Share this article on:

Visual Fatigue and Accommodative Dynamics in Asymptomatic Individuals

Thiagarajan, Preethi*; Ciuffreda, Kenneth J.

doi: 10.1097/OPX.0b013e31827a233e
Original Articles

Purpose To evaluate accommodative dynamics immediately before and after visual fatigue-inducing tasks with congruent (C) and noncongruent (NC) visual stimuli.

Methods Accommodative dynamics for a 2-diopter (D) step (2.5 ↔ 4.5 D) stimulus were assessed using the WAM-5500 open-field autorefractor in 10 visually normal, asymptomatic, young-adult subjects before and after C and NC tasks performed on different days in a counterbalanced manner. For the C task, subjects altered binocular fixation (50 cycles) between reduced Snellen charts at 50 and 20 cm every 3 seconds to the beat of a metronome. For the NC task, subjects performed 50 cycles of accommodative flipper (±1.5 D) as rapidly as possible while binocularly fixating on a reduced Snellen chart at 40 cm.

Results Mean steady-state (SS) response level reflecting its accuracy was significantly (p ≤ 0.05) reduced post-NC task for both increasing (4.5-D level) and decreasing (2.5-D) accommodation. Although the SS response variability was significantly (p ≤ 0.05) increased post-NC task at the 4.5-D stimulus level, it did not change significantly (p > 0.05) for the C task. There were no significant differences between the pretask and posttask group mean initial amplitude, time constant, and peak velocity for either the C or the NC task for both increasing and decreasing accommodation.

Conclusions The accommodative system exhibited consistent fatigue effects with respect to SS variability and response accuracy. There was no objective evidence of oculomotor learning for these specific tasks.



Department of Biological and Vision Sciences, SUNY State College of Optometry, New York, New York.

Preethi Thiagarajan Department of Biological and Vision Sciences SUNY State College of Optometry 33 West 42nd St New York, New York 10036 e-mail:

Fatigue is a complex, nonspecific symptom involving multiple biological components (e.g., central mechanisms, cellular mechanisms, neuroendocrine factors, and muscle properties) whose underlying mechanism is not yet clearly and fully defined.1,2 Fatigue is likely to be produced by a number of factors, including stress, medical conditions, and medications, as well as environmental factors, such as work load, improper diet, and lack of rest.1–3 The absence of established methods to quantify fatigue objectively and, furthermore, to correlate it with presumed related symptoms makes its diagnosis as well as its treatment a challenge for clinicians.

Although general muscle fatigue has been extensively researched,4,5 there is growing interest in visual fatigue given the increasingly greater amount of nearwork imposed by modern society.6–10 “Visual fatigue” (or asthenopia) is the most common symptom found in individuals with binocular vision oculomotor dysfunctions, such as accommodative insufficiency and convergence insufficiency.11 It presents symptomatically as blur, diplopia, skipping lines while reading, headache, and dizziness, to name a few.11 The onset of symptoms ranges from a few minutes to a few hours after sustained nearwork. In addition, with the increased usage of computers, visual fatigue is becoming a common symptom among such users (e.g., computer-vision syndrome),8 along with correlated ergonomically related symptoms (e.g., neck pain, backache, etc) in such a compromised and nonoptimal work environment.9 However, the basic mechanisms underlying its oculomotor-related component remain somewhat elusive. Lastly, although visual symptom rating scales have been well established,12,13 there is a paucity of objective recordings of the possible oculomotor components that may reflect “fatiguing of the visual system.”

The accommodative system, which represents one of the critical components of the near triad, has been studied extensively to evaluate its role on causation of nearwork-related visual fatigue.11 Studies have identified reduced accommodative amplitude (i.e., accommodative insufficiency) to be one of the major causes of visual discomfort and related near vision symptoms in young adults14–16 and in children.17,18 However, it is critical to distinguish accommodative insufficiency from accommodative fatigue, with the latter being the focus of the present investigation. Although accommodative insufficiency is defined as a consistently reduced maximum amplitude of accommodation (>2 diopters [D] from the age-matched normative value),11 accommodative fatigue frequently manifests as a progressive increase in the accommodative lag (i.e., near point recession) with sustained near viewing.16,19,20 This is caused by an inability to maintain the steady-state (SS) response, thus resulting in ill-sustained accommodation. Hence, assessment of the accommodative response during specific periods is warranted to unmask accommodative fatigue in individuals with a normal amplitude of accommodation, as assessed conventionally in the clinic.21 Studies on accommodative fatigue using a provocative test, such as a sustained or a repetitive accommodative demand, exist from the early 20th century. The principle of assessing general muscle fatigue using ergographic methods was applied to evaluate the “ocular” muscles (i.e., the ciliary muscle and extraocular muscle) with respect to fatigue.22,23 Studies have reported progressive recession in the accommodative amplitude, as assessed subjectively with repetitive accommodative tracking in young individuals who were otherwise asymptomatic and having a normal amplitude of accommodation.22,24–27

Fatigue related to a sustained accommodative near task has been studied by both subjective and objective means. Although a subjective clinical measure such as reduced accommodative amplitude has been used as an indicator of accommodative fatigue in numerous much earlier studies,22,24–27 other more recent studies evaluated objectively different parameters of accommodation. For example, Takeda et al.28 found reduction in the area under the accommodative stimulus/response curve assessed objectively after 3 hours of nearwork, which reflected an overall decrease in accommodative accuracy (i.e., increase in accommodative SS error). In contrast, Ciuffreda et al.29 and Simmers et al.30 did not find any difference in the accommodative stimulus/response function between visually symptomatic and asymptomatic groups. However, these two studies did not incorporate a period of extended nearwork. Both the high-31 and low-30 frequency components of accommodative microfluctations were found to be increased in the symptomatic individuals only. Lastly, tonic accommodation was found to reduce after a repetitive lens flipper task,32 and this likely reflected fatigue-induced reduction in the baseline neural firing to the ciliary muscle controlling accommodation according to these investigators.

However, there have been some very recent investigations in this area. Chase et al.16 recorded objective accommodative responses in college students with and without nearwork-related symptoms (e.g., blur, headache). Monocular accommodative responses were continuously recorded for 2 minutes at each of five dioptric demand levels (0, 2, 3, 4, and 5 D), with a 1-minute rest period after testing at each dioptric level. The accommodative responses averaged during the 2-minute test period revealed increased accommodative lag at the higher dioptric levels (i.e., 4 and 5 D) in the symptomatic group only. Furthermore, the symptoms were correlated with the progressively increased accommodative lag over time of viewing. In a second study by the same group, Tosha et al.20 also found a progressive increase in the accommodative lag at closer viewing distances (e.g., 5-D stimulus) during the 2-minute test period, but only in the high-discomfort group when categorized based on the Conlon Visual Discomfort Survey13; the low-discomfort group exhibited stable accommodative responses. It is important to note that the high-discomfort group demonstrated an accommodative lag equal to the low-discomfort group during the first 10 seconds of the recordings, with the lag then progressively increasing thereafter during the next several seconds. Lastly, a recent study33 on accommodative dynamics in two visually normal young adults showed lack of reduction in peak velocity and, hence, lack of fatigue based on this velocity criterion, with repeated accommodative step tracking for either every 5 seconds (alternating between 0.16 and 5 D) or every 2 seconds (alternating between 0.16 and 6 D) measured on two different days during a continuous 30-minute period.

With the above considerations in mind, the present study evaluated the dynamics of accommodation using two different repetitive accommodative tasks developed with the intention of being fatigue inducing. This included a C (congruent) task in which changes in the accommodative and vergence stimulus were the same (i.e., congruent), with this simulating viewing conditions under most naturalistic settings; and an NC (noncongruent) task in which the vergence stimulus was maintained constant, whereas the accommodative stimulus was altered (i.e., noncongruent), with this simulating more nonnaturalistic and less common viewing conditions, as during binocular viewing of three-dimensional movies and virtual reality settings. Thus, the objective of the present study was to ascertain whether the accommodative system could be fatigued, and if so, what accommodative parameters would reflect such fatigue effects, and furthermore whether there was a task-related difference.

Back to Top | Article Outline



Ten visually asymptomatic (by self-report and detailed case history), young adults were recruited from the student/staff population of the SUNY State College of Optometry. Ages ranged from 22 to 31 years, with a mean (±1 SD) of 26 ± 3 years. None had a history or diagnosis of vergence and/or accommodative dysfunction. None had systemic, ocular, and/or neurological disease, nor were they taking any drugs or medications that would adversely affect vergence, accommodation, and/or their interaction or attentional status. Subjects included both emmetropes (n = 4) and myopes (n = 6). Emmetropes had a spherical equivalent refractive range of +0.5 to −0.5 D, with a mean (±SE) of −0.25 ± 0.37 D. Myopes had a spherical equivalent refractive range of −1.5 to −5.00 D, with a mean (±SE) of −2.75 ± 0.75 D. Astigmatism was less than 1 D. All had 20/20 or better Snellen visual acuity in the distance and near in each eye with either their current spectacle or contact lens prescription. Binocular motor findings, including the distance and near phoria, positive and negative fusional vergence ranges at distance and near, near point of convergence, positive relative accommodation, negative relative accommodation, and near point of accommodation, were found to lie within Morgan norms, as assessed in the conventional clinical manner.34 Informed consent was obtained from each subject after the nature of the study was explained. The research followed the tenets of the Declaration of Helsinki and was approved by the college’s internal review board.

Back to Top | Article Outline


Accommodative step responses were obtained using the commercially available WAM-5500 objective, infrared, open-field autorefractor (Grand Seiko, Hiroshima, Japan). In the dynamic mode, continuous measurements of the refractive state were sampled approximately every 200 milliseconds (5 Hz). It provides a reliable measure of accommodation and overall refractive state dynamically.35 The spherical dioptric range is −22 to +22 D, with a company reported resolution of 0.01 D. Up to 10 D of cylindrical refractive error can be measured, with a company reported resolution of 0.01 D and with an axis resolution of 1 degree. Accommodative response traces, data tables, graphical displays, and statistical analyses were completed using Microsoft Excel (Microsoft Corporation, Redmond, Wash) and GraphPad Prism (GraphPad Software, La Jolla, Calif).

Back to Top | Article Outline

Test Procedure

Congruent and NC visual tasks36,37 of equal accommodative dioptric demand (3 D) were tested on two different sessions on separate days. For the C condition, the accommodative and vergence stimulus demands were the same, whereas they differed for the NC condition. The tasks were counterbalanced among the subjects. Each session had three phases: pretask accommodative dynamics assessment, fatigue-inducing task (either C or NC), and immediate posttask accommodative dynamics assessment. No objective measurements were taken during the actual fatiguing tasks.

Back to Top | Article Outline

Pretask Accommodative Dynamics

Accommodative dynamics were recorded to a 2-D step change in stimulus amplitude for both increasing and decreasing accommodation from the right eye.35 Subjects binocularly viewed a line of high contrast (>90%) 20/30 Snellen letters having a luminance of 36 cd/m2 positioned at 40 cm (2.5 D) that were on a white background and a high contrast (>90%) 20/60 word with a luminance of 36 cd/m2 at 22.2 cm (4.5 D) against a transparent background. These target stimulus levels did not intrude into the subjects’ upper nonlinear region of accommodative responsivity.21 The autorefractor was aligned along the line of sight of the right eye, as well as with both accommodative stimuli. The subject was instructed to change focus every 4 to 5 seconds as rapidly as possible between the far and near targets on command from the experimenter. There were approximately 14 changes in focus during the test period, seven increasing and seven decreasing accommodative responses. Targets were presented with temporal randomization to minimize prediction. Subjects were encouraged to blink minimally during testing to reduce the chance of an artifact occurring in the actual response.

Back to Top | Article Outline

Fatigue-Inducing Tasks

Congruent Task

In the C task, the accommodative and vergence stimulus levels were always the same value; that is, either 2 D and 2 meter angles (MA) or 5 D and 5 MA, respectively, for the far and near target changes. Such stimulus congruency is the same as that which occurs in one’s natural viewing conditions. Subjects altered bifixation between the far and near targets precisely every 3 seconds to the beat of a metronome in this predictable task. Subjects were instructed to maintain the letters in focus and fused at all times. Fifty cycles (50 increasing and 50 decreasing) of accommodative stimulation were presented in a continuous manner.

Back to Top | Article Outline

Noncongruent Task

In the NC task, the accommodative and vergence stimulus levels always differed. That is, the vergence stimulus was maintained at 2.5 MA (40 cm), whereas the accommodative stimulus was varied in a step manner using accommodative lens flippers of a magnitude ±1.5 D (i.e., either 1 or 4 D). This flipper magnitude was chosen to match the mean stimulus amplitude of 3 D (50 cm ↔ 20 cm) as used in the C task. Subjects performed 50 continuous cycles (i.e., 50 increasing and 50 decreasing) of rapid accommodative flipper alternation while bifixating the target at 2 MA. Subjects were instructed to alter the flipper lenses as rapidly as possible once they achieved accurate focus and fusion of the Snellen chart through the flipper lenses.

Back to Top | Article Outline

Posttask Accommodative Dynamics

Immediately after the above fatigue-inducing tasks, the posttask accommodative dynamics were recorded. Target setup and test procedures were the same as for the Pretask Accommodative Dynamics.

Back to Top | Article Outline

Data Acquisition and Analysis

The recorded video files were saved. They were then transferred into Microsoft Excel for detailed analysis. Three artifact-free (e.g., blink-free) increasing and three decreasing accommodative responses were selected for analysis from the right eye traces for each subject. There were approximately seven increasing and seven decreasing responses in total for each subject. In most cases, the middle three blink-free responses were used for analysis. For those who did not have such artifact-free middle three responses, either the second or the sixth artifact-free response was used for the analysis. An exponential decay function was fit to the dynamic trajectory, and the response amplitudes and time constants were obtained using the GraphPad Prism software. The goodness of fit was assessed from the r 2 values of each fit. The mean r 2 value for both increasing and decreasing steps was greater than 0.8 for each subject. The peak velocities were derived from first-order differentiation of the exponential equation. The mean amplitude, time constant, and the peak velocity of the responses for pretask and posttask accommodative dynamics were compared statistically using GraphPad software. In addition, SS response variability immediately before and after the fatigue-inducing tasks was compared by calculating the mean and the SD of the response level within the measured window of time (4 to 5 seconds) after they reached SS (at 2.5 and at 4.5 D). For each subject, the mean for each parameter was calculated, and then the overall group mean was computed.

Back to Top | Article Outline

Subjective Rating

In addition to the objective recordings, pre-to-post subjective responses to the fatiguing tasks were recorded. On a scale of 1 to 10 (1 being absolutely no symptoms and 10 being maximum symptoms), subjects were asked to report their overall symptom rating (e.g., eye strain, headache, blur, diplopia, etc) before and after the fatigue-inducing tasks. The subjects were classified as “fatigued” when the scale was greater than or equal to 5 and as “nonfatigued” when it was less than 5.

Back to Top | Article Outline


Objective Findings

Accommodative SS Response Level

The mean accommodative SS response levels are presented in TTable 1. After the NC task, the response level reduced significantly from the pretask baseline at both the 4.5-D (t 9 = 2.57, p = 0.03) and the 2.5-D (t 9 = 2.56, p = 0.02) stimulus levels. This decrease in posttask accommodative response level was approximately 13% at the 2.5-D level (0.26 D) and approximately 6% at the 4.5-D level (0.21 D). Thus, there was an increase in accommodative lag at both stimulus levels after the NC task (i.e., reduced accommodative accuracy). In contrast, there were no significant changes in the accommodative SS response levels after the C task at either the 2.5-D (t 9 = 1.51, p = 0.16) or the 4.5-D (t 9 = 0.32, p = 0.75) stimulus levels.



Back to Top | Article Outline

Accommodative SS Response Variability

The accommodative SS variability levels are presented in Table 1. After the NC task at the 4.5-D stimulus level, the response variability increased significantly (t 9 = 2.3, p = 0.04), by approximately 36%, as compared with the pretask baseline. The posttask response variability increased by approximately 17% as compared with the pretask 2.5-D stimulus level, but it was not statistically significant (t 9 = 1.70, p = 0.12). Fig. 1 presents unprocessed traces of the accommodative responses before and after the NC task in a typical subjectively based fatigued subject. Increases in both the accommodative lag and SS response variability were more pronounced at the higher versus lower stimulus level. In contrast, after the C task, the SS response variability did not significantly change at either the 2.5-D (t 9 = 0.52, p = 0.61) or the 4.5-D (t 9 = 0.17, p = 0.87) stimulus levels. Fig. 2 presents unprocessed traces of the accommodative responses before and after the C task in a typical subjectively based nonfatigued subject. There were no obvious differences in the pre-to-post SS response level and SS response variability at both 2.5-D and 4.5-D stimulus levels.





Back to Top | Article Outline

Other Parameters (Response Amplitude, Peak Velocity, and Time Constant)

The accommodative response amplitudes were not significantly different for either increasing or decreasing accommodation (t test, p > 0.05 for all comparisons). The mean values of response amplitude, peak velocity, and time constant before and after the NC task are given in Table 2. Similarly, after the C task, there were no significant changes in initial response amplitude, peak velocity, and time constant for either increasing or decreasing accommodation (Table 3) (t test, p > 0.05 for all comparisons).





Back to Top | Article Outline

Subjective Findings

The subjective findings were mixed. Six of the 10 subjects reported difficulty attaining focused imagery through the plus lenses only during the NC task, as well as the sensation of “eye strain” immediately after the NC task. The mean subjective fatigue rating scale was 7 in these “fatigued” subjects. Despite this initial problematic plus lenses aspect, four of these six subjects reported the task to “get easier” with repeated flipper cycles; the other two reported increased difficulty attaining focused imagery through the plus lenses. The remaining four of the 10 subjects were “nonfatigued” with the mean subjective scale of 1.3. However, the C task did not provoke any such symptoms suggestive of visual fatigue in any of the 10 subjects, with the mean subjective scale of 1.5. Subgroup (fatigued vs. nonfatigued) data analysis did not reveal any significant differences in the parameters assessed (t test, p > 0.05 for all comparisons). In addition, subjects who reported the NC task to become easier with flipper repetition did not exhibit any significant change in the objective recordings suggestive of an oculomotor learning effect (i.e., increased peak velocity) (t test, p > 0.05 for all comparisons).

Back to Top | Article Outline


Our primary hypothesis was that repetitive tasks would fatigue the accommodative system and, furthermore, that the fatigue-related changes would be reflected in the dynamics of the accommodative system. Overall, there were no significant differences in the initial response amplitude, peak velocity, and time constant of accommodation after either the C or NC task from the pretask level for both increasing and decreasing steps of accommodation. However, a trend for a reduction in peak velocity (∼7%) was found for both increasing and decreasing accommodation after the NC task only. This latter observation is consistent with the findings reported from testing of other oculomotor systems, such as for the vergence and saccadic systems, demonstrating a rather subtle (≤20%) but consistent fatigue-related reduction in peak velocity,38,39 as will be discussed later.

Our secondary hypothesis was that a mismatch in the stimulus demand40 under the nonnaturalistic NC task viewing condition would fatigue the accommodative system more readily than the naturalistic C task. This is based on the fact that under naturalistic binocular viewing conditions, the accommodative and vergence stimuli, as well as their respective changes, are equal in magnitude (i.e., congruent).36,37 This factor readily allows each system to coordinate and cross couple to produce synchronous time-optimal responses. However, this synchronization may be disturbed when the stimulus demands are unequal in magnitude (i.e., noncongruent).36,40,41 When the stimulus for one of the oculomotor systems is fixed (e.g., vergence) and the companion oculomotor system (e.g., accommodation) is changed rapidly and periodically (i.e., step input changes), as with the NC task paradigm used in the present study, visually-related symptoms (e.g., “visual” fatigue/asthenopia) are common.41 As predicted, whereas none of the subjects reported fatigue after the C task, 60% of the subjects reported visual fatigue after the NC task. This was evident from their significantly reduced mean accommodative response level (i.e., increased lag of accommodation), which was observed for both increasing (4.5-D level) and decreasing (2.5-D level) step changes in accommodation. In addition, for the NC task, the SS response variability was significantly increased at the 4.5-D level (∼36%), with a trend for increased variability observed at the 2.5-D level (∼17%). However, after the C task, neither the mean accommodative response level nor the SS response variability changed significantly from the pretask baseline at either stimulus level. Thus, there was lack of any fatigue-induced effects on the objective measures of accommodation only with the C task.

What might be the oculomotor-based mechanism related to the visual fatigue aspects uncovered in the present study? In the present investigation, this task was performed under binocular viewing conditions. For the C condition, vergence drive would be primary, and accommodative drive would be secondary, as blur drive gain initially from the eccentric retinal regions is low.42,43 However, no fatigue effects were observed under the C condition. In contrast, for the NC condition, the initial lens-induced blur drive would be primary, and the subsequent vergence drive would be secondary.37,44 Since fatigue-related changes in the objective measures were observed only under NC condition, the fatigue could be primarily attributed to result from the accommodative system. However, secondarily, the vergence system could also be involved due to the repetitive fusional convergence/divergence dynamic responses via the accommodative-vergence crosslink interaction.45,46

Back to Top | Article Outline

Comparison with Other Studies

In earlier studies, symptomatic subjects demonstrated a progressive increase in accommodative lag/error with attempted sustained accommodation (i.e., 2 minutes) at near (20 to 50 cm) using objective recording techniques,16,20 with this being a variant of the increased SS variability found in the present study related to visual fatigue. Whereas these two previous studies measured accommodation monocularly at five different dioptric distances, the present study recorded accommodative responses for 2-D step stimulus changes under binocular-viewing conditions only in initially asymptomatic individuals. Hence, the present study involved a strong and normal vergence interaction with accommodation that could potentially influence the results and related fatigue factors, as previously described, when compared with the earlier related studies. Furthermore, the present study evaluated the influence of C versus NC tasks on accommodative dynamics, whereas the other studies were purely congruently based. In addition, whereas the above studies16,20 monitored accommodative SS variability for at least 2 minutes continuously parsed into 10 second epochs, the present study recorded it for only a total of four to five continuous seconds. Lastly, the present study results are consistent with the findings of Vilupuru et al.33 in two young-adult emmetropic human subjects demonstrating relatively unchanged accommodative dynamics with repetitive step accommodative tracking. However, their study did not monitor the SS accommodative response over time in humans. Thus, there were many methodological differences between the various studies. A comprehensive study for both short- and long-term stimulus intervals, under a range of viewing conditions, needs to be conducted to understand more fully this important area of visual fatigue and its objective correlates in both symptomatic and asymptomatic young adults and children.

Back to Top | Article Outline

Fatigue of Other Oculomotor Systems

Several researches during the past century have been conducted to evaluate the effect of provocative and strenuous visual tasks on the oculomotor system.6,16,20,22,38,39 The most common paradigms used in these studies involved repetitive tasks under presumed maximum stimulus demand/“load.”

For example, Yuan and Semmlow38 recorded vergence dynamics to repetitive 4-degree step (100 convergence only responses were assessed; their subjects only casually gazed back to the far target without formal stimulus change and response, as was done for convergence) and 4-degree sinusoidal (0.25 Hz; 100 cycles) disparity vergence stimuli, as well as to 100, 250, and 500 repetitive saccades (10-degree amplitude) for comparative system purposes, in four visually normal, asymptomatic, adult subjects. All four subjects exhibited approximately a 20% decrease in peak velocity both with the step convergence task and with the 500 repetitive saccade task; however, the results were not statistically assessed. Whereas 100 repetitive saccades produced no change, 250 repetitive saccades produced at least a 10% decrease in peak velocity, thus suggesting a progressive fatigable saccadic system phenomenon. In addition, an increased occurrence of overlapping double-vergence responses was found with 100 repetitive vergence steps but not with the same number of repetitive saccades, in their study. In contrast, the presumably easier, more slowly varying sinusoidal task did not produce any consistent changes in vergence dynamics suggestive of fatigue. The reduction in step-related peak velocity and the increased frequency of double responses were attributed to fatigue of the vergence system caused by a defective pulse, but not step, of the overall oculomotor signal component (i.e., overall pulse-step) controlling the vergence neurons and vergence response. They also did not investigate SS vergence variability or any other dynamic parameters. Furthermore, symptoms of visual fatigue were not quantitatively assessed. In addition, similar to vergence, self-reported fatigued subjects demonstrated slowed saccades with a repetitive saccadic task.39,42 Thus, all of the earlier findings using a controlled and repetitive paradigm suggest that oculomotor fatigue may also be reflected by a relatively small and subtle, but consistent, decrease in peak velocity. Lack of significant reduction in peak velocity in the present study warrants detailed investigation with more target repetitions, as well as more sophisticated and/or sensitive quantitative objective and subjective assessment techniques. However, the finding of a 7% decrease in accommodative peak velocity approaches the fatigue-based values of these earlier studies (i.e., 20% of Yuan and Semmlow38).

Back to Top | Article Outline

Neurophysiological Implications

Similar to vergence, neurons have been identified in the midbrain that increase their firing rate during accommodation.47 Analogous to the vergence pulse-step signal,48 given their similar dynamic response characteristics, there are accommodation burst cells that fire in relation to accommodative velocity, producing a transient pulse-like response. Furthermore, there are accommodation tonic cells that fire in relation to the accommodative magnitude to produce the sustained step response component. The neural pulse is responsible for the actual accommodative trajectory and its related dynamics (e.g., peak velocity), whereas the neural step is responsible for the maintenance of accommodation per modeling studies based on human response findings.49 In the present study, whereas mean response amplitude and peak velocity did not change significantly, accommodative SS level was not maintained well over the several second measurement window; the accommodative SS level exhibited significantly increased response variability (as evident from the increased accommodative lag) after the NC task at both the 2.5-D and 4.5-D levels. This finding suggests that the neural site of the fatigue resides primarily in the “step” controller neural signal (i.e., tonic cell activity), as the step is responsible for accurately maintaining the accommodative SS level. Fatigue may act to impair gain accuracy of the neural integrator, thus resulting in a more variable SS accommodative response in addition to an increased lag. This has been demonstrated by Vilupuru et al.,33 who found that monkeys could not maintain the SS accommodative level with repeated and direct stimulation of the Edinger-Westphal nucleus. Whereas the above study involved direct central stimulation of accommodation, neurophysiological experiments involving sensory, motor, perceptual, and active attention in awake monkeys under more naturalistic conditions should be conducted in the future to study accommodative fatigue in detail.

Back to Top | Article Outline

Is Oculomotor Fatigue a Central or a Peripheral Phenomenon?

Whereas central fatigue involves higher level neurological control mechanisms as described in the Introduction,50 peripheral fatigue results from a decline in external muscle function associated with increased physical activity derived from non–central nervous system mechanisms.51 However, the long-standing debate as to the actual site of muscle fatigue seems to have been only recently resolved. When an individual reports “visual fatigue,” it has remained unclear whether the fatigue involved central or peripheral neural control mechanisms, or both.

This critical question was addressed in a recent study52 that tested macaque monkeys using single motor unit recordings from the oculomotor-based abducens nucleus. Responsivity was evaluated for repeated visually guided horizontal saccades (500 saccades) of varying amplitudes (8 to 24 degrees) tested for a total of 62 sessions. While saccadic accuracy was maintained, peak velocity decreased, and furthermore, saccadic duration increased (as predicted). Thus, the responses were slowed but of normal amplitude. The decrease in peak velocity of the saccades is consistent with the finding of Yuan and Semmlow,38 who reported a similar decrease in peak velocity in human subjects with 500 repetitive saccades. However, the finding of Prsa et al.52 of an increase in saccadic duration was attributed to a cerebellar adaptive maneuver to prevent the occurrence of saccadic inaccuracies, that is, multiple reduced-amplitude saccades, but with each having a normal peak velocity (i.e., main sequence).53 Thus, the monkeys executed one slowed but accurate saccade rather than several smaller and inaccurate, but dynamically rapid and normal, saccades. A model simulation using the firing rate of abducens neurons and eye position based on the viscoelastic properties of the eye muscle demonstrated that the eye muscle dynamics remained unaltered before and after the fatigue task. Thus, the decrease in peak velocity was associated with a decrease in peak discharge of the neurons and not caused by changes in the peripheral extraocular muscles. Hence, Prsa et al.52 concluded that the fatigue was centrally mediated. Furthermore, they performed an additional experiment to support the above notion. The abducens nucleus was stimulated using a constant intensity and duration of electrical current over a range of frequencies (300 to 600 Hz). If this fatigue-inducing task were successful, it would result in saccades that were significantly slowed with reduced amplitudes. However, there were no significant differences in either saccadic amplitude or peak velocity immediately before, immediately after, and after an unspecified rest period, with the task. This finding suggests a central nonperipheral site of the correlated eye movement fatigue. Although no consistent reduction in peak velocity has been shown in rhesus monkeys with repeated accommodation,33 detailed neurophysiological investigation under additional test conditions is lacking. We speculate on the presence of a similar kind of underlying mechanism in the present experiment in humans that acts during the repeated accommodative task to cause fatigue of a central origin. This supports the view of Yuan and Semmlow,38 who believed that fatigue resulting from repeated vergence was caused by central neural processes that controlled the pulse signal of the response and not by any change in the extraocular muscles per se. These findings suggest that the neural site of the fatigue resides primarily in the “pulse” controller neural signal (i.e., burst cell activity), as the pulse is responsible for the actual velocity of the dynamic trajectory.

Finally, the repeated congruent and the noncongruent tasks used in the present study were developed in an attempt to induce fatigue and to evaluate its related changes in the objective recordings and subjective ratings. As discussed in the Results section under Subjective Findings, six of the 10 subjects reported difficulty with plus lenses during the NC task. Initially, these subjects took extra time (ranging from 2 to 5 seconds) to clear the plus side. Four of these six subjects reported the task to get easier with increasing flipper cycles, and they cleared the target within a second or so. The sense of a subjective improvement reported by these four subjects may be attributed to “motor learning.”54–56 However, the objective measures did not demonstrate any consistent changes related to motor learning. Further experiments are warranted to determine the threshold for motor learning versus fatigue induced by the repetitive task. This may be especially important in dealing with oculomotor therapy in visually symptomatic individuals.56

Back to Top | Article Outline


The present findings have demonstrated that fatigue induced by a repetitive accommodative task is best and most consistently reflected by an increase in both the SS accommodative response level and its associated variability. Further experiments will be required to study accommodative dynamics in visually symptomatic individuals manifesting binocular dysfunction. Lastly, experiments should be conducted in the future to study the basic underlying mechanism involved in accommodative fatigue in monkeys using multiunit recording and in humans perhaps using noninvasive impedance cyclography.57

Preethi Thiagarajan

Department of Biological and Vision Sciences

SUNY State College of Optometry

33 West 42nd St

New York, NY 10036


Back to Top | Article Outline


The authors thank the College of Optometrists in Vision Development for providing funding. The authors also thank Dr. George Hung for his helpful comments.

This work was presented as a poster at the Academy meeting in November 2010 in San Francisco.

Received February 20, 2012; accepted October 1, 2012.

Back to Top | Article Outline


1. Torres-Harding S, Jason LA. What is fatigue? History and epidemiology. In: DeLuca J, ed. Fatigue as a Window to the Brain. Cambridge, MA: MIT Press; 2005: 3–18.
2. Tiesinga LJ, Dassen TW, Halfens RJ. Fatigue: a summary of the definitions, dimensions, and indicators. Nurs Diagn 1996; 7: 51–62.
3. Jason LA, Evans M, Brown M, Porter N. What is fatigue? Pathological and nonpathological fatigue. PM R 2010; 2: 327–31.
4. Al-Mulla MR, Sepulveda F, Colley M. A review of noninvasive techniques to detect and predict localised muscle fatigue. Sensors (Basel) 2010; 11: 3545–94.
5. Vøllestad NK. Measurement of human muscle fatigue. J Neurosci Methods 1997; 74: 219–27.
6. Berens C, Hardy le GH, Pierce HF. Studies in ocular fatigue. II. Convergence fatigue in practice. Trans Am Ophthalmol Soc 1926; 24: 262–87.
7. Carmichael L, Dearborn WF. Reading and Visual Fatigue. Boston, MA: Houghton Mifflin; 1947.
8. Sheedy JE, Hayes JN, Engle J. Is all asthenopia the same? Optom Vis Sci 2003; 80: 732–9.
9. Blehm C, Vishnu S, Khattak A, Mitra S, Yee RW. Computer vision syndrome: a review. Surv Ophthalmol 2005; 50: 253–62.
10. Chu C, Rosenfield M, Portello JK, Benzoni JA, Collier JD. A comparison of symptoms after viewing text on a computer screen and hardcopy. Ophthalmic Physiol Opt 2011; 31: 29–32.
11. Scheiman M, Wick B. Clinical Management of Binocular Vision, 3rd ed. Philadelphia, PA: Wolters Kluwer Health/Lippincott Williams & Wilkins; 2008.
12. Rouse M, Borsting E, Mitchell GL, Cotter SA, Kulp M, Scheiman M, Barnhardt C, Bade A, Yamada T. Validity of the convergence insufficiency symptom survey: a confirmatory study. Optom Vis Sci 2009; 86: 357–63.
13. Conlon EG, Lovegrove WJ, Chekaluk E, Pattison PE. Measuring visual discomfort. Vis Cogn 1999; 6: 637–63.
14. Daum KM. Accommodative dysfunction. Doc Ophthalmol 1983; 55: 177–98.
15. Scheiman M, Blaskey P, Ciner EB, Gallaway M, Parisi M, Pollack K, Selznick R. Vision characteristics of individuals identified as Irlen Filter candidates. J Am Optom Assoc 1990; 61: 600–5.
16. Chase C, Tosha C, Borsting E, Ridder WH 3rd. Visual discomfort and objective measures of static accommodation. Optom Vis Sci 2009; 86: 883–9.
17. Borsting E, Rouse MW, Deland PN, Hovett S, Kimura D, Park M, Stephens B. Association of symptoms and convergence and accommodative insufficiency in school-age children. Optometry 2003; 74: 25–34.
18. Marran LF, De Land PN, Nguyen AL. Accommodative insufficiency is the primary source of symptoms in children diagnosed with convergence insufficiency. Optom Vis Sci 2006; 83: 281–9.
19. Duane A. Subnormal accommodation. Trans Am Ophthalmol Soc 1925; 23: 159–83.
20. Tosha C, Borsting E, Ridder WH 3rd, Chase C. Accommodation response and visual discomfort. Ophthalmic Physiol Opt 2009; 29: 625–33.
21. Ciuffreda KJ. Accommodation, the pupil, and presbyopia. In: Benjamin WJ, Borish IM, eds. Borish’s Clinical Refraction, 2nd ed. Oxford, UK: Butterworth-Heinemann; 2006: 93–144.
22. Howe L. The fatigue of accommodation. JAMA 1916; 67: 130–42.
23. Howe L. On varieties of the fatigue of accommodation as registered by the ergograph. Trans Am Ophthalmol Soc 1917; 15: 145–53.
24. Berens C, Stark EK. Studies in ocular fatigue. IV. Fatigue of accommodation, experimental and clinic observations. Am J Ophthalmol 1932; 15: 527–42.
25. Berens C, Sells SB. Experimental studies on fatigue of accommodation. I. Plan of research and observations on recession of near point of accommodation following a period of interpolated work on the ophthalmic ergograph. Arch Ophthalmol 1944; 31: 148–59.
26. Koch CC, Kurtz JI. An experimental study of ocular fatigue. I. General fatigue. Am J Optom 1938; 5: 86–117.
27. Malmstrom FV, Randle RJ, Murphy MR, Reed LE, Weber RJ. Visual fatigue: the need for an integrated model. Bull Psychonom Soc 1981; 17: 183–6.
28. Takeda T, Ostberg O, Fukui Y, Iida T. Dynamic accommodation measurements for objective assessment of eyestrain and visual fatigue. J Hum Ergol (Tokyo) 1988; 17: 21–35.
29. Ciuffreda KJ, Scheiman M, Ong E, Rosenfield M, Solan HA. Irlen lenses do not improve accommodative accuracy at near. Optom Vis Sci 1997; 74: 298–302.
30. Simmers AJ, Gray LS, Wilkins AJ. The influence of tinted lenses upon ocular accommodation. Vision Res 2001; 41: 1229–38.
31. Kajita M, Ono M, Suzuki S, Kato K. Accommodative microfluctuation in asthenopia caused by accommodative spasm. Fukushima J Med Sci 2001; 47: 13–20.
32. Hasebe S, Graf EW, Schor CM. Fatigue reduces tonic accommodation. Ophthalmic Physiol Opt 2001; 21: 151–60.
33. Vilupuru AS, Kasthurirangan S, Glasser A. Dynamics of accommodative fatigue in rhesus monkeys and humans. Vision Res 2005; 45: 181–91.
34. Benjamin WJ, Borish IM, eds. Borish’s Clinical Refraction. 2nd ed. Oxford, UK: Butterworth-Heinemann; 2006.
35. Green W, Ciuffreda KJ, Thiagarajan P, Szymanowicz D, Ludlam DP, Kapoor N. Accommodation in mild traumatic brain injury. J Rehabil Res Dev 2010; 47: 183–99.
36. Kran BS, Ciuffreda KJ. Noncongruent stimuli and tonic adaptation. Am J Optom Physiol Opt 1988; 65: 703–9.
37. Ciuffreda KJ. Components of clinical near vergence testing. J Behav Optom 1992; 3: 3–13.
38. Yuan W, Semmlow JL. The influence of repetitive eye movements on vergence performance. Vision Res 2000; 40: 3089–98.
39. Bahill AT, Stark L. Overlapping saccades and glissades are produced by fatigue in the saccadic eye movement system. Exp Neurol 1975; 48: 95–106.
40. Eadie AS, Gray LS, Carlin P, Mon-Williams M. Modeling adaptation effects in vergence and accommodation after exposure to a simulated virtual reality stimulus. Ophthalmic Physiol Opt 2000; 20: 242–51.
41. Hoffman DM, Girshick AR, Akeley K, Banks MS. Vergence-accommodation conflicts hinder visual performance and cause visual fatigue. J Vis 2008; 8: 33.1–30.
42. Hung GK, Ciuffreda KJ. Accommodative responses to eccentric and laterally-oscillating targets. Ophthalmic Physiol Opt 1992; 12: 361–4.
43. Hung GK, Semmlow JL, Ciuffreda KJ. Identification of accommodative vergence contribution to the near response using response variance. Invest Ophthalmol Vis Sci 1983; 24: 772–7.
44. Semmlow J, Wetzel P. Dynamic contributions of the components of binocular vergence. J Opt Soc Am 1979; 69: 639–45.
45. Hung GK, Ciuffreda KJ. Sensitivity analysis of relative accommodation and vergence. IEEE Trans Biomed Eng 1994; 41: 241–8.
46. Fuchs AF, Binder MD. Fatigue resistance of human extraocular muscles. J Neurophysiol 1983; 49: 28–34.
47. Gamlin PD. Subcortical neural circuits for ocular accommodation and vergence in primates. Ophthalmic Physiol Opt 1999; 19: 81–9.
48. Mays LE. Neural control of vergence eye movements: convergence and divergence neurons in midbrain. J Neurophysiol 1984; 51: 1091–108.
49. Schor CM, Bharadwaj SR. A pulse-step model of accommodation dynamics in the aging eye. Vision Res 2005; 45: 1237–54.
50. Leavitt VM, DeLuca J. Central fatigue: issues related to cognition, mood and behavior, and psychiatric diagnoses. PM R 2010; 2: 332–7.
51. Keyser RE. Peripheral fatigue: high-energy phosphates and hydrogen ions. PMR 2010; 2: 347–58.
52. Prsa M, Dicke PW, Thier P. The absence of eye muscle fatigue indicates that the nervous system compensates for non-motor disturbances of oculomotor function. J Neurosci 2010; 30: 15834–42.
53. Bahill AT, Clark MR, Stark L. The Main Sequence, a tool for studying human eye movements. Math Biosci 1975; 24: 191–204.
54. Halsband U, Lange RK. Motor learning in man: a review of functional and clinical studies. J Physiol Paris 2006; 99: 414–24.
55. Abernathy B, Kipper V, Mackinnon LT, Neal RJ, Hanrahan S. The Biophysical Foundations of Human Movement. Champaign, IL: Human Kinetics; 1997.
56. Ciuffreda KJ. The scientific basis for and efficacy of optometric vision therapy in nonstrabismic accommodative and vergence disorders. Optometry 2002; 73: 735–62.
57. Saladin JJ, Usui S, Stark L. Impedance cyclography as an indicator of ciliary muscle contraction. Am J Optom Physiol Opt 1974; 51: 613–25.

accommodation; vergence; visual fatigue; asthenopia; accommodative dynamics; oculomotor system; eye movements

© 2013 American Academy of Optometry