) by compliance levels (good, mediocre, and poor) after 8 weeks of binocular perceptual learning. Error bars show mean ± standard error. Binocular reading speed was the only factor that significantly varied with compliance level (93.5, 16.5, and −21.5 wpm for good, mediocre, and poor compliance, respectively). Compliance level did not have an effect on visual function after monocular perceptual learning.

Figure 3.

Visual function by compliance group. Change in visual acuity (VA), binocular summation score (BSS), and reading speed (both eyes [OU], amblyopic eye [AE]) by compliance levels (good, mediocre, and poor) after 8 weeks of binocular perceptual learning. Error bars show mean ± standard error. Binocular reading speed was the only factor that significantly varied with compliance level (93.5, 16.5, and −21.5 wpm for good, mediocre, and poor compliance, respectively). Compliance level did not have an effect on visual function after monocular perceptual learning.

When comparing treatment groups within the good compliance level, there was no statistically significant difference between binocular perceptual learning, monocular perceptual learning, or patching on reading speed (both eyes, amblyopic eye) (F = 3.2, P = .20) (Figure 4). However, improvement in reading speed (both eyes) was highest for binocular perceptual learning (94 wpm [95% CI = −7 to 180]), followed by monocular perceptual learning (13 wpm [95% CI = −58 to 83]), then patching (0 wpm [95% CI = −122 to 122]) (F = 3.2, P = .20). Similarly, change in reading speed (amblyopic eye) was highest for binocular perceptual learning (39 wpm [95% CI = −37 to 114]), followed by monocular perceptual learning (26 wpm [95% CI = −35 to 88]), then patching (−8 [95% CI = −114 to 98]), (F = 0.7, P = .60). In addition, within the good compliance group, the binocular summation score improved with binocular perceptual learning (ΔBSSmean = 5), whereas it decreased with monocular perceptual learning (ΔBSSmean = −2) and patching (ΔBSSmean = −9). This difference of change in binocular summation score among treatment groups was statistically significant (F = 47, P < .01).

Visual function in good compliance group. Change in visual acuity (VA), binocular summation score (BSS), and reading speed (both eyes [OU], amblyopic eye [AE]) with 8 weeks of treatment (binocular perceptual learning [2], monocular perceptual learning [4], patching [1]) in the good compliance group. Error bars show mean ± spherical equivalent (SE). The only variable that significantly varied with treatment group was BSS (ΔBSS = 5, −2, and −9 letters for binocular perceptual learning, monocular perceptual learning, and patching, respectively).

Figure 4.

Visual function in good compliance group. Change in visual acuity (VA), binocular summation score (BSS), and reading speed (both eyes [OU], amblyopic eye [AE]) with 8 weeks of treatment (binocular perceptual learning [2], monocular perceptual learning [4], patching [1]) in the good compliance group. Error bars show mean ± spherical equivalent (SE). The only variable that significantly varied with treatment group was BSS (ΔBSS = 5, −2, and −9 letters for binocular perceptual learning, monocular perceptual learning, and patching, respectively).

Effect of Time

To determine whether there was stability of effects over time, only the participants in binocular (n = 7) and monocular (n = 8) perceptual learning groups during the entirety of the study were included for the following analysis. Visual acuity (both eyes, amblyopic eye, unaffected eye) improved during the first 8 weeks of treatment by 1.1, 5.1, and 1.6 letters, respectively, then decreased by 15.5, 6.9, and 15.1 letters, respectively, during the second half of the study (F = 7.9, 4.9, and 8.1; P = .009, .04, and .009, for visual acuity of both eyes, amblyopic eye, and unaffected eye, respectively).

Discussion

It was previously believed that amblyopia treatment was only effective during the critical period of vision development in early childhood, but some amblyopic adults show improvement with binocular therapy, suggesting that the adult visual cortex retains some plasticity.13

There has been growing evidence that penalizing and occlusion interventions such as patching or atropine drops may not fully treat all aspects of amblyopia,20 and new techniques such as dichoptic therapy and perceptual learning have generated excitement as potential solutions for patch-resistant patients. A proposed mechanism of resistance to conventional amblyopia treatment in adults is that the unaffected eye sends inhibitory signals that continue to suppress the amblyopic eye.13 It is hypothesized that dichoptic therapy can decrease the gating mechanism of the unaffected eye, allowing the amblyopic eye and brain to learn to see.13 The theory of these dichoptic therapies and perceptual learning is that binocular visual training, aimed at stimulating the amblyopic eye, can ameliorate or eliminate the interocular suppression.1,12,13,21

Existing data regarding the effectiveness of binocular therapy for treatment of amblyopia are controversial. Randomized controlled trials have not demonstrated the superiority of binocular therapy compared to patching, and some suggest that it may be inferior to patching.8,9 However, there is some historical evidence suggesting the promising potential of binocular therapy. For instance, the use of I-BiT, a computer-based interactive system that presents background scenery to both eyes with additional images (ie, coins and obstacles) only for the amblyopic eye resulted in improvement of visual acuity in children with amblyopia.10 Similarly, some smaller studies of dichoptic games resulted in improved amblyopic visual acuity and stereopsis.6 It is possible that dichoptic treatment may be effective in a subgroup of amblyopia that is yet to be elucidated.

The current study has not shown strong evidence that self-administered, at-home use of a perceptual learning–based video game provides additional therapeutic benefits in patients who are resistant to patching. Age was a significant factor only in the monocular perceptual learning group, with older children performing better than younger children, likely due to bias from the small sample size (n = 1 in the older age group). The positive effect of binocular perceptual learning on reading speed for the good but not the poor compliance group also suggests that there may be a threshold of therapy use at which this treatment starts to have therapeutic effect. Similarly, the improvement in binocular summation score in binocular perceptual learning compared to monocular perceptual learning and patching groups only in those with good compliance also raises the possibility that binocular perceptual learning may have a functional benefit when used more consistently. However, 8 additional weeks of therapy for a total of 16 non-consecutive weeks did not have a therapeutic benefit compared to 8 weeks, which may suggest that length of time may not be as essential as frequent and compliant use.

The main limitation of this study is the small sample size, which precludes the ability to meaningfully analyze subgroups of amblyopia types and ages. Power calculation was performed using means and standard deviation from previous studies of binocular summation scores in patients with strabismus to detect a 0.2 difference in binocular summation ratio scores between treatment groups for α = 0.05 and β = 20%. It was determined that each treatment group would need 25 patients, which would imply that this study was too small to detect significant results. However, it was not feasible to include 75 patients in this pilot study, which was set up mainly to determine whether there was a significant improvement that could be attributable to the use of perceptual learning, which might motivate further research. In addition, this is a single center study, and may represent a homogeneous population because the majority of patients reported to be white. There was also difficulty in obtaining certain types of data due to various reasons (illiteracy, poor cooperation, etc). In addition, compliance data were available for only approximately half of participants, which further decreased the sample size when compliance level was incorporated into the analysis. Furthermore, overall compliance of the study participants was low, with only 4 of 12 participants classified as having a good compliance level.

It is important to note that this study looked at only those who were resistant to patching therapy. It would be important to understand how perceptual learning would affect patients with amblyopia prior to patching therapy or in combination with patching. There is also a need for further investigation with a randomized clinical trial with a larger sample size to better determine whether binocular therapy with perceptual learning techniques can benefit all, or only a certain subgroup of patients with amblyopia resistant to patching therapy. Also, more studies need to be conducted to detect if there is a certain threshold of length of time and frequency of treatment to affect visual function, and whether these effects are lasting.

Studies have shown that certain gamified training approaches both increase compliance with training and give rise to improved training outcomes.14–16 This current model of game-play was not successful in motivating children to participate consistently. Although we do not have consistent reports regarding the extent to which children of different ages enjoyed the game, subjective feedback suggests that there are substantial individual differences where some children enjoyed the game and parents saw value in the game, whereas others did not. The factors that determine this are unclear and it would be valuable to obtain feedback from users of different ages and demographics to understand individual needs to better determine who may or may not be the best targets of such an intervention and also how to improve the games to better motivate a larger demographic. In future studies, factors such as enrollment of only those with high compliance during a trial period, daily reminders, encouraging feedback associated with both compliance and success that gives players a sense of progression across sessions, visual aesthetics, various designs to target different age groups, and incentives for better compliance may be incorporated in efforts to improve compliance. Further, there are broader questions regarding what types of vision exercises are most appropriate for different etiologies of amblyopia, the ideal durations of such exercises, and how interventions for these may be best combined and used across children's development trajectories. For example, is it better to start with a contrast sensitivity training such as that used here and then follow with a binocular treatment, or vice versa?

References

  1. Levi DM, Li RW. Perceptual learning as a potential treatment for amblyopia: a mini-review. Vision Res. 2009;49(21):2535–2549. doi:10.1016/j.visres.2009.02.010 [CrossRef]
  2. Repka MX, Kraker RT, Holmes JM, et al. Atropine vs patching for treatment of moderate amblyopia: follow-up at 15 years of age of a randomized clinical trial. 2014;132(7):799–805. doi:10.1001/jamaophthalmol.2014.392 [CrossRef]
  3. Bhola R, Keech RV, Kutschke P, Pfeifer W, Scott WE. Recurrence of amblyopia after occlusion therapy. Ophthalmology. 2006;113(11):2097–2100. doi:10.1016/j.ophtha.2006.04.034 [CrossRef]
  4. Wallace DK, Lazar EL, Melia M, et al. Pediatric Eye Disease Investigator Group. Stereoacuity in children with anisometropic amblyopia. J AAPOS. 2011;15(5):455–461. doi:10.1016/j.jaapos.2011.06.007 [CrossRef]
  5. Kelly KR, Jost RM, De La Cruz A, Birch EE. Amblyopic children read more slowly than controls under natural, binocular reading conditions. J AAPOS. 2015;19(6):515–520. doi:10.1016/j.jaapos.2015.09.002 [CrossRef]
  6. Hess RF, Thompson B, Baker DH. Binocular vision in amblyopia: structure, suppression and plasticity. Ophthalmic Physiol Opt. 2014;34(2):146–162. doi:10.1111/opo.12123 [CrossRef]
  7. Narasimhan S, Harrison ER, Giaschi DE. Quantitative measurement of interocular suppression in children with amblyopia. Vision Res. 2012;66(66):1–10. doi:10.1016/j.visres.2012.06.007 [CrossRef]
  8. Kelly KR, Jost RM, Dao L, Beauchamp CL, Leffler JN, Birch EE. Binocular iPad game vs patching for treatment. JAMA Ophthalmol. 2016;134(12):1402–1408. doi:10.1001/jamaophthalmol.2016.4224 [CrossRef]
  9. Holmes JM, Manh VM, Lazar EL, et al. Pediatric Eye Disease Investigator Group. Effect of a binocular ipad game vs part-time patching in children aged 5 to 12 years with amblyopia a randomized clinical trial. JAMA Ophthalmol. 2016;134(12):1391–1400. doi:10.1001/jamaophthalmol.2016.4262 [CrossRef]
  10. Herbison N, Cobb S, Gregson R, et al. I-BiT study group. Interactive binocular treatment (I-BiT) for amblyopia: results of a pilot study of 3D shutter glasses system. Eye (Lond). 2013;27(9):1077–1083. doi:10.1038/eye.2013.113 [CrossRef]
  11. Deveau J, Seitz AR. Applying perceptual learning to achieve practical changes in vision. Front Psychol. 2014;5:1166. doi:10.3389/fpsyg.2014.01166 [CrossRef]
  12. Ooi TL, Su YR, Natale DM, He ZJ. A push-pull treatment for strengthening the ‘lazy eye’ in amblyopia. Curr Biol. 2013;23(8):R309–R310. doi:10.1016/j.cub.2013.03.004 [CrossRef]
  13. Li J, Thompson B, Deng D, Chan LYL, Yu M, Hess RF. Dichoptic training enables the adult amblyopic brain to learn. Curr Biol. 2013;23(8):R308–R309. doi:10.1016/j.cub.2013.01.059 [CrossRef]
  14. Risenhuber M. An action video game modifies visual processing. Trends Neurosci. 2004;27(2):72–74. doi:10.1016/j.tins.2003.11.004 [CrossRef]
  15. Green CS, Bavelier D. Action-video-game experience alters the spatial resolution of vision. Psychol Sci. 2007;18(1):88–94. doi:10.1111/j.1467-9280.2007.01853.x [CrossRef]
  16. Li RW, Ngo C, Nguyen J, Levi DM. Video-game play induces plasticity in the visual system of adults with amblyopia. PLoS Biol. 2011;9(8):e1001135. doi:10.1371/journal.pbio.1001135 [CrossRef]
  17. Deveau J, Ozer DJ, Seitz AR. Improved vision and on-field performance in baseball through perceptual learning. Curr Biol. 2014;24(4):R146–R147. doi:10.1016/j.cub.2014.01.004 [CrossRef]
  18. Deveau J, Lovcik G, Seitz AR. Broad-based visual benefits from training with an integrated perceptual-learning video game. Vision Res. 2014;99:134–140. doi:10.1016/j.visres.2013.12.015 [CrossRef]
  19. Wallace DK, Edwards AR, Cotter SA, et al. Pediatric Eye Disease Investigator Group. A randomized trial to evaluate 2 hours of daily patching for strabismic and anisometropic amblyopia in children. Ophthalmology. 2006;113(6):904–912. doi:10.1016/j.ophtha.2006.01.069 [CrossRef]
  20. Loudon SE, Simonsz HJ. The history of the treatment of amblyopia. Strabismus. 2005;13(2):93–106. doi:10.1080/09273970590949818 [CrossRef]
  21. Polat U, Ma-Naim T, Belkin M, Sagi D. Improving vision in adult amblyopia by perceptual learning. Proc Natl Acad Sci USA. 2004;101(17):6692–6697. doi:10.1073/pnas.0401200101 [CrossRef]

Participant Characteristics

Characteristicn
Amblyopia type
  Anisometropic10
  Mixed4
  Strabismic11
Age (y)
  8 to 1319
    8 to 99
    10 to 114
    12 to 136
  14 to 186
    14 to 152
    16 to 184
Patching history
  Yes19
  No6
Ethnicity
  White12
  Hispanic5
  Asian1
  Unknown7
Sex
  Female10
  Male15

t Test Before and After Treatment

VariableEyePretreatment (Mean)Posttreatment (Mean)|t|P
VA (letters)OU52.748.81.9.06
AE24.124.30.1.90
UE50.947.31.7.09
LA (letters)OU33.534.71.3.20
AE12.916.63.3.002a
UE30.431.71.2.20
BSS (letters)3.02.60.4.70
Reading speed (wpm)OU186.5197.51.2.20
AE143.9154.81.2.20
Stereoacuity (near) (arcsec)OU73.270.70.051.00
Stereoacuity (distance) (arcsec)OU57.173.80.7.50

One-way ANOVA Comparison Among Three Treatment Groups for Difference of Means Before and After Treatment

Δ Variable (Post–Pre)EyeBinocular PL (Mean)Monocular PL (Mean)Patching (Mean)FP
VA (letters)OU−5.1−5.82.21.1.30
AE1.1−2.02.70.6.60
UE−5.1−5.94.11.8.20
LA (letters)OU−0.82.42.71.6.20
AE3.73.44.10.01.00
UE−1.82.15.93.9.03a
BSS (letters)0.80.1−3.21.7.20
Reading speed (wpm)OU10.84.220.30.2.90
AE5.026.09.80.4.70
Near stereoacuity (arcsec)OU50.0−52.90.00.4.60
Distance stereoacuity (arcsec)OU−5.958.8−25.01.2.30
Authors

From the Departments of Ophthalmology/Stein Eye Institute (YHL, FV, JLD, SP) and Neurology (JLD), David Geffen School of Medicine, University of California Los Angeles, Los Angeles, California; the Department of Psychology, University of California Riverside, Riverside, California (MM, ARS); the Department of Ophthalmology, Duke University School of Medicine, Durham, North Carolina (FV); and Doheny Eye Institute, Los Angeles, California (FV).

The authors have no financial or proprietary interest in the materials presented herein.

Supported by the National Institute of Health National Eye Institute (R01EY023582-03) and the Research to Prevent Blindness Disney Award for Amblyopia Research.

Correspondence: Stacy Pineles, MD, Stein Eye Institute, University of California Los Angeles, 100 Stein Plaza, Los Angeles, CA 90095. E-mail: pineles@jsei.ucla.edu

Received: October 01, 2019
Accepted: February 19, 2020

10.3928/01913913-20200306-01

Sign up to receive

Journal E-contents