Original Contribution
Eye Movements During Saccadic and Fixation Tasks in Patients With Homonymous Hemianopia Jens I. Reinhard, MD, Ingelene Damm, MD, Iliya V. Ivanov, PhD, Susanne Trauzettel-Klosinski, MD
Background: The aim of our study was to quantify ocular motor performance in patients with homonymous hemianopia and in healthy controls during saccadic and fixation tasks and to detect potential spontaneous adaptive mechanisms in the hemianopic patients. Methods: Eye movements were recorded in 33 hemianopic patients (15 right, 18 left; disease duration, 0.2–29 years) and 14 healthy subjects by scanning laser ophthalmoscope allowing determination of the absolute fovea position relative to the stimulus without calibration. Landing accuracy of saccades was determined for 5° saccades, indicated by the number of dysmetric saccades (DS), and fixation stability (FS) after landing. In addition, during continuous fixation of a central cross, FS, and distribution of fixational eye movements (FEMs) were measured. Size of macular sparing was determined using custom microperimetry software (stimulus grid, 0.5°). Results: Compared with controls, landing accuracy was decreased in hemianopic patients, indicated by significantly more DS (hypometric and hypermetric) to the blind side compared with the seeing side. The number of DS was greater in patients with macular sparing of ,4°. DS were not correlated with age and disease duration. FS after landing was lower after saccades to the blind side. Distribution of FEM during continuous fixation was asymmetrically shifted to the blind side, especially in cases of macular sparing of ,4°. Conclusions: Number of DS was not correlated with disease duration indicating insufficient spontaneous long-term adaptation. Increased number of DS and decreased FS after landing in patients with small or absent macular sparing stresses the importance of intact parafoveal vision. Asymmetric FEMs during
Vision Rehabilitation Research Unit, Center for Ophthalmology, University of Tuebingen, Germany. Partly supported by the Kerstan Foundation (to J. R.) and the Intramural Research Program AKF (Research Grant Number 296-0-0) of the Medical Faculty University of Tuebingen (to I. V. I). The authors report no conflicts of interest. Address correspondence to Iliya V. Ivanov, PhD, Center for Ophthalmology, Vision Rehabilitation Research Unit, University of Tuebingen, Schleichstr, 12, D-72076 T} ubingen, Germany; E-mail: [email protected]
354
continuous fixation indicate an advantageous adaptive mechanism to shift the visual field border towards the hemianopic side. Journal of Neuro-Ophthalmology 2014;34:354–361 doi: 10.1097/WNO.0000000000000146 © 2014 by North American Neuro-Ophthalmology Society
S
accadic eye movements are used to position the image of a visual target directly onto the fovea. Saccades are accurate within 0.5° of the fovea (1), and accuracy can be improved by motor learning (2). Saccade adaptation has been described in patients with unilateral sixth nerve palsy (3,4) and in studies of healthy subjects by advancing the target stepwise during saccades (5–7). For patients with homonymous hemianopia, adaptation of landing accuracy of saccades to a defined target position has been reported (8,9). However, these studies dealt with shortterm adaptation. Much less is known about long-term adaptation mechanisms, which seem to have different characteristics (10). Previous studies reported long-term saccadic adaptation after training of several weeks in healthy monkeys (10,11). In patients with hemianopia, long-term adaptation to more complex tasks such as visual search, persists after the end of training (12,13). However, potential adaptation regarding saccadic accuracy has not yet been investigated systematically. We focused on a purely saccadic task to the visual target to determine whether spontaneous adaptation can occur over time. If long-term saccadic adaptation were present in patients with hemianopia (without specific training), we would expect disease duration to be a factor on the eye movement pattern. We also examined fixation in hemianopic patients. In previous studies of smaller groups of hemianopic patients using the scanning laser ophthalmoscope (SLO), fixational eye movements (FEMs) were found to be asymmetric towards the blind hemifield (14,15). Reinhard et al: J Neuro-Ophthalmol 2014; 34: 354-361
Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.
Original Contribution METHODS Patients and Control Subjects We studied 33 patients, 15 with right, 18 with left homonymous visual field defects: 30 with hemianopias, 2 with quadrant defects, and 1 with paracentral scotomas. The etiology of the brain damage was 1) infarction of the posterior cerebral artery (12), 2) traumatic (6), 3) hemorrhage (4), 4) postoperative complications (7), 5) arteriovenous malformation (2), and 6) unknown (2). Median age was 47 years (±26.5); interquartile range (IQR) 21–80 years. Time since onset was documented for 22 patients (11 left and 11 right homonymous defects) and ranged from 0.2 to 29 years (median duration, 2.15 years ± 3.8 IQR). Additionally, 14 healthy control subjects were examined (median age, 30.5 ± 5.5 years; IQR range, 22–82 years). All participants gave informed consent in accordance with the tenets of the declaration of Helsinki and the approval of the Ethics Committee of the Medical Faculty of Tuebingen University.
Eye Movement Recording An SLO (model 101; Rodenstock, Munich, Germany) was used to track the eye movements during fixation and saccades. The SLO simultaneously displayed an image of the retina and the stimuli on a video screen showing the absolute position of the fovea (Fig. 1). This method does not require calibration, which is a great advantage, especially in patients with unstable and eccentric fixation. The spatial resolution is at least 12 arcmin, the temporal resolution is defined by the line frequency of 50 Hz, that is, 25 complete video frames per second. The SLO allows only monocular examinations. We performed them in both eyes, but because there were no functional differences between the eyes, the recordings of the right eye were analyzed.
Fixation Fixation stability (FS) after a saccade (i.e., after “landing”) and during continuous fixation of a single cross (36 arcmin diameter) in the center of the screen for 20 seconds (i.e., 1,000 video fields) was measured and quantified by a Fixation Stability Index (FSI) (16). This index does not depend on a normal distribution of the fixation data and quantifies the number of all new locations on the retina that are used for fixation. Theoretically, “perfect” fixation uses only 1 location and, thus, has a FSI of 100%. The number of different pixel coordinates that occurred during a fixation trial divided by the total number of pixel coordinates that were successfully tracked defines the FSI. The FSI is calculated as follows: FSI ¼ ð1 2 N =T Þ · 100%; where N represents number of different pixel coordinates and T is the number of tracked pixel coordinates. Fixation locus was assessed by the position of the fovea relative to the fixation cross, which can be directly seen in the SLO-image.
Saccades In the SLO image, 3 black crosses of 36 arcmin diameters with a distance of 5° between each other were displayed on a bright red background (Fig. 1). First, the patients were asked to fixate the cross in the middle. Subsequently, they were asked to fixate the right cross, middle, left cross, middle. This cycle was repeated at least 3 times.
Microperimetry To measure the size of a macular sparing, custom microperimetry software was used with a spatial grid of 0.5° under direct fixation control, which guarantees central fixation during stimulus presentation (15,17).
Data Analysis The SLO images were recorded on SVHS video tape and were later digitized as AVI files. These files were analyzed by custom software that tracks the retinal movements (i.e., the position of a vessel branching) 50 times a second. These data were converted into a Microsoft Excel file and displayed as an eye position curve over time. The number of hypometric and hypermetric saccades per gaze direction were counted and related to the side of the field defect. FS after landing: FEM from the left, the middle, and the right cross were separated, and their FSI was calculated. Additionally, FS during continuous fixation (20 seconds) of a fixation cross was recorded, and the FSI was calculated.
Ophthalmic Examination FIG. 1. Scanning laser ophthalmoscope image of the fundus together with the stimuli. The 3 crosses are 5° apart each. The image shows the absolute position of the fovea related to the stimulus, which allows direct fixation control without calibration. Reinhard et al: J Neuro-Ophthalmol 2014; 34: 354-361
All patients underwent a routine ophthalmological and neuro-ophthalmological examination. Standardized perimetry with a Tuebingen Automated Perimeter was performed to make certain that the only visual field abnormality was the hemianopia. 355
Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.
Original Contribution
FIG. 2. Eye movements during 5° gaze shifts: 5° saccades between 3 fixation crosses (left cross 25°, central cross 0°, and right cross +5°). A. control subject, (B) patient with left hemianopia. The control subject shows accurate landing on the crosses (with 2 slightly hypermetric saccades to the right) and relatively stable fixation after landing. B. The hemianopic subject shows inaccurate landing to the left: first gaze shift—4 hypometric and 1 hypermetric saccades, with slightly decreasing number of DS during the following gaze shifts. Fixation after landing is unstable.
Statistical Methods
5° Saccades
A language and environment for statistical computing (R, version 2.14.1; R Foundation for Statistical Computing, Vienna, Austria) was used for statistical analysis. Univariate statistical methods were used because the number of participants was too small for multiple regression analysis. Welch 2 sample t test, F Test, Wilcoxon signed rank test, and descriptive statistics were used to assess accuracy of landing and FS after landing and during continuous fixation. For quantitative variables, we applied a confidence interval of 95%. For additional analyses, data were extracted and analyzed by descriptive statistics.
Accuracy of Landing Figure 2A shows the eye movement recording in a control subject performing 5° saccades. Landing is accurate on each cross indicated by only 2 hypermetric saccades to the right, and fixation after landing is stable. Figure 2B displays the horizontal eye movements of a patient with left homonymous hemianopia with an increased number of steps during gaze shift to the left. During the first gaze shift to the left, there are 4 hypometric and 1 hypermetric saccades. During the following gaze shifts, number of DS decreases. Fixation after landing is unstable. The mean values of DS for the patient groups were separated into right and left hemianopia, depending on the direction of the gaze shift and the side of the field defect. Results show that the number of saccades is increased in the direction of the field defect (Fig. 3A–D). The mean number of hypometric saccades during gaze shift to the right is 1.2 (±0.2 SEM) for patients with right heminaopia and 0.7 (±0.1 SEM) for those with left hemianopia (Fig. 3A). Correspondingly, the mean number of hypometric saccades during gaze shift to the left is much higher in patients with left hemianopia (1.6 ± 0.2 SEM) compared with those with right heminaopia (1.0 ± 0.2 SEM) (Fig. 3B). The mean number of hypermetric saccades during gaze shift to the
RESULTS Size of Macular Sparing Twenty-five patients (13 with left, 12 with right homonymous hemianopia) had either no macular sparing or of a very small size (,4°). The other 8 patients had macular sparing of $4°, of those only 4 had a sparing of $5° (3 had 4°, 1 had 4.5°, 1 had 5°, 2 had 8°, 1 had 10°). For comparison of patient data with small or absent sparing with those with larger sparing, we used as separation ,4° vs $4° to obtain favorable group size for statistical analysis. 356
Reinhard et al: J Neuro-Ophthalmol 2014; 34: 354-361
Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.
Original Contribution
FIG. 3. Mean number of dysmetric saccades (DS) per gaze shift. A. Hypometric saccades during gaze shift to the right: higher number for right hemianopia. B. Hypometric saccades during gaze shift to the left: higher number for left hemianopia. C. Hypermetric saccades during gaze shift to the right: nearly 2.5 times higher for right hemianopia. D. Hypermetric saccades during gaze shift to the left: nearly 2.5 times higher for left hemianopia. E. Mean number of all DS (per gaze shift) during gaze shift to the blind and the seeing hemifield in all patients and in control subjects. Gaze shift to the blind hemifield leads to an increase of hypermetric and, more pronounced, hypometric saccades. The number of DS to the blind side is markedly higher compared with the seeing side and to control subjects. The values for the seeing hemifield show no difference compared with control subjects. F. Mean number of DS during gaze shift to the blind side for patients with macular sparing $4° vs ,4°. Patients with smaller sparing had more DS. DS, dysmetric saccades.
right is nearly 2.5 times higher in right hemianopia (1.2 ± 0.2 SEM) compared with left hemianopia (0.5 ± 0.3 SEM) (Fig. 3C). Accordingly, the mean number of hypermetric saccades during gaze shift to the left is approximately 2.5 times higher in left hemianopia (1.2 ± 0.2 SEM) compared with right hemianopia (0.5 ± 0.1 SEM) (Fig. 3D). The analysis of all DS considering the gaze shift to either the blind or seeing hemifield for all patients together, as well as for healthy subjects, is displayed in Figure 3E. The mean number Reinhard et al: J Neuro-Ophthalmol 2014; 34: 354-361
of hypermetric saccades is much higher during gaze shifts to the blind hemifield (1.2 ± 0.2 SEM) than to the seeing hemifield (0.5 ± 0.2 SEM) (P , 0.003) and compared with control subjects (0.26 ± 0.08 SEM) (P , 0.0001). Also, the mean number of hypometric saccades is much higher during gaze shifts to the blind hemifield (1.4 ± 0.2 SEM) than to the seeing hemifield (0.5 ± 0.1 SEM) (P = 0.001). The mean numbers of hypermetric and hypometric saccades for the seeing hemifield show no difference compared with healthy 357
Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.
Original Contribution
FIG. 4. Landing accuracy during 5° saccades: Dots show fovea positions in the scanning laser ophthalmoscope video sequence. A. The control subject shows dense clouds of landing positions at 25°, 0°, and +5°. B. The patient with right homonymous hemianopia shows more scattered clouds for the central and right cross indicating inaccurate landing during gaze shift to the right. C. The patient with left homonymous hemianopia develops very scattered clouds for the left and central cross, indicating inaccurate landing when the gaze shift is directed to the blind hemifield.
subjects (P = 0.1 and P = 0.3, respectively). These data show that accuracy of saccades during gaze shift to the blind hemifield is much lower compared with those to the seeing hemifield and compared with control subjects. In patients with small or absent macular sparing (,4°), the number of DS is higher compared with those with larger sparing (Fig. 3F). If the number of DS to the blind side was compared between patients with a small or no sparing (,4°, n = 25) and patients with a sparing of $4° (n = 8), we found a slight difference for hypometric saccades (1.5 ± 0.2 SEM vs 1.2 ± 0.3 SEM). This latter difference did not reach statistical significance, because the group with larger sparing consisted of only 8 patients. For hypermetric saccades, there was a statistically significant difference (1.4 ± 0.2 SEM vs 0.5 ± 0.2 SEM; P = 0.008). The landing accuracy of saccades is also shown for 3 examples of subjects in an X/Y plot of the horizontal and vertical eye movements. Figure 4A shows the plot for a control subject with dense clouds of dots at the landing positions of the 3 crosses. Figure 4B shows more scattered clouds of landing positions in a patient with right hemianopia and shows that scattering is much higher when gaze shifted to the right. Figure 4C demonstrates very scattered clouds of landing positions in a patient with left hemianopia, with the gaze shift directed to the left. 358
Correlations The number of DS did not show any correlations with age and duration of disease.
Fixation Stability After Landing Fixation after landing in left and right hemianopia patients was less stable on the blind side compared with the seeing side with the mean FSI of 67.5% vs 74.5%, respectively. Mean FSI after gaze shifts to the seeing side was significantly higher than the FSI after gaze shifts to the nonseeing side (P = 0.013). Variance of the FSI to the seeing side was also less than to the nonseeing side (F[1,32] = 2.55, P = 0.01). When the size of macular sparing was also considered, this effect of the side of the field defect was only present in patients with a sparing of ,4° (P = 0.0167), whereas there was no difference between seeing and nonseeing side if the sparing was $4°.
Continuous Fixation of a Single Cross Fixation Stability FEMs were asymmetric towards the blind side (a positive mean value corresponds to a shift of the distribution to the right, a negative mean value to a shift to the left): In right Reinhard et al: J Neuro-Ophthalmol 2014; 34: 354-361
Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.
Original Contribution
FIG. 5. Distribution of fixational eye movements (FEMs) during continuous fixation. Asymmetric distribution of horizontal FEMs during continuous fixation of a single cross (20 seconds) in 25 patients with macular sparing of ,4°. A positive mean value corresponds to a shift of the distribution to the right, a negative mean value corresponds to a shift to the left. In right homonymous hemianopia (A), the distribution is centered at +2.6 (mean), in left hemianopia (B), the distribution is centered at 21.3° (mean). The distributions are significantly different from a normal distribution centered at 0° (P , 0.0001). On the right side, individual examples of asymmetric FEMs towards the blind side are shown for a patient with right (A) and left (B) homonymous hemianopia.
hemianopia, fixations are described by a distribution centered at 0.51° (mean). In left hemianopia, the distribution of the fixations is centered at 21.03° (mean). The Wilcoxon signed rank test showed that both distributions are significantly different (P , 0.0001) from a normal distribution centered at 0° (mean), which would suggest no asymmetric FEM. If we include only those patients with a macular sparing of ,4° (n = 25, where 13 had left and 12 had right hemianopia), the asymmetry of FEM becomes more pronounced. In patients with left homonymous hemianopia, the mean is shifted to 21.312° (left) and Reinhard et al: J Neuro-Ophthalmol 2014; 34: 354-361
the distribution is not normal (P , 0.0001). In patients with right hemianopia, the mean is shifted to 2.6° (right), and the distribution is also not normal (P , 0.0001). The asymmetric distribution of the FEM is shown in Figure 5 (patients with sparing of ,4°) on the right side an individual example each.
Fixation Locus Fixation was eccentric in 6 patients (18%) with an amount of 0.5–1.5°, in 2 of them only occasionally. 359
Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.
Original Contribution DISCUSSION Eye Movements During Goal-Directed Gaze Shifts Our data show reduced landing accuracy for gaze shifts to the blind side compared with the seeing side, indicated by an increased number of DS (Fig. 3E). The higher number of DS in patients with small or absent macular sparing (,4°) compared with those with a larger sparing ($4°) indicates the importance of parafoveal vision for accurate saccade landing. In addition to the perceptual information, this parafoveal seeing area favors attentional mechanisms that are important for saccade preparation (18). The influence of the size of sparing has not been examined in previous studies. The number of DS did not show any correlations with age and duration of disease. With increasing disease duration, we would expect a decrease of hypometric saccades as an adaptive strategy. Previous studies reported 3 stages of saccadic patterns in patients with hemianopia (8,9): first a staircase pattern (hypometric saccades), then an overshoot (hypermetric), and finally predictive saccades. However, this represented short-term adaptation of landing accuracy during repeated gaze shifts to the target. Because we did not find a correlation with disease duration, we conclude that long-term adaptation with more effective strategies does not or only insufficiently occur. The increased number of DS to the blind side indicates early adaptive mechanisms: Hypometric saccades represent a strategy to approach the target in small steps, which is safe, but slow. Hypermetric saccades can make finding the target easier. However, a more effective adaptation does not occur spontaneously. For more complex tasks, such as visual search, spontaneous adaptive strategies were rarely observed [1 of 9 patients over 18 months (19)]. Previous studies involving training of saccades (20,21) and of search tasks (12,13,22) showed improvement of reaction times and scanpaths (23). However, those studies did not examine whether the accuracy of saccade landings also improved. Additionally, FS after landing was decreased during gaze shift to the blind side, but only in patients with small or absent macular sparing (,4°). This shows the importance of a small seeing field around the fovea. The vertical splitting of the field is associated with normal visual acuity. Therefore, it is not only the fixation locus (foveal), but also perception in the surrounding parafoveal area, which contributes to FS. Our study demonstrates that for macular splitting, where foveal vision is intact, it is the size of the parafoveal visual field that contributes to stable fixation. It has been reported that patients with different degrees of visual impairment had reduced FS in acquired foveal visual loss and that eye movements are optimized by visual feedback (24,25).
FS During Continuous Fixation of a Cross FEMs are a well-known physiological phenomenon in healthy subjects to prevent fading and maintain continuous 360
vision during fixation (26–28). A recent study (29) in normal observers reports that microsaccades alter perception. In our study, we showed that lack of perception (field defect) affects fixational eye movements. The directions of FEM during continuous fixation of a cross were not normally distributed but showed significant asymmetry towards the blind side. This effect was more pronounced in patients with small or absent macular sparing (,4°, Fig. 5). This asymmetry indicates an advantageous spontaneous adaptive mechanism that shifts the visual field border towards the hemianopic side (16), which can be misinterpreted as improvement of the visual field (30–32). We demonstrated that FEMs were asymmetric towards the hemianopic side indicated by the abnormal distribution with the mean shifted to the right in patients with right hemianopia and to the left in patients with left hemianopia (Fig. 5). The fixation locus in this study was slightly (0.5–1.5°) eccentric in 18% of the patients, less frequent than in our previous study (33). Eccentric fixation shifts the vertical field border towards the hemianopic side and creates a small perceptual area along the vertical midline. This is an adaptive strategy for enlarging the reading visual field (15,31,33). In summary, this study showed: 1) intact parafoveal vision is important for accuracy of saccade landing and stability of fixation, 2) there was no spontaneous long-term adaptation regarding DS, and 3) an asymmetry of the directions of fixational eye movements towards the hemianopic side.
ACKNOWLEDGMENTS The authors thank Manfred Mackeben, PhD, The Smith Kettlewell Eye Research Institute, San Francisco, CA, for critical comments and editorial help.
REFERENCES 1. Leigh RJ, Zee DS, eds. The Neurology of Eye Movements, 4th edition. New York, NY: Oxford University Press, 2006. 2. Iwamoto Y, Kaku Y. Saccade adaptation as a model of learning in voluntary movements. Exp Brain Res. 2010;204:145–162. 3. Kommerell G, Olivier D, Theopold H. Adaptive programming of phasic and tonic components in saccadic eye movements. Investigations of patients with abducens palsy. Invest Ophthalmol. 1976;15:657–660. 4. Abel LA, Schmidt D, Dell’Osso LF, Daroff RB. Saccadic system plasticity in humans. Ann Neurol. 1978;4:313–318. 5. McLaughlin SC. Parametric adjustment in saccadic eye movements. Percept Psychophys. 1967;2:359–362. 6. Frens MA, van Opstal AJ. Transfer of short-term adaptation in human saccadic eye movements. Exp Brain Res. 1994;100:293–306. 7. Panouillères M, Urquizar C, Salemme R, Pélisson D. Sensory processing of motor inaccuracy depends on previously performed movement and on subsequent motor corrections: a study of the saccadic system. PLoS One. 2011;6:e17329. 8. Meienberg O, Zangemeister WH, Rosenberg M, Hoyt WF, Stark L. Saccadic eye movement strategies in patients with homonymous hemianopia. Ann Neurol. 1981;9:537–544. 9. Meienberg O. Augenbewegungs-muster bei homonymer hemianopsie und visuellem hemineglekt. Okulographische abgrenzungs-kriterien anhand von 19 fällen und diagnostische bedeutung [Article in German]. Klin Monbl Augenheilkd. 1988;192:108–112. Reinhard et al: J Neuro-Ophthalmol 2014; 34: 354-361
Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.
Original Contribution 10. Robinson FR, Soetedjo R, Noto C. Distinct short-term and longterm adaptation to reduce saccade size in monkey. J Neurophysiol. 2006;96:1030–1041. 11. Kuki Y, Hirata Y, Blazquez PM, Heiney SA, Highstein SM. Memory retention of vestibuloocular reflex motor learning in squirrel monkeys. Neuroreport. 2004;15:1007–1011. 12. Pambakian AL, Mannan SK, Hodgson TL, Kennard C. Saccadic visual search training: a treatment for patients with homonymous hemianopia. J Neurol Neurosurg Psychiatry. 2004;75:1443–1448. 13. Roth T, Sokolov AN, Messias A, Roth P, Weller M, TrauzettelKlosinski S. Comparing explorative saccade and flicker training in hemianopia: a randomized controlled study. Neurology. 2009;72:324–331. 14. Bischoff P, Lang J, Huber A. Macular sparing as a perimetric artifact. Am J Ophthalmol. 1995;119:72–80. 15. Trauzettel-Klosinski S, Reinhard J. The vertical field border in hemianopia and its significance for fixation and reading. Invest Ophthalmol Vis Sci. 1998;39:2177–2186. 16. Reinhard J, Messias A, Dietz K, Mackeben M, Lakmann R, Scholl HP, Apfelstedt-Sylla E, Weber BH, Seeliger MW, Zrenner E, Trauzettel-Klosinski S. Quantifying fixation in patients with Stargardt disease. Vision Res. 2007;47:2076–2085. 17. Reinhard J, Trauzettel-Klosinski S. Nasotemporal overlap of retinal ganglion cells in humans: a functional study. Invest Ophthalmol Vis Sci. 2003;44:1568–1572. 18. Deubel H, Schneider WX. Saccade target selection and object recognition: evidence for a common attentional mechanism. Vision Res. 1996;36:1827–1837. 19. Machner B, Sprenger A, Sander T, Heide W, Kimmig H, Helmchen C, Kömpf D. Visual search disorders in acute and chronic homonymous hemianopia: lesion effects and adaptive strategies. Ann N Y Acad Sci. 2009;1164:419–426. 20. Kerkhoff G, Munssinger U, Haaf E, Eberle-Strauss G, Stogerer E. Rehabilitation of homonymous scotomata in patients with postgeniculate damage of the visual system: saccadic compensation training. Restor Neurol Neurosci. 1992;4:245–254. 21. Zihl J. Visual scanning behavior in patients with homonymous hemianopia. Neuropsychologia. 1995;33:287–303.
Reinhard et al: J Neuro-Ophthalmol 2014; 34: 354-361
22. Mannan SK, Pambakian ALM, Kennard C. Compensatory strategies following visualsearch training in patients with homonymous hemianopia: an eye movement study. J Neurol. 2010;257:1812–1821. 23. Trauzettel-Klosinski S. Current methods of visual rehabilitation. Dtsch Arztebl Int. 2011;108:871–878. 24. Schneider RM, Thurtell MJ, Eisele S, Lincoff N, Bala E, Leigh RJ. Neurological basis for eye movements of the blind. PLoS One. 2013;8:e56556. 25. Leigh RJ, Thurston SE, Tomsak RL, Grossman GE, Lanska DJ. Effect of monocular visual loss upon stability of gaze. Invest Ophthalmol Vis Sci. 1989;30:288–292. 26. Troxler D. Über das Verschwinden gegebener Gegenstände innerhalb unseres Gesichtskreises. In: Himly K, Schmidt JA. Ophthalmologische Bibliothek (German). 1804;2:1–56. 27. Martinez-Conde S, Macknik SL, Troncoso XG, Dyar TA. Microsaccades counteract visual fading during fixation. Neuron. 2006;49:297–305. 28. McCamy MB, Otero-Millan J, Macknik SL, Yang Y, Troncoso XG, Baer SM, Crook SM, Martinez-Conde S. Microsaccadic efficacy and contribution to foveal and peripheral vision. J Neurosci. 2012;32:9194–9204. 29. Hafed ZM. Alteration of visual perception prior to microsaccades. Neuron. 2013;77:775–786. 30. Kasten E, Wüst S, Behrens-Baumann W, Sabel BA. Computerbased training for the treatment of partial blindness. Nat Med. 1998;4:1083–1087. 31. Reinhard J, Schreiber A, Schiefer U, Kasten E, Sabel BA, Kenkel S, Vonthein R, Trauzettel-Klosinski S. Does visual restitution training change absolute homonymous visual field defects? A fundus controlled study. Br J Ophthalmol. 2005;89:30–35. 32. Kasten E, Bunzenthal U, Sabel BA. Visual field recovery after vision restoration therapy (VRT) is independent of eye movements: an eye tracker study. Behav Brain Res. 2006;175:18–26. 33. Trauzettel-Klosinski S. Eccentric fixation in hemianopic field defects: a valuable strategy to improve reading ability and an indication for cortical plasticity. Ophthalmol. 1997;18:117–131.
361
Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.
Eye Movements During Saccadic and Fixation Tasks in Patients With Homonymous Hemianopia Jens I. Reinhard, MD, Ingelene Damm, MD, Iliya V. Ivanov, PhD, Susanne Trauzettel-Klosinski, MD
Background: The aim of our study was to quantify ocular motor performance in patients with homonymous hemianopia and in healthy controls during saccadic and fixation tasks and to detect potential spontaneous adaptive mechanisms in the hemianopic patients. Methods: Eye movements were recorded in 33 hemianopic patients (15 right, 18 left; disease duration, 0.2–29 years) and 14 healthy subjects by scanning laser ophthalmoscope allowing determination of the absolute fovea position relative to the stimulus without calibration. Landing accuracy of saccades was determined for 5° saccades, indicated by the number of dysmetric saccades (DS), and fixation stability (FS) after landing. In addition, during continuous fixation of a central cross, FS, and distribution of fixational eye movements (FEMs) were measured. Size of macular sparing was determined using custom microperimetry software (stimulus grid, 0.5°). Results: Compared with controls, landing accuracy was decreased in hemianopic patients, indicated by significantly more DS (hypometric and hypermetric) to the blind side compared with the seeing side. The number of DS was greater in patients with macular sparing of ,4°. DS were not correlated with age and disease duration. FS after landing was lower after saccades to the blind side. Distribution of FEM during continuous fixation was asymmetrically shifted to the blind side, especially in cases of macular sparing of ,4°. Conclusions: Number of DS was not correlated with disease duration indicating insufficient spontaneous long-term adaptation. Increased number of DS and decreased FS after landing in patients with small or absent macular sparing stresses the importance of intact parafoveal vision. Asymmetric FEMs during
Vision Rehabilitation Research Unit, Center for Ophthalmology, University of Tuebingen, Germany. Partly supported by the Kerstan Foundation (to J. R.) and the Intramural Research Program AKF (Research Grant Number 296-0-0) of the Medical Faculty University of Tuebingen (to I. V. I). The authors report no conflicts of interest. Address correspondence to Iliya V. Ivanov, PhD, Center for Ophthalmology, Vision Rehabilitation Research Unit, University of Tuebingen, Schleichstr, 12, D-72076 T} ubingen, Germany; E-mail: [email protected]
354
continuous fixation indicate an advantageous adaptive mechanism to shift the visual field border towards the hemianopic side. Journal of Neuro-Ophthalmology 2014;34:354–361 doi: 10.1097/WNO.0000000000000146 © 2014 by North American Neuro-Ophthalmology Society
S
accadic eye movements are used to position the image of a visual target directly onto the fovea. Saccades are accurate within 0.5° of the fovea (1), and accuracy can be improved by motor learning (2). Saccade adaptation has been described in patients with unilateral sixth nerve palsy (3,4) and in studies of healthy subjects by advancing the target stepwise during saccades (5–7). For patients with homonymous hemianopia, adaptation of landing accuracy of saccades to a defined target position has been reported (8,9). However, these studies dealt with shortterm adaptation. Much less is known about long-term adaptation mechanisms, which seem to have different characteristics (10). Previous studies reported long-term saccadic adaptation after training of several weeks in healthy monkeys (10,11). In patients with hemianopia, long-term adaptation to more complex tasks such as visual search, persists after the end of training (12,13). However, potential adaptation regarding saccadic accuracy has not yet been investigated systematically. We focused on a purely saccadic task to the visual target to determine whether spontaneous adaptation can occur over time. If long-term saccadic adaptation were present in patients with hemianopia (without specific training), we would expect disease duration to be a factor on the eye movement pattern. We also examined fixation in hemianopic patients. In previous studies of smaller groups of hemianopic patients using the scanning laser ophthalmoscope (SLO), fixational eye movements (FEMs) were found to be asymmetric towards the blind hemifield (14,15). Reinhard et al: J Neuro-Ophthalmol 2014; 34: 354-361
Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.
Original Contribution METHODS Patients and Control Subjects We studied 33 patients, 15 with right, 18 with left homonymous visual field defects: 30 with hemianopias, 2 with quadrant defects, and 1 with paracentral scotomas. The etiology of the brain damage was 1) infarction of the posterior cerebral artery (12), 2) traumatic (6), 3) hemorrhage (4), 4) postoperative complications (7), 5) arteriovenous malformation (2), and 6) unknown (2). Median age was 47 years (±26.5); interquartile range (IQR) 21–80 years. Time since onset was documented for 22 patients (11 left and 11 right homonymous defects) and ranged from 0.2 to 29 years (median duration, 2.15 years ± 3.8 IQR). Additionally, 14 healthy control subjects were examined (median age, 30.5 ± 5.5 years; IQR range, 22–82 years). All participants gave informed consent in accordance with the tenets of the declaration of Helsinki and the approval of the Ethics Committee of the Medical Faculty of Tuebingen University.
Eye Movement Recording An SLO (model 101; Rodenstock, Munich, Germany) was used to track the eye movements during fixation and saccades. The SLO simultaneously displayed an image of the retina and the stimuli on a video screen showing the absolute position of the fovea (Fig. 1). This method does not require calibration, which is a great advantage, especially in patients with unstable and eccentric fixation. The spatial resolution is at least 12 arcmin, the temporal resolution is defined by the line frequency of 50 Hz, that is, 25 complete video frames per second. The SLO allows only monocular examinations. We performed them in both eyes, but because there were no functional differences between the eyes, the recordings of the right eye were analyzed.
Fixation Fixation stability (FS) after a saccade (i.e., after “landing”) and during continuous fixation of a single cross (36 arcmin diameter) in the center of the screen for 20 seconds (i.e., 1,000 video fields) was measured and quantified by a Fixation Stability Index (FSI) (16). This index does not depend on a normal distribution of the fixation data and quantifies the number of all new locations on the retina that are used for fixation. Theoretically, “perfect” fixation uses only 1 location and, thus, has a FSI of 100%. The number of different pixel coordinates that occurred during a fixation trial divided by the total number of pixel coordinates that were successfully tracked defines the FSI. The FSI is calculated as follows: FSI ¼ ð1 2 N =T Þ · 100%; where N represents number of different pixel coordinates and T is the number of tracked pixel coordinates. Fixation locus was assessed by the position of the fovea relative to the fixation cross, which can be directly seen in the SLO-image.
Saccades In the SLO image, 3 black crosses of 36 arcmin diameters with a distance of 5° between each other were displayed on a bright red background (Fig. 1). First, the patients were asked to fixate the cross in the middle. Subsequently, they were asked to fixate the right cross, middle, left cross, middle. This cycle was repeated at least 3 times.
Microperimetry To measure the size of a macular sparing, custom microperimetry software was used with a spatial grid of 0.5° under direct fixation control, which guarantees central fixation during stimulus presentation (15,17).
Data Analysis The SLO images were recorded on SVHS video tape and were later digitized as AVI files. These files were analyzed by custom software that tracks the retinal movements (i.e., the position of a vessel branching) 50 times a second. These data were converted into a Microsoft Excel file and displayed as an eye position curve over time. The number of hypometric and hypermetric saccades per gaze direction were counted and related to the side of the field defect. FS after landing: FEM from the left, the middle, and the right cross were separated, and their FSI was calculated. Additionally, FS during continuous fixation (20 seconds) of a fixation cross was recorded, and the FSI was calculated.
Ophthalmic Examination FIG. 1. Scanning laser ophthalmoscope image of the fundus together with the stimuli. The 3 crosses are 5° apart each. The image shows the absolute position of the fovea related to the stimulus, which allows direct fixation control without calibration. Reinhard et al: J Neuro-Ophthalmol 2014; 34: 354-361
All patients underwent a routine ophthalmological and neuro-ophthalmological examination. Standardized perimetry with a Tuebingen Automated Perimeter was performed to make certain that the only visual field abnormality was the hemianopia. 355
Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.
Original Contribution
FIG. 2. Eye movements during 5° gaze shifts: 5° saccades between 3 fixation crosses (left cross 25°, central cross 0°, and right cross +5°). A. control subject, (B) patient with left hemianopia. The control subject shows accurate landing on the crosses (with 2 slightly hypermetric saccades to the right) and relatively stable fixation after landing. B. The hemianopic subject shows inaccurate landing to the left: first gaze shift—4 hypometric and 1 hypermetric saccades, with slightly decreasing number of DS during the following gaze shifts. Fixation after landing is unstable.
Statistical Methods
5° Saccades
A language and environment for statistical computing (R, version 2.14.1; R Foundation for Statistical Computing, Vienna, Austria) was used for statistical analysis. Univariate statistical methods were used because the number of participants was too small for multiple regression analysis. Welch 2 sample t test, F Test, Wilcoxon signed rank test, and descriptive statistics were used to assess accuracy of landing and FS after landing and during continuous fixation. For quantitative variables, we applied a confidence interval of 95%. For additional analyses, data were extracted and analyzed by descriptive statistics.
Accuracy of Landing Figure 2A shows the eye movement recording in a control subject performing 5° saccades. Landing is accurate on each cross indicated by only 2 hypermetric saccades to the right, and fixation after landing is stable. Figure 2B displays the horizontal eye movements of a patient with left homonymous hemianopia with an increased number of steps during gaze shift to the left. During the first gaze shift to the left, there are 4 hypometric and 1 hypermetric saccades. During the following gaze shifts, number of DS decreases. Fixation after landing is unstable. The mean values of DS for the patient groups were separated into right and left hemianopia, depending on the direction of the gaze shift and the side of the field defect. Results show that the number of saccades is increased in the direction of the field defect (Fig. 3A–D). The mean number of hypometric saccades during gaze shift to the right is 1.2 (±0.2 SEM) for patients with right heminaopia and 0.7 (±0.1 SEM) for those with left hemianopia (Fig. 3A). Correspondingly, the mean number of hypometric saccades during gaze shift to the left is much higher in patients with left hemianopia (1.6 ± 0.2 SEM) compared with those with right heminaopia (1.0 ± 0.2 SEM) (Fig. 3B). The mean number of hypermetric saccades during gaze shift to the
RESULTS Size of Macular Sparing Twenty-five patients (13 with left, 12 with right homonymous hemianopia) had either no macular sparing or of a very small size (,4°). The other 8 patients had macular sparing of $4°, of those only 4 had a sparing of $5° (3 had 4°, 1 had 4.5°, 1 had 5°, 2 had 8°, 1 had 10°). For comparison of patient data with small or absent sparing with those with larger sparing, we used as separation ,4° vs $4° to obtain favorable group size for statistical analysis. 356
Reinhard et al: J Neuro-Ophthalmol 2014; 34: 354-361
Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.
Original Contribution
FIG. 3. Mean number of dysmetric saccades (DS) per gaze shift. A. Hypometric saccades during gaze shift to the right: higher number for right hemianopia. B. Hypometric saccades during gaze shift to the left: higher number for left hemianopia. C. Hypermetric saccades during gaze shift to the right: nearly 2.5 times higher for right hemianopia. D. Hypermetric saccades during gaze shift to the left: nearly 2.5 times higher for left hemianopia. E. Mean number of all DS (per gaze shift) during gaze shift to the blind and the seeing hemifield in all patients and in control subjects. Gaze shift to the blind hemifield leads to an increase of hypermetric and, more pronounced, hypometric saccades. The number of DS to the blind side is markedly higher compared with the seeing side and to control subjects. The values for the seeing hemifield show no difference compared with control subjects. F. Mean number of DS during gaze shift to the blind side for patients with macular sparing $4° vs ,4°. Patients with smaller sparing had more DS. DS, dysmetric saccades.
right is nearly 2.5 times higher in right hemianopia (1.2 ± 0.2 SEM) compared with left hemianopia (0.5 ± 0.3 SEM) (Fig. 3C). Accordingly, the mean number of hypermetric saccades during gaze shift to the left is approximately 2.5 times higher in left hemianopia (1.2 ± 0.2 SEM) compared with right hemianopia (0.5 ± 0.1 SEM) (Fig. 3D). The analysis of all DS considering the gaze shift to either the blind or seeing hemifield for all patients together, as well as for healthy subjects, is displayed in Figure 3E. The mean number Reinhard et al: J Neuro-Ophthalmol 2014; 34: 354-361
of hypermetric saccades is much higher during gaze shifts to the blind hemifield (1.2 ± 0.2 SEM) than to the seeing hemifield (0.5 ± 0.2 SEM) (P , 0.003) and compared with control subjects (0.26 ± 0.08 SEM) (P , 0.0001). Also, the mean number of hypometric saccades is much higher during gaze shifts to the blind hemifield (1.4 ± 0.2 SEM) than to the seeing hemifield (0.5 ± 0.1 SEM) (P = 0.001). The mean numbers of hypermetric and hypometric saccades for the seeing hemifield show no difference compared with healthy 357
Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.
Original Contribution
FIG. 4. Landing accuracy during 5° saccades: Dots show fovea positions in the scanning laser ophthalmoscope video sequence. A. The control subject shows dense clouds of landing positions at 25°, 0°, and +5°. B. The patient with right homonymous hemianopia shows more scattered clouds for the central and right cross indicating inaccurate landing during gaze shift to the right. C. The patient with left homonymous hemianopia develops very scattered clouds for the left and central cross, indicating inaccurate landing when the gaze shift is directed to the blind hemifield.
subjects (P = 0.1 and P = 0.3, respectively). These data show that accuracy of saccades during gaze shift to the blind hemifield is much lower compared with those to the seeing hemifield and compared with control subjects. In patients with small or absent macular sparing (,4°), the number of DS is higher compared with those with larger sparing (Fig. 3F). If the number of DS to the blind side was compared between patients with a small or no sparing (,4°, n = 25) and patients with a sparing of $4° (n = 8), we found a slight difference for hypometric saccades (1.5 ± 0.2 SEM vs 1.2 ± 0.3 SEM). This latter difference did not reach statistical significance, because the group with larger sparing consisted of only 8 patients. For hypermetric saccades, there was a statistically significant difference (1.4 ± 0.2 SEM vs 0.5 ± 0.2 SEM; P = 0.008). The landing accuracy of saccades is also shown for 3 examples of subjects in an X/Y plot of the horizontal and vertical eye movements. Figure 4A shows the plot for a control subject with dense clouds of dots at the landing positions of the 3 crosses. Figure 4B shows more scattered clouds of landing positions in a patient with right hemianopia and shows that scattering is much higher when gaze shifted to the right. Figure 4C demonstrates very scattered clouds of landing positions in a patient with left hemianopia, with the gaze shift directed to the left. 358
Correlations The number of DS did not show any correlations with age and duration of disease.
Fixation Stability After Landing Fixation after landing in left and right hemianopia patients was less stable on the blind side compared with the seeing side with the mean FSI of 67.5% vs 74.5%, respectively. Mean FSI after gaze shifts to the seeing side was significantly higher than the FSI after gaze shifts to the nonseeing side (P = 0.013). Variance of the FSI to the seeing side was also less than to the nonseeing side (F[1,32] = 2.55, P = 0.01). When the size of macular sparing was also considered, this effect of the side of the field defect was only present in patients with a sparing of ,4° (P = 0.0167), whereas there was no difference between seeing and nonseeing side if the sparing was $4°.
Continuous Fixation of a Single Cross Fixation Stability FEMs were asymmetric towards the blind side (a positive mean value corresponds to a shift of the distribution to the right, a negative mean value to a shift to the left): In right Reinhard et al: J Neuro-Ophthalmol 2014; 34: 354-361
Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.
Original Contribution
FIG. 5. Distribution of fixational eye movements (FEMs) during continuous fixation. Asymmetric distribution of horizontal FEMs during continuous fixation of a single cross (20 seconds) in 25 patients with macular sparing of ,4°. A positive mean value corresponds to a shift of the distribution to the right, a negative mean value corresponds to a shift to the left. In right homonymous hemianopia (A), the distribution is centered at +2.6 (mean), in left hemianopia (B), the distribution is centered at 21.3° (mean). The distributions are significantly different from a normal distribution centered at 0° (P , 0.0001). On the right side, individual examples of asymmetric FEMs towards the blind side are shown for a patient with right (A) and left (B) homonymous hemianopia.
hemianopia, fixations are described by a distribution centered at 0.51° (mean). In left hemianopia, the distribution of the fixations is centered at 21.03° (mean). The Wilcoxon signed rank test showed that both distributions are significantly different (P , 0.0001) from a normal distribution centered at 0° (mean), which would suggest no asymmetric FEM. If we include only those patients with a macular sparing of ,4° (n = 25, where 13 had left and 12 had right hemianopia), the asymmetry of FEM becomes more pronounced. In patients with left homonymous hemianopia, the mean is shifted to 21.312° (left) and Reinhard et al: J Neuro-Ophthalmol 2014; 34: 354-361
the distribution is not normal (P , 0.0001). In patients with right hemianopia, the mean is shifted to 2.6° (right), and the distribution is also not normal (P , 0.0001). The asymmetric distribution of the FEM is shown in Figure 5 (patients with sparing of ,4°) on the right side an individual example each.
Fixation Locus Fixation was eccentric in 6 patients (18%) with an amount of 0.5–1.5°, in 2 of them only occasionally. 359
Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.
Original Contribution DISCUSSION Eye Movements During Goal-Directed Gaze Shifts Our data show reduced landing accuracy for gaze shifts to the blind side compared with the seeing side, indicated by an increased number of DS (Fig. 3E). The higher number of DS in patients with small or absent macular sparing (,4°) compared with those with a larger sparing ($4°) indicates the importance of parafoveal vision for accurate saccade landing. In addition to the perceptual information, this parafoveal seeing area favors attentional mechanisms that are important for saccade preparation (18). The influence of the size of sparing has not been examined in previous studies. The number of DS did not show any correlations with age and duration of disease. With increasing disease duration, we would expect a decrease of hypometric saccades as an adaptive strategy. Previous studies reported 3 stages of saccadic patterns in patients with hemianopia (8,9): first a staircase pattern (hypometric saccades), then an overshoot (hypermetric), and finally predictive saccades. However, this represented short-term adaptation of landing accuracy during repeated gaze shifts to the target. Because we did not find a correlation with disease duration, we conclude that long-term adaptation with more effective strategies does not or only insufficiently occur. The increased number of DS to the blind side indicates early adaptive mechanisms: Hypometric saccades represent a strategy to approach the target in small steps, which is safe, but slow. Hypermetric saccades can make finding the target easier. However, a more effective adaptation does not occur spontaneously. For more complex tasks, such as visual search, spontaneous adaptive strategies were rarely observed [1 of 9 patients over 18 months (19)]. Previous studies involving training of saccades (20,21) and of search tasks (12,13,22) showed improvement of reaction times and scanpaths (23). However, those studies did not examine whether the accuracy of saccade landings also improved. Additionally, FS after landing was decreased during gaze shift to the blind side, but only in patients with small or absent macular sparing (,4°). This shows the importance of a small seeing field around the fovea. The vertical splitting of the field is associated with normal visual acuity. Therefore, it is not only the fixation locus (foveal), but also perception in the surrounding parafoveal area, which contributes to FS. Our study demonstrates that for macular splitting, where foveal vision is intact, it is the size of the parafoveal visual field that contributes to stable fixation. It has been reported that patients with different degrees of visual impairment had reduced FS in acquired foveal visual loss and that eye movements are optimized by visual feedback (24,25).
FS During Continuous Fixation of a Cross FEMs are a well-known physiological phenomenon in healthy subjects to prevent fading and maintain continuous 360
vision during fixation (26–28). A recent study (29) in normal observers reports that microsaccades alter perception. In our study, we showed that lack of perception (field defect) affects fixational eye movements. The directions of FEM during continuous fixation of a cross were not normally distributed but showed significant asymmetry towards the blind side. This effect was more pronounced in patients with small or absent macular sparing (,4°, Fig. 5). This asymmetry indicates an advantageous spontaneous adaptive mechanism that shifts the visual field border towards the hemianopic side (16), which can be misinterpreted as improvement of the visual field (30–32). We demonstrated that FEMs were asymmetric towards the hemianopic side indicated by the abnormal distribution with the mean shifted to the right in patients with right hemianopia and to the left in patients with left hemianopia (Fig. 5). The fixation locus in this study was slightly (0.5–1.5°) eccentric in 18% of the patients, less frequent than in our previous study (33). Eccentric fixation shifts the vertical field border towards the hemianopic side and creates a small perceptual area along the vertical midline. This is an adaptive strategy for enlarging the reading visual field (15,31,33). In summary, this study showed: 1) intact parafoveal vision is important for accuracy of saccade landing and stability of fixation, 2) there was no spontaneous long-term adaptation regarding DS, and 3) an asymmetry of the directions of fixational eye movements towards the hemianopic side.
ACKNOWLEDGMENTS The authors thank Manfred Mackeben, PhD, The Smith Kettlewell Eye Research Institute, San Francisco, CA, for critical comments and editorial help.
REFERENCES 1. Leigh RJ, Zee DS, eds. The Neurology of Eye Movements, 4th edition. New York, NY: Oxford University Press, 2006. 2. Iwamoto Y, Kaku Y. Saccade adaptation as a model of learning in voluntary movements. Exp Brain Res. 2010;204:145–162. 3. Kommerell G, Olivier D, Theopold H. Adaptive programming of phasic and tonic components in saccadic eye movements. Investigations of patients with abducens palsy. Invest Ophthalmol. 1976;15:657–660. 4. Abel LA, Schmidt D, Dell’Osso LF, Daroff RB. Saccadic system plasticity in humans. Ann Neurol. 1978;4:313–318. 5. McLaughlin SC. Parametric adjustment in saccadic eye movements. Percept Psychophys. 1967;2:359–362. 6. Frens MA, van Opstal AJ. Transfer of short-term adaptation in human saccadic eye movements. Exp Brain Res. 1994;100:293–306. 7. Panouillères M, Urquizar C, Salemme R, Pélisson D. Sensory processing of motor inaccuracy depends on previously performed movement and on subsequent motor corrections: a study of the saccadic system. PLoS One. 2011;6:e17329. 8. Meienberg O, Zangemeister WH, Rosenberg M, Hoyt WF, Stark L. Saccadic eye movement strategies in patients with homonymous hemianopia. Ann Neurol. 1981;9:537–544. 9. Meienberg O. Augenbewegungs-muster bei homonymer hemianopsie und visuellem hemineglekt. Okulographische abgrenzungs-kriterien anhand von 19 fällen und diagnostische bedeutung [Article in German]. Klin Monbl Augenheilkd. 1988;192:108–112. Reinhard et al: J Neuro-Ophthalmol 2014; 34: 354-361
Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.
Original Contribution 10. Robinson FR, Soetedjo R, Noto C. Distinct short-term and longterm adaptation to reduce saccade size in monkey. J Neurophysiol. 2006;96:1030–1041. 11. Kuki Y, Hirata Y, Blazquez PM, Heiney SA, Highstein SM. Memory retention of vestibuloocular reflex motor learning in squirrel monkeys. Neuroreport. 2004;15:1007–1011. 12. Pambakian AL, Mannan SK, Hodgson TL, Kennard C. Saccadic visual search training: a treatment for patients with homonymous hemianopia. J Neurol Neurosurg Psychiatry. 2004;75:1443–1448. 13. Roth T, Sokolov AN, Messias A, Roth P, Weller M, TrauzettelKlosinski S. Comparing explorative saccade and flicker training in hemianopia: a randomized controlled study. Neurology. 2009;72:324–331. 14. Bischoff P, Lang J, Huber A. Macular sparing as a perimetric artifact. Am J Ophthalmol. 1995;119:72–80. 15. Trauzettel-Klosinski S, Reinhard J. The vertical field border in hemianopia and its significance for fixation and reading. Invest Ophthalmol Vis Sci. 1998;39:2177–2186. 16. Reinhard J, Messias A, Dietz K, Mackeben M, Lakmann R, Scholl HP, Apfelstedt-Sylla E, Weber BH, Seeliger MW, Zrenner E, Trauzettel-Klosinski S. Quantifying fixation in patients with Stargardt disease. Vision Res. 2007;47:2076–2085. 17. Reinhard J, Trauzettel-Klosinski S. Nasotemporal overlap of retinal ganglion cells in humans: a functional study. Invest Ophthalmol Vis Sci. 2003;44:1568–1572. 18. Deubel H, Schneider WX. Saccade target selection and object recognition: evidence for a common attentional mechanism. Vision Res. 1996;36:1827–1837. 19. Machner B, Sprenger A, Sander T, Heide W, Kimmig H, Helmchen C, Kömpf D. Visual search disorders in acute and chronic homonymous hemianopia: lesion effects and adaptive strategies. Ann N Y Acad Sci. 2009;1164:419–426. 20. Kerkhoff G, Munssinger U, Haaf E, Eberle-Strauss G, Stogerer E. Rehabilitation of homonymous scotomata in patients with postgeniculate damage of the visual system: saccadic compensation training. Restor Neurol Neurosci. 1992;4:245–254. 21. Zihl J. Visual scanning behavior in patients with homonymous hemianopia. Neuropsychologia. 1995;33:287–303.
Reinhard et al: J Neuro-Ophthalmol 2014; 34: 354-361
22. Mannan SK, Pambakian ALM, Kennard C. Compensatory strategies following visualsearch training in patients with homonymous hemianopia: an eye movement study. J Neurol. 2010;257:1812–1821. 23. Trauzettel-Klosinski S. Current methods of visual rehabilitation. Dtsch Arztebl Int. 2011;108:871–878. 24. Schneider RM, Thurtell MJ, Eisele S, Lincoff N, Bala E, Leigh RJ. Neurological basis for eye movements of the blind. PLoS One. 2013;8:e56556. 25. Leigh RJ, Thurston SE, Tomsak RL, Grossman GE, Lanska DJ. Effect of monocular visual loss upon stability of gaze. Invest Ophthalmol Vis Sci. 1989;30:288–292. 26. Troxler D. Über das Verschwinden gegebener Gegenstände innerhalb unseres Gesichtskreises. In: Himly K, Schmidt JA. Ophthalmologische Bibliothek (German). 1804;2:1–56. 27. Martinez-Conde S, Macknik SL, Troncoso XG, Dyar TA. Microsaccades counteract visual fading during fixation. Neuron. 2006;49:297–305. 28. McCamy MB, Otero-Millan J, Macknik SL, Yang Y, Troncoso XG, Baer SM, Crook SM, Martinez-Conde S. Microsaccadic efficacy and contribution to foveal and peripheral vision. J Neurosci. 2012;32:9194–9204. 29. Hafed ZM. Alteration of visual perception prior to microsaccades. Neuron. 2013;77:775–786. 30. Kasten E, Wüst S, Behrens-Baumann W, Sabel BA. Computerbased training for the treatment of partial blindness. Nat Med. 1998;4:1083–1087. 31. Reinhard J, Schreiber A, Schiefer U, Kasten E, Sabel BA, Kenkel S, Vonthein R, Trauzettel-Klosinski S. Does visual restitution training change absolute homonymous visual field defects? A fundus controlled study. Br J Ophthalmol. 2005;89:30–35. 32. Kasten E, Bunzenthal U, Sabel BA. Visual field recovery after vision restoration therapy (VRT) is independent of eye movements: an eye tracker study. Behav Brain Res. 2006;175:18–26. 33. Trauzettel-Klosinski S. Eccentric fixation in hemianopic field defects: a valuable strategy to improve reading ability and an indication for cortical plasticity. Ophthalmol. 1997;18:117–131.
361
Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.
