Effects of Parkinson Disease on Blur-Driven and Disparity-Driven Vergence Eye Movements

Synchronous movements of the 2 eyes in the opposite direction, disconjugate movements such as ver- gence, facilitate depth perception. The vergence eye move- ments are affected in Parkinson disease (PD). Visual blur (accommodation) and fusion (retinal disparity) are important triggers for the vergence. The neural circuit responsible for blur-driven and disparity-driven vergence is tightly coupled. We investigated the effect of PD on these 2 vergence paradigms. In the experiment involving 14 patients with PD and 6 healthy controls, substantial differences between blur- driven and disparity-driven vergence were found. The gain (ratio of actual vs desired eye movements) was reduced in patients with PD in case of disparity-driven vergence but not in blur-driven vergence. The latency of disparity-driven ver- gence onset was significantly longer for patients with PD compared with healthy controls.

Basic and Translational Research Section Editors: Jeffrey L. Bennett, MD, PhD Kenneth S. Shindler, MD, PhD Effects of Parkinson Disease on Blur-Driven and Disparity-Driven Vergence Eye Movements Palak Gupta, BS, Sinem Beylergil, PhD, Jordan Murray, PhD, Jonathan Jacobs, PhD, Camilla Kilbane, MD, Aasef G. Shaikh, MD, PhD, Fatema F. Ghasia, MD Abstract: Synchronous movements of the 2 eyes in the opposite direction, disconjugate movements such as vergence, facilitate depth perception. The vergence eye movements are affected in Parkinson disease (PD). Visual blur (accommodation) and fusion (retinal disparity) are important triggers for the vergence. The neural circuit responsible for blur-driven and disparity-driven vergence is tightly coupled. We investigated the effect of PD on these 2 vergence paradigms. In the experiment involving 14 patients with PD and 6 healthy controls, substantial differences between blurdriven and disparity-driven vergence were found. The gain (ratio of actual vs desired eye movements) was reduced in patients with PD in case of disparity-driven vergence but not in blur-driven vergence. The latency of disparity-driven vergence onset was significantly longer for patients with PD compared with healthy controls. Four strategies were used to drive disparity-driven vergence: a) pure disconjugate vergence, b) conjugate saccadic movements, c) disconjugate vergence followed by saccadic movements, and d) conjugate saccades followed by disconjugate vergence movements. Blur-driven vergence had only 2 strategies: a) conjugate Department of Biomedical Engineering (PG, SB, AGS), Case Western Reserve University, Cleveland, Ohio; Daroff-Dell’Osso Ocular Motility Laboratory (PG, SB, JJ, AGS, FFG), Louis Stokes Cleveland VA Medical Center, Cleveland, Ohio; Cole Eye Institute (JM, FFG), Cleveland Clinic, Cleveland, Ohio; and Department of Neurology (CK, AGS), Neurological Institute, University Hospitals, Cleveland, Ohio. F. F. Ghasia received Cleveland Brain Health Initiative and Research to Prevent Blindness Disney Award and was supported in part by the NIH-NEI P 30 Core Grant (IP30EY025585), Unrestricted Grants from the Research to Prevent Blindness Inc., and Cleveland Eye Bank Foundation Grant awarded to the Cole Eye Institute. J. Murray was supported by NIH T32 fellowship. A. G. Shaikh received a Dystonia Medical Research Foundation (DMRF) Clinical Fellowship, DMRF/ Dystonia Coalition Career Development Award, American Academy of Neurology Career Development Award, American Parkinson Disease Association George C. Cotzias Memorial Fellowship, Department of Veterans Affairs Merit Review grant (I01 CX00208601A2), and philanthropic funds to the Department of Neurology at University Hospitals (Allan Woll Fund and Sanford Fox Family Fund). A. G. Shaikh has received Penni and Stephen Weinberg Chair in Brain Health. The authors thank Peggy Skelly, PhD for technical assistance. The remaining authors report no conflicts of interest. A. G. Shaikh and F. F. Ghasia contributed equally to this work. Address correspondence to Fatema F. Ghasia, MD, 2022 E 105th Street, Cleveland, OH 44106; E-mail: [e-mail] 442 saccades followed by disconjugate vergence and b) conjugate saccadic movements only. The results are consistent with the prediction that PD primarily affects disparity-driven vergence, but there are some effects on the strategies to execute blur-driven vergence. We speculate that the deep cerebellar nuclei and the supraoculomotor area of the midbrain that carry the disparity-driven and blur-driven vergence are affected in PD. It is possible to modulate their function through projections to the subthalamic nuclei. Journal of Neuro-Ophthalmology 2021;41:442–451 doi: 10.1097/WNO.0000000000001422 © 2021 by North American Neuro-Ophthalmology Society P arkinson disease (PD), the second most common neurodegenerative condition, affects 10 million people worldwide. Visual impairments, frequently seen in PD, can be due to multiple etiologies. Impaired gaze shifts, such as saccades, cause difficulty in reading and scanning the surroundings (1–5). Abnormalities in simultaneous disconjugate movements of the 2 eyes in opposite directions, vergence, affect the perception of depth and dimension. Such deficit, seen in about 70% of patients with PD, leads to impaired fusion of the visual signal and debilitating double vision (4–7). Abnormalities in visuomotor function result in negative social impact and a higher rate of symptoms related to depression and anxiety that are already common in PD (8). Most studies describing vergence abnormalities in PD are based on visual function questionnaires and clinical examination. Vision-targeted health status as measured with composite visual function questionnaire is much lower in patients with PD than that in healthy controls; the scores are lower on almost every subscale, most notably those for near vision and eye movements (1,4). Convergence insufficiency improves on administration of levodopa (9) and with deep brain stimulation (DBS), but the improvement is not complete (1). A study on vergence quantified using infrared oculography in PD reported increased latency of converGupta et al: J Neuro-Ophthalmol 2021; 41: 442-451 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Basic and Translational Research gence and divergence movements with reduced speed compared with healthy controls during binocular viewing (10). There are 2 key sensory drives for vergence: fusional vergence driven by retinal disparity during binocular viewing and accommodation-driven vergence due to visual blur requiring focal length adjustment of the intraocular lens to project a clear image on the fovea (11–15). Fusional and blurdriven vergence systems are cross-coupled, that is, change in blur or disparity alone triggers both accommodation and vergence, as they do when presenting together (16). Figure 1 depicts the simplified concept of vergence and accommodation interactions. The structures carrying the velocity commands (pulse signals) (e.g., disparity and blur controllers, Fig. 1) are upstream from where the cross-coupling between the 2 systems occurs, and they use the retinal image disparity and retinal blurinduced stimuli. The tonic adaptive systems that encode eye position signals to maintain the eyes in a particular vergence angle after gaze shift or those maintaining the accommodative state after the initial accommodative response are further downstream after the cross-coupling happens (e.g., nodes “A” and “V” in Fig. 1). The tonic adaptive system uses the motor output of the pulse fusional vergence and the output of the accommodative systems, respectively, as its error signals (17,18). PD affects vergence (10,19). We evaluated the effects of PD on blur-driven and disparity-driven vergence. We suggest 3 possibilities, each with a distinct set of predictions. (1) PD affects the disparity-driven vergence substrates downstream after the cross-coupling of the vergence and accommodation systems (Fig. 1). According to this possibility, the disparity-driven vergence will be different between patients with PD and healthy controls, but blur-driven vergence will be comparable in the 2 groups. (2) PD affects the blur-driven vergence substrates downstream after the cross-coupling of the vergence and accommodation systems (Fig. 1). Accordingly, the FIG. 1. Schematic model of the disparity-driven and blurdriven vergence system. Each arm, one leading to vergence and the other leading to accommodation, is coupled. There are 4 nodes of controllers. Blur and disparity controllers are upstream before cross-coupling occurs while nodes “A” and “V” are downstream after cross-coupling has occurred. The upstream controllers are sensory while downstream nodes are motor. Node A is Edinger–Westphal nucleus while node V is supraoculomotor nucleus. Gupta et al: J Neuro-Ophthalmol 2021; 41: 442-451 blur-driven vergence will differ between patients with PD and healthy controls, but the disparity-driven vergence will be comparable in the 2 groups. (3) PD affects the disparity-driven or blur-driven vergence substrates upstream of the cross-coupled response, in which case both disparity-driven and blur-driven vergence will be compromised. To test these possibilities, we measured blur-driven vergence under monocular viewing and disparity-driven vergence under binocular viewing of the targets at different depths in patients with PD and healthy controls. METHODS Subjects and Experimental Protocol Disparity-driven and blur-driven vergence responses were examined in 14 patients with PD and 6 age-matched healthy controls. Table 1 presents clinical and demographic details. All subjects provided written informed consent approved by the Cleveland Clinic Institutional Review Board. The experimental protocol was consistent with the standards of the Declaration of Helsinki. The subjects were seated in a chair with their head stabilized on a chinrest. The subjects wore their corrective lenses for distance and near vision for the experiments. The experiments were performed in a binocular viewing condition with best-corrected vision facilitated measurement of disparity-driven vergence. Monocular viewing condition recordings eliminating the retinal disparity measured blur-driven vergence. Horizontal and vertical eye movements were measured with a highresolution video oculography (0.1° spatial resolution and 500 Hz temporal resolution, EyeLink 1000, SR Research, Ontario, Canada). The calibration was performed during right and left eye viewing monocular conditions. An infrared permissive filter allowed precise measurements of the eye position under cover, hence facilitating binocular recordings during monocular viewing conditions (20–22). Convergence and divergence eye movements were measured while viewing stationary light emitting diode (LED) targets placed at 30 and 240 cm distance along the sagittal plane and at the level of the eyes, presented as a step stimulus. The vergence response for each viewing condition was measured. We defined normal vergence as a near-point convergence (NPC) of less than 10 cm. Of the 6 healthy controls, 5 had intact vergence, whereas one control had mild increase in NPC at 14 cm. The eye position data were further processed and analyzed with MATLAB (Mathworks, Natick, MA) using the algorithms developed in our laboratory (20–23). Data Analysis We analyzed the positions of right and left eyes. The signal was denoised with a Savitzky–Golay filter (polynomial order: 3 and frame length: 51) and then differentiated to compute eye velocity. The vergence onset was determined when the eye position shifted approximately 0.5° away from the steady baseline after target shift. The onset and offset of vergence was 443 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Basic and Translational Research TABLE 1. Clinical and demographic details of PD patients studied. ID Age Sex 1 2 3 4 5 6 7 8 9 10 11 12 13 14 76 53 52 71 69 64 71 79 69 64 59 66 65 68 M M M M M M M M M M M M M F ID LED UPDRS-III (off) 1 2 3 4 5 6 7 8 9 1040 60 1505 1100 0 0 610 500 500 10 11 12 13 14 348 600 1000 1200 1000 BMI Dominant Hand/Leg Disease Duration (y) Affected Side #Falls/6 mo 24.98 34.14 34.19 29.58 29.17 25.00 27.08 20.42 27.27 23.01 38.20 30.70 25 26 R/R R/R R/R R/R R/R R/R R/R R/R R/R R/R R/R R/R R/R R/R 11 5 12 21 9 18 14 17 8 10 7 2 6 8 R R R R L L L L&R L L R L L&R R 7 2 0 900 4 0 60 2 0 26 4 6 2 3 Ethnicity White White White White White White White White White White White White White White Visual Acuity Objective Convergence Subjective Convergence OD OS RAF (cm) PEN (cm) DIV Eye RAF (cm) PEN (cm) DIV Eye 51 12 40.5 36.5 48.5 56 41 54 25.5 20/25 20/25 20/20 20/50 20/50 20/30 20/30 20/25 20/40 20/30 20/20 20/20 20/30 20/30 20/40 20/25 20/25 20/40 22 33.5 16 28 23 31 24 24 R R L R L L R L 31 22 33.5 14 25.5 26 31 24 24 20/20 20/20 20/20 20/30 20/20 20/20 20/25 20/25 19 8 14 9 13 26 33 27 30 15 28 25 28 29 (w/glasses) 65 woc 20 13 13 13 (RE) 16 R R R R R L 16 R R 23.5 30 23 34 30 33 27 29 19 28 25 28 29 (w/glasses) 65 woc 24 15 13 13 (RE) 16 R L L L L 20 14 11 13 R L L L L BMI, Body Mass Index; DIV, Eye, Diverging eye; F, female; OD, right eye; L, left; LED, LED (Light Emitting Diode) lights used as visual stimulus presentations; M, male; mo, months; OS, left eye; R, right; RAF, Royal Air Force rule 50cm long with a slider and a rotating four-sided cube with different targets used in clinical exams for convergence measurement; UPDRS-III, Unified Parkinson's Disease Rating Scale part 3; woc, Without correction, PEN, pen target; y, years. interactively validated and classified into different strategies used to facilitate gaze shift. MATLAB toolboxes and SPSS were used for statistical analyses. A random-effects model was used to account for using monocular data from both right and left eyes. We used this test with an eye (right or left) as the random variable and gain of blur-driven vergence. We conducted a similar analysis for latencies of blur-driven vergence. The gain and latency responses of disparity-driven vergence obtained during binocular viewing were tested using the unpaired Student t test. We tested the percentage distribution of our hypothesized eye movement strategies in PD and healthy control cohorts using the x2 test. RESULTS This study evaluates the effects of PD on vergence and describes the mechanistic underpinning of such effects on 444 blur-driven and disparity-driven vergence. Figure 2 shows the horizontal position of the right and left eyes and their difference during binocular (Fig. 2A, B) and monocular (Fig. 2C, D) viewing conditions in healthy control (Fig. 2B, D) and a patient with PD (Fig. 2A, C). The figure depicts the eye positions, conjugate (average of right and left eye position), and disconjugate response (left eye 2 right eye/2) between healthy controls and patients with PD. The disconjugate response in binocular viewing condition is reduced in a patient with PD as compared with healthy controls. Conversely, the disconjugate response during monocular viewing condition in both patient with PD and healthy control is compromised. Figure 3A depicts mean eye position trajectory and standard deviation around the mean in patients with PD and healthy controls who shifted the gaze from distant target to near target, making convergence eye movement in binocular Gupta et al: J Neuro-Ophthalmol 2021; 41: 442-451 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Basic and Translational Research FIG. 2. Comparison of sample plots of eye positions and measured binocular and monocular gaze shift responses in patients with PD and healthy controls. The horizontal eye position is plotted on y axis while the x axis depicts corresponding time. Gray dotted lines depict desired target position for each eye. The left eye and right eye traces are represented by solid green and red lines, respectively. The vergence and conjugate responses are represented by solid blue and cyan lines, respectively. A, B. The eye position, conjugate, and disconjugate responses under binocular viewing condition for healthy controls (A) and patients with PD (B). C, D. The same for monocular viewing condition (C: healthy controls, D: patients with PD). condition triggering disparity-driven vergence. Although healthy subjects were able to perform disparity-driven convergence eye movements adequately, there was a substantial limitation in patients with PD. Figure 3B depicts the same analysis under monocular viewing conditions triggering blur-driven vergence. Compared with disparity-driven vergence, blur-driven vergence was limited in healthy controls (compare Fig. 3A, B). Such impairment in blur-driven vergence was also seen in patients with PD. The impairment was comparable with healthy controls (blue and black, Fig. 3B). Figure 3C and D depicts the summary of disparity and blur-driven vergence gain (actual gaze shift/desired gaze shift) in 14 patients with PD (black boxplots) and 6 healthy Gupta et al: J Neuro-Ophthalmol 2021; 41: 442-451 controls (blue boxplots). In healthy controls, the median gain (and interquartile interval [IQR]) of disparity-driven convergence and divergence was 0.88 (IQR = 0.2) and 0.87 (IQR = 0.243), respectively. The values were significantly higher compared with the disparity-driven convergence and divergence in PD, 0.473 (IQR = 0.319) and 0.458 (IQR = 0.399), respectively (unpaired t test, P , 0.001). The values of gain for blur-driven convergence and divergence in healthy control were 0.145 (IQR = 0.057) and 0.133 (IQR = 0.037); not significantly different from patients with PD (convergence: 0.204 [IQR = 0.116]; divergence: 0.188 [IQR = 0.096]) (convergence: mixed analysis of variance [ANOVA], F = 5.5 P = 0.25 with no statistically significant interaction between eye and subgroup F= 0.8, 445 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Basic and Translational Research FIG. 3. Comparison of blur-driven and disparity-driven vergence measured from patients with PD and healthy controls. A, B. The difference between right and left (vergence) eye positions are plotted on y axis and corresponding time is plotted on the x axis. Black line depicts mean position in patients with PD while blue line depicts the control subjects. The light shades (gray and light blue) depict the standard deviation of spread around the mean. The gaze shift in binocular viewing condition depicting disparity-driven vergence (A) and (B) blur-driven vergence. The overlapping values depict lack of difference between healthy controls and patients with PD in case of blur-driven vergence. C, D. Box plots depict gain (actual vergence position/ desired vergence position) during binocular viewing condition (disparity-driven) and monocular viewing condition (blur-driven) vergence. The gain values are plotted on the y axis, the length of the box depict interquartile difference, the whiskers are the range, and horizontal line in the center of the box depicts median value. E, F. The summary of latency of disparity-driven and blur-driven vergence in patients with PD and healthy subjects. P =0.37; divergence: mixed ANVOA, F = 5.9, P =0.24 with no statistically significant interaction between eye and subgroup F = 0.01, P = 0.9). In both PD and healthy controls, the gain of blur-driven vergence was significantly smaller than disparity-driven vergence (ANOVA, for both controls and PD: P , 0.001). Subsequent analysis compared the latency, the time difference between the target onset and gaze shift, of disparity-driven and blur-driven vergence (Fig. 3E, F). The median latency and IQR of disparity-driven convergence and divergence in healthy controls were 0.2 (IQR = 0.03) and 0.218 (IQR = 0.033), respectively. The latencies 446 of disparity-driven vergence for healthy controls were significantly shorter compared with the patients with PD (convergence: 0.372 (IQR = 0.247) and divergence: 0.307 (IQR = 0.078), unpaired t test, P , 0.001). Blur-driven convergence and divergence latency in healthy control group was 0.29 (IQR = 0.049) and 0.32 (IQR = 0.08) respectively; it was not significantly different from the PD group (convergence: 0.326 (IQR = 0.112) and divergence 0.3 (IQR = 0.078), unpaired t test, for convergence P = 0.09, for divergence P = 0.523) (mixed ANOVA: convergence F = 11.4, P = 0.18 with no statistically significant interaction between eye and subgroup F = 1.5, P = 0.22; divergence: F = 4.4, P = Gupta et al: J Neuro-Ophthalmol 2021; 41: 442-451 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Basic and Translational Research 0.28 with no statistically significant interaction between eye and subgroup F = 0.8, P =0.3). The overall latency of blurdriven vergence was significantly higher than that of disparity-driven vergence in healthy controls, whereas in PD difference in latency between blur-driven vergence and disparity-driven vergence was not statistically significant (unpaired t test, controls P , 0.001; PD P = 0.102). The patients with PD followed 4 or classes to accomplish disparity-driven vergence: the first was the normal disconjugate vergence movements (Class 1), the second had only conjugate saccadic movements (Class 2), and the third and fourth combined both conjugate and disconjugate movements (Class 3 and 4). Figure 4A– D depicts classes mapping to Classes 1 through 4 respec- tively for disparity-driven vergence. Figure 4E and F depicts classes to accomplish blur-driven vergence. Blur-driven vergence only had “Class 4” and “Class 2” movements (Fig. 4E, F). Notably, blur-driven vergence was never initiated by a disconjugate component (Classes 1 and 3). We examined the prevalence of classes of movements in blur-driven and disparity-driven vergence in patients with PD compared with healthy controls (Fig. 5). In the case of disparity-driven vergence (Fig. 5A, B, E, F), healthy controls performed “Class 1” movements 76.36% of the times for convergence and 65.76% for divergence. By contrast, for the same condition, in PD, “Class 1” movements were seen for 15.25% for convergence and 7.69% for divergence. FIG. 4. Different gaze-shifting strategies to compensate for disparity-driven and blur-driven vergence deficits in patients with PD. Binocular viewing accounts for disparity-driven vergence while monocular viewing condition accounts for blur-driven or accommodation-driven vergence. Each panel depicts one strategy; the vergence position is plotted on y-axis while corresponding time is plotted on x-axis. Gray line depicts desired target position. A. Disconjugate movement under binocular viewing condition. B. Conjugate movement under binocular viewing condition. C. Disconjugate initiation followed by conjugate movement under binocular viewing condition. D. Conjugate initiation followed by disconjugate movement under binocular viewing condition. E. Conjugate initiation followed by disconjugate movement under monocular viewing condition. F. Disconjugate initiation followed by conjugate movement under monocular viewing condition. Gupta et al: J Neuro-Ophthalmol 2021; 41: 442-451 447 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Basic and Translational Research FIG. 5. Gaze shift strategies used by patients with PD and healthy controls grouped as convergence and divergence under binocular and monocular viewing conditions presented as a series of pie charts. BinoView: binocular viewing condition (disparity-driven); MonoView: monocular viewing condition (blur-driven); Conj.: conjugate movement; Disconj.: disconjugate movement; Conj. f/b Disconj.: conjugate initiation followed by disconjugate movement; Disconj. f/b conj.: disconjugate initiation followed by conjugate movement; No Mvmt: no discernible movement. In disparity-driven vergence, subjects with PD used “Class 4” movements 27.96% of the times in convergence 15.38% of the times in divergence; healthy controls did not have such strategy. Subjects with PD did not make any movement 29.66% of the times for the disparity-driven convergence and 33.3% for the disparity-driven divergence conditions; such trend was not present in healthy controls. The x2 test showed statistical independence in both convergence and divergence eye movements between subjects with PD and healthy controls (disparity-driven convergence: x2 = 1.75e220; disparity-driven divergence: x2 = 3.1e225). Figure 5E–H depicts distributions of trends in blur-driven vergence. Two classes of movements were seen, “Class 2” and “Class 4.” In healthy subjects, the “Class 2” movements were present 18.52% of the times for blurdriven convergence and 12.76% of the times in blurdriven divergence. The difference was not significant compared with PD (36.67% blur-driven convergence and 26.67% blur-driven divergence (Class 2: x2 = 0.902, Class 4: x2 = 0.646). “Class 4” movements were seen 81.48% of the times for blur-driven convergence and 81.54% for blur-driven divergence in healthy subjects; in PD it was present for 56.67% of the times for blur-driven convergence and 63.33% for blur-driven divergence. We found a statistically significant difference in both convergence and divergence eye movements between the strategy used by patients with PD and healthy controls (blur-driven convergence x2 = 1.9e25; blur-driven divergence x2 = 0.015). 448 DISCUSSION Vergence abnormalities are common in PD (1,2,4,5,10,19). Vergence movements are driven by 2 types of signals, visual blur and retinal disparity. Accommodation-driven and disparity-driven vergence happen in synchrony. Their triggers, blur and disparity, are also coupled. In patients with PD and age-matched healthy controls, we measured blurdriven vergence during monocular viewing of targets at different depths. The disparity-driven vergence was measured during binocular viewing when the subjects viewed the same target depths with their best-corrected distance and near vision. We found fundamental differences in blur-driven and disparity-driven vergence in PD. The vergence gain in both conditions was reduced in patients with PD. Such reduction was significantly different than controls in the case of disparity-driven vergence but not so in blurdriven vergence. The latency of vergence in those with PD was also increased in disparity-driven vergence compared with healthy controls. To compensate for impaired vergence, the patients with PD adopted four classes: the first was the normal disconjugate vergence movements (Class 1), the second had only conjugate saccadic movements (Class 2), and the third and fourth combined both conjugate and disconjugate movements (Class 3 and 4). The blur-driven vergence had only two classes both were associated with preceding saccades: one was conjugate saccadic movements only (Class 2) while the other had preceding conjugate saccade followed by a disconjugate vergence component Gupta et al: J Neuro-Ophthalmol 2021; 41: 442-451 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Basic and Translational Research (Class 4). Together with our results suggested that disparitydriven vergence in PD is markedly abnormal compared with the healthy controls. By contrast, blur-driven vergence, although abnormal regarding the strategies used, had similar gains in PD and age-matched healthy controls. The gain of blur-driven vergence was lower compared with disparitydriven vergence in both controls and PD group. The smaller amplitude disconjugate vergence response under monocular viewing is multifactorial from subjects wearing their refractive correction and from the expected accommodation response is reduced in elderly subjects (both PD and controls) due to presbyopia. In light of the cross-coupled model of blur-driven and disparity-driven vergence, one would anticipate whether PD affects the neurons after the vergence accommodation coupling, that is, in location “A” or “V” in Fig 1 the effects of PD would only reduce blur-driven vergence and disparity-driven vergence, respectively. By contrast, if it is before the coupling, that is, the disparity controller in Fig 1, the blur-driven vergence will be mildly affected in PD compared with disparitydriven vergence, and if blur controller is affected, then disparity-driven vergence will be mildly affected compared with blur-driven vergence. We found a mild reduction in gain parameters of blur-driven vergence with primarily conjugate saccadic component strategies more commonly used than combined conjugate disconjugate strategies in PD compared with healthy controls. On the other hand, the gain and latencies of disparity-driven vergence were significantly reduced in PD than those in healthy controls. Thus, our results were consistent with the possibility that impaired vergence in PD is due to the involvement of the disparity controller. In the subsequent section, we will discuss our results, particularly the influence of PD, considering the crosscoupled vergence accommodation model and physiologically realistic connections of cerebro–brainstem–cerebellar components of the vergence control. The fundamental question we want to address is how PD affects this circuit. The neural correlates of vergence and accommodation are depicted in Fig 6. In brief, the frontal and supplementary frontal eye fields have vergence neurons, and the mediotemporal and mediosuperior-temporal of the cerebral cortex. These areas important for the perception of depth and project to the vergence sensitive supraoculomotor area in the midbrain. The supraoculomotor area projects to the medial rectus motor neurons that drive the disparity-driven vergence. The supraoculomotor area has projections to the nucleus reticularis tegmenti pontis (NRTP) and also has reciprocal connections with the deep cerebellar nuclei (24). The NRTP receives projections from the frontal eye fields and project to the oculomotor vermis of the cerebellum (24,25). The deep cerebellar nuclei have neurons that discharge in relation to vergence and accommodation (18). The cerebral cortex, namely, area V1, and the parietotemporal areas are involved in accommodation (25–27). These areas project to the Edinger–Westphal nucleus, which formulates the motor response that is transmitted to the ciliary muscles responsible for accommodation (28). Such anatomical organization when viewed in context of Fig 1 (16,25) could indicate that the fastigial nucleus and FIG. 6. Hypothetical schematic for the control of gaze shift, detailing possible mechanisms for activation of disparity-driven and blur-driven vergence. Premotor commands for vergence movements involving recruitment of slow extraocular muscle fibers are generated in the midbrain in a region called the mesencephalic reticular formation (MRF) which has projections to the supraocular motor area (SOA), which is a key region for vergence control. The SOA also receives projections from the superior colliculus (SC), the frontal eye field (FEF), and the supplementary eye field (SEF). It also has reciprocal connections to the cerebellar nuclei, caudal fastigial nuclei (CFN), and the posterior interpositus nuclei (PIN). The SOA contains vergence burst, tonic, and burst–tonic neurons which send excitatory projections to medial rectus motor neurons (MRMNs) to generate convergence eye movement (lateral rectus for divergence). The FEF also project to the basilar pontine nuclei medial NRTP which has connections to the cerebellar nuclei, and it has direct projections to the CFN and projects to the PIN through the dorsal vermis. The SC also receives feedback from FEF and tonic inhibition from the substantia nigra pars compacta (SNpr). The middle temporal (MT) and medial superior temporal (MST) areas located in the cerebral cortex play a key role in coding of visual motion. They project directly to the accessory oculomotor nucleus Edinger–Westphal nucleus (EWN) which inturn innervates the ciliary muscle fibers. Gupta et al: J Neuro-Ophthalmol 2021; 41: 442-451 449 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Basic and Translational Research interpositus are the disparity controllers while the cerebral cortex is the blur controller. As noted in the previous paragraph, our results suggest that PD affects disparity controller. This conceptual scheme is consistent with anatomical realism. There is increasing evidence that the cerebellum is affected in PD, with anatomical findings showing that the cerebellum and basal ganglia have substantial two-way communication between each other (29,30). The cerebellum receives a dopaminergic projection for the ventral tegmental area/substantia nigra (31–33). A recent study has shown that subthalamic DBS modulates the activity of the deep cerebellar nuclei (34). The fastigial nuclei and interpositus of the cerebellum, the putative locations of the disparity controller, receive input from the perirubral region that is modulated by the subthalamic nucleus (34,35). We speculate that the abnormally increased discharge rate and pathological firing pattern of the subthalamic nucleus in PD can affect the activity of the fastigial and interpositus nuclei and subsequently the vergence sensitive neurons found further downstream in the supraoculomotor region (24,36– 38). The consequence of such modulation is abnormal disparity-driven vergence in PD. Although our hypothesis is based on indirect experimental evidence, it provides important insights into neuromodulation or pharmacotherapy affecting the corticobasal ganglionic network to improve vergence in those with PD. STATEMENT OF AUTHORSHIP Category 1: a. Conception and design: A. G. Shaikh and F. F. Ghasia; b. Acquisition of data: A. G. Shaikh, F. F. Ghasia, J. Jacobs, P. Gupta, S. Beylergil, C. Kilbane, and J. Murrey; c. Analysis and interpretation of data: A. G. Shaikh, F. F. Ghasia, and P. Gupta. Category 2: a. Drafting the manuscript: A. G. Shaikh, F. F. Ghasia, and P. Gupta; b. Revising it for intellectual content: A. G. Shaikh, F. F. Ghasia, J. Jacobs, P. Gupta, S. Beylergil, C. Kilbane, and J. Murrey; Category 3: a. Final approval of the completed manuscript: A. G. Shaikh, F. F. Ghasia, J. Jacobs, P. Gupta, S. Beylergil, C. Kilbane, and J. Murrey. REFERENCES 1. Almer Z, Klein KS, Marsh L, Gerstenhaber M, Repka MX. Ocular motor and sensory function in Parkinson’s disease. Ophthalmology. 2012;119:178–182. 2. Armstrong RA. Oculo-visual dysfunction in Parkinson’s disease. J Parkinsons Dis. 2015;5:715–726. 3. Antoniades CA, Demeyere N, Kennard C, Humphreys GW, Hu MT. Antisaccades and executive dysfunction in early drug-naive Parkinson’s disease: the discovery study. Mov Disord. 2015;30:843–847. 4. Lepore FE. Parkinson’s disease and diplopia. Neuroophthalmology. 2006;30:37–40. 5. Urwyler P, Nef T, Killen A, Collerton D, Thomas A, Burn D, McKeith I, Mosimann UP. Visual complaints and visual hallucinations in Parkinson’s disease. Parkinsonism Relat Disord. 2014;20:318–322. 6. Bloem BR, Hausdorff JM, Visser JE, Giladi N. Falls and freezing of gait in Parkinson’s disease: a review of two interconnected, episodic phenomena. Mov Disord. 2004;19:871–884. 7. Davidsdottir S, Wagenaar R, Young D, Cronin-Golomb A. Impact of optic flow perception and egocentric coordinates on veering in Parkinson’s disease. Brain. 2008;131:2882–2893. 450 8. Cumurcu T, Cumurcu BE, Ozcan O, Demirel S, Duz C, Porgalı E, Doganay S. Reprint of: social phobia and other psychiatric problems in children with strabismus. Can J Ophthalmol. 2015;50(suppl 1):S7–S11. 9. Racette BA, Gokden MS, Tychsen LS, Perlmutter JS. Convergence insufficiency in idiopathic Parkinson’s disease responsive to levodopa. Strabismus. 1999;7:169–174. 10. Hanuska J, Bonnet C, Rusz J, Sieger T, Jech R, Rivaud-Péchoux S, Vidailhet M, Gaymard B, R u zi cka E. Fast vergence eye movements are disrupted in Parkinson’s disease: a videooculography study. Parkinsonism Relat Disord. 2015;21:797– 799. 11. Fincham EF, Walton J. The reciprocal actions of accommodation and convergence. J Physiol. 1957;137:488– 508. 12. Krishnan VV, Phillips S, Stark L. Frequency analysis of accommodation, accommodative vergence and disparity vergence. Vis Res. 1973;13:1545–1554. 13. Krishnan VV, Stark L. A heuristic model for the human vergence eye movement system. IEEE Trans Biomed Eng. 1977;24:44–49. 14. Rashbass C, Westheimer G. Independence of conjugate and disjunctive eye movements. J Physiol. 1961;159:361–364. 15. Hung GK, Semmlow JL. Static behavior of accommodation and vergence: computer simulation of an interactive dual-feedback system. IEEE Trans Biomed Eng. 1980;27:439–447. 16. Westheimer G. Amphetamine, barbiturates, and accommodation-convergence. Arch Ophthalmol. 1963;70:830–836. 17. SCHOR C. Influence of accommodative and vergence adaptation on binocular motor disorders. Optom Vis Sci. 1988;65:464–475. 18. Zee RJLDS. Vergence eye movements. In: The Neurology of Eye Movements. New York, NY: Oxford University Press, 2015:544–550. 19. Repka MX, Claro MC, Loupe DN, Reich SG. Ocular motility in Parkinson’s disease. J Pediatr Ophthalmol Strabismus. 1996;33:144–147. 20. Ghasia FF, Otero-Millan J, Shaikh AG. Abnormal fixational eye movements in strabismus. Br J Ophthalmol. 2018;102:253– 259. 21. Scaramuzzi M, Murray J, Otero-Millan J, Nucci P, Shaikh AG, Ghasia FF. Fixation instability in amblyopia: oculomotor disease biomarkers predictive of treatment effectiveness. Prog Brain Res. 2019;249:235–248. 22. Shaikh AG, Otero-Millan J, Kumar P, Ghasia FF. Abnormal fixational eye movements in amblyopia. PLoS One. 2016;11:e0149953. 23. Ghasia FF, Shaikh AG, Jacobs J, Walker MF. Cross-coupled eye movement supports neural origin of pattern strabismus. Invest Ophthalmol Vis Sci. 2015;56:2855–2866. 24. Joshi AC, Das VE. Muscimol inactivation of caudal fastigial nucleus and posterior interposed nucleus in monkeys with strabismus. J Neurophysiol. 2013;110:1882–1891. 25. Gamlin PD, Yoon K, Zhang H. The role of cerebro-pontocerebellar pathways in the control of vergence eye movements. Eye (Lond). 1996;10:167–171. 26. Jampel RS. Representation of the near-response on the cerebral cortex of the macaque. Am J Ophthalmol. 1959;48:573–582. 27. Harrison RJ. Loss of fusional vergence with partial loss of accommodative convergence and accommodation following head injury. Bino Vis. 1987;2:93–100. 28. Gamlin PD. Subcortical neural circuits for ocular accommodation and vergence in primates. Ophthalmic Physiol Opt. 1999;19:81–89. 29. Middleton FA, Strick PL. Basal ganglia and cerebellar loops: motor and cognitive circuits. Brain Res Brain Res Rev. 2000;31:236–250. 30. Wu T, Hallett M. The cerebellum in Parkinson’s disease. Brain. 2013;136:696–709. 31. Ikai Y, Takada M, Shinonaga Y, Mizuno N. Dopaminergic and non-dopaminergic neurons in the ventral tegmental area of the Gupta et al: J Neuro-Ophthalmol 2021; 41: 442-451 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Basic and Translational Research rat project, respectively, to the cerebellar cortex and deep cerebellar nuclei. Neuroscience. 1992;51:719–728. 32. Panagopoulos NT, Papadopoulos GC, Matsokis NA. Dopaminergic innervation and binding in the rat cerebellum. Neurosci Lett. 1991;130:208–212. 33. Melchitzky DS, Lewis DA. Tyrosine hydroxylase- and dopamine transporter-immunoreactive axons in the primate cerebellum. Neuropsychopharmacology. 2000;22:466–472. 34. May PJ, Porter JD, Gamlin PD. Interconnections between the primate cerebellum and midbrain near-response regions. J Comp Neurol. 1992;315:98–116. Gupta et al: J Neuro-Ophthalmol 2021; 41: 442-451 35. Bostan AC, Strick PL. The basal ganglia and the cerebellum: nodes in an integrated network. Nat Rev Neurosci. 2018;19:338–350. 36. Das VE. Responses of cells in the midbrain near-response area in monkeys with strabismus. Invest Ophthalmol Vis Sci. 2012;53:3858–3864. 37. Das VE. Cells in the supraoculomotor area in monkeys with strabismus show activity related to the strabismus angle. Ann N Y Acad Sci. 2011;1233:85–90. 38. Das VE, Leigh RJ, Swann M, Thurtell MJ. Muscimol inactivation caudal to the interstitial nucleus of Cajal induces hemi-seesaw nystagmus. Exp Brain Res. 2010;205:405–413. 451 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.