| OCR Text |
Show ORIGINAL CONTRIBUTION Perceptual Distortion in Homonymous Paracentral Scotomas Nikolaos A. Mavrakanas, MD, Nathalie P. L. Dang-Burgener, MD, Erika N. Lorincz, PhD, Theodor Landis, MD, and Avinoam B. Safran, MD In the adult cerebral cortex, cortical remapping has been found after altered inputs in normal subjects (1,2) and in patients with retinal (3) and cortical lesions (4). Reorgan-ization after peripheral or central visual deafferentation alters visual perception mainly by filling in the impaired visual field with information derived from areas neighbor-ing the scotoma (4,5) and by generating spatial distortion (5,6). However, it is not yet clear whether such cortical reorganization affects perception and impinges on areas remote from the scotoma. There is evidence in normal subjects and patients that perceptual alterations can occur at some distance from scotoma borders. In normal subjects, Ramachandran and Gregory (2) reported that when a cross made of two lines of different length fell on the physiologic blind spot, only the longer one appeared continuous. Kapadia et al (1) induced an artificial scotoma in normal individuals and observed a perceptual shift that pulled the perceived images surrounding the scotoma by a few minutes of arc toward the center of the masked area. The shift increased with eccentricity from the fixation point. These observations suggest that cortical reorganization can involve visual information functionally related to areas distant from the scotoma. In a previous clinical report (6), we described patients with homonymous paracentral scotomas who perceived one of their interlocutor's elbows as being narrower than it actually was. We named this form of perceptual spatial distortion the ‘‘thin man phenomenon.'' To explore the nature of this phenomenon, we devised a test that we named the ‘‘parallel line test.'' It consists of asking subjects to fixate a point located between two isometric vertical parallel lines, one of which crosses the scotoma, and to describe any perceived alteration on a displayed figure. Patients with homonymous paracentral scotomas perceived the line crossing the scotoma as being shorter than the mirrored one. The effect of distance from the scotoma border on perceptual distortion was not explored. The purpose of the present study was to investigate whether spatial distortion occurs only in the vicinity of scotoma borders or also in more distant areas of the visual Background: Cortical remapping after peripheral or central visual deafferentation alters visual perception, but it is unclear whether such a phenomenon impinges on areas remote from a scotoma. To investigate this question, we studied variations of perceptual spatial distortion in the visual field of patients with hom-onymous paracentral scotoma. Methods: Two patients with right inferior homon-ymous paracentral scotoma were asked to describe their perception of a series of figures showing two isometric vertical lines symmetrically located on either side of a fixation point. In each figure, the fixation point varied by steps of 2° along a hypothet-ical vertical line equidistant between the test lines. The lines subtended 20° of visual angle, and the right line passed through the scotoma in both cases. Time for spatial distortion to manifest was recorded. Results: Both subjects reported that the right line was perceived as shorter than the left one. The line shortening varied in magnitude with the distance of the fixation point from the end of the line and was more pronounced when the distance increased. More-over, perceptual line shortening appeared 5-10 sec-onds after steady fixation, but values of shortening varied during the following 10 seconds. In addition, the right line appeared uninterrupted or slightly blurred in the scotoma region. Conclusions: These observations reflect long-range cortical reorganization after brain damage. Larger receptive fields in the periphery of the visual map could explain why perceptual shortening is more pronounced with increased eccentricity. (J Neuro-Ophthalmol 2009;29:37-42) Ophthalmology (NAM, NPLD-B, ENL, ABS) and Neurology (TL) Services, Department of Clinical Neurosciences, University of Geneva Hospitals, Geneva, Switzerland. This study was supported by the Fondation en Faveur des Aveugles, Geneva, Switzerland. Address correspondence to Avinoam B. Safran, MD, Ophthalmology Service, Department of Clinical Neurosciences, University of Geneva Hospitals, 1211 Geneva 14, Switzerland; E-mail: a.b.safran@hcuge.ch J Neuro-Ophthalmol, Vol. 29, No. 1, 2009 37 J Neuro-Ophthalmol, Vol. 29, No. 1, 2009 Mavrakanas et al field and whether spatial distortion varies as a function of the distance from the scotoma border. This investigation was performed in patients with homonymous paracentral scotomas, as they are able to fixate centrally, hence preventing the occurrence of possible perceptual bias and instability induced by eccentric fixa-tion. Moreover, proximity of the fixation point to the visual field defect facilitates the analysis of scotoma-induced perceptual distortion. In addition, selecting patients with cortical rather than retinal lesions prevents interference from metamorphopsias generated by retinal tissue defor-mation. In our previous investigation (6), we found that the parallel line test was suitable to evaluate perceptual altera-tions in patients with homonymous paracentral scotomas and we therefore tailored this test to this investigation. METHODS Patients Two patients were included in the study. The first patient (NG) was a 60-year-old woman who had had an ischemic injury in the left occipital lobe 11 years earlier, resulting in an inferior right homonymous para-central scotoma (Fig. 1 and Fig. 2) (6). No major changes were noted when her clinical assessment was repeated. The second patient (CS) was a 53-year-old woman who had attempted suicide 6 months earlier by injecting 100 mg morphine and 45 mg midazolam intravenously. She was found at a Glasgow Coma Scale score of 3 and referred to the intensive care unit. Brain CT and MRI performed a few hours later showed an ischemic lesion in the left occipital lobe (Figs. 1 and 2). The attempt presumably provoked an episode of reduced blood pressure, which affected the occipital watershed area (7), resulting in a right inferior homonymous paracentral scotoma. After explanation of the nature and possible conse-quences of the study, patients gave their informed consent in accordance with the Declaration of Helsinki for research with human subjects. Experimental Conditions The scotoma borders were plotted for each patient on a tangent screen and by microperimetry using a scanning laser ophthalmoscope (Rodenstock, Munich, Germany). The vertical size of the scotoma was measured along the right vertical line, taking into account the scotoma area overlying the line (Fig. 3). We used a series of 11 figures showing two black isometric parallel vertical lines on a white background. Viewed from a distance of 30 cm, vertical lines subtended 20° of visual angle. The lines were symmetrically located at a distance of 3° on either side of the fixation point. The 38 FIG. 1. A. Axial T2 MRI of patient NG shows high signal extending from the left occipital pole to the lateral occipital gyri, consistent with infarction. B. Axial T2 MRI of patient CS shows high signal involving the left occipital lobe and the occipitotemporal junction, consistent with infarction. position of the fixation point differed in each figure, varying by 2° steps, so that altogether 11 figures with different fixation locations were presented. When the patient was fixating, the scotoma partly masked the right line. Cards, each showing a different figure, were held secure in a stand and the subject's head was stabilized at a distance of 30 cm. Testing Procedure Figures were shown in random order. Patients were asked to indicate by pen on the left line the levels corre-sponding to the perceived upper and lower ends of the right FIG. 2. Tangent screen examination (2w/1000) shows right inferior homonymous paracentral scotomas in patients NG and CS. q 2009 Lippincott Williams & Wilkins Homonymous Paracentral Scotomas J Neuro-Ophthalmol, Vol. 29, No. 1, 2009 FIG. 3. The parallel line test. A. The three figures shown to the patients. The specific location of the fixation point is indicated by an X character on each test sheet. B. The gray Xs show the location of various fixation points on different parallel line test cards. The bold X is the actual fixation point for this particular test sheet. C. Schematic representation of perceptual shortening of the right (R) line (upper and lower arrows), corresponding to the test sheet shown in Figure 3B (patient NG). The right line appeared uninterrupted or only slightly blurred in the scotoma area. line, which crossed the scotoma. The levels were marked on the figures by the examiner. The distance between the perceived and the actual end of the line was measured. This distance will be referred to as ‘‘line shortening.'' Entire card sets were presented 4 times to patient NG and 5 times to patient CS, providing a total of 44 and 55 values for each of the upper and lower estimates. Patients were asked to report when line shortening started to manifest and to indicate the levels of perceived line shortening when this visual phenomenon became stable. These times were recorded by the examiner. Data Analysis We analyzed the correlation between the right line shortening and the distance from the fixation point with SPSS software. The correlation between the right line shortening and the distance of the fixation point from the end of the line was analyzed separately for the ‘‘upper'' and ‘‘lower'' estimates, that is, for perceived upper and lower line shortening. This approach is justified because the scotoma occluded the line ends for each location of the fixation point differently. For each subject, the relationship between line shortening and distance from the fixation was investigated using Spearman's rank order correlation coef-ficient. Each r value was converted into a z score and the observed value of z (zobs) was calculated for the upper and lower measurement, using the following formula: zobs z 1 - z 2 1 - N 1-3 ' N2-3 with z1 and z2 being the z scores of the line shortening for the lower and upper measurements and N1 and N2 the respective number of trials. A nonstatistical difference would be concluded if -1.96 < zobs < 1.96. To disentangle mechanisms responsible for line shortening, the latter was plotted in terms of eccentricity in the visual field on the one hand and in terms of distance between the upper or lower line end and the closest scotoma border on the other hand. RESULTS A strong positive correlation between line shortening and the distance of the fixation point from the end of the line was found for both lower and upper estimates (Spearman's correlations: subject NG: lower, r = 0.797, n = 55, P < 0.0005; upper, r = 0.895, n = 55, P < 0.0005; subject CS: lower, r = 0.887, n = 44, P < 0.005, upper r = 0.915, n = 44, P < 0.0005). The farther the end of the line from the fixation point, the more it was perceived as 39 J Neuro-Ophthalmol, Vol. 29, No. 1, 2009 Mavrakanas et al Distance between the fixation point and the lower end of the line [°] Distance between the inferior end of the line and the closer scotoma border [°] FIG. 4. Perceptual shortening of the right line (crossing the scotoma) according to fixation position for patient NG (A) and patient CS (B). The upper parts show the relative position of the scotoma for each location of the fixation point for patients NG and CS. shortened (Fig. 4). In addition, no statistical difference in the strength of the correlation was found between the distance of the fixation point and the shortening for lower and upper measurements (subject NG: zobs lower/upper = 21.846, not significant (NS); subject CS: zobs lower/upper = 20.72, NS). For identical degrees of eccentricity, the magnitude of line shortening appeared similar for the upper or lower line ends. In contrast, this was not the case when identical dis-tances between the upper or lower line end and the closest scotoma border were considered (Fig. 4). When asked whether the lines appeared interrupted, both patients replied that they appeared continuous, and the right line appeared slightly blurred in the area passing through the scotoma. Furthermore, line shortening appeared an average of 5 seconds after steady fixation in patient NG and an average of 10 seconds in patient CS. Values of perceived line shortening were variable during the sub-sequent seconds, and this perceptual phenomenon became stable 15 seconds after steady fixation (for each card presentation) for patient NG and 20 seconds after steady fixation for patient CS. The values of perceptual shortening (y coordinates) for patients NG and CS are plotted for lower and upper segments of the right line in terms of eccentricity in the visual field (x coordinates: red fonts) on the one hand and 40 q 2009 Lippincott Williams & Wilkins Homonymous Paracentral Scotomas J Neuro-Ophthalmol, Vol. 29, No. 1, 2009 distance from the scotoma border on the other hand (x coordinates: blue fonts). In Figure 4, note that the further the end of the line is from the fixation point, the more it is perceived to be shortened. Moreover, for identical degrees of eccentricity, the magnitude of line shortening was similar for the upper and lower line ends. In contrast, this was not the case when identical distances between the upper or lower line end and the closest scotoma border were considered. On the shaded areas of the graphs, the line shortening probably results from the actual masking of the line end by the scotoma. Lower line segments have larger shaded areas than the upper line segments, as the scotoma occludes the lower and upper ends of the right line differently. Specifically, the inferior end of the right line is not visible by patient NG for the successive fixation positions 0, 2, and 4 and by patient CS for the fixation positions 0 and 2. For both patients, the upper end of the right line is occluded by the scotoma when fixating at the highest fixation position. DISCUSSION In this study, we used the model of homonymous paracentral scotomas to investigate alterations in visual perception after an occipital lesion. We analyzed spatial distortion and filling-in phenomena and considered possible neural mechanisms involved in the generation of these perceptual alterations. Perceptual Spatial Distortion The right line, partly masked by the scotoma, was perceived as shorter than the left line. Perceptual line shortening varied in magnitude, depending on the location of the fixation point along the midline. Moreover, the farther the end of the line from the fixation point, the more it was perceived to be shortened. Two factors may be involved in the variations of this perceptual phenomenon. The amount of line shortening could be related either to eccentricity in the visual field or to the effect of the scotoma overlap on the right line. Interestingly, for a given patient and eccentricity, the amount of perceptual shortening appeared similar for the lower and upper line ends, even though the scotoma occluded the lower and upper ends differently. This finding makes us believe that although the occurrence of perceptual line shortening results from scotoma overlap of the right line, eccentricity in the visual field is the predominant factor in determining the amount of this perceptual illusion. Considering that both filling-in and spatial distortion phe-nomena have been interpreted as an expression of cortical plasticity and that plasticity in the visual system is related to changes in the receptive fields (20), larger receptive fields in the periphery of the visual map could explain why perceptual shortening is more pronounced with increased eccentricity. Different neural mechanisms could be involved in such perceptual alterations. It has been postulated that neurons preferentially shift their connections to the nondeprived areas of the visual field (8). It is conceivable that illusory line shortening is related to receptive field (RF) changes induced by the scotoma. RF expansion was originally observed by Gilbert and Wiesel (9) in cats with focal binocular retinal lesions. These authors noted an immediate increase in RF size for some deafferented cortical cells with receptive fields near the edge of the projected retinal scotoma. After a few months, cortical areas, which were initially silenced by the lesion, recovered visual activity. Therefore, some of the deprived cortical neurons started to become reactivated and acquired new RFs, suggesting that subthreshold inputs reached threshold at formerly unresponsive synapses. This result indicates that long-term mechanisms can induce plasticity changes in initially silenced neurons, probably through long-range horizontal connections (10). Long-range horizontal connections between pyrami-dal cells in the visual cortex allow integration of information over cortical distances that are much larger than the classic model of the RF expansion and may serve as a relay for contextual information between local subsets of cells (10,11). For the RFs in the scotoma to shift, the horizontal connections must be strengthened. They do this by sprouting axon collaterals and by synaptogenesis, whereby clusters within the existing framework of horizontal connections are reinforced by adding collaterals and synaptic buttons (12). Various molecular, cellular, and physiological mechanisms play a role in visual ‘‘rewiring'' and plasticity of cortical networks (13). Whereas changes in cortical RFs have initially been described in retinal lesions, more recent studies (14,15) have shown similar changes in RFs of cortical neurons after cortical lesions. Production of focal visual cortex lesions in adult cats (14,15) has yielded an increase in the spatial extent of the RF of neurons located at the border of the lesion after a training procedure. It is therefore reasonable to postulate that RF expansion could be involved in perceptual spatial distortion reported by our patients. Larger changes in RF size could also be related to feedback or feedforward projections from other brain structures to the visual cortex. Indeed, neurons in deep areas of V1 project intracortically, as well as subcortically, to several structures such as the pulvinar and the lateral geniculate nucleus of the thalamus (LGN), superior colli-culus, and claustrum (16). Thus, neurons in these brain areas receive signals from the reorganized V1 region but interpret them according to the original retinotopic map 41 J Neuro-Ophthalmol, Vol. 29, No. 1, 2009 Mavrakanas et al as if the signals originated from visual input within the scotoma. Such mislocalization of visual signals might cause perceptual filling-in as well as distortion of visual space. Perceptual Filling-in Patients reported that the right line appeared un-interrupted or only slightly blurred in the scotoma area. The line completion can be attributed to perceptual interpolation that allowed visual stimuli to be perceived where there was actually no visual input (17). Similar results are obtained by the use of Amsler grids in patients with central scotomas due to macular disorders. In these patients, the area of alteration in the grid pattern is much smaller than the surface of the scotoma delineated by perimetry (18). We did not develop this specific point further, as we already had done so in earlier studies (6,18,19). Time Course Line shortening appeared 5-10 seconds after steady fixation was started but evolved during the following 10 seconds. These observations are in accord with our previous findings in patients with homonymous paracentral scotomas (6). Specifically, in one of our previous studies (6), in which patient NG was also tested, filling-in of the scotoma appeared 5-10 seconds after initiation of steady fixation. In the current experiment, the perceptual filling-in phenomenon was not specifically investigated, because during the testing procedure the patient was not able to pay attention and describe simultaneously two distinct percep-tual phenomena. As a result, in the single patient in whom the two phenomena were investigated, a similar delay was observed in both perceptual filling-in and spatial distortion. Additional investigations using discrete stimuli (points rather than lines) could determine to what extent the two phenomena are linked. As noted above, changes in visual afferents result in cortical reorganization, inducing perceptual alterations such as filling-in and spatial distortion. Filling-in has been widely studied and two different types, instantaneous and delayed, have been proposed, occurring after prolonged fixation (20). Rapid appearance of these phenomena can be attributed to structural changes after long-standing scoto-mas, whereas delayed distortion essentially happens as a result of functional and synaptic adaptation. Similar delays in the occurrence of perceptual visual alterations have been described in patients with cerebral metamorphopsia. Progressive distortion of images shown to these patients takes place a few seconds after steady fixation. Mechanisms implied in these phenomena are probably responsible for the delay of spatial distortion found in this study. In this study, we have presented a simple way to quantify spatial distortion in the visual field after cortical lesions. Our observations show that perceptual spatial distortion extends up to 20° from the border of the scotoma and is more pronounced as eccentricity from the scotoma border increases. These observations probably reflect ‘‘long-range'' cortical reorganization after brain damage. Additional investigations could broaden our understanding of mechanisms involved in this perceptual visual alteration. REFERENCES 1. Kapadia MK, Gilbert CD, Westheimer G. A quantitative measure for short-term cortical plasticity in human vision. J Neurosci 1994;14: 451-7. 2. Ramachandran VS, Gregory RL. Perceptual filling in of artificially induced scotomas in human vision. Nature 1991;350:699-702. 3. Gerrits HJ, Timmerman GJ. The filling-in process in patients with retinal scotomata. Vision Res 1969;9:439-42. 4. Nelles G, Widman G, de Greiff A, et al. Brain representation of hemifield stimulation in poststroke visual field effects. Stroke 2002; 33:1286-93. 5. Gilbert CD. Circuitry, architecture, and functional dynamics of visual cortex. Cereb Cortex 1994;5:373-86. 6. Safran AB, Achard O, Duret F, et al. The ‘‘thin man'' phenomenon: a sign of cortical plasticity following inferior homonymous paracentral scotomas. Br J Ophthalmol 1999;83:137-42. 7. Caine D, Watson JD. Neuropsychological and neuropathological sequelae of cerebral anoxia: a critical review. J Int Neuropsychol Soc 2000;6:86-99. 8. Cynader M. Perspectives: neuroscience. Strengthening visual connections. Science 2000;17:1943-4. 9. Gilbert CD, Wiesel TN. Receptive field dynamics in adult primary visual cortex. Nature 1992;12:150-2. 10. Gilbert CD, Wiesel TN. Morphology and intracortical projections of functionally characterized neurones in the cat visual cortex. Nature 1979;12:120-5. 11. Gilbert CD, Wiesel TN. Columnar specificity of intrinsic horizontal and corticocortical connections in cat visual cortex. J Neurosci 1989; 9:2432-4. 12. Darian-Smith C, Gilbert CD. Axonal sprouting accompanies func-tional reorganization in adult cat striate cortex. Nature 1994;21:737-40. 13. Horng SH, Sur M. Visual activity and cortical rewiring: activity-dependent plasticity of cortical networks. Prog Brain Res 2006;157: 3-11. 14. Eysel UT, Schweigart G. Increased receptive field size in the surround of chronic lesions in the adult cat visual cortex. Cereb Cortex 1999;9:101-9. 15. Schweigart G, Eysel UT. Activity-dependent receptive field changes in the surround of adult cat visual cortex lesions. Eur J Neurosci 2002;15:1585-96. 16. Komatsu H. The neural mechanisms of perceptual filling-in. Nat Rev Neurosci 2006;7:220-31. 17. Ramachandran VS. Blind spots. Sci Am 1992;266:86-91. 18. Achard OA, Safran AB, Duret FC, et al. Role of the completion phenomenon in the evaluation of Amsler grid results. Am J Ophthalmol 1995;120:322-9. 19. Safran AB, Landis T. The vanishing of the sun: a manifestation of cortical plasticity. Surv Ophthalmol 1998;42:449-52. 20. Spillmann L, De Weerd P. Mechanisms of surface completion: perceptual filling-in of texture. In: Pessoa L, De Weerd P, eds. Filling-in: From Perceptual Completion to Cortical Reorganization. Oxford, UK: Oxford University Press; 2003:81-105. 42 q 2009 Lippincott Williams & Wilkins |
