Rehabilitation Strategies for Patients with Homonymous Visual Field Defects

STATE OF THE ART

Rehabilitation Strategies for Patients With Homonymous Visual Field Defects

Alidz Pambakian, MRCP, Jon Currie, MBBS, PhD, and Christopher Kennard, PhD, FRCP, F Med Sci

Abstract: Homonymous visual field defects (HVFDs) are raise optimism regarding the potential for success with a among the most common disorders that occur in brain dam variety of techniques. age, particularly after stroke. They lead to considerable dis abilities, particularly with reading and visual exploration. A variety of different approaches, including optical aids and PRINCIPLES UNDERLYING visual training techniques, have been examined for the VISUAL REHABILITATION rehabilitation of these HVFDs. Despite the considerable in Visual rehabilitation of patients with HVFDs rests on genuity that has been applied and anecdotal evidence that certain physiologic principles and observations. has accumulated, rigorously controlled trials that clearly es First, the traditional view that lesions of the geniculos- tablish efficacy of any method are lacking. triate pathway in humans result in complete and permanent (JNeuro-Ophthalmol 2005;25: 136-142) visual loss in the topographically related area of visual field has been challenged as it has become clear that the "blind" hemifields of hemianopes may retain certain visual functions (13). omonymous visual field defects (HVFDs) are among Second, some patients eventually experience varying H the most frequent consequences of brain damage. degrees of spontaneous recovery of their visual defect, de Thirty percent of all patients with stroke (1) and 70% of pending on the underlying pathology and location of the those with stroke involving the posterior cerebral artery have causative lesion (14,15). Studies using 18-fluoro-2-deoxy- hemianopias (2). Subarachnoid hemorrhages, intracerebral glucose positron emission tomography have demonstrated hematomas, and trauma add to this figure (3,4). Patients with a reduction in the size of the metabolic lesion and improve HVFDs have particular difficulties with reading and visual ment in striate cortex metabolism in patients with revers exploration that have far-reaching repercussions on their ible homonymous field defects, but not in patients with domestic and vocational lives (3,4). In addition, HVFDs are persisting field defects (16). HVFDs caused by ischemia associated with a reduced prognosis for successful rehabil show recovery of full visual field in fewer than 10% of itation from stroke (5-8), particularly when combined with cases (5,17). In patients with complete homonymous visual neglect or visual sensory inattention (9-12). This arti hemianopias caused by ischemia, field recovery is largely cle updates a previous review of rehabilitation strategies for complete within the first 10 days. Recovery of a partial defect patients with HVFDs (1). The topic has also been compre is usually maximal within the first 48 hours and complete hensively reviewed by Kerkhoff (3) and Zihl (4). within 10 to 12 weeks (5,17). In general, vision returns to the affected field in a sequence starting with perception of Although rehabilitation strategies for aphasia, motor light, motion, form, color, and finally stereognosis (5,17). dysfunction, and cognitive dysfunction are widely used in Functional brain reorganization may allow recovery of clinical practice, there are no uniformly accepted treatments function after the resolution of peri-infarct edema and for HVFDs, and surprisingly little systematic, evidence- reperfusion of the ischemic penumbra. However, the strict based research in this area. As a result, therapeutic nihilism unilateral retinotopic representation of the primary visual is often the response to rehabilitation of patients with such cortex probably limits the degree of reorganization that has problems. However, a number of clinical observations should been observed in other neural networks that are organized in a more extended and overlapping fashion (18-20). Department of Visual Neuroscience (AP,CK), Imperial College, Third, patients with established HVFDs perform London, UK and Westmead Hospital (JC), Sydney, Australia. maneuvers designed to compensate for their visual loss. Address correspondence to Christopher Kennard, Department of Visual Neuroscience, Faculty of Medicine, Charing Cross Campus, Imperial When viewing simple patterns, hemianopic patients con College, London, St. Dunston's Road, London W68RP, UK. E-mail: centrate their gaze towards the blind hemifield rather than [email protected] toward the center of the pattern, demonstrating what is

136 J Neuro-Ophthalmol, Vol. 25, No. 2, 2005 Homonymous Visual Field Defects J Neuro-Ophthalmol, Vol. 25, No. 2, 2005

thought by some to be a compensatory strategy (21-23). By primary gaze position. Their use in HVFDs in one con deviating their fixation point into the blind field, more of the trolled trial (38) concluded that the prism-treated group visual scene is brought into the intact hemifield. They also performed significantly better on visuospatial tests but did adopt a number of compensatory strategies when refixat- not demonstrate any functional improvement in activities ing between simple visual targets presented unpredictably of daily living. Monocularly fitted sectoral prisms serve to (1,24). When performing visual search tasks, patients with expand the field but cause unacceptable central diplopia long-standing hemianopias scan very differently than do (40), which may be overcome by trimming the central area normal subjects and demonstrate numerous refixations and (41). Because of these drawbacks, Fresnel prisms used in this inaccurate saccades that result in disorganized scanning, way have never really entered clinical practice. However, two longer search times, and the omission of relevant objects recent modifications offer promise: vision multiplexing and (22,25). In a recent study, we recorded the eye movements prism adaptation. of patients with HVFDs while they viewed images of real scenes (26) and found that those with lesions of less than Vision Multiplexing 6 months' duration made patterns of fixations that approx Multiplexing refers to the transmission of two or imated those of normal subjects, whereas those with older more signals simultaneously over the same communication lesions demonstrated fixation patterns different from those channel so that all the information can be separated and of normal subjects. This finding may reflect the evolution of used at the receiving end. Vision multiplexing is the gen a spontaneous compensatory eye movement strategy and eral engineering approach used by Peli (31,32) to develop raises the interesting question of whether patients can be a range of visual aids. Under normal circumstances, the taught to scan more efficiently (25,27-30). visual system allows a wide field of view of almost 180° at Rehabilitation of patients with HVFDs is based on very high resolution, which is achieved by using "temporal maximizing whatever visual function remains. Two ap multiplexing" and variable spatial resolution (33). The pe proaches have been used: optical aids and visual training. ripheral visual fields are continually monitored at low spa tial resolution for targets of interest, which are then fixated by the high-resolution fovea via saccadic eye movements. OPTICAL AIDS In this way, high-resolution information from several tar Many optical aids have been promoted but never gets of interest is temporally multiplexed and used to gen subjected to clinical trial in the treatment of patients with erate a high-resolution view over a wide field, even though HVFDs (31-33). These include mirrors attached to spectacle at any instant only a fraction of the field is seen at such high frames (34), partially reflecting mirrors (beam splitters), and resolution. dichroic mirrors (which reflect a red image and transmit Any visual disorder, such as hemianopia, disrupts a green one) (35), reversed telescopes (36), prisms, wide- the interplay between central and peripheral vision and angle lenses, and closed-circuit television monitors (37). interrupts multiplexing. A visual multiplexing aid would Optical aids function in one of two ways: (1) to relo aim to multiplex the missing component with the remain cate the image to a part of the visual field not within the ing one. For hemianopia, the missing peripheral field should scotoma; or (2) to expand the visual field. Relocation re be multiplexed with the remaining high-resolution central places one scotoma with another, because the relocated field in such a way that the visual system can access and image causes another previously visible part of the visual interpret both. field to become invisible. Field expansion is therefore the An ideal optical aid for patients with HVFDs would preferable option. This has been performed most success therefore: (1) expand the visual field rather than relocate fully with prisms. it; (2) function in all positions of gaze; (3) avoid central diplopia; and (4) reconstruct the interplay of central and Standard Prisms peripheral vision. Peli (31-33) has developed a visual aid Prisms have been applied to spectacle lenses to treat that uses monocular sectoral prisms that are restricted to the HVFDs for more than 30 years with varying anecdotal peripheral fields (superior, inferior, or both). The prisms are success. Fresnel prisms work by displacing the images of placed across the whole width of the spectacle lens on the objects toward their apex and are commercially available side of the hemianopia, above and below the pupil, and as plastic press-on lenses that can easily be applied to provide peripheral field information that is effective in all spectacle lenses with their bases toward the hemianopic lateral positions of gaze. The prism expands the field by side. Binocular sectoral prisms applied to the hemianopic causing peripheral diplopia, which is a common feature of half of each lens appear to be the most commonly used normal vision and is therefore tolerated far better than (38,39). They relocate the image when the patient looks central diplopia. The effect is to cause an artificial exotropia sideways through the prism but cause a scotoma in the in peripheral but not central vision, which is reminiscent of

137 J Neuro-Ophthalmol, Vol. 25, No. 2, 2005 Pambakian et al

the exotropia that some congenital hemianopes develop as (57,58) demonstrated by the successful forced choice a natural compensatory mechanism (42). A visual field ex discrimination of positive versus negative facial expres pansion of approximately 20° measurable by standard sions projected into the blind hemifield. Further functional perimetry has been described in the first 30 patients treated studies with these patients have demonstrated successful with a 40-diopter prism by Peli (32). Patients have reported fear-conditioning (59-61) whereby the presentation of improved walking and avoidance of obstacles and appear to a bland visual stimulus to the blind field is not consciously accurately perceive the location of objects detected using perceived but produces a fear response in the amygdala. the device. No formal testing has been conducted, but such This raises a further potential route for rehabilitation: could an engineering approach offers considerable prospects. blindsight serve to condition useful, adaptive responses in patients? Prism Adaptation Binocular prisms, fitted the full width of the spectacle Compensatory Oculomotor Strategies for lens, function to relocate the whole field of view (43). A 20- Visual Exploration diopter prism, for example, shifts the image by approxi Working with the concept that hemianopes naturally mately 10° as long as patients do not neutralize the effect by make strategic eye movement adaptations to overcome field making a compensatory eye movement of 10°. There is loss, several groups have attempted to develop training evidence that subjects with normal vision can adapt to the programs aimed at systematically reinforcing these com effects of the prism (44). Rosetti et al (45) have recently pensatory eye movements, thereby fortifying and enlarging reported using active prism adaptation in patients with ne the functional field of visual search. Normal daily activities glect rather than hemianopia. It remains to be seen whether apparently do not achieve the same effect. this technique will provide an aid to visual rehabilitation in There is limited evidence that patients can success patients with neglect let alone in those with HVFDs. fully adapt to their hemianopias with training in visual search (1,25,27,30,62). Training programs comprise three major steps: (1) practicing large saccades into the blind VISUAL TRAINING field in place of the inappropriately small saccades Blindsight normally made by hemianopes; (2) practicing visual search Despite total destruction of the striate cortex, or even on projected slides to improve the spatial organization of hemispherectomy, 20% of patients with HVFDs remain eye movements; and (3) applying both techniques to real- able to discriminate, without consciously perceiving, attri life scenarios. Various parameters have been used to judge butes of visual targets that are presented tachistoscopically the success of these techniques, including improvements in to their blind hemifields in a forced choice context (46,47). response time and error rates during visual search, enlarge There are several theories concerning the mechanism of this ments in the visual field and visual search field (defined as "blindsight" (48,49). One theory is that "spared islands" of the perimetrically measured area that a patient can actively striate cortical neurons that have survived injury retain their scan using eye movements without head movements when regular projections to extrastriate cortical regions and searching for a suprathreshold stimulus), and improved pro mediate residual visual function (50). But when no func ficiency in activities of daily living assessed objectively and tional striate cortex remains, as after hemispherectomy, this with questionnaires. explanation cannot be valid. In that case, blindsight may be After training 30 patients with HVFDs in daily mediated by extrageniculostriate pathways projecting to sessions, Zihl (62) initially reported that the visual search subcortical structures, including the superior colliculus and field expanded from 10° to 30° within four to eight ses the pulvinar, and ultimately to the ipsilesional extrastriate sions. Kerkhoff et al (30) validated these results with 92 cortex via tecto-tectal pathways (48,51,52). An alternative hemianopic patients and an additional 30 patients with explanation invokes cortical plasticity. hemineglect, whose search field increased from 15° to 35°. Hemianopes have been taught to detect targets in In a further study, Kerkhoff et al (27) quantified the func their defective hemifields using blindsight retraining tech tional benefit of this training program. After 25 treatment niques in which they practice discriminating visual stimuli sessions, 22 patients showed a 50% reduction in the time presented as part of forced-choice paradigms (17,53-56). taken to find objects on a table, complementing the subjec The consensus had been that this training merely induced tive improvement elicited in a questionnaire that rated dis patients to make larger saccades into their blind fields, rather ability. Error rates and response times in visual search had than using blindsight. However, recent fMRI (functional significantly improved. Significant visual field and visual magnetic resonance imaging) studies have highlighted an search field expansion were also confirmed. After treat affective component to blindsight in mediating the perception ment, 91% of this group returned to part-time work. Zihl of unconscious emotional expression via the amygdala (25) proceeded to record the eye movements of eight

hemianopes before and after training in visual search, and To explain the mechanism underlying visual field ex highlighted how the improvements in the organization of pansion, Sabel et al (71,74) have hypothesized that regular their scanpaths contributed to the shortening of their re visual stimulation at the damaged visual field border region sponse times (25). by VRT could reactivate surviving neurons, providing More recently, Nelles et al (29) have trained 21 small islands of residual vision in the cortical region that patients in visual search and found that training improved subserves that field. Considering that survival of a mere detection and reaction to visual stimuli without expansion 10% to 15% of neurons in a damaged region has been of the visual field. Patients reported significant functional postulated to be sufficient for recovery of basic visual improvements in a daily living questionnaire. functions (75), repetitive stimulation may lead to reac tivation of these neurons, possibly with expanded receptive fields and improved synaptic connectivity (74). Visual Field Recovery However, a recently published study by Sabel's group Enormous controversy has surrounded the question and scientists in Tubingen, Germany (76), in which the of whether any training techniques can significantly reverse visual field was tested independently using a scanning laser visual field loss in hemianopes (1). In a series of experi ophthalmoscope that controls for visual fixation, failed to ments in patients with HVFDs, Zihl (63-65) used a technique find any significant improvement in the visual field defects, that involved repeatedly determining light thresholds at the although many of the patients noted a subjective impression visual field border (64-66). In most cases, the visual field that they had benefited from the VRT. In an accompanying enlargement did not exceed 5° but there were individual editorial, Horton (77) suggests that during VRT visual cases with remarkable recovery. Kerkhoff et al (27,30) and fixation is inadequately controlled, so the reported mean 5° Pommerenke and Markowitsch (67) observed minor expan field recovery could be explained by the frequent saccades sions of the field border by 1° to 6.7° in analogous experi hemianopic patients make toward their scotoma to maintain ments. Balliet et al (68), however, were unable to reproduce surveillance of blind areas in their visual fields (24,26). The Zihl's results (see Zihl's response) (69). visual field improvement therefore may be a function of the Sabel et al (72) have reopened the debate by pub method of visual field assessment and fixation control. In lishing a series of experiments based on sounder methodol an earlier study in which the visual fields were tested before ogies developed in response to the deficiencies highlighted and after VRT using a Tubingen automatic perimeter, no field in previous publications (70-73). Using a domiciliary PC- improvement was noted, although this was not the case when based stimulus system, they randomized patients with HVFDs the visual fields were assessed using the same software pro to receive an active visual restitution training (VRT) stim gram as had been used for the VRT (73). ulation program or a placebo stimulus program for 1 hour daily, 6 days per week, for 6 months. In the active VRT Reading program, patients were required to respond with a key press Patients with hemianopias have reading difficulties to hundreds of repetitive visual stimuli presented in the that reflect the laterality of the visual field defect. Left transition zone between the intact and damaged visual field hemianopias cause difficulties with the return eye move sector. Responses were monitored by the computer and the ments required to find the beginning of a new line. Right stimulus pattern was modulated in accordance with the hemianopias cause more severe reading difficulties, with responses. Stimuli for the placebo program were presented loss of the anticipatory parafoveal scanning process and a only at fixation. In the study, 95% of the actively treated characteristic reading disorder termed "hemianopic dys patients demonstrated a central visual field gain of approx lexia" (78,79). This disordered reading is reflected in the imately 5° into the hemianopic field, an enlargement that disruption of reading eye movement scanpaths, which show restores the parafoveal field that is functionally significant prolonged fixations, inappropriately small amplitude sac- for reading. None of the control subjects improved. Of the cades to the right, and many saccadic regressions. Patients patients receiving the active VRT program, 72% reported develop individual tricks to overcome these problems, using subjective improvement in vision, compared with 17% of rulers to keep to the correct line of text or turning a page the control group. The authors demonstrated that VRT of text by 90° so that left-to-right reading becomes above- using a white dot stimulus generalized to color and pattern to-below reading (80). recognition (70), although additional treatment with spe Several groups have postulated that by retraining and cific color or shape recognition training resulted in a more improving the disordered eye movements, reading perfor pronounced improvement of those functions (69). This mance will improve. Computer-based systems should have technique has now been commercialized (NovaVision AG) many advantages in delivering and monitoring VRT pro and received approval from the United States Food and grams. A computer-based VRT system designed by Zihl Drug Administration in 2003. (1,81) was used to train a group of 96 patients with HVFDs.

139 J Neuro-Ophthalmol, Vol. 25, No. 2, 2005 Pambakian et al

At the end of training, patients were able to read faster and visual search in the laboratory setting but also perform with fewer errors, and eye movement recordings demon significantly better in validated activities of daily living. strated fewer fixations, shorter fixation periods, and larger Patients rate themselves as less disabled in a visual disorder saccadic jumps. Patients with right HVFDs were more questionnaire (63). However, there was no evidence for any disabled and required more training sessions than did visual field recovery. The program is now available to patients with left HVFDs (33 sessions compared with 26), patients on video but still needs validation in a placebo- and at the end of training they had not improved as much as controlled trial. those with left HVFDs (1). Kerkhoff et al (82) used an With respect to computer-based VRT, the domiciliary identical protocol to train a group of 56 hemianopic patients PC-based design of the system certainly increases the pro for a period of approximately 3 weeks (mean =13 sessions). gram's convenience and cost efficiency. Although the mag This group also demonstrated improved reading, and for nitude of visual field expansion appears small (5°), such both patient groups the improvement was maintained at a gains in the central field would unquestionably be useful to follow-up of 6 months to 2 years (1). patients exploring their immediate environment. However, the most recent study (76) casts doubt on its effectiveness. There have been great advances in the field of optical CONCLUSIONS aids with Peli's pioneering multiplexing technique. However, Although the benefits of rehabilitation in patients all optical aids require expert fitting in specialist centers and with HVFDs are often perceived as offering marginal gain, are available only to a minority. Although definitive evidence it is likely that this perception may underestimate their real of their clinical effectiveness is still lacking, computer-based value because the confidence and boost in self-esteem that or video-based visual training programs are potentially more they give patients is not recognized. The rehabilitation readily available and free of side-effects. programs currently available or undergoing trial are com plex and labor-intensive, and in general they require rela tively specific facilities for their implementation. Therefore, the emphasis is shifting to the development of simple, user- REFERENCES friendly techniques that patients can practice in their own 1. Pambakian AL, Kennard C. Can visual function be restored in patients with homonymous hemianopia? Br J Ophthalmol homes, causing minimal disruption to their daily lives. 1997;81:324-8. Whether saccadic visual search training provides 2. Brandt T, Thie A, Caplan LR, Hacke W. [Infarcts in the brain areas a successful treatment of patients with HVFDs remains supplied by the posterior cerebral artery. Clinical aspects, pathogen esis and prognosis]. Nervenarzt 1995;66:261-1 A. unanswered, largely because many of the various studies 3. Kerkhoff G. Neurovisual rehabilitation: recent developments and have not been adequately controlled, and have not future directions. J Neurol Neurosurg Psychiatry 2000;68:691-706. convincingly demonstrated improvements in visual search 4. Zihl J. Rehabilitation of visual disorders after brain injury. Hove, UK: Psychology Press; 2000. performance in activities of daily living. In these studies, 5. Gray CS, French JM, Bates D, Cartlidge NE, Venables GS, James OF. patients have acted as their own controls (within-subject Recovery of visual fields in acute stroke: homonymous hemianopia repeated measure design) and there are no control groups or associated with adverse prognosis. Age Ageing 1989;18:419-21. placebo treatments. Given that patients can improve their 6. Kalra L, Smith DH, Crome P. Stroke in patients aged over 75 years: outcome and predictors. Postgrad Med J 1993;69:33-6. performance in visual search with training, it is now nec 7. Patel AT, Duncan PW, Lai SM, Studenski S. The relation between essary to address the specificity of the technique against impairments and functional outcomes poststroke. Arch Phys Med a placebo condition. A randomized controlled trial with Rehabil 2000;81:1357-63. 8. Han L, Law-Gibson D, Reding M. Key neurologic impairments patients who are not trained is needed to assess therapeutic influence function-related group outcomes after stroke. Stroke efficacy. However, because of the heterogeneity of under 2002;33:1920^1. lying lesions, it would be difficult to ensure that control and 9. Kalra L, Perez I, Gupta S, Wittink M. The influence of visual neglect on stroke rehabilitation. Stroke 1997;28:1386-91. experimental groups are comparable. A study with a cross 10. Cassidy TP, Bruce DW, Lewis S, Gray CS. The association of visual over design may be preferable technically but unacceptably field deficits and visuo-spatial neglect in acute right-hemisphere long for patients. stroke patients. Age Ageing 1999;28:257-60. 11. Cassidy TP, Lewis S, Gray CS. Recovery from visuospatial neglect in In response to clinical need, visual search training stroke patients. J Neurol Neurosurg Psychiatry 1998;64:555-7. techniques are now available. For example, we have devel 12. Jehkonen M, Ahonen JP, Dastidar P et al. Visual neglect as a pre oped a domiciliary rehabilitation tool for patients with dictor of functional outcome one year after stroke. Acta Neurol Scand 2000;101:195-201. HVFDs based on a visual search paradigm that is inexpen 13. Stoerig P, Cowey A. Blindsight in man and monkey. Brain 1997; sive and easy to use (63). We have trained 29 hemianopic 120(Pt 3):535-59. subjects in their own homes on a dedicated television 14. Poppelreuter W. Die Storungen der niederen und horeren Sehleis- tungen durch Verletzungen des Okzipitalhirns. Die psychischen monitor using a portable computer. Our data show that most Schddigungen durch Kopfschuli im Kriege 1914/16. Leipzig: patients not only become significantly faster at performing Leopold Voss; 1917.

15. Hine ML. The recovery of fields of vision in concussion injuries of 42. Levy Y, Turetz J, Krakowski D, Hartmann B, Nemet P. the occipital cortex. Br J Ophthalmol 1918;2:12-25. Development of compensating exotropia with anomalous reti 16. Bosley TM, Dann R, Silver FL, et al. Recovery of vision after nal correspondence after early infancy in congenital homony ischemic lesions: positron emission tomography. Ann Neurol mous hemianopia. J Pediatr Ophthalmol Strabismus 1995;32: 1987;21:444-50. 236-8. 17. Zihl J. "Blindsight": improvement of visually guided eye movements 43. Cohen JM, Waiss B. Visual field remediation. In: Cole RG, Rosenthal by systematic practice in patients with cerebral blindness. Neuro BP, eds. Remediation and management of low vision. St Louis: psychologic! 1980;18:71-7. Mosby; 1996:1-25. 18. Weiller C, Chollet F, Friston KJ, Wise RJ, Frackowiak RS. Functional 44. Welch RB, Bridgeman B, Anand S, Browman KE. Alternating prism reorganization of the brain in recovery from striatocapsular infarction exposure causes dual adaptation and generalization to a novel in man. Ann Neurol 1992;31:463-72. displacement. Percept Psychophys 1993;54:195-204. 19. Nelles G, Spiekermann G, Jueptner, M et al. Reorganization of 45. Rossetti Y, Rode G, Pisella L, et al. Prism adaptation to a rightward sensory and motor systems in hemiplegic stroke patients. A positron optical deviation rehabilitates left hemispatial neglect. Nature emission tomography study. Stroke 1999;30:1510-6. 1998;395:166-9. 20. Nelles G, Widman G, de Greiff A, et al. Brain Representation of 46. Weiskrantz L, Warrington EK, Sanders MD, Marshall J. Visual Hemifield Stimulation in Poststroke Visual Field Defects. Stroke capacity in the hemianopic field following a restricted occipital 2002;33:1286-93. ablation. Brain 1974;97:709-28. 21. Ishiai S, Odajima N, Kotera M, Furukawa T, Tsukagoshi H. [Eye 47. Zeki S, ffytche DH. The Riddoch syndrome: insights into the fixation-patterns in homonymous hemianopia and unilateral spatial neurobiology of conscious vision. Brain 1998;121(Pt 1):25—45. neglect: a study with an eye camera]. Rinsho Shinkeigaku 1987; 48. Ptito A, Fortin A, Ptito M. 'Seeing' in the blind hemifield following 27:634-41. hemispherectomy Prog Brain Res 2001;134:367-78. 22. Chedru F, Leblanc M, Lhermitte F. Visual searching in normal and 49. Schoenfeld MA, Noesselt T, Poggel D, et al. Analysis of pathways brain-damaged subjects (contribution to the study of unilateral mediating preserved vision after striate cortex lesions. Ann Neurol inattention). Cortex 1973;9:94-111. 2002;52:814-24. 23. Gassel MM, Williams D. Visual function in patients with 50. Fendrich R, Wessinger CM, Gazzaniga MS. Residual vision in homonymous hemianopia. Part II. Oculomotor mechanisms. Brain a scotoma: implications for blindsight. Science 1992;258:1489-91. 1963;86:1-36. 51. Barbur JL, Watson JD, Frackowiak RS, Zeki S. Conscious visual 24. Meienberg O, Zangemeister WH, Rosenberg M, Hoyt WF, Stark L. perception without VI. Brain 1993;116(Pt 6):1293-302. Saccadic eye movement strategies in patients with homonymous 52. ffytche DH, Guy CN, Zeki S. The parallel visual motion inputs into hemianopia. Ann Neurol 1981;9:537^44. areas VI and V5 of human cerebral cortex. Brain 1995;118(Pt 6): 25. Zihl J. Visual scanning behavior in patients with homonymous 1375-94. hemianopia. Neuropsychologia 1995;33:287-303. 53. Zihl J, von Cramon D. Registration of light stimuli in the cortically 26. Pambakian AL, Wooding DS, Patel N, Morland AB, Kennard C, blind hemifield and its effect on localization. Behav Brain Res 1980; 1: Mannan SK. Scanning the visual world: a study of patients with 287-98. homonymous hemianopia. J Neurol Neurosurg Psychiatry 2000; 54. Zihl J, Werth R. Contributions to the study of "blindsight"-II. The 69:751-9. role of specific practice for saccadic localization in patients with 27. Kerkhoff G, Munssinger U, Meier EK. Neurovisual rehabilitation in postgeniculate visual field defects. Neuropsychologia 1984;22:13-22. cerebral blindness. Arch Neurol 1994;51:474-81. 55. Mohler CW, Wurtz RH. Role of striate cortex and superior colliculus 28. Zihl J. [Homonymous hemianopsia and its rehabilitation]. Klin in visual guidance of saccadic eye movements in monkeys. Monatsbl Augenheilkd 1988;192:555-8. J Neurophysiol 1977;40:74-94. 29. Nelles G, Esser J, Eckstein A, Tiede A, Gerhard H, Diener HC. Compensatory visual field training for patients with hemianopia after 56. Weiskrantz L, Cowey A, Passingham C. Spatial responses to brief stroke. Neurosci Lett 2001;306:189-92. stimuli by monkeys with striate cortex ablations. Brain 1977;100: 30. Kerkhoff G, Munssinger U, Haaf E, et al. Rehabilitation of homo 655-70. nymous scotomatas in patients with postgeniculate damage of the 57. de Gelder B, Vroomen J, Pourtois G, Weiskrantz L. Non-conscious visual system: saccadic compensation training. Restorative Neurol recognition of affect in the absence of striate cortex. Neuroreport Neurosci 1992;4:245-54. 1999;10:3759-63. 31. Peli E. Field expansion for homonymous hemianopia by optically 58. de Gelder B, Pourtois G, Weiskrantz L. Fear recognition in the voice induced peripheral exotropia. Optom Vis Sci 2000;77:453-64. is modulated by unconsciously recognized facial expressions but not 32. Peli E. Treating with spectacle lenses: a novel idea!? Optom Vis Sci by unconsciously recognized affective pictures. Proc Natl Acad Sci U 2002;79:569-80. SA 2002;99:4121-6. 33. Peli E. Vision multiplexing: an engineering approach to vision 59. Morris JS, DeGelder B, Weiskrantz L, Dolan RJ. Differential rehabilitation device development. Optom Vis Sci 2001;78:304-15. extrageniculostriate and amygdala responses to presentation of 34. Walsh TJ, Lawton Smith JL. Hemianopic spectacles. Am J emotional faces in a cortically blind field. Brain 2001;124:1241-52. Ophthalmol 1966;61:914-5. 60. Hamm AO, Weike Al, Schupp HT, Treig T, Dressel A, Kessler C. 35. Duszynski LR. Hemianopsia dichroic mirror device. Am J Affective blindsight: intact fear conditioning to a visual cue in a Ophthalmol 1955;39:876-8. cortically blind patient. Brain 2003;126:267-75. 36. Goodlaw EI. Assessing field defects of the low-vision patient. Am J 61. Weiskrantz L. Mind-the gap, after 65 years: visual conditioning in Optom Physiol Opt 1981;58:486-91. cortical blindness. Brain 2003;126:265-6. 37. Peli E. Vision multiplexing: an engineering approach to vision 62. Zihl J. Neuropsychologische Rehabilitation. In: von Cramon D, rehabilitation device development. Optom Vis Sci 2001;78:304-15. Zihl J, eds. Neuropsychologische rehabilitation: grudlagen, diagnostic, 38. Rossi PW, Kheyfets S, Reding Ml Fresnel prisms improve visual behandlungsverfahren. Berlin: Springer-Verlag; 1988:105-31. perception in stroke patients with homonymous hemianopia or 63. Pambakian AL, Mannan SK, Hodgson TL, Kennard C. Saccadic unilateral visual neglect. Neurology 1990;40:1597-9. visual search training: A treatment of patients with homonymous 39. Lee AG, Perez AM. Improving awareness of peripheral visual field hemianopia. J Neurol Neurosurg Psychiatry 2004;75:1443-8. using sectorial prism. J Am Optom Assoc 1999;70:624-8. 64. Zihl J, von Cramon D. Restitution of visual function in patients with 40. Gottlieb DD. Method of using a prism in lens for the treatment of cerebral blindness. J Neurol Neurosurg Psychiatry 1979;42:312-22. visual field loss. [4,779,972]. 1998. US. Ref Type: Patent. 65. Zihl J. Recovery of visual functions in patients with cerebral 41. Lawton Smith JL, Weiner IG, Lucero AJ. Hemianopic fresnel prisms. blindness. Effect of specific practice with saccadic localization. Exp J Clin Neuroophthalmol 1982;2:19-22. Brain Res 1981;44:159-69.

141 J Neuro-Ophthalmol, Vol. 25, No. 2, 2005 Pambakian et al

66. Zihl J, von Cramon D. Visual field recovery from scotoma in patients patient with a gunshot lesion. J Neurol Neurosurg Psychiatry 2001; with postgeniculate damage. A review of 55 cases. Brain 1985; 70:236-9. 108(Pt 2):335-65. 75. Sabel BA, Kasten E, Kreutz MR. Recovery of vision after partial 67. Pommerenke K, Markowitsch HJ. Rehabilitation training of visual system injury as a model of postlesion neuroplasticity. In: homonymous visual field defects in patients with postgeniculate Freund HJ, Sabel BA, Witte OW, eds. Brain Plasticity. Philadelphia: damage of the visual system. Restorative Neurol Neurosci 1989;1: Lippincott-Raven; 1997:251-76. 47-63. 76. Reinhart J, Schreiber A, Schiefer U, Kasten E, Sabel BA, Kenkel S, 68. Balliet R, Blood KM, Bach-y-Rita P. Visual field rehabilitation Vonthein R, Trauzettel-Klosinski S. Does visual restitution training in the cortically blind? J Neurol Neurosurg Psychiatry 1985;48: change absolute homonymous visual field defects? A fundus con 1113-24. trolled study. Br J Ophthmol 2005;89:30-5. 69. Zihl J, von Cramon D. Visual field recovery in the cortically blind? 77. Horton JC. Disappointing results from Nova Vision's visual J Neurol Neurosurg Psychiatry 1986;49:965-7. restitution therapy. Br J Ophthmol 2005;89:1-2. 70. Kasten E, Sabel BA. Visual-field enlargement after computer-training 78. Leff AP, Scott SK, Crewes H, et al. Impaired reading in patients with in brain-damaged patients with homonymous deficits-an open pilot right hemianopia. Ann Neurol 2000;47:171-8. trial. Restorative Neurol Neurosci 1995;8:113-27. 79. Leff AP, Crewes H, Plant GT, Scott SK, Kennard C, Wise RJ. The 71. Kasten E, Poggel DA, Sabel BA. Computer-based training of stimulus functional anatomy of single-word reading in patients with hemi detection improves color and simple pattern recognition in the defective anopic and pure alexia. Brain 2001;124:510-21. field of hemianopic subjects. J Cogn Neurosci 2000;12:1001-12. 80. Wang MK. Reading with a right homonymous haemianopia. Lancet 72. Kasten E, Wuest S, Sabel BA. Residual vision in transition zones in 2003;361:1138. patients with cerebral blindness. J Clin Exp Neuropsychol 1998;20: 81. Zihl J, Kennard C. Disorders of higher visual function. In: Brandt T, 581-98. Caplan LR, Dichigans J, Diener HC, Kennard C, eds. Neurological 73. Kasten E, Wust S, Behrens-Baumann W, Sabel BA. Computer-based Disorders: course and treatment. San Diego: Academic Press; training for the treatment of partial blindness. Nat Med 1998;4: 1996:201-12. 1083-87. 82. Kerkhoff G, Munssinger U, Eberle-Strauss G, Stogerer E. Re 74. Poggel DA, Kasten E, Muller-Oehring EM, Sabel BA, Brandt SA. habilitation of hemianopic dyslexia in patients with postgeniculate Unusual spontaneous and training induced visual field recovery in a field disorders. Neuropsychol Rehabil 1992;2:21^2.