Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 Sep 9.
Published in final edited form as: Curr Biol. 2013 Aug 15;23(17):1663–1669. doi: 10.1016/j.cub.2013.06.056

Rapid and persistent adaptability of human oculomotor control in response to simulated central vision loss

MiYoung Kwon 1,4, Anirvan S Nandy 2, Bosco S Tjan 1,3
PMCID: PMC3773263  NIHMSID: NIHMS502321  PMID: 23954427

SUMMARY

The central region of the human retina, the fovea, provides high-acuity vision. The oculomotor system continually brings targets of interest into the fovea via ballistic eye movements (saccades). The fovea thus serves both as the locus for fixations and as the oculomotor reference for saccades. This highly automated process of foveation is functionally critical to vision and is observed from infancy [1, 2]. How would the oculomotor system adjust to loss of foveal vision (central scotoma)? Clinical observations of patients with central vision loss [3, 4] suggest a lengthy adjustment period [5], but the nature and dynamics of this adjustment remain unclear. Here we demonstrate that the oculomotor system can spontaneously and rapidly adopt a peripheral locus for fixation and can re-reference saccades to this locus, in normally sighted individuals whose central vision is blocked by an artificial scotoma. Once developed, the fixation locus is retained over weeks in the absence of the simulated scotoma. Our data reveal a basic guiding principle of the oculomotor system that prefers control simplicity over optimality. We demonstrate the importance of a visible scotoma on the speed of the adjustment and suggest a possible rehabilitation regimen for patients with central vision loss.

RESULTS

We simulated a central scotoma in six normally-sighted young adults using a gaze-contingent display while the subjects performed a demanding visual-search task. The simulated scotoma appeared as a visible gray disc and blocked the screen content at and around the center of gaze. A separate group of six normally-sighted subjects participated in a control experiment, in which they performed the identical task without a simulated scotoma. The task alternated between two main components (Fig. 1a): (1) object following, in which subjects maintained fixation on an object as it was randomly repositioned on the screen against either a uniform or cluttered background, and (2) visual search, in which subjects searched for the target object shown in (1). Subjects were instructed to perform the search task as quickly and as accurately as possible during the visual-search phase. During the initial period of the experiment, subjects were not told to use a particular eccentric location to guide their eye movements (free exploration).

Figure 1. Task and experiment conditions.

Figure 1

Open in a new tab

a. Task sequence during the free-exploration period. This sequence is an example of one trial. There were 30 trials per block. Each trial was comprised of four phases: (i) object following against a gray background; (ii) centering of gaze; (iii) searching for the target object presented in the object-following phase; (iv) object following amidst a cluttered background with an array of non-target distracters. b. Example of object following during the explicit training period. A small white cross (~0.7° in height) appeared at each subject’s estimated PRL. Subjects were instructed to use the cross as a gaze reference point for fixation and saccadic eye movements. c. Task sequence of the invisible scotoma experiment. The first phase was identical to the first object-following phase of the main experiment except that the luminance of the simulated scotoma was matched to that of the background. As a result, the simulated central scotoma was invisible to the subjects. Subjects then searched for the target object amongst an array of distracters presented against a uniform gray background. For ease of visibility in the figure, target objects and the gaze reference cross were rendered at 2.5 times their sizes used in the experiment, relative to the rest of the displayed elements.

Rapid development of preferred retinal locus (PRL) for fixation

After having performed the experimental task with simulated central scotoma for approximately 3 hours, spread over 2 to 3 days (free exploration, Table S1), all six subjects in the experimental condition spontaneously followed the jumping target with eccentric fixation, keeping it out of the central scotoma (Fig. S1).

Five of the six subjects (two shown in Fig. 2a) spontaneously formed a single preferred retinal locus (PRL) [6] for fixation near the border of the scotoma, while a sixth subject developed two PRLs at roughly opposite sides of the scotoma (S5 in Fig. 2a). For the five single-PRL subjects, the variance of the fixational PRL decreased rapidly during this free-exploration period (Fig. 2b). The variance (defined as Bivariate Contour Ellipse Area (BCEA)) measured at the end of this period was significantly smaller than that at the beginning (54% reduction, t(4) = 2.97, p < 0.05). Fixation stability continued to improve with practice in the presence of the simulated scotoma.

Figure 2. Development and retention of preferred retinal locus (PRL) for fixation.

Figure 2

Open in a new tab

a. Probability density maps of the retinal positions of a target object at fixation are shown for three of the six subjects. (Refer to supplemental Table S1 for the specific number of blocks each subject performed in the different stages of the experiment.) Rows, from top to bottom: densities estimated from the first and last blocks of the free-exploration period, density estimated from the last block of the explicit-training period, density estimated from all 5–7 blocks of the retention assessment, and density estimated from all 5 blocks when the scotoma was made invisible. The density of the fovea-viewing control experiment, averaged across all control subjects, is also shown. Data from the object-following components with blank background and cluttered background were combined. Each polar plot represents the visual field. The gray patch depicts the central scotoma. The red dot marks the location of peak density, which we took as the estimated location of the fixational PRL. All six subjects exhibited a rapid and spontaneous emergence of fixational PRLs near the border of the scotoma within 3 hours (8–9 blocks). All subjects were able to further refine the precision of their fixational PRLs during a subsequent explicit training period that lasted at least 15 hours (40 blocks). Not only was the PRL retained after being established, but the same PRL was also used when the scotoma was made invisible. b. Variance of the fixational PRL as a function of block number. Variance was defined as the bivariate contour ellipse area (BCEA) that encompassed 68% of fixations around the mean. The blue solid line indicates the average BCEA across all five subjects who had spontaneously developed a single PRL (one subject, S5 shown here, developed two PRLs); shaded blue area indicates ±1 standard deviation of the BCEA, showing inter-subject variability. The gray rectangular box represents the distribution of BCEA from the control subjects in the fovea-viewing condition (±1 standard deviation). A gap of 5 blocks was inserted on the abscissa between free exploration and explicit training to denote a break of 1–2 days from the experiment in-between these two stages. The BCEA of the fixation PRLs from the five subjects with a single PRL decreased rapidly within a short period of time without explicit training. Explicit training reduced the BCEA to the range of foveal fixation. It also eliminated one of the two PRLs of S5, as demonstrated in the retention test.

Rapid development of re-referenced saccades near the fixational PRL

More importantly, for the five subjects with a single PRL, their first saccade after each target movement placed the target at or near their fixational PRL (Fig. 3a, Fig. S2) and away from the fovea, demonstrating a shift in oculomotor reference from the fovea to the PRL. Similar to the fixational PRL, the variance of the first saccade-landing site (i.e. target location on the retina after the first saccade following each target movement) decreased rapidly (Fig. 3b). The variance (BCEA) of the first saccade-landing site during the last block of free exploration was reduced by 28% compared to that of the first block (t(4) = 5.02, p < 0.05). This noticeable change in the first saccade-landing site after about 3 hours of exposure to the simulated central scotoma demonstrates a remarkable adaptability of the oculomotor system.

Figure 3. Development and retention of re-referenced saccades measured by the first saccade-landing site.

Figure 3

Open in a new tab

a. Probability density maps of the retinal position of a target object at the completion of the first saccade following target movement (same format as in Fig. 2a). b. Variance (BCEA) of the first saccade-landing site as a function of block number (same format as in Fig. 2b). The first saccade-landing site after each target movement was near the fixational PRL and distant from the fovea. Similar to the fixational PRL, the variance of the first saccade landing sites decreased rapidly within a short period of time without explicit training, demonstrating a spontaneous shift in oculomotor reference from the fovea to the PRL. With explicit training, the BCEA was further reduced to the normal range of intact foveal vision.

The variance of the first saccade-landing site was significantly larger than that of the fixational PRL even for the last block of the free exploration (38.77 deg2 vs. 16.01 deg2, t(4) = −2.98, p < 0.05). Oculomotor re-referencing may require a longer time to fully develop. The slower time course of saccadic re-referencing has been observed in adult monkeys with bilateral foveal lesions [7]. It thus appears that the refinement of the fixational PRL preceded the refinement of saccade re-reference.

Refinement of the fixational PRL and saccadic re-reference following explicit training

Despite the rapid emergence of both fixational PRL and saccade re-reference with simulated central scotoma during the free exploration, we observed that the variances of both remained higher than those of the foveal-viewing control subjects (fixational PRL: t(8) = 2.70, p < 0.05; saccade re-reference: t(8) = 2.69, p < 0.05). We asked whether it would be possible to further reduce the variances with explicit training. We displayed a small white cross at the retinal location of each subject’s emerged fixational PRL (Fig. 1b) and instructed subjects to follow the target with this gaze marker (explicit training). Otherwise, subjects performed the identical task as before. With 15 to 25 hours of such explicit training (Table S1), all six subjects were able to refine their oculomotor control such that the variances of the PRL and the first saccade-landing site became comparable to those of the control subject (Figs 2b & 3b). Contrary to conventional wisdom [8], we found that with an effective training regimen, the oculomotor control with peripheral vision can be as precise and accurate as that with foveal vision.

Retention of the learned PRL in the absence of the simulated scotoma

To investigate whether the learned fixational PRL and saccade re-reference could be retained over an extended period of time without practice, subjects were recalled for a retention assessment (retention, Table S1) at least a week (up to a month) after completion of the explicit training. Subjects performed the same task (visual search, object-following in clutter) as they did during the free-exploration period of the experiment, without any gaze marker. To further assess the robustness of the adaptation, we also tested with an invisible scotoma (Fig 1c, in separate sessions) during the object-following phase by matching the color and luminance of the scotoma to that of the uniform gray background (invisible scotoma, Table S1). All subjects retained the same PRLs, even in the invisible central scotoma condition (Figs. 2a, 3a). However, the variances of both the fixational PRL and first saccade-landing site for the retention and invisible scotoma conditions were larger compared to those measured at the end of the explicit-training period (p < 0.05, both comparisons).

Changes in the characteristics of eye movements during PRL development

The characteristics of eye movements covaried with the development of PRL. We examined the time course of changes in saccade latency, the number of saccades immediately succeeding each object movement, and fixation duration during the object-following component of the task. We found a significant decrease in both saccade latency (42% reduction, Fig. 4a) and the number of saccades (60% reduction, Fig. 4b). The saccade latency and number of saccades at the end of explicit training became comparable to those of controls (saccade latency: t(8) = 1.36, p = 0.21; number of saccades: t(8) = 1.30, p = 0.23). Although there were no significant changes in fixation duration, there was a slight upward trend during the course of the experiment (Fig. 4c). The rate of change in both saccade latency and the number of saccades mirrored those of the refinement of the fixational PRL and the saccadic re-reference: a rapid decrease during the free exploration followed by a persistent decrease during explicit training to levels that were comparable to controls.

Figure 4. Changes in eye movement characteristics during object following.

Figure 4

Open in a new tab

a. Median saccade latency (the time taken to make the first saccade upon object movement) as a function of block number (same format as in Fig. 2b). b. Mean number of saccades per object movement (i.e. the number of corrective saccades plus one) as a function of block number. c. Mean duration of fixation in-between object movement as a function of block number. Error bounds are ±1 SD.

Changes in visual search performance during PRL development

Performance during the visual search component of the task also improved over the course of PRL development (Fig. 5). Search accuracy was high (~89 ± 2.1% SD, relative to the chance level of 50%, Fig. 5a) and remained unchanged throughout the experiment. The average search time showed a considerable decrease (42% reduction, Fig. 5b) during the free exploration period followed by a gradual decrease during explicit training. A corresponding decrease was observed in the number of saccades that was required to find the target (60% reduction, Fig. 5c). At the end of explicit training, subjects were performing the search task as fast as (t(8) = −0.94, p = 0.38) and with as few saccadic eye movements as (t(8) = 1.30, p = 0.23) the foveal-viewing controls.

Figure 5. Changes in visual search performance.

Figure 5

Open in a new tab

a. Visual search accuracy (percent correct) as a function of block number (same format as in Fig. 2b). The mean accuracy level (89% ± 2.1% SD) was similar to that of the control group and was constant through the course of the experiment. b. Time taken to complete a search trial (either target present or target absent) as a function of block number. c. Number of saccades required for completion of a search trial. Only correct trials were included for b and c. Error bounds are ±1 SD

DISCUSSION

We found that a stable PRL spontaneously emerges within hours of performing a visual task with a simulated central scotoma. Saccades were re-referenced to the PRL. The acquired PRL was retained for at least a week, during which subjects went about their daily lives with normal central vision. The same PRL was used even when the simulated scotoma was not visible. With explicit training using a gaze marker, the fixation stability at the PRLs and the precision of the targeting saccades became as good as those with normal foveal vision. These findings imply a flexible and adaptive oculomotor system. Akin to learning a motor skill such as riding a bicycle, the system can rapidly develop a new motor plan even with limited and sporadic exposure. Moreover, the motor plan improves with use and is retained without continuous practice.

Our findings suggest that the oculomotor system prefers developing a motor plan that is simple over one that may be optimal with respect to saccade amplitude and accuracy. With intact central vision, the fovea has the highest acuity and is therefore a unique point in the visual field. The oculomotor system is presented with the simple goal of bringing the high-resolution fovea to the target of interest. Losing central vision eliminates this unique point. It is conceivable that there exists a unique “best” point in the spared peripheral retina for optimal form vision, and the location of this point may be partly determined by oculomotor factors, such as fixational drift, that contribute to spatiotemporal sensitivity [9, 10]. However, the difference in functional acuity between the best peripheral point and the next best is likely to be small, and certainly less than between normal fovea and parafovea. Indeed, preliminary data from a recent study showed that local acuity did not predict the location of PRL in MD patients [11]. Hence, given a target of interest, there are likely multiple points (or contours) on the peripheral retina that are equally adequate. As such, the oculomotor system could choose a point that is closest to the target in order to minimize saccade amplitude and thus improve accuracy. Alternatively, it could choose a point on the peripheral retina that avoids occluding the informative parts of the target with the scotoma (e.g. for reading English text, placing the current word below the scotoma could be advantageous). Despite these possibilities for an on-demand variable-fixation strategy that is perhaps more efficient in terms of accuracy or information gain, the oculomotor system opts for a strategy that is exceedingly simple from a control perspective: using a single point in the periphery as the preferred locus for fixation. This amounts to adding a constant vector offset to the existing oculomotor control system.

Our observations are qualitatively different from what is generally known as saccadic adaptation [12]. Saccadic adaptation is a continuous recalibration process that minimizes the perceived saccade error; it is thought to be necessary for adapting to an ever-changing oculomotor plan due to growth, aging and diseases. Saccadic adaptation, while rapid in humans (in the order of a couple of hundreds of saccades), is magnitude and direction specific [13, 14, 15]. Establishing a PRL using saccadic adaptation would require one to adapt to a large, if not infinite, number of magnitudes and directions. More importantly, saccadic adaption affects the current state of the oculomotor controller. It takes time to adapt and to recover from adaptation. This contradicts the observations that (1) our subjects’ normal foveation behavior was not affected by performing the task using their PRL, and (2) subjects retained their PRL between the daily experimental sessions and after weeks of not performing the task. Hence, the development of PRL that we observed is unlikely to be due to saccadic adaptation.

Our results with simulated central scotoma echo some of the clinical findings. A majority of macular degeneration (MD) patients with bilateral central vision loss utilize a single PRL for viewing or fixating [16, 17, 18], while some patients develop two or three task dependent PRLs [19]. A portion of these patients make saccades that are re-referenced to their fixational PRLs [20]. Patients’ visual performance is closely correlated with the establishment of a stable PRL [21, 22, 23], further suggesting that variable fixation is not an option used by the oculomotor system. However, unlike MD patients who apparently take months to develop PRLs and to re-reference the oculomotor system to the PRLs [5], we found that PRL development was rapid with a (visible) simulated central scotoma. Our result is consistent with the report of a similarly rapid development in the case of pursuit eye movements with a sinusoidally moving target [24] and qualitative accounts of fast emergence of visual-field preference for target identification [25, 26]. In contrast, while the natural development of PRLs may occur in some patients within weeks of vision loss, saccade re-referencing to PRLs in macular patients is a slow process, with a median time ranging from 1 to 6 months depending on the age of MD onset [5]. The difference in the time courses between simulated and real scotoma is even more remarkable when considering the fact that the simulated scotoma in our experiment was imposed for only about an hour a day. It is quite plausible that the gradual deterioration of vision in MD patients interferes with the development of adaptive oculomotor behavior. Age may also be a contributing factor to the slow development. Patients have better fixation stability if their central vision loss occurred at an early age [20]. In a recent study, we found that older adults were slower and used excessive eye movements during a visual search task with simulated central scotoma [27], although the same study did not examine PRL development. Another likely cause for the slow development is that real scotomas are often invisible or with unclear boundary [28, 29, 30]. A visible scotoma can provide accurate positional feedback and may thus speed up PRL development. Given our results of a robust development of PRL and oculomotor re-referencing with a visible scotoma and the persistence of a PRL even when the scotoma was rendered invisible, it might be possible to speed up the process of PRL development in patients by superimposing a simulated scotoma on a real one, thereby creating a visible border.

The remarkable ability of the oculomotor system to rapidly adjust to an occluded central vision reveals a basic guiding principle of oculomotor control. Given the time pressure underlying saccadic eye movements, the system strives to maintain simplicity – a new motor plan is formed by adding a constant vector offset to the existing and well-practiced motor plan. Our finding opens up the rich possibility of exploring plasticity in the cortical and sub-cortical structures associated with oculomotor control, and in the visual areas that are retinotopically associated with the newly formed PRL.

EXPERIMENTAL PROCEDURES

Participants

Twelve normally sighted subjects were recruited from the University of Southern California campus (ages ranging from 19 to 30 years, one female). Six subjects participated in the simulated scotoma experiments. The other six subjects participated in the control experiment. They all had normal or corrected-to-normal vision without known cognitive or neurological impairments. The mean acuity (Lighthouse distance acuity chart) was −0.1 logMar (Snellen 20/15). Subjects received monetary compensation for their participation. The experimental protocols were approved by the Internal Review Board (IRB) of the University of Southern California and written informed consent was obtained from all subjects prior to the experiment.

Stimuli and Apparatus

All stimuli were high contrast 24-bit RGB color images of indoor scenes and objects. Images of 49 indoor scenes (1024 × 768 pixels) were selected from an image database [31]. Images of 140 objects (average diagonal size of 2.4°, range 2° to 2.7°) were selected from a commercially available set of photographs of real objects at www.photos.com (now at www.thinkstockphotos.com). The stimuli were generated and controlled using MATLAB (version 7.9) and Psychophysics Toolbox extensions (Windows 7 OS) [32, 33], running on a Dell PC computer. The display was a 19″ CRT monitor (refresh rate: 85 Hz; resolution: 1024 × 768). The stimuli were presented at a viewing distance of 57 cm. The displayed scenes subtended visual angles of 39° (width) by 29° (height).

Eye Movement Recoding and Simulated Scotoma

Subjects’ eye movements were monitored using an infrared video-based eye-tracker sampled at 2000 Hz (EyeLink 1000/Tower Mount monocular eyetracker, SR Research Ltd., Ontario, Canada) with a maximum spatial resolution of 0.02°. A 9-point calibration/validation sequence was performed at the beginning of every block and drift correction was made at the beginning of every trial. Calibration and/or validation were repeated until the validation error was smaller than 1° on average. The average gaze error was 0.5°, ranging from 0.1° to 1°. A gaze-contingent visual display was used to simulate central visual field loss in normally-sighted subjects. This paradigm, referred to as “artificial scotoma” [34], has been used in previous studies to investigate various issues related to central field loss [24, 35, 36]. The real-time gaze position was sent to the display computer through a high speed Ethernet link. The continuous gaze information was used to draw a scotoma on the display screen at a refresh and update rate of 85Hz. The size and shape of the scotoma were derived from the visual field measurement obtained from a patient with age-related macular degeneration [37], as shown in Fig 1a. The scotoma was a nearly circular disc and subtended about 10° of visual angle and was rendered as a uniform gray patch (luminance 18 cd/m2) on the screen. We also ran an invisible scotoma condition with a slightly lighter gray patch, the luminance of which was matched to that of the gray background (22 cd/m2). The average delay between actual eye movement and screen update was estimated to be approximately 10 ms (range from 2 ms to 15 ms, based on the eye-data to frame-time latencies we measured on the stimulus computer and on the manufacturer’s data on minimum eye data latency). The gaze position error, i.e., the difference between target position and computed gaze position was estimated during the 9-points validation process. A chin-and-forehead rest was used throughout the experiment in order to minimize head movements and trial-to-trial variability in the estimate of gaze position. The emergence of an eccentric retinal locus with re-referenced saccades observed in the free exploration period of the experiment, in contrast to the normal foveating behavior observed in the control experiment, served as confirmation that the slight amount of spatiotemporal imprecision in the eye tracking system was not sufficient to interfere with the effectiveness of the simulated central scotoma.

Procedure

Subjects performed a demanding visual task in uninterrupted blocks of 30 trials each. On average, subjects took 30 hours to complete the entire experiment (excluding breaks and calibration time). This was spread into several sessions spanning two months. Each block took approximately 25 minutes to complete. Subjects performed the task in a dimly lit room while they were seated in a comfortable position with chin and forehead supports. Each block started with the calibration/validation sequence described earlier (~5 minutes), followed by drift correction. The trial started with an auditory beep immediately after drift correction. For each trial, subjects had to complete four task phases (Fig 1a):

  1. Object following, in which subjects followed a target object as it was randomly repositioned four times against a gray background (Fig 1a.i). The center of the object was uniformly distributed within the central 31° by 22° region of the display. Each move was initiated only when the onscreen position of the subjects’ scotoma did not occlude the target object for at least 1.5 seconds. This was done to encourage eccentric fixation. Subjects were told to examine the target object as accurately as possible since it was the search target for the upcoming visual-search phase. Otherwise, subjects did not receive any explicit instructions on how to use their gaze.

  2. Centering of gaze, in which subjects centered their gaze in the middle of the screen so that their scotoma was placed inside a black rectangular box for 1.5 s (Fig 1a.ii). This was done right before the onset of visual search to minimize any positional bias.

  3. Visual search in which subjects searched for the target object (same object as in Phase 1) amidst a cluttered background with an array of non-target distracters (Fig 1a.iii). Both target and non-target distracters were superimposed on the scene rather than being part of the scene. This was done in order to minimize any contextual effects on search performance. Subjects were given an unlimited amount of time to perform the search, after which they indicated the presence or absence of the search target by a key-press (‘p’ for presence, ‘a’ for absence). The probability of the target being present was 0.5. Subjects were instructed to perform the search task as quickly and as accurately as possible. An auditory feedback was provided for correct response.

  4. Object following similar to the first object following phase, except that the target was a randomly drawn object and that it appeared in a cluttered background (Fig 1a.iv). Oculomotor performance (fixational PRL and saccadic re-reference) was assessed during the two object following phases (Phases 1 and 4), under the assumption that the target object was the intended target of gaze.

Subjects performed at least two or three blocks per day and completed 9–30 blocks of the task during free exploration. Subjects then received explicit training (42–72 blocks) to refine the emerged PRL. This was done by having the subjects performing the same task as in the free-exploration period except that their PRLs were marked with a small 0.7° white cross (Fig. 1b), and subjects were instructed to use the cross as the point of gaze. This effectively encouraged the subjects to rely on the emerged PRLs for the task and possibly improve fixation accuracy with the peripheral locus. For normal foveal vision, a significant decrease in the accuracy of eye positions has been reported in the absence of a fixation marker [38]. Upon completion of explicit training, subjects took a break for at least one week before being recalled for another round of the task for 5–7 blocks (retention assessment). After the retention assessment, subjects performed 5 additional blocks of the object following and visual search tasks against a gray background, identical to the first object-following phase and the third visual search phase of the free exploration period of the experiment but with one exception: the luminance of the simulated scotoma was matched to that of the background (Fig. 1c). As a result, the simulated central scotoma was not visible to the subjects (invisible scotoma). The invisible scotoma experiment, thus, consisted of two phases: object following (Fig. 1c.i) and visual search (Fig. 1c.ii). A separate group of six subjects participated in a control experiment (total 5 blocks) identical to free exploration but without the simulated scotoma (control experiment). The specific numbers of blocks each subject performed in the different stages of the experiment are listed in supplemental Table S1.

Data Analysis

Gaze position data was first preprocessed by an edge-preserving median filter with a 200 ms window to remove transient noise. A modified version of a standard parsing algorithm [39, 40] was then applied to the preprocessed data to robustly classify saccades and fixations while excluding microsaccades [41]. We defined a period of eye movement as a saccade if the following conditions were met: (1) the eye velocity exceeded 20 deg/s during the entire period, (2) the peak velocity during this period was in the top 25 percentile of the recording phase, (3) the beginning and end of this period had a velocity of at least 15% of the peak velocity, and (4) the amplitude of the eye movement during this period exceeded 0.9°. Periods that were not saccades were candidates for fixations. A candidate period was classified as a fixation if it started with at least 50 ms of stable gaze (standard derivation in gaze position did not exceed 0.5° in any direction) and ended with one of the following conditions: (1) a period of missing data (a blink) exceeding 500 ms, (2) the start of a saccade, or (3) the start of a reliable drift that exceeded 1°.

Fixation density maps were derived from the retinal positions of the target objects during periods of fixation via density estimation with a bivariate Gaussian kernel [42]. The PRL was defined as the peak of the density. Fixation stability has traditionally been quantified by calculating the area of the ellipse that encompasses a given proportion (p) of eye positions during fixation [43]. This area is termed the Bivariate Contour Ellipse Area (BCEA) [44]. A more stable fixation corresponds to a smaller BCEA. In the current study, the variance of fixation was defined as the BCEA that encompassed 68% of fixations around the mean [43] and was used as an indicator of fixation stability. BCEAs were calculated from the density maps. (The fixational BECAs measured from the current experiment are likely to be overestimations because we instructed our subjects to simply “follow and examine” an object, 2°–2.7° in size, without specifying where on the object a subject must fixate.) The density maps for the first saccade-landing site were obtained in a similar manner from the retinal positions of the target objects at the endpoint of the first saccade after object movement.

Supplementary Material

01

HIGHLIGHTS.

  • Fast (~hours) emergence of eccentric fixation locus following occluded central vision

  • Rapid re-referencing of saccades to the eccentric fixation locus

  • Learned oculomotor plan is retained for weeks with normal (unoccluded) central vision

Acknowledgments

This work was supported by NIH grants R01EY017707, R01EY016093. The authors would like to thank Anita Disney, Gordon Legge, George Timberlake and two anonymous reviewers for their helpful comments and discussion of the manuscript, and Jessica Leigh Gonzales and Eric Trinh for their help with subject recruitment and data collection.

Footnotes

*

The authors declare that they have no competing financial interests.

Author Contributions

M.K., A.S.N. and B.S.T. conceived the experiments. M.K. collected the data. M.K. and B.S.T. analyzed the data. M.K., A.S.N. and B.S.T wrote the manuscript.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • 1.Slater AM, Findlay JM. The measurement of fixation position in the newborn baby. J Exp child psychol. 1972;14:349–364. doi: 10.1016/0022-0965(72)90056-2. [DOI] [PubMed] [Google Scholar]
  • 2.Aslin RN, Salapatek P. Saccadic localization of visual targets by the very young human infant. Percept Psychophys. 1975;17:293–302. [Google Scholar]
  • 3.Cummings RW, Whittaker SG, Watson GR, Budd JM. Scanning characters and reading with a central scotoma. Am J Optom Physiol Opt. 1985;62:833–843. doi: 10.1097/00006324-198512000-00004. [DOI] [PubMed] [Google Scholar]
  • 4.Schuchard RA, Fletcher DC. Preferred retinal locus: a review with applications in low vision rehabilitation. Ophthalmo Clin North Am. 1994;7:243–256. [Google Scholar]
  • 5.Crossland MD, Culham LE, Kabanarou SA, Rubin GS. Preferred retinal locus development in patients with macular disease. Ophthalmology. 2005;112:1579–1585. doi: 10.1016/j.ophtha.2005.03.027. [DOI] [PubMed] [Google Scholar]
  • 6.Crossland MD, Engel SA, Legge GE. The Preferred Retinal Locus in macular disease: toward a consensus definition. Retina. 2011;31:2109–2114. doi: 10.1097/IAE.0b013e31820d3fba. [DOI] [PubMed] [Google Scholar]
  • 7.Heinen SJ, Skavenski AA. Adaptation of saccades and fixation to bilateral foveal lesions in adult monkey. Vision Res. 1992;32:365–373. doi: 10.1016/0042-6989(92)90145-9. [DOI] [PubMed] [Google Scholar]
  • 8.Sansbury RV, Skavenski AA, Haddad GM, Seinman RM. Normal fixation of eccentric targets. J Opt Soc Amer A. 1973;63:612–614. doi: 10.1364/josa.63.000612. [DOI] [PubMed] [Google Scholar]
  • 9.Kuang X, Poletti M, Victor JD, Rucci M. Temporal encoding of spatial information during active visual fixation. Curr Biol. 2012;22:510–514. doi: 10.1016/j.cub.2012.01.050. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Rucci M, Iovin R, Poletti M, Santini F. Miniature eye movements enhance fine spatial detail. Nature. 2007;447:852–855. doi: 10.1038/nature05866. [DOI] [PubMed] [Google Scholar]
  • 11.Chung S, Bernard J-B. Does the location of the PRL correspond to the retinal location with the best acuity? Invest Ophthalmol Vis Sci. 2013;54:E-Abstract 2183. [Google Scholar]
  • 12.Hopp JJ, Fuchs AF. The characteristics and neuronal substrate of saccadic eye movement plasticity. Prog Neurobiol. 2004;72:27–53. doi: 10.1016/j.pneurobio.2003.12.002. [DOI] [PubMed] [Google Scholar]
  • 13.Albano J. Adaptive changes in saccade amplitude: oculocentric or orbitocentric mapping. Vision Res. 1996;36:2087–2098. doi: 10.1016/0042-6989(96)89627-1. [DOI] [PubMed] [Google Scholar]
  • 14.Abel L, Schmidt D, Dell’osso L, Daroff R. Saccadic system plasticity in human. Ann Neurol. 1978;4:313–318. doi: 10.1002/ana.410040405. [DOI] [PubMed] [Google Scholar]
  • 15.Moidell B, Bedell H. Changes in oculocentric visual direction induced by the recalibration of saccades. Vision Res. 1988;28:329–336. doi: 10.1016/0042-6989(88)90161-7. [DOI] [PubMed] [Google Scholar]
  • 16.von Noorden GK, Mackensen G. Phenomenology of eccentric fixation. Am J Ophthalmol. 1962;53:642–661. doi: 10.1016/0002-9394(62)91987-6. [DOI] [PubMed] [Google Scholar]
  • 17.Whittaker SG, Budd J, Cummings RW. Eccentric fixation with macular scotoma. Invest Ophthalmol Vis Sci. 1988;29:268–278. [PubMed] [Google Scholar]
  • 18.Fletcher DC, Schuchard RA. Preferred retinal loci relationship to macular scotomas in a low-vision population. Ophthalmology. 1999;104:632–638. doi: 10.1016/s0161-6420(97)30260-7. [DOI] [PubMed] [Google Scholar]
  • 19.Timberlake GT, Mainster MA, Peli E, Auglier RA, Essock EA, Arend LE. Reading with a macular scotoma, I. Retinal location of scotoma and fixation area. Invest Ophthalmol Vis Sci. 1986;27:1137–1147. [PubMed] [Google Scholar]
  • 20.White JM, Bedell HE. The oculomotor reference in humans with bilateral macular disease. Invest Ophthalmol Vis Sci. 1990;31:1149–1161. [PubMed] [Google Scholar]
  • 21.Whittaker SG, Budd J, Cummings RW. Eccentric fixation with macular scotoma. Invest Ophthalmol Vis Sci. 1988;29:268–278. [PubMed] [Google Scholar]
  • 22.Falkenber HK, Rubin GS, Bex PJ. Acuity, crowding, reading and fixation stability. Vision Res. 2007;47:126–135. doi: 10.1016/j.visres.2006.09.014. [DOI] [PubMed] [Google Scholar]
  • 23.Tarita-Nistor L, Gonzalez EG, Markowitz SN, Steinbach MJ. Plasticity of fixation in patients with central vision loss. Visual Neurosci. 2009;26:487–494. doi: 10.1017/S0952523809990265. [DOI] [PubMed] [Google Scholar]
  • 24.Pidecoe PE, Wetzel PA. Oculomotor tracking strategy in normal subjects with and without simulated scotoma. Invest Ophthalmol Vis Sci. 2006;47:169–178. doi: 10.1167/iovs.04-0564. [DOI] [PubMed] [Google Scholar]
  • 25.Ness JW, Zwick H, Molchany JW. Preferred retinal location induced by macular occlusion in a target recognition task. SPIE Proceedings of Laser-Inflicted Eye Injuries: Epidemiology, Prevention, and Treatment. 1996;2674:131–135. [Google Scholar]
  • 26.Varsori M, Perez-Fornos A, Safran AB, Whatham AR. Development of a viewing strategy during adaptation to an artificial central scotoma. Vision Res. 2004;44:2691–2705. doi: 10.1016/j.visres.2004.05.027. [DOI] [PubMed] [Google Scholar]
  • 27.Kwon M, Ramachandra C, Mel BW, Tjan BS. Contour enhancement benefits peripheral vision task for older adults. Optometry Vision Sci. 2012;89:1374–1384. doi: 10.1097/OPX.0b013e3182678e52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Schuchard RA. Validity and interpretation of Amsler grid reports. Arch Ophthalmol. 1993;111:776–780. doi: 10.1001/archopht.1993.01090060064024. [DOI] [PubMed] [Google Scholar]
  • 29.Schuchard RA. Preferred retinal loci and macular scotoma characteristics in patients with age-related macular degeneration. Can J of Ophthalmology. 2005;40:303–312. doi: 10.1016/S0008-4182(05)80073-0. [DOI] [PubMed] [Google Scholar]
  • 30.Fletcher DC, Schuchard RA, Renninger LW. Patient awareness of binocular central scotoma in age-related macular degeneration. Optom Vis Sci. 2012;89:1395–1398. doi: 10.1097/OPX.0b013e318264cc77. [DOI] [PubMed] [Google Scholar]
  • 31.Luo G, Satgunam P, Peli E. Visual search performance of patients with vision impairment: Effect of JPEG image enhancement. Ophtal Physl Opt. 2012;32:421–428. doi: 10.1111/j.1475-1313.2012.00908.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Brainard DH. The psychophysics toolbox. Spatial Vision. 1997;10:433–436. [PubMed] [Google Scholar]
  • 33.Pelli DG. The VideoToolbox software for visual psychophysics: Transforming numbers into movies. Spatial Vision. 1997;10:437–442. [PubMed] [Google Scholar]
  • 34.Aguilar C, Castet E. Gaze-contingent simulation of retinopathy: Some potential pitfalls and remedies. Vision Res. 2011;51:997–1012. doi: 10.1016/j.visres.2011.02.010. [DOI] [PubMed] [Google Scholar]
  • 35.Fine EM, Rubin GS. Effects of cataract and scotoma on visual acuity: A simulation study. Optometry Vision Sci. 1999;76:468–473. doi: 10.1097/00006324-199907000-00022. [DOI] [PubMed] [Google Scholar]
  • 36.Bernard JB, Scherlen AC, Castet E. Page mode reading with simulated scotomas: A modest effect of interline spacing on reading speed. Vision Res. 2007;47:3447–3459. doi: 10.1016/j.visres.2007.10.005. [DOI] [PubMed] [Google Scholar]
  • 37.Chung STL. Enhancing reading speed in people with central vision loss through perceptual learning. Invest Ophthalmol Vis Sci. 2011;52:1164–1170. doi: 10.1167/iovs.10-6034. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Cherici C, Kuang X, Poletti M, Rucci M. Precision of sustained fixation in trained and untrained observers. Journal of Vis. 2012;31:1–16. doi: 10.1167/12.6.31. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Fischer B, Biscaldi M, Otto P. Saccadic eye movements of dyslexic adults. Neuropsychologia. 1993;31:887–906. doi: 10.1016/0028-3932(93)90146-q. [DOI] [PubMed] [Google Scholar]
  • 40.Gitelman DR. ILAB: a program for postexperimental eye movement analysis. Behav Res Meth, Ins C. 2002;34:605–612. doi: 10.3758/bf03195488. [DOI] [PubMed] [Google Scholar]
  • 41.Engbert R, Mergenthaler K. Microsaccades are triggered by low retinal image slip. Proc Nat Acad Sci USA. 2006;103:7192–7197. doi: 10.1073/pnas.0509557103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Botev ZI, Grotowski JF, Kroese DP. Kernel density estimation via diffusion. Annals of Statistics. 2010;38:2916–2957. [Google Scholar]
  • 43.Crossland MD, Sims M, Galbraith RF, Rubin GS. Evaluation of a new quantitative technique to assess the number and extent of preferred retinal loci in macular disease. Vision Res. 2004;44:1537–1546. doi: 10.1016/j.visres.2004.01.006. [DOI] [PubMed] [Google Scholar]
  • 44.Steinman RM. Effect of target size, luminance, and color on monocular fixation. J Opt Soc Amer A. 1965;55:1158–1165. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01
NIHMS502321-supplement-01.doc (5.2MB, doc)

RESOURCES