The human visual system achieves clear vision of the environment by modulating the eye’s accommodative state in response to sensory cues of retinal blur, disparity, image size, and target proximity.1–3 Under naturalistic binocular viewing conditions, all these sensory cues are consistent with each other for signaling a change in the eye’s accommodative state. For instance, a change in fixation from distance to near causes hyperopic blur in the retina, exo fixation disparity, and an increase in retinal image size and proximity, all of which signal an increase in accommodation to reinstate clear vision; the reverse happens when fixation changes from a near to distant target.1,4
The visual system is subject to many physiological and pathological scenarios where the sensory cues to accommodation are not all present and/or consistent with each other. Specifically, physiological scenarios like binocular viewing of targets that are placed at secondary or tertiary gaze angles at unequal distances from the two eyes5,6 and pathological conditions like strabismus7–9 result in the monocular blur/proximity cues to be mismatched and inconsistent with binocular disparity cues. There has been considerable interest in the field to understand how the accommodative system performs in the presence of such conflicting or cue impoverished conditions.10–13 For instance, it has been shown that the accommodative gain of typically developing children and adults decreases in the absence of binocular cues, relative to cue-consistent binocular viewing conditions.10,14,15 The accommodative gain also decreases when the disparity cue is placed in conflict with blur (e.g. when accommodation is stimulated with a pair of negative lenses while both eyes continue to converge on a target), relative to when both cues are consistent.11,13 A similar decrease in accommodative gain has also been reported in children with strabismus during periods of decompensation of their distance exotropia.7–9
Loss of functional vision in one eye is another scenario in which the visual system has to accommodate in the absence of some of the aforementioned sensory cues—specifically, in the absence of a binocular disparity cue. Children with this condition are either born with only one functional eye or they may be rendered uniocular during early development because of trauma or ocular pathology. These children are permanently devoid of binocularity and they perform all their habitual visual tasks monocularly.16 Accommodative responses of these children will also be driven solely by monocular cues without any support from binocular vergence-driven accommodation.1,4 Thus far, little is known about the accommodative behavior of these uniocular children. This is despite the fact that many such children are routinely followed up in the clinic to ensure that the affected eye’s orbital socket and fascia are free of recurrent pathology and the remaining functional eye is free from disease. Accommodation and other visuomotor functions of these children are seldom examined in follow-up visits, and it is generally assumed that the functional eye is not optically deficient in any significant manner. Several important questions have remained unanswered in these children. First, do the accommodative responses of these children resemble those of typically developing children who are purposely made monocular for transient periods of time? Alternatively, does the permanent loss of binocularity in these children modify the utilization of monocular cues to accommodation and/or the way in which their optics function to achieve clear vision? For instance, there could be increased pupil miosis in these children to expand optical depth of focus, thereby reducing the demand on blur-driven accommodation.17,18 Such a permanent loss of binocularity has been shown in the past to enhance performance in psychophysical tasks that require processing of spatial details in adults with uniocular vision.16 Overall, data acquired in this study will provide greater insights and novel information on how the developing visual system copes with the permanent absence of a binocular disparity cue to generate accommodative responses, vis-à-vis, age-matched controls.
This study systematically evaluated the accommodative behavior of a cohort of uniocular cases to ramp changes in near vision demand, relative to the binocular and monocular accommodative behavior of age-matched controls. For the purposes of this study, the phrases binocular viewing and monocular viewing refer to control subjects performing the task with both eyes open or with one eye open (fellow eye purposely occluded using an infrared filter), respectively, whereas the phrase uniocular viewing refers to cases performing the task with their healthy eye.
A total of 69 cases with uniocular vision (4 months–24 years of age) and 63 controls with normal binocular vision (3–28 years of age) were recruited for the study. There was maximum overlap in the ages of cases and controls between 3 and 14 years. Data of 41 cases and 43 controls from within this age range is therefore analyzed and presented here in this report. The study protocol adhered to the tenets of the Declaration of Helsinki, and it was approved by the Institutional Review Board of the L V Prasad Eye Institute (LVPEI), Hyderabad, India. The study commenced after parents of all children provided written informed consent for the study. The cases were recruited from the patient pool of LVPEI. Recruitment of cases depended on their clinical appointment at LVPEI and based on their availability and interest to participate in the study. Controls were recruited from the children and relatives of staff members of LVPEI. There was no pre-selection bias in recruiting the cases for this study, and no specific strategy was followed to match the ages of cases and controls at the time of recruitment. Recruitment of controls was an ongoing process and the subjects were recruited for the study as long as they were in the pediatric age group, met the inclusion criteria (noted below), and were available for participation in the study. The socioeconomic strata and the education level of adults and the parents of children who formed the cases and control cohorts were similar to each other. All controls underwent a cursory eye examination to rule out any organic pathology. All cases underwent a detailed eye examination in the routine clinics of LVPEI, and only those subjects in whom the functional eye was deemed completely free of any pathology were recruited for the study. Of the 69 cases recruited for this study, 44, 14, and 11 of them had become uniocular because of trauma, retinoblastoma tumor, and because of other causes (e.g. unilateral phthisis bulbi), respectively. Standard clinical management was followed in all the cases and there was no influence of the current study in their clinical care. The uncorrected and best-corrected distance and near visual acuity of both cases and controls were also recorded using standard clinical paradigms (i.e. logMAR acuity charts for older children and Lea symbols for toddlers). Subjects with logMAR acuity equivalent of 0.2 or better were included in the study (Table 1). Spherocylindrical refractive error of all subjects was obtained using non-cycloplegic retinoscopy (Table 1). Subjects with uncorrected myopia of ≥0.5D were excluded from the study.
Three experiments were performed in this study. The first experiment evaluated the accommodative gain of all cases and controls to ramp changes in near vision demand. The second experiment determined the magnitude of pupil miosis of cases and controls while performing the near vision task in the first experiment. The third experiment determined the impact of cognitive load imposed by the near vision task on the accommodative responses of cases and controls to ramp changes in near vision demand. The overall paradigm used in all these experiments were very similar to that used by Bharadwaj and Candy to demonstrate differences in the binocular and monocular accommodative and pupil performance of typically developing children.10,19 Briefly, all subjects watched a colored, high-contrast cartoon movie with broadband spatial frequency content that was displayed on an LCD screen while it ramped six times between 90 and 30 cm (2.2D change in near vision demand) before the subject at 0.2D/s speed, with 4 seconds of stable position at each viewing distance. Subject’s accommodative and pupil responses were recorded at 50 fps in sync with the stimulus motion using the PlusOptix PowerRef3 photorefractor (Plusoptix, Nuremberg, Germany).10,20 The photorefractor was positioned midline between the two eyes of the subject at a distance of 1 m, orthogonal to the stimulation channel (Fig. 1). The machine obtained photorefraction images that were reflected off an infrared (IR) reflecting hot-mirror (Tower Optical Corporation, Florida) aligned at 45 degrees to the subject (Fig. 1). All controls performed this task twice—first under the binocular viewing condition and a second time under the monocular viewing condition. In the monocular viewing condition, the right eye was occluded with an Optcast IR transmitting filter (Edmund Optics, NT43-954) that blocked most of the visible wavelengths while allowing the photorefractor to collect data. Thus, even though viewing was monocular, valid data was collected by the photorefractor from both eyes in this paradigm. Infants and toddlers tend to become fussy and do not cooperate with the task when one eye is occluded. This loss of cooperation has been observed in the past for tasks related to visual acuity,21 accommodation, and vergence.10,14,22 Therefore, to maximize subject cooperation and reduce the chances of no data being collected from the subject (especially from toddlers), the binocular viewing paradigm was always performed before the monocular viewing paradigm in controls. All cases performed the task once under the uniocular viewing condition.
Raw traces of accommodation and pupil diameter recorded by the photorefractor were then smoothed using a 200-millisecond averaging window to improve the signal-to-noise ratio of the data.10 The stimulus and response data were smoothed using the same window to maintain the temporal relationship between them. The stimulus profile was then divided into epochs, each containing a 4-second stable stimulus period plus the change in stimulus before and after this period. The raw data within each epoch were then subjected to several quality checks, and they were included in the final data only when they met all of the following quality checks10: one, the accommodation data were within the linear operating range of the instrument (+4.0 to −7.0D) and the pupil diameters were between 3 and 9 mm (the manufacturer of the PowerRef3 photorefractor specifies 4–9 mm as the optimal operating range for pupil diameter for the instrument to collect data; we have however empirically observed that valid data can be collected with the instrument even with a 3-mm pupil diameter)20; two, the data collected were within 15-degree gaze eccentricity from the pupillary axis to minimize peripheral refraction effects23; and, three, the subject generated an eye movement to track the stimulus (i.e. they generated a convergence/divergence eye movement that is correlated with the target position in controls and an adduction/abduction eye movement of the functional eye that is correlated with the target position in cases), suggesting that the accommodative and pupil responses represented a valid attempt by the subject to follow the target moving in front of them.10 The data epochs of some subjects neither clearly met nor clearly failed all the aforementioned inclusion criteria. Such data epochs were manually analyzed by the investigators to determine their inclusion in the analysis. A conservative approach of rejecting the doubtful epoch was usually taken in these cases. Taken together, a minimum of two usable epochs had to be present in a given subject for their data to be included in the final analyses. Data from seven controls and one case was rejected because they failed (either clearly or doubtfully) to meet the inclusion criteria. Of these, data from two controls and one case were rejected because of the pupil size criterion and data from six controls and one case were rejected because of the eye movement criterion (this case had two of the three inclusion criteria violated). The accommodative response amplitude from all epochs that passed these quality checks was subsequently calculated for each subject as the difference in average (calculated over the central 2 s of stable viewing) accommodative states for the 2.2D accommodative demand.10 Similarly, the pupil miosis was calculated as the difference in the average pupil diameter over the same 2 seconds of stable viewing between the two stable viewing distances.19
In the first experiment, all subjects watched a cartoon movie passively and there was no explicit demand on the subject for resolving the high spatial frequency content of the target. Given that Bharadwaj and Candy previously observed that the monocular accommodative responses of typically developing 4- to 6-year-old children increased in magnitude when they were engaged in a text-reading task (requiring resolution of high spatial frequency information) versus a passive movie watching task,10 the third experiment in the present study was performed to determine whether a similar task-specific difference in accommodative performance existed in the uniocular subjects as well. Thirty-two cases (11.4 ± 5.8 years) and 14 controls (15.3 ± 6.9 years) who participated in the first experiment repeated the ramp task twice, once while watching the movie under the monocular viewing condition and a second time while reading aloud 20/40-sized alphabets displayed on the LCD screen also under the monocular viewing condition. All other details were identical to the first experiment.
Overall, only the binocular disparity cue was manipulated (present or absent) in all three experiments of this study whereas the proximity cue was present and remained consistent with blur cue throughout.
The photorefractor technique measures accommodation by converting the luminance slope formed across the pupil into units of diopters using a defocus calibration factor.20,24–26 Previous studies have found a significant inter-subject variability of this defocus calibration factor and have suggested the use of individual or ethnicity-specific calibration factors for improving the accuracy of refraction estimates.20,24–26 Manufacturers of the PowerRef3 photorefractor recommend the instrument to automatically adjust for the luminance gain formed across the pupil to reduce this variability. The gain was however set to a constant value of 50% in this study against the manufacturer recommendations to have complete control over the change in luminance profile per diopter change in eye’s refractive power. An attempt was therefore made to obtain individual-specific defocus calibration factors on all controls and cases using a protocol described previously.20,24,25 All controls fixated on the cartoon movie at a fixed viewing distance of 1 m with their right eye while the left eye was occluded using the IR transmitting filter. Trial lenses from −4D to +4D were placed before the occluded eye in 1D steps at 10 to 14 mm vertex distance for ~5 seconds each. No lens was placed before the right eye. Two seconds of stable data within this epoch were averaged in the left eye for each induced lens power. This was plotted against the induced lens power and a linear regression equation was fit to the linear portion of the data using custom-written Matlab software. The slope of this linear regression equation gave the defocus calibration factor. For cases, their only functional eye was occluded using the IR filter and the trial lenses were placed before this eye. The rest of the procedure was identical to controls. Effectively, the cases were devoid of form vision during the calibration process, and they were therefore constantly reminded to keep “looking straight towards the direction of the movie” in their vernacular language. In general, cases were less cooperative with this task than controls and the procedure was attempted multiple times on some of them to obtain a calibration factor before it was abandoned. The accommodation data of each subject was then scaled using their individual calibration factor, if available, or using the Indian population-average calibration factor obtained previously by Sravani et al. using a similar potocol.20 Usually, the defocus calibration factor is obtained by plotting the difference in refractive power between the two eyes (i.e. anisometropia) against the induced lens power to account for changes in the accommodative state of the eye during the calibration routine.20,24,25 This was however not possible for the uniocular cases who participated in this study and hence only changes in the eye’s refraction over which the trial lenses were placed was considered for calculating the defocus calibration factor in both cases and controls. For both groups, the eye’s refractive power was ensured to remain stable within ±0.5D throughout the calibration routine by monitoring the refractive power of the eye that fixated on the target (for controls) or by monitoring the refractive power of the eye in between periods of lens stimulation (for cases).20,24,25 In total, individual calibration factors were available from 19 of the 41 cases and all 43 controls who participated in the study.
The Kolmogorov-Smirnov test indicated that the primary output variables in this study (accommodative gain and pupil miosis) did not follow a normal distribution. Nonparametric statistics were therefore used for all statistical analyses. A P value of ≤.05 was considered statistically significant in all analyses reported below. Table 1 shows descriptive statistics of age, spherical equivalent refractive error (astigmatism was ≤0.75D with axis 90 ± 20° in all subjects; anisometropia was ≤0.50D in all controls), logMAR visual acuity, duration of uniocular status (for cases), accommodative gain, and pupil miosis obtained from the controls and cases in this study along with relevant p values.
The accommodative responses of controls to ramp changes in near vision demand were more robust under the binocular than the monocular viewing condition. The uniocular accommodative responses of cases largely resembled the monocular accommodative responses of controls obtained in this study. The binocular accommodative gain of controls was higher than the corresponding monocular accommodative gain, with a median accommodative gain of 0.95 [interquartile range (IQR): 0.81–1.11)] and 0.56 (IQR: 0.47–0.79) under the binocular and the monocular viewing condition, respectively (U = 882, z = 5.8, P < .001) (Fig. 2A, Table 1). The median ratio of monocular to binocular accommodative gain of controls was 0.62 (IQR: 0.45–0.88) (Fig. 2B). The median uniocular accommodative gain of cases was 0.73 (IQR: 0.60–0.85), and this was significantly higher than the monocular accommodative gain of controls (U = 861, z = 2.2, P = .03) (Fig. 2C, Table 1) and significantly lower than the binocular accommodative gain of controls (U = 861, z = 4.5, P < .001) (Table 1; comparison data not shown directly in a figure). The subject’s age (for controls and cases) and the duration of uniocular status (for cases only) were poorly correlated with the binocular and monocular accommodative gain of controls (Spearman’s rank correlation coefficient; r = −0.01; P = .8 for age) and with the uniocular accommodative gain of cases (r ≤ 0.25; P ≥ .1; for both age and time of deficit). This result suggested that the accommodative performance observed in this study was not associated with the subject’s age or with when cases became functionally uniocular.
The accommodative data of all controls were scaled using the subject’s individual defocus calibration factor, as described in the methods section. However, the data of only 19 out of 41 cases were scaled using the subject’s individual defocus calibration factor; the data of the remaining cases (n = 22) were scaled using Indian population average calibration factor. To determine if there was a difference in the characteristics of these cases, the median (25th–75th IQR) age and the accommodative gain were calculated for the subcategory of cases whose accommodative data were scaled using individual calibration factor and of cases whose data were scaled using population average calibration factor. The median age of the former subcategory [12 years (IQR: 10–13.5 years)] was statistically significantly higher than the latter subcategory [7.5 years (IQR: 5–11 years)] (U = 823, z = 2.8, P < .001). This is somewhat expected given that the calibration protocol involved occluding the only functional eye of cases, leading to younger children becoming more uncooperative than older ones. The accommodative gain of the former subcategory [0.73 (IQR: 0.62–0.79)] was however not significantly different from the latter subcategory [0.73 (IQR: 0.55–0.85)] (U = 23, z = 0.01, P = .8), suggesting that the calibration factor used for data scaling did not have a strong influence on the accommodative performance reported here. The IQR of the latter subcategory was larger than the former subcategory, indicating that the inter-subject variability of accommodative gain reduced when data were scaled using the individual’s own calibration factor. This result follows the expectation of an increase in precision of photorefraction-based refraction estimates obtained using the individual’s defocus calibration.20,24
Results of the second experiment conducted in this study from all cases and controls who participated in the first experiment are presented below. As observed in previous studies,19 the pupil diameter for distance (90 cm) and near (30 cm) viewing and the magnitude of near-pupil miosis showed a large inter-subject variability in controls and cases (Fig. 3, Table 1). The median (25th–75th IQR) binocular pupil diameter of controls was 6.4 mm (IQR: 5.9–6.8 mm) and 6.1 mm (IQR: 5.7–6.6 mm) for distance and near, respectively (U = 882, z = 1.4, P = .15), with a median pupil miosis of 0.23 mm (IQR: 0.14–0.34 mm) (Fig. 3A and D, Table 1). The median (25th–75th IQR) monocular pupil diameter of controls was 6.4 mm (IQR: 5.8–6.4 mm) and 6.2 mm (IQR: 5.8–6.6 mm) for distance and near, respectively (U = 882, z = 1.01, P = .3), with a median pupil miosis of 0.12 mm (IQR: 0.05–0.29 mm) (Fig. 3B and D, Table 1). The median (25th–75th IQR) uniocular pupil diameter of cases was 6.1 mm (IQR: 5.7–6.7 mm) and 5.9 mm (IQR: 5.5–6.6 mm) for distance and near, respectively (U = 840.5, z = −0.01, P = .99), with a median pupil miosis of 0.14 mm (IQR: 0.06–0.24 mm) (Fig. 3C and D, Table 1). The median (25th–75th IQR) pupil miosis of controls was larger under binocular than monocular viewing conditions (U = 882, z = 2.5, P = .01) (Fig. 3D, Table 1). The median uniocular pupil miosis of cases was not statistically significantly different from the monocular pupil miosis of controls (U = 861, z = 0.40, P = .69), but it was significantly smaller than the binocular pupil miosis of controls (U = 861, z = 2.3, P = .02) (Fig. 3D, Table 1). The subject’s age (for controls and cases) and the duration of uniocular status (for cases only) were poorly correlated with the binocular and monocular pupil miosis of controls (Spearman’s rank correlation coefficient; r = −0.31; P = .33 for age) and with the uniocular pupil miosis of cases (r ≤ −0.22; P ≥ .1; for both factors).
Results of the third experiment conducted in this study from a subset of 32 cases and 14 controls who participated in the first experiment are presented below. The median monocular accommodative gain of controls when watching the movie was 0.41 (IQR: 0.23–0.65), and it increased to 0.74 (IQR: 0.60–0.97) when reading alphabets on an LCD screen (Fig. 4A, Table 1). Similarly, the median uniocular accommodative gain of cases when watching the movie was 0.67 (IQR: 0.58–0.79) and it increased to 0.71 (IQR: 0.65–0.87) when reading the alphabets (Fig. 4B, Table 1). If the data were normally distributed, the impact of visual task on the accommodative performance of cases and controls would be analyzed using a two-factor ANOVA test. However, these data were not normally distributed and there is no established equivalent of a two-factor ANOVA test in nonparametric statistics for performing this analysis. Hence, a standard Mann–Whitney–Wilcoxon test was applied to the data of controls and cases separately to analyze the impact of visual task on the accommodative gain. The results indicated that the increase in accommodative gain from movie watching to reading was statistically significant in controls (U = 32, z = 2.54, P = .02) but not in cases (U = 180.5, z = 1.44, P = .15). The difference in the accommodative gain between cases and controls was statistically significant only for movie watching (U = 76, z = 1.93, P = .05) but not for alphabet reading (U = 76, z = 0.0001, P > .99).
The overall purpose of this study was to understand how human accommodation varies with the availability or consistency of sensory cues used for generating these responses. Most previous studies have addressed this issue by comparing the accommodative performance of typically developing children under habitual binocular viewing conditions to experimental conditions in which one of the cues to accommodation is purposely removed or made inconsistent with other cues for transient periods of time.10–13,15,22 Although this paradigm offers good control over the characteristics of sensory cues that are manipulated, the results obtained may not represent the habitual behavior of children under cue impoverished/cue conflict situations—they may merely represent the behavior of typically developing children when encountered with such a novel viewing experience. This study addressed the aforementioned issue by comparing the responses of uniocular children who accommodate in the permanent absence of a binocular cue to the habitual binocular and monocular responses of age-matched controls.
Previous studies have observed that the accommodative gains of the typically developing visual system for ramp changes in near vision demand are smaller in the absence of binocular cues than under cue-consistent binocular viewing conditions.10,12,15,22 The results of the present study replicated these findings and also showed that the uniocular accommodative gain of cases were significantly higher than the monocular accommodative gain of controls (median difference in accommodative gain was 0.17) and significantly smaller than the binocular accommodative gain of controls (median difference was 0.22) (Fig. 2, Table 1). These results indicate that children with only one functional eye are as capable or may be slightly better than age-matched controls at utilizing monocular sensory cues to drive their accommodative response. These results are in line with previous literature reporting that the performance of the uniocular cohort in spatial vision tasks (e.g. contrast sensitivity) exceeds the monocular performance of age-matched controls with normal binocular vision or sometimes equals the binocular performance of age-matched controls.16 The present study also found that uniocular pupil miosis of cases was not significantly different from the monocular pupil miosis of controls whereas they were significantly different from the binocular pupil miosis of controls (Fig. 3, Table 1). This suggests that the pupillary system does not in any way alter the near vision demand experienced by uniocular children relative to that of age-matched controls.
The uniocular gains reported here represent the habitual accommodative behavior of cases whereas the monocular gains of controls represent behavior only during periods of transient monocularity—the binocular gains represent the habitual accommodative behavior of controls, instead. The accommodative responses of both cohorts are likely to represent a state that optimizes retinal image quality for the given viewing condition. Increase in the accommodative gain of the control cohort with the cognitive demand of the task is certainly supportive of this notion (Fig. 4).10 The smaller gain of the uniocular accommodation of cases and the monocular accommodation of controls may reflect a response state that maximizes the retinal image quality for that subject and for that viewing condition. Any further increase in response magnitude may not yield any benefit to image quality as the responses may be within the eye’s optical depth of focus. A similar logic has been applied to explain the reduction in accommodative gain of adults with an increase in the higher-order wavefront aberrations of the eye.27,28 The larger gain of binocular accommodation may reflect an obligatory input from the vergence crosslink (vergence-accommodation to vergence ratio) that may place the accommodative state within the eye’s optical depth of focus. Whether such an obligatory vergence-driven accommodative response results in a subtle loss of image quality under binocular viewing conditions remains to be evaluated. A detailed through-focus analysis of how the eye’s image quality varies with optical focus under monocular and binocular viewing conditions is necessary to answer this question in detail.
The smaller accommodative gains observed in cases and in monocular viewing of controls may reflect performance to a naturalistic and passive near task that requires limited mental effort and processing of high spatial frequencies (movie watching, in this study). Both groups showed an improvement in the accommodative gain when reading 20/40-sized alphabets [although this difference was statistically significant only for controls and not for cases (Fig. 4, Table 1)], reflecting the accommodative state that provides the most optimal image quality when performing a near task that required increased mental effort and processing of high spatial frequency information (Fig. 4). Even though the monocular movie-watching accommodative gain of controls that participated in the cognitive task [0.41 (IQR: 0.23–0.65)] was significantly smaller than the corresponding uniocular accommodative gain of cases [0.67 (IQR: 0.58–0.79)], the gains of both groups increased to similar levels with the alphabet reading task [controls: 0.74 (IQR: 0.60–0.97); cases: 0.71 (IQR: 0.65–0.87)] (Fig. 4). Although the reason for the difference in movie watching gain between cohorts is unknown, the greater increase in accommodative gain of controls is likely caused by their relatively smaller accommodative gain during movie watching. The monocular near acuity of all subjects on whom this data could be collected was better than 0.2 logMAR units, also implying that the accommodative response should have been better than what was observed during movie watching for resolving the near-vision optotypes. Taken together, these observations suggest that both cases and controls are capable of using monocular blur and proximity cues to drive accommodation under visually demanding situations, but the controls do not seem to do so under more naturalistic viewing.
This study had three limitations. First, the binocular viewing condition was always performed before the monocular viewing condition in controls. The familiarity of control subjects with the accommodative task is likely to have been somewhat better by the time the monocular viewing paradigm was presented. This added advantage of familiarity may not be present in cases who performed the task only once under the uniocular viewing condition. It is therefore possible that the difference in accommodative gain between the monocular viewing of controls and the uniocular viewing of cases could be larger than what was observed had task familiarity been equalized in the two cohorts. Second, even while it was qualitatively ensured that form vision was not present through the IR filter used in this study, the filter itself is not completely opaque to visible light but transmits some light in the visible spectrum (~20% at 700 nm; http://www.edmundoptics.com/optics/optical-filters/longpass-edge-filters/optical-cast-infrared-ir-longpass-filters/1918/). The level of visibility through the filter may also depend on the spectral content of the target presented for accommodation (movie vs. alphabet target), which, again, was not tested in this study. The viewing experience of the control cohort was therefore strictly not monocular in the present study. The influence of this on the binocular and monocular accommodation and pupil data presented in this study remains unknown and needs further exploration. Third, the accommodative gains reported in this study cannot be used to make inferences about the lag of accommodation and therefore on the quantum of blur experienced by cases and controls during the near task. The accommodative lag represents the difference between the magnitude of accommodative demand experienced by the subject and the magnitude of accommodative response generated to overcome this demand. The accommodative gain represents the magnitude of change in accommodation for a given change in near vision demand. None of the subjects wore any refractive correction during data collection to mimic their habitual viewing experience, and therefore any influence of their uncorrected refractive error on the absolute accommodative demand and therefore the accommodative lag cannot be ascertained in this study. An analysis of accommodative lag also requires the photorefractor used in this study to be calibrated for absolute refraction values25—only calibration of the relative changes in the magnitude of refraction was however performed here.
In conclusion, the accommodative gain of children with permanent loss of binocularity is in between the binocular and monocular gains of typically developing children. Their accommodative gains do not show any significant increase with a cognitively demanding task even while such a behavior is observed in controls. Pupil responses of uniocular children are similar to the monocular responses of age-matched controls.
Shrikant R. Bharadwaj
Brien Holden Institute of Optometry and Vision Sciences
L V Prasad Eye Institute
Road no. 2, Banjara Hills
Hyderabad 500034 Telangana
The authors would like to thank all subjects who participated in the study. The authors would also like to thank Bill Monette and Tom Kemerley from the Indiana University School of Optometry for designing the instrument used in this study.
This research was supported by a Fast-Track Grant for Young Scientists scheme, Science and Engineering Research Board, Department of Science and Technology, Government of India to SB, and a Champalimaud Foundation grant to the Prof. Brien Holden Eye Research Centre, L V Prasad Eye Institute.
Commercial relationship: None. Financial interest: None.
Received June 20, 2015; accepted August 11, 2016.
1. Eadie AS, Carlin PJ. Evolution of control system models of ocular accommodation
and their interaction. Med Biol Eng Comput 1995;33:517–24.
2. Heath GG. Components of accommodation
. Am J Optom Arch Am Acad Optom 1956;33:569–79.
3. Mays LE, Gamlin PD. Neuronal circuitry controlling the near response. Curr Opin Neurobiol 1995;5:763–8.
4. Schor CM, Alexander J, Cormack L, et al. Negative feedback control model of proximal convergence and accommodation
. Ophthalmic Physiol Opt 1992;12:307–18.
5. Charman WN. Aniso-accommodation
as a possible factor in myopia development. Ophthalmic Physiol Opt 2004;24:471–9.
6. Nguyen D, Vedamurthy I, Schor C. Cross-coupling between accommodation
and convergence is optimized for a broad range of directions and distances of gaze. Vision Res 2008;48:893–903.
7. Horwood AM, Riddell PM. Decreased accommodation
during decompensation of distance exotropia. Br J Ophthalmol 2012;96:508–13.
8. Horwood AM, Riddell PM. Accommodation
response gains to different near cues characterize specific esotropias. Strabismus 2013;21:155–64.
9. Horwood AM, Riddell PM. Disparity
-driven vs blur
-driven models of accommodation
and convergence in binocular
vision and intermittent strabismus. J AAPOS 2014;18:576–83.
10. Bharadwaj SR, Candy TR. Cues for the control of ocular accommodation
during postnatal human development. J Vis 2008;8(14):1–6.
11. Bharadwaj SR, Candy TR. Accommodative and vergence
responses to conflicting blur
stimuli during development. J Vis 2009;9:1–18.
12. Horwood AM, Riddell PM. Developmental changes in the balance of disparity
, and looming/proximity cues to drive ocular alignment and focus. Perception 2013;42:693–715.
13. Ramsdale C, Charman WN. Accommodation
and convergence: effects of lenses and prisms in ‘closed-loop’ conditions. Ophthalmic Physiol Opt 1988;8:43–52.
14. Currie DC, Manny RE. The development of accommodation
. Vision Res 1997;37:1525–33.
15. Horwood AM, Riddell PM. The use of cues to convergence and accommodation
in naïve, uninstructed participants. Vision Res 2008;48:1613–24.
16. Steeves JK, González EG, Steinbach MJ. Vision with one eye: a review of visual function following unilateral enucleation. Spat Vis 2008;21:509–29.
17. Charman WN, Whitefoot H. Pupil diameter and the depth-of-field of the human eye as measured by laser speckle. Opt Acta 1977;24:1211–6.
18. Ward PA, Charman WN. Effect of pupil size
on steady state accommodation
. Vision Res 1985;25:1317–26.
19. Bharadwaj SR, Wang J, Candy TR. Pupil responses to near visual demand during human visual development. J Vis 2011;11(4):1–6.
20. Sravani NG, Nilagiri VK, Bharadwaj SR. Photorefraction
estimates of refractive power varies with the ethnic origin of human eyes. Sci Rep 2015;5:7976.
21. Atkinson J, Braddick O, Pimm-Smith E. ‘Preferential looking’ for monocular
acuity testing of infants. Br J Ophthalmol 1982;66:264–8.
22. Turner JE, Horwood AM, Houston SM, et al. Development of the response AC/A ratio over the first year of life. Vision Res 2002;42:2521–32.
23. Seidemann A, Schaeffel F, Guirao A, et al. Peripheral refractive errors in myopic, emmetropic, and hyperopic young subjects. J Opt Soc Am (A) 2002;19:2363–73.
24. Bharadwaj SR, Sravani NG, Little JA, et al. Empirical variability in the calibration of slope-based eccentric photorefraction
. J Opt Soc Am (A) 2013;30:923–31.
25. Blade PJ, Candy TR. Validation of the PowerRefractor for measuring human infant refraction. Optom Vis Sci 2006;83:346–53.
26. Schaeffel F, Wilhelm H, Zrenner E. Inter-individual variability in the dynamics of natural accommodation
in humans: relation to age and refractive errors. J Physiol 1993;461:301–20.
27. Buehren T, Collins MJ. Accommodation
stimulus-response function and retinal image quality. Vision Res 2006;46:1633–45.
28. Plainis S, Ginis HS, Pallikaris A. The effect of ocular aberrations on steady-state errors of accommodative response. J Vis 2005;5:466–77.