Journal of Neuro-Ophthalmology, September 1997, Volume 17, Issue 3

Postgeniculate Afferent Visual System and Visual Higher Cortical Function 1995-1996

Journal of Neuro- Ophthalmology 17( 3): 209- 217, 1997. © 1997 Lippincott- Raven Publishers, Philadelphia Postgeniculate Afferent Visual System and Visual Higher Cortical Function, 1995- 1996 Michael Wall, M. D. In this review of articles published in 1995 and 1996, advances in our knowledge of the postgeniculate visual pathways, optic radiations and visual cortical areas VI through V5, are discussed. Syndromes of higher visual cortical disturbances are reviewed. The article is initially organized anatomically by flow of sensory information ( Fig. 1) and later by topic. Basic studies continue to clarify the parallel process­ing of information within the parallel pathways of the sensory visual system. At least two major streams of information flow within the visual sensory system ( Fig. 1). As noted in the last review, this figure is an oversim­plification: there is considerable overlap of the functions ascribed to the two major systems. The small fiber, slower conducting P system originates in small retinal ganglion cells and travels by small axons to synapse in the parvocellular layers of the lateral ge­niculate nucleus ( LGN). Fibers then course through the optic radiations to synapse in the 4c( 3 layer of area VI. Secondary projections pass ventromedially through area V2 to synapse in area V4 of the inferior temporal cortex. This pathway is important for information processing related to fine spatial resolution, color, size discrimina­tion, and fine stereopsis; hence, object recognition. Le­sions of area V4 cause prosopagnosia and cerebral ach­romatopsia. The large fiber, faster conducting M system synapses in the magnocellular layers of the LGN. Fibers arise in the LGN and pass through the optic radiations to synapse in the 4ca layer of the occipital cortex ( VI). Fibers then project bypassing area V2 and flow through cortical area V3. They then course dorsolateral^ in the temporal lobe to area V5. This pathway is important for perception of motion and flicker at high temporal frequencies. Lesions of area V5 are associated with Balint's syndrome. Publications in 1995 and 1996 further define the hu­man cortical visual areas. Sereno et al. ( 1) used func- Manuscript received May 1, 1997; accepted June 23, 1997. From the Department of Neurology, University of Iowa Hospital and Clinics, Iowa City, Iowa. Address correspondence and reprint requests to Dr. Michael Wall, University of Iowa Hospital and Clinics, Department of Neurology, Iowa City, IA 55242. tional magnetic resonance imaging ( MRI) to map the borders of human visual areas VI, V2, V3, V4, and VP during phase- encoded retinal stimulation. Sereno et al. found evidence for cortical magnification factor in both striate and extrastriate cortical areas. ( Cortical magnifi­cation factor is a scaling factor that describes the pro­gressive increase in cortical area devoted to visual pro­cessing as one moves from the periphery to the central visual field. Multiplication by the cortical magnification factor produces equal cortical stimulus areas from the different retinal locations.) Importantly, Sereno et al. documented many parallels between human, macaque, and owl monkeys with regard to the location of these visual cortical areas. Their many color figures are in­structive. They confirmed that area VI in humans is shifted medially around the occipital pole when com­pared with the location in subhuman primates and dem­onstrated the human extrastriate areas to be relatively larger than those of the monkey. Figure 1 shows the proposed segregation of the paral­lel pathways of human vision. In recent years this simple dichotomy of parallel processing has been continually challenged. Although early visual processing is charac­terized by two independent parallel pathways- the mag­nocellular and parvocellular streams- Sawatari and Cal­loway ( 2) documented some convergence of these two pathways. Using biocytin labeling with laser scanning photostimulation and intracellular recording, these inves­tigators found that most striate cortex 4B neurons re­ceived strong input from both magnocellular and parvo­cellular neurons. Thus, the classic figure published herein requires modification: the magnocellular stream appears to have significant parvocellular input. Shipp ( 3) reviewed this relationship of two distinct visual pathways and concluded that we should not think of the functions ascribed to the parvocellular pathway as pure because there is no part of the extrastriate cortex outside VI does the parvocellular system maintain ex­clusive access. However, Shipp pointed out that the mag­nocellular pathway outside of VI does appear to have exclusive access. He adds that the W pathway ( subserved by W retinal ganglion cells) also may be important as a third parallel stream of information. Shipp noted that the W system is represented by small neurons found within 209 210 M. WALL wru- nr XJfT lii^ litr Yifiuul urvto I ••' ••• fdm i UIILLLATLM ' I I'riniiiry \ iMlill O n i f < c Fig. 1. Functional segregation in the primate visual system ( from Livingstone67). Overlap in the functions of these systems is a theme found in the 1994 literature. the gaps between the M and P layers of the lateral ge­niculate nucleus. These cells have large receptive fields and long latencies and therefore have intermediate prop­erties between the parvocellular and magnocellular sys­tems. Others have proposed that these W cells may be important in the pupillary light reflex. Shipp continued by suggesting that the close proximity of the two path­ways, when they do appear to be separate, may promote or facilitate their interaction. He concluded that segrega­tion is an efficient way of coding for a feature, whereas interaction is important for association of the various features. Bullier and Nowak ( 4) reviewed the interactions be­tween these parallel streams of visual information from a different viewpoint. They pointed out that the hierarchi­cal processing model of Hubel and Weisel- which re­quires serial processing within distinct parallel streams- is flawed: it is based solely on anatomical observation ( 5). Functional studies such as those described above provide little support for this type of hierarchical for­ward- flowing serial processing. Buller and Nowak noted that although lesions of VI silence higher ( extrastriate visual) areas, lesions of cortical area V2 do not. They argued that the hierarchical serial processing model that predicts latencies of higher order areas in a hierarchical system should be longer than those of lower order ones. However, what is found is that the magnocellular system carries information much faster than the parvocellular system, and this information reaches area V2 in the mag­nocellular stream before parvocellular information des­tined for V2 has left V1. More importantly, it has been shown that the average latency of area V5 is 9 ms longer than that of VI and that the shortest latencies in area V5 are even shorter than those in VI or V2 ( 6). Lastly, Buller and Nowak pointed out that although there is par­allel organization up to VI in the extrastriate pathways, there is much evidence for distributed processing. In other words, there is not one but several cortical regions specifically activated by a stimulus feature, be it color, motion, or form. It is believed that through interactions between these modules, perceptual decisions are made. A major problem with automated perimetry is vari­ability. This variability is due to a combination of intrin­sic subject factors, stimulus properties, test mechanics, and testing algorithm. The intrinsic subject factors are attention, training of patient and technician, motivation, and " noise" generated by the sensory visual system neu­rons. Several articles have investigated the neural noise of the sensory visual system. It is known that visual pro­cessing reduces noise, that is, there is more noise when recording from neurons than there is with the actual sub­ject's responses. What is the purpose of this neural noise? Stemmler et al. ( 7) proposed an answer in their article entitled " Lateral Interactions in Primary Visual Cortex: A Model Bridging Physiology and Psychophys-ics." They postulated that lateral interactions are impor­tant for two related but opposing effects. One effect oc­curs when the visual scene is homogeneous except for one feature: the psychophysical effect of feature " pop-out." Here the effect of visual context is to suppress weak signals so that the strong signal is quickly pro- J Neuro- Ophthahnol, Vol. 17, No. 3, 1997 POSTGENICULATE AFFERENT VISUAL SYSTEM AND VISUAL HIGHER CORTICAL FUNCTION 211 cessed and pops out. Second, in contrast, if the visual scene is composed of a smudged photograph, the visual system uses neighboring visual landmarks to complete parts of the scene. Here enhancement of the weak stimuli, rather than suppression of the strong stimuli, occurs. Stemler et al. explained this phenomena in terms of the classic receptive field center- surround organiza­tion. They suggested that for stimuli that " pop out," the addition of similar stimuli outside of the classic receptive field leads to a suppression of the response. Conversely, when subthreshold stimuli are present inside the classic receptive field ( the smudged photograph), the addition of multiple similar stimuli in the surround produces a weak increase in the response. Their model suggests that noise increases the firing rate as a function of contrast and via lateral interactions in primary visual cortex and that neu­ral noise enhances both pop- out and contour completion. The investigators provide ample evidence in support of their model. Arilei et al. ( 8) also studied the effects of neural noise but on the evoked activity of mammalian cortex and the resulting behavioral responses. They documented large variability in these evoked responses to repeated presen­tations of the same stimulus. Using real- time optical im­aging in cat primary visual cortex, these investigators found that the coherent ongoing activity was often as large as the evoked activity. In addition, they reported that the evoked activity was highly correlated to the ini­tial state of the system. Thus, when the initial state had a low noise level and lower activity, there was not much likelihood of a response occurring. When the noise and activity level was high, a response occurrence was likely and predictable. They concluded that the processing of sensory input in the visual cortex involves the combina­tion of a deterministic response coupled with the state of the ongoing neural network dynamics. MOTION PERCEPTION ( AREA V5, AREA MT) In the last two reviews, data were presented suggesting that area V5 is the human homologue of area MT of the monkey. The evidence suggested that the location of this motion- processing center was the junction of the inferior temporal and lateral occipital sulci. Significant advances have been made in the past 2 years, further clarifying the location of this important motion- processing center in humans. Tootell and Taylor ( 9) stained human visual cortex for myelin, cytochrome oxidase, and the monoclonal anti­body CAT- 301 in an attempt to anatomically map area V5. The myelin stain was used because the dorsal layers of MT are highly myelinated. CAT- 301 labels MT and many other magnocellular input areas in the monkey. Distinctive cytochrome oxidase staining also occurs in area MT of the monkey. These investigators found in comparing the three stains that human area V5 was an oval region - 1.2 x 2.0 cm located 5- 6 cm anterior and dorsal to the foveal Vl- 2 border. This anatomic local­ization correlates well with the physiologic site of func­tional activity in humans. In a related study, Tootell et al. ( 10) used functional MRI to reinvestigate retinotopy in area V5. They dem­onstrated evidence for retinotopy in V1 but much less in area V5. Their evidence for functionally localizing V5 corresponds to their anatomic evidence. These findings localizing V5' s anterolateral occipital location have been continued by Howard et al. ( 11), Anderson et al. ( 12), and Barton et al. ( 13) using both functional MRI and magnetoencephalography. Other evidence for the human homologue of area MT comes from lesion analysis in patients. Barton et al. ( 14) studied motion direction discrimination in peripheral and central vision in 23 patients with unilateral cerebral hemispheric lesions documented by computed tomogra­phy ( CT) or MRI. They found that five of the 23 patients with focal cerebral lesions had an ipsidirectional defect in motion direction discrimination suggesting a lesion in the area V5. The lesions of these patients overlapped in white matter underlying the lateral temporal occipital cortex at the junction of Brodmann areas 19 and 37. Patients without directional deficits did not have involve­ment of this region. Rizzo et al. ( 15) restudied patient LM, a famous pa­tient first reported by Zihl et al. in 1983. This stroke patient had a severe disturbance of movement vision af­ter damage to the posterior region of both hemispheres. The lesions involved dorsolateral visual association cor­tex and spared striate cortex. They found, using random dot motion stimuli, that LM could perceive global co­herent motion and could discriminate motion direction but these abilities failed at moderate levels of added background noise. The patient could also perceive two-dimensional shape and three- dimensional structure from motion, however, these abilities again were disrupted at moderate levels of noise in the display. They suggested that the role of area V5 may be to reduce the effects of stationary and moving noise to " sharpen" our picture of the world. Nawrot and Rizzo ( 16) studied 16 patients with acute cerebellar lesions with motion direction discrimination testing. They found deficits in motion direction discrimi­nation with acute midline cerebellar lesions. The deficit was similar to those reported after cortical area V5 le­sions in primates. They hypothesized that the motion perception deficit resulted from damage to a cerebellar mechanism involved in perceptual stabilization even though visual cortical motion processing mechanisms re­mained intact. Ffytche et al. ( 17) studied the parallel visual motion inputs into area VI and V5 of human cerebral cortex using visually evoked electroencephalography ( EEG) coupled to magnetoencephalography. They found that a movement stimulus with the speed of 22 degrees per second induced an increase in area V4 activity before it was found in area V1. With speeds slower than 6 degrees per second, signals arrived first in area VI. They con­cluded that in addition to the classical hierarchical input to V5 through VI there is also a fast parallel input that bypassed VI. The investigators suggested that this may explain the residual motion perception of patients with lesions of VI or V5. .1 Neuro- Ophlhalmol, Vol. 17, No. J, 1997 212 M. WALL It has been suggested that dyslexics have a selective deficit in the magnocellular system on the basis of both psychophysical and electrophysiologic evidence. In ad­dition, postmortem examination of the lateral geniculate nucleus in brains of dyslexic patients shows a decrease in magnocellular neuron size with normal parvocellular structure. Eden et al. ( 18) used functional MRI to study six dyslexic patients and eight controls. In all the dys­lexics, motion stimuli failed to produce the same task-related functional activation in area V5 that was found in controls. Presentation of stationary patterns resulted in equivalent activations in VI and V2 in extrastriate cortex in both groups. They concluded that this deficit in sys­tems that process temporal properties of stimuli can manifest as disorders of phonologic awareness, rapid naming, rapid visual processing, or motion detection. In addition, this technique may serve as a biologic marker of dyslexia ( 19). For the reader who seeks a more basic review of recent advances in visual motion perception, Albright and Stoner ( 20) provided an excellent and detailed article about current motion mechanisms and models. COLOR PROCESSING ( AREA V4) Area V4, the purported color center of extrastriate cor­tex, also functions in visual form discrimination as dis­cussed in the annual reviews of the past 2 years. Fur­thermore, insight was acquired from the works of Meri-gan ( 21), who made ibotenic acid lesions in four macaque monkeys in the region of cortical area V4 that corresponded to the lower quadrant of one hemifield. He found a twofold reduction of luminance contrast sensi­tivity and red- green chromatic contrast sensitivity using stationary contrast gratings as a stimulus. However, there was little or no loss found for contrast sensitivity detec­tion or direction discrimination with 10 Hz drifting grat­ings, nor was there a change in visual acuity. Hue and luminance matching could not be learned in the visual field locus corresponding to the V4 lesion but could be learned in the other quadrants. In contrast, if the animal was previously trained in the hue or luminance matching task, this capability was not damaged by the lesion. Merigan found major effects on form discrimination tasks such as discriminating the orientation of collinear groups of dots on a background of randomly placed dots and discriminating the orientation of a group of three line segments surrounded by line segments of a different ori­entation. Therefore, in addition to color processing, area V4 is of major importance for form discrimination. Heywood et al. ( 22) studied cerebral achromatopsia in monkeys. In human cases of cerebral achromatopsia, ex­trastriate cortical damage produces a severe or complete loss of color vision with relative sparing of luminance vision. The critical lesion in these cases appears to be a medial occipitotemporal one, involving the lingual and caudal fusiform gyri. These investigators argued that positron emission tomography ( PET) investigations in humans have shown this cortical region to be one of several that are activated in normal humans during color vision tasks. Attempts to find analogous regions in mon­key cortex have been unsuccessful. They stated that de­struction of cortical area V4 in the monkey produced mild impairments in color discrimination. They therefore tested the color vision of monkeys after cortical ablations that spared area V4. Heywood's group found that mon­keys with ablations in the temporal lobe anterior to area V4 had a produced color vision impairment with relative sparing of luminance vision. They concluded that the monkeys' behavior was indistinguishable from that of a human patient with total cerebral achromatopsia tested on the same tasks. Their data suggest that area V4 in macaque monkeys is probably not homologous to the human color center. Instead, the area of the monkey brain corresponding to color area in the humans may be anterior to area V4. CORTICAL VISUAL PROCESSING Frances Crick, of Watson and Crick fame, has more recently become interested in mechanisms of visual awareness ( 23). He has speculated that cortical layers V and VI express the results of computations from higher layers. Activity in layers V and VI produced a reverber­ating circuit from layer VI back to the thalamus. Using such a feed- forward connection, the thalamus may syn­chronize various sets of cortical processing areas. There­fore, the thalamus could be a structure that " binds" fea­tures of objects, possibly by coordinated firing in the 40- Hz range, although Crick voiced many reservations about this model ( 23). Mounting data in support of this model of visual awareness are becoming available. Gray and McCormick ( 24) identified a subset of neu­rons in the striate and prestriate cortex that displays syn­chronous rhythmic firing in the 20- to 70- Hertz range. They identified the source of this activity in " chattering cells," a class of pyramidal neuron. Crick and Koch ( 25) reviewed the neuroanatomic data in the macaque and psychophysical evidence in the humans and made a strong argument that humans are not aware of neural activity taking place in striate cortex. They cited evi­dence that there have been no direct connections dem­onstrated between striate cortex and frontal areas to act as a substrate for awareness. This is unlike higher visual areas such as areas V2 through V5 that do project to frontal locations. Psychophysical^, they noted that it has now been shown that stimuli of which we are unaware, because they have detail too fine for us to see, can in­fluence the activity of higher cortical areas. More support for high- frequency oscillations to ex­plain the " binding problem" ( of features of visual stimuli) comes from Brosch et al. ( 26) These investiga­tors studied area 18, akin to area V2, of the cat visual cortex. They found correlations of multiunit activity in the frequency range of 35- 80 Hertz that were related to the response properties of cortical neurons. This suggests that correlated oscillatory activity provides a potential neural code to support sensory information processing. Further evidence for area V2 playing a major role in binding color, motion, and form within an object comes J Neiiro- Oplilhalmol, Vol. 17, No. 3, 1997 POSTGENICULATE AFFERENT VISUAL SYSTEM AND VISUAL HIGHER CORTICAL FUNCTION 213 from the work of Gegenfurther et al. in an anatomic/ physiologic study ( 27). For many years, neural activity has been studied with basic spike frequencies. Analysis of the temporal pattern of the spike train, or spike timing, is now receiving analysis. Mainen and Sejnowski ( 28) have shown that rat neocortical neurons, like neurons in the peripheral audi­tory system, can encode information on the basis of the timing of the individual spikes. This lends support to the " multiplex filter hypothesis" of sensory information processing ( 29), which states that information is multi­plexed like an FM signal onto neurons based on its tem­poral sequence and is decoded at higher levels of the visual system. It has been shown that within minutes after making a retinal lesion, receptive field size can increase substan­tially. This has led to a new view of the receptive field that is characterized by response specificity highly influ­enced by feature context. The receptive field is therefore dynamically changed by sensory experience. Vochan and Gilbert ( 30) studied this dynamic receptive field effect by creating artificial scotomas and measuring the size of the related cortical receptive fields. Specifically, they looked at the intraocular transfer effect in binocular re­ceptive fields. They showed that the corresponding re­ceptive field expansion is due to mechanisms intrinsic to visual cortex. Anatomic support for this conclusion also comes from a recent article by Darian- Smith and Gilbert ( 31). Chino et al. ( 32) also studied dynamically changing receptive fields and found that newly activated units have normal orientation tuning, direction selectivity, and spatial frequency tuning. However, contrast thresholds of most of the related neurons were elevated and their maxi­mum response amplitude was reduced. Kujala et al. ( 33) provided a provocative study of the plasticity within the visual system. They studied five blind humans, deprived of visual input since early in­fancy, with magnetic recordings. They recorded mag­netic responses to pitch changes in a sound sequence while subjects either counted the changes or ignored the stimuli. Interestingly, in these subjects the magnetic re­sponses to pitch changes were located in visual and tem­poral cortices, whereas ignoring pitch changes activated only the temporal cortex. The investigators concluded that the visual cortex of blind humans can participate in auditory discrimination. HEMIANOPIAS AND CORTICAL BLINDNESS Aulhorn introduced noise field campimetry after an observation by one of her patients that while looking at a television screen of white noise he was able to observe and draw his visual field defect ( 34,35). Kolb et al. ( 36) studied scotoma perception using white noise field campimetry in patients with postchiasmal visual pathway lesions. They examined 59 patients with homonymous hemianopia with white noise field campimetry and con­ventional perimetry. Eighteen of the 56 patients ( 32%) were incapable of perceiving any white noise field sc­otoma. These patients all had lesions in their optic ra­diations. The remaining patients had lesions of striate cortex. The investigators concluded that white noise field campimetry was insensitive to visual field defects due to the lesions of the optic radiations but was sensitive for those due to lesions of striate cortex. They explained their findings by the necessity of intact striate cortex ( VI) function for the filling in phenomena. Cunningham et al. ( 37) examined 15 women over a 14- year period with cortical blindness due to preeclamp­sia. The blindness lasted from 4 hours to 8 days and resolved in all patients. Of the 13 women who underwent CT, eight had low- density lesions localized predomi­nantly in the occipital lobes. Five of these patients un­derwent MRI and two showed corresponding hyperin-tense lesions in the occipital area. Joseph and Louis ( 38) reported a case of transient ictal cortical blindness. They culled 44 cases from the litera­ture and divided them into three groups: ( a) patients with prolonged seizures or status epilepticus who likely had cellular damage of the neural parenchyma causing blind­ness, ( b) those with ictal blindness in association with bilateral bursts of seizure activity or persistent spike and wave activity in the occipital lobes, and ( c) those with focal seizure activity originating in or adjacent to an occipital lobe. In this group, the blindness was either ictal or postictal. It is well known that the effect of a cortical lesion is not only manifest in the function ascribed to the area damaged, but dysfunction also occurs related to long-range interactions or disconnection of structures influ­enced by the input. Rizzo and Robin ( 39) reported an example of this. They found bilateral effects of unilateral visual cortex lesions in humans. They studied 12 patients with unilateral lesions of the visual cortex with corre­sponding visual field defects. In addition to the expected homonymous hemianopic deficit, their patients also had a small defect in the " spared" homonymous visual field. Their study demonstrated that this deficit was task de­pendent and could be explained as a global reduction in the capacity of visual attention. BLINDSIGHT Blindsight is thought to be residual visual function of which the subject is not consciously aware after striate cortex damage. In the last annual review of higher cor­tical function, a new standard for cases of blindsight was discussed that required image stabilization to eliminate eye motion artifacts while testing subjects' vision ( 40). This study suggested that some if not all cases of blind-sight in humans were due to the presence of small re­sidual islands of vision remaining. Wessinger et al. ( 41) looked for residual vision with­out awareness in the homonymous visual field contralat­eral to a partial or complete functional hemispherectomy. They used image stabilization to eliminate eye move­ment artifacts and found a narrow zone of retained vision along the vertical meridian of the visual field of each patient. The lateral edge of the zone was generally within J Neiiiv- Oplillialiniil. Vol. 17. No. J, 1997 214 M. WALL 3.5 degrees of the vertical midline. It extended outward toward the periphery but not beyond 6 degrees at any one field location in each subject. Within their zones of re­sidual vision, the patients could detect stimuli and per­form simple shape discriminations. They were aware of their vision within this zone. No residual vision with or without awareness was found in areas tested outside these zones. The investigators attributed this finding of vision along the vertical meridian to a zone of nasal-temporal overlap as reported by developmental studies in the cat and monkey. This explanation fits their observa­tions much better than attributing the findings to blind-sight. The existence of blindsight in humans has not yet been definitively proven. Other studies suggesting blind-sight in humans were not reviewed here because image stabilization techniques were used. Blindsight in monkeys was studied by Cowey and Sto-erig ( 42). They investigated four macaque monkeys, three of whom had ablation of their left striate cortex along with complete severance of the splenium of the corpus callosum several years before the experiments. MRI confirmed that the ablation of striate was complete. The monkeys then were trained to touch a screen at the position where a target had appeared. Although their thresholds were elevated, the monkeys could detect stimuli in the hemifield related to the ablated hemi­sphere. However, when the monkeys were asked to clas­sify stimuli as to whether actual stimuli had been pre­sented or a blank screen had been shown, they correctly identified all the blank screens. The investigators inter­preted this as a demonstration of visual abilities in the blind hemifield of the monkey without awareness. VISUAL HALLUCINATIONS AND OCCIPITAL SEIZURES Charles Bonnet, a Swiss philosopher- naturalist, in 1769 reported a case of visual hallucinations occurring in an elderly person. He gave an account of a patient's hallucination of a pleasant or neutral nature that occurred after visual loss. The hallucinations were described as vivid and detailed. The patient had sufficient insight to realize that the hallucinations were not real. Cogan called these " release hallucinations." Although this is the term used most often by neuroophthalmologists, this phenom­enon is most commonly referred to in the literature as the " Charles Bonnet syndrome." Many articles using this moniker have appeared over the past 2 years. Tueth et al. ( 43) reported two cases of Charles Bonnet syndrome. In both cases, ocular pathology was present. Maricele et al. ( 44) reported a case of a woman who tested positive for the human immunodeficiency virus but who had no other organic or psychiatric manifesta­tions. Teunisse et al. ( 45) sought to determine the preva­lence of Charles Bonnet syndrome in low- vision patients. They gave a semistructured interview about visual hal­lucinations to 300 adult low- vision patients and 200 el­derly general ophthalmic patients. Their diagnostic cri­teria for the Charles Bonnet syndrome were the presence of complex persistent or repetitive visual hallucinations with full or partial retention of insight. They required that no hallucinations were present in other modalities and no delusions were present. They found that Charles Bonnet type visual hallucinations were present in 11% of low-vision patients. They were significantly associated with age over 64 years and visual acuity in the best eye of 20- 60 or less. They concluded that Charles Bonnet syn­drome was associated with sensory deprivation and ad­vanced age. Adachi ( 46) reported a case of the Charles Bonnet syndrome in leprosy related to multisensory deprivation. Reyes- Ortiz et al. ( 47) described Charles Bonnet syn­drome in a 103- year- old woman with glaucoma in the right eye and cataract in the left eye. Occupational therapy and promoting communication and was said to reduce the frequency of the hallucinations. Pliskin et al. ( 48) suggested that Charles Bonnet syndrome may be an early marker for dementia. Cassinello et al. ( 49) reported visual hallucinations after medication with oral mida­zolam. Kolev ( 50) found elementary type visual hallucina­tions evoked by caloric vestibular stimulation in normal humans. The investigator studied 37 healthy subjects of both sexes from the ages of 18 to 40 years with normal vision selected at random from a population of normal individuals. After caloric stimulation, 25 of the 37 sub­jects had visual hallucinations. Their latency was - 30 s and they lasted for about 20- 30 s. Single images ap­peared in 18 of the 25 subjects, and 10 detected a set of images. The images were usually centrally located. Three subjects saw the hallucinatory image with one eye only. Various geometric shapes were reported. Moving images occurred in the same direction as the self motion perception in seven subjects and were independent of the direction of self motion in three subjects. Various poten­tial mechanisms are suggested, but none stand out as a good explanation for the phenomena. Walker and colleagues ( 51) reported an interesting case of a 31- year- old man with occipital lobe epilepsy misdiagnosed as migraine. The patient had dull right frontal headaches associated with shimmering ellipsoid silver lights " like sunlike shimmering on the sea." The positive phenomena lasted a fraction of a second and occurred only in the left visual field. They initially re­curred every hour and then increased in frequency to every 2 min. The visual phenomena were associated with seizures at 3 years of age but had been controlled for 5 years before the diagnosis of migraine. The perception of the flash of light in the left visual field corresponded to a single sharp/ slow- wave discharge over the right occipi­tal lobe. During the aura, right occipital lobe epileptiform activity occurred. The patient also had an abnormal re­sponse to photic stimulation with occipital lobe dis­charges during 3- to 4- Hz stimulation that was time locked to the stimulus. MRI showed that the right hemi­sphere and right caudate nucleus were smaller than the left with an abnormal gyral pattern noted over the right parietal region. To establish the nature of their perceptual hallucina- J Neuro- Ophthalmol, Vol. 17. No. 3, 1997 POSTGENICULATE AFFERENT VISUAL SYSTEM AND VISUAL HIGHER CORTICAL FUNCTION 215 tions associated with alcohol- induced delirium tremens, Platz et al. ( 52) studied 64 patients with a semistandard-ized investigation. Visual hallucinations were most com­mon followed by auditory and then tactile hallucinations. Hallucinated cats, dogs, and snakes were most frequent. The classic " white mice" hallucination was only re­ported in one instance. The visual hallucinations oc­curred in 94% of the patients. One patient appropriately hallucinated 10 bottles of vodka. Sanchez- Ramos studied visual hallucinations in 214 consecutive patients interviewed during routine visits to a Parkinson's disease clinic. The hallucinations were present in one of four patients. Dementia, age, duration of disease, history of depression, and history of sleep disorder were strongly correlated with the hallucinations. Vaphiades et al. ( 53) gave a questionnaire to 32 pa­tients with ischemic infarction of the retrochiasmal vi­sual pathways. Thirteen patients ( 41%) reported positive spontaneous visual phenomena in their blind hemifield. These positive phenomena were phosphenes, photopsias, visual hallucinations, and, less commonly, palinopsias. They were not associated with other sensory hallucina­tory phenomena as long as the ischemic infarction was uncomplicated. The various types of positive visual phe­nomena were not predictive of the neuroimaging lesions. Larger lesions destroying anteriorly located visual asso­ciated areas were not associated with the development of positive visual phenomena. Nagaratnam et al. ( 54) reported an 84- year- old woman with a right occipital meningioma who had both musical and visual hallucinations. The visual hallucinations were formed and the musical hallucinations were ringing bells of the same Dutch Christmas carol. This case, like other cases with musical hallucinations, had a right- sided le­sion. MIGRAINE Liu et al. ( 55) reported ten patients with migraine who developed persistent positive visual phenomena lasting months to years. The positive phenomena usually con­sisted of a few small particles and involved the entire visual field. Examples of the particles were television static, snow, lines of ants, dots, and rain. Seven of the patients had typical migraine with aura before the con­tinuous phenomena developed. Results of neurologic, neuroimaging, and electroencephalographic examina­tions were normal, except in one patient with a nonspe­cific biparietal white matter lesion and another with a small venous angioma. Treatment of these symptoms was unsuccessful. A subsequent letter to the editor re­ported success in two patients who had persistent posi­tive visual phenomena using valproate ( Depalcote) ( 56). The investigators speculated that their patients' positive visual phenomena resulted from spontaneous cortical discharges. Wray et al. studied 12 migraine patients with aura and 12 age- matched control subjects between their migraine attacks. They found that the migraine group had a faster response time at two low- level visual processes: orien­tation detection and temporal order judgment. High- level processes of picture naming and word priming showed no differences between the groups. The authors con­cluded that the migraneurs' apparent response time ad­vantage in the low- level tasks provided a psychophysical corroboration to their known enhanced sensitivity to vi­sual stimuli. PALINOPSIA Palinopsia is usually thought of as a persistence of a visual image that was just seen. Occasionally, investiga­tors lump palinopsia together with visual perseveration. In the latter, the image was recently but not necessarily immediately or just seen. Kawasaki and Purvin ( 57) re­ported persistent " palinopsia" ( visual perseveration or persistent recurring visual images not just seen but seen fairly recently) after ingestion of LSD in three patients. Results of neuro- ophthalmologic and neurologic exami­nations and electrophysiologic studies were normal. The investigators noted that LSD closely resembles seroto­nin. They cited a recent theory proposing that LSD binds and destroys small serotonin- responsive cortical inter-neurons. These neurons have inhibitory output that are usually GABA mediated. They speculated that the visual predominance of LSD- related hallucinations may come from the high density of serotonin receptors in the pri­mary somatosensory cortex. The visual persistence might be related to a failure of the normal inhibitory messages on an activated visual circuit. Miiller et al. ( 58) reported three patients with occipital lesions in the right hemisphere and palinopsia. The pa­tient's EEGs were abnormal, and one patient had a right occipital seizure. The visual activity in the one patient occurred as a seizure phenomenon. In all three patients palinopsia disappeared after anticonvulsant therapy. Lastly, Robert et al. ( 59) reported palinopsia as a mi­graine accompaniment. PROSOPAGNOSIA A study by Takahashi et al. ( 60) of four cases of prosopagnosia lends support to the view that prosopag­nosia can occur from a unilateral lesion of the right hemi­sphere with involvement to the lingual and fusiform gyri. In addition to the three types of prosopagnosia previ­ously reported, these investigators suggested a fourth type in a patient with no deficit either in face perception or in recalling visual images of familiar faces. However, the patient could not identify the faces of family mem­bers, friends, and famous people. He was also incapable of visually distinguishing among fish species that he was familiar with and could tell them apart only by touch. They suggested a disconnection between face processing and face memory as an explanation. Three other studies investigated details of prosopag­nosia. Carlesimo and Kalpagirone ( 61) reported two pa­tients with face agnosia. Their study also lended support to the above anatomic localization. The one patient they studied with PET scanning showed hypometabolism J Neiiro- Ophthalmol, Vol. 17, No. 3, 1997 216 M. WALL confined to the right hemisphere. These investigators noted that their subjects were unable to attribute the cor­rect approximate age to unknown faces. Kosslyn et al. ( 62) evaluated curved and straight ver­sions of various stimuli in a patient with prosopagnosia along with a group of age- and education- matched con­trol subjects. The patient consistently required more time to perceive curved and straight lines relative to the con­trols. In addition, he had a deficit when comparing curved lines that were simultaneously visible along with various other tasks requiring perception of curvature. The patient also made more errors when he named pic­tures of curved objects. The investigators related their findings to the role of the end- stopped cells of striate cortex. Farah et al. ( 63) attempted to answer the question whether the human visual system has a specialized sys­tem for facial recognition not used for the recognition of other objects by using the " face inversion effect." This effect is the loss of a normal proficiency of face percep­tion when faces are inverted. They found that a prosopagnosia subject paradoxically performed better at matching inverted than upright faces. They suggested this as evidence of a localized module for upright face recognition in humans. ALEXIA WITHOUT AGRAPHIA ( PURE ALEXIA) Letter- by- letter reading and very slow but mostly ac­curate performance characterize alexia without agraphia ( when this type of reading is present). Patients are greatly slowed by word length and may need 3- 4 s to name or classify common three- letter words. Performance is slowed by - 2- 3 s for every additional letter increase in word length. Bub and Argun ( 64) studied a patient with alexia without agraphia who was able to make rapid and accurate lexical decisions for common words but did not have the ability to fully recover their identity. Unlike most patients with pure alexia, this patient did not need to analyze individual letter forms unless he needed to fully identify a word. The investigators related these findings to a neural network model of lexical processing. Feinberg et al. ( 65) studied an interesting patient with visual agnosia and pure alexia. The patient was a 72- year- old woman with a left occipital infarction who had object agnosia and pure alexia. She was unable to ex­plicitly identify visual stimuli such as objects or words. However, she was able to make reliable judgments on forced choice matching tasks. In other words, the patient could not explicitly demonstrate knowledge that she ob­viously had retained based on the results of the forced-choice matching tasks. The authors concluded that this pattern and performance suggests limited or partial ac­cess to preserved semantic knowledge, which although degraded is not conscious. Price and Humphreys ( 66) studied the reading behav­ior of two pure alexic patients. Both were severely im­paired at reading single words and long words. For both patients, letter identification was better for widely spaced letters than typically spaced letter strings. This was most pronounced for the central letters of the string. Although letter identification was better for widely spaced letters, there were mixed effects for letter spacing with reading. In one patient, widely spaced letters gave rise to im­proved reading; in the other, word reading was disrupted. The investigators attributed this to strategies used by the patients. They proposed a common functional impair­ment that produces both true alexia and simultanagnosia. That is, impaired discrimination of visual features pro­cessed in parallel. Both of the patients studied by these authors also had simultanagnosia. The highlights of advances in higher cortical function over the past 2 years were that receptive fields can change dynamically, even within minutes. No cases of blindsight have yet been proven when small islands of visual function are sought and image stabilization tech­niques are used with the sensory visual testing. Dyslexics have functional MRI abnormalities in area V5 ( the mo­tion- processing center), and persistent positive visual phenomena may respond to valproate. REFERENCES 1. Sereno MI, Dale AM, Reppas JB, et al. Borders of multiple visual areas in humans revealed by functional magnetic resonance imag­ing. Science 1995; 268: 889- 93. 2. Sawatari A, Callaway EM. Convergence of magno- and parvocel-lular pathways in layer 4B of macaque primary visual cortex. Na­ture 1996; 380: 442- 6. 3. Shipp S. Visual processing. The odd couple. Curr Biol 1995; 5: 116- 9. 4. Bullier J, Nowak LG. Parallel versus serial processing: new vistas on the distributed organization of the visual system. Curr Opin Neurobiol 1995; 5: 497- 503. 5. Shipp S, Zeki S. Segregation and convergence of specialised path­ways in macaque monkey visual cortex. J Anat 1995; 187: 547- 62. 6. Raiguel SE, Lagae L, Gulyas B, Orban GA. Response latencies of visual cells in macaque areas VI, V2 and V5. Brain Res 1989; 493: 155- 9. 7. Stemmler M, Usher M, Niebur E. Lateral interactions in primary visual cortex: a model bridging physiology and psychophysics. Science 1995; 269: 1877- 80. 8. Arieli A, Sterkin A, Grinvald A, Aertsen A. Dynamics of ongoing activity: explanation of the large variability in evoked cortical responses. Science 1996; 273: 1868- 71. 9. Tootell RB, Taylor JB. Anatomical evidence for MT and additional cortical visual areas in humans. Cerebral Cortex 1995; 5: 39- 55. 10. Tootell RB, Reppas JB, Kwong KK, et al. Functional analysis of human MT and related visual cortical areas using magnetic reso­nance imaging. J Neurosci 1995; 15: 3215- 30. 11. Howard RJ, Bullmore E, Brammer M, et al. Activation of area V5 by visual perception of motion demonstrated with echoplanar MR imaging. Magn Reson Imaging 1995; 13: 907- 9. 12. Anderson SJ, Holliday IE, Singh KD, Harding GFA. Localization and functional analysis of human cortical area V5 using magneto-encephalography. Proc R Soc Lond 1996; B: 423- 31. 13. Barton JJ, Simpson T, Kiriakopoulos E, et al. Functional MRI of lateral occipitotemporal cortex during pursuit and motion percep­tion. Ann Neurol 1996; 40: 387- 98. 14. Barton JJ, Sharpe JA, Raymond JE. Retinotopic and directional defects in motion discrimination in humans with cerebral lesions. Ann Neurol 1995; 37: 665- 75. 15. Rizzo M, Nawrot M, Zihl J. Motion and shape perception in ce­rebral akinetopsia. Brain 1995; 118: 1105- 27. 16. Nawrot M, Rizzo M. Motion perception deficits from midline cer­ebellar lesions in human. Vision Res 1995; 35: 723- 31. J Neuw- Ophthalmol, Vol. 17, No. 3, 1997 POSTGENICULATE AFFERENT VISUAL SYSTEM AND VISUAL HIGHER CORTICAL FUNCTION 217 17. ffytche DH, Guy CN, Zeki S. The parallel visual motion inputs into areas VI and V5 of human cerebral cortex. Brain 1995; 118: 1375- 94. 18. Eden GF, VanMeter JW, Rumsey JM, Maisog JM, Woods RP, Zeffiro TA. Abnormal processing of visual motion in dyslexia revealed by functional brain imaging. Nature 1996; 382: 66- 9. 19. Frith C, Frith U. A biological marker for dyslexia [ news; com­ment]. Nature 1996; 382: 19- 20. 20. Albright TD, Stoner GR. Visual motion perception. Proc Natl AcadSci USA 1995; 92: 2433- 40. 21. Merigan WH. Basic visual capacities and shape discrimination after lesions of area V4 in macaques. Vis Neurosci 1996; 13: 51- 60. 22. Heywood CA, Gaffan D, Cowey A. Cerebral achromatopsia in monkeys. Eur J Neurosci 1995; 7: 1064- 73. 23. Crick F. 77ie Astonishing Hypothesis; The Scientific Search for the Soul. New York: Scribner's; 1994. 24. Gray CM, McCormick DA. Chattering cells: superficial pyramidal neurons contributing to the generation of synchronous oscillations in the visual cortex. Science 1996; 274: 109- 13. 25. Crick F, Koch C. Are we aware of neural activity in primary visual cortex? Nature 1995; 375: 121- 3. 26. Brosch M, Bauer R, Eckhorn R. Synchronous high- frequency os­cillations in cat area 18. Eur J Neurosci 1995; 7: 86- 95. 27. Gegenfurtner KR, Kiper DC, Fenstemaker S. Processing of color, form and motion in macaque area V2. Vis Neurosci 1996; 13: 161 - 72. 28. Mainen ZF, Sejnowski TJ. Influence of dendritic structure on firing pattern in model neocortical neurons. Nature 1996; 382: 363- 6. 29. Richmond BJ, Optican LM. Temporal encoding of two-dimensional patterns by single units in primate primary visual cortex. II. Information transmission. J Neurophysiol 1990; 64: 370- 80. 30. Volchan E, Gilbert CD. Interocular transfer of receptive field ex­pansion in cat visual cortex. Vision Res 1995; 35: 1- 6. 31. Darian- Smith C, Gilbert CD. Topographic reorganization in the striate cortex of the adult cat and monkey is cortically mediated. / Neurosci 1995; 15: 1631- 47. 32. Chino YM, Smith EL 3rd, Kaas JH, Sasaki Y, Cheng H. Recep­tive- field properties of deafferentated visual cortical neurons after topographic map reorganization in adult cats. J Neurosci 1995; 15: 2417- 33. 33. Kujala T, Huotilainen M, Sinkkonen I, et al. Visual cortex activa­tion in blind humans during sound discrimination. Neurosci Lett 1995; 183: 143- 6. 34. Aulhorn E, Kost G. Noise field campimetry. A new perimetric method ( snow campimetry). In: Heijl A, ed. Perimetry Update 1988 89. Proceedings of the VHIth International Perimetric Society Meeting, Amstelveen, Netherlands. 1989: 336. 35. Schiefer U, Ulrich WD, Ulrich C, Wilhelm H, Aulhorn E. White-noise field campimetry before and after artificial increase of intra­ocular pressure. Possible applications in diagnosis and evaluation of glaucoma. Ophthalmologe 1992; 89: 477- 88. 36. Kolb M, Petersen D, Schiefer U, Kolb R, Skalej M. Scotoma perception in white- noise- field campimetry and postchiasmal vi­sual pathway lesions. German J Ophthalmol 1995; 4: 228- 33. 37. Cunningham FG, Fernandez CO, Hernandez C. Blindness associ­ated with preeclampsia and eclampsia. Am J Obstet Gynecol 1995; 172: 1291- 8. 38. Joseph JM, Louis S. Transient ictal cortical blindness during middle age. A case report and review of the literature. / Neurooph-thalmol 1995; 15: 39^ t2. 39. Rizzo M, Robin DA. Bilateral effects of unilateral visual cortex lesions in human. Brain 119: 951- 63, 1996. 40. Fendrich R, Wessinger CM, Gazzaniga MS. Residual vision in a scotoma: implications for blindsight. Science 1992; 258: 1489- 91. 41. Wessinger CM, Fendrich R, Ptito A, Villeure JG, Gazzaniga MS. Residual vision with awareness in the field contralateral to a partial or complete functional hemispherectomy. Neuropsychology 1996; 34: 1129- 37. 42. Cowey A, Stoerig P. Blindsight in monkeys. Nature 1995; 373: 247- 9. 43. Tueth MJ, Cheong JA, Samander J. The Charles Bonnet syndrome: a type of organic visual hallucinosis. J Geriatr Psychiatr Neurol 1995; 8: 1- 3. 44. Maricle RA, Turner LD, Lehman KD. The Charles Bonnet syn­drome: a brief review and case report. Psychiatr Serv 1995; 46: 289- 91. 45. Teunisse RJ, Cruysberg JR, Verbeek A, Zitman FG. The Charles Bonnet syndrome: a large prospective study in The Netherlands. A study of the prevalence of the Charles Bonnet syndrome and as­sociated factors in 500 patients attending the University Depart­ment of Ophthalmology at Nijmegen. Br J Psychiatry 1995; 166: 254- 7. 46. Adachi N. Charles Bonnet syndrome in leprosy; prevalence and clinical characteristics. Acta Psychiatr Scand 1996; 93: 279- 81. 47. Reyes- Ortiz CA, Camacho ME, Mulligan T. Charles Bonnet syn­drome in a centenarian. JAMA 1996; 276: 451- 2. 48. Pliskin NH, Kiolbasa TA, Towle VL, et al. Charles Bonnet syn­drome: an early marker for dementia? J Am Geriatr Soc 1996; 44: 1055- 61. 49. Cassinello F, de Pablos G, de Stefano J. Visual hallucinations in a girl following premedication with oral midazolam. Rev Esp Anes-tesiol Reanim 1995; 42: 76- 7. 50. Kolev OI. Visual hallucinations evoked by caloric vestibular stimulation in normal humans. J Vestib Res 1995; 5: 19- 23. 51. Walker MC, Smith SJ, Sisodiya SM, Shorvon SD. Case of simple partial status epilepticus in occipital lobe epilepsy misdiagnosed as migraine: clinical, electrophysiological, and magnetic resonance imaging characteristics. Epilepsia 1995; 36: 1233- 6. 52. Platz WE, Oberlaender FA, Seidel ML. The phenomenology of perceptual hallucinations in alcohol- induced delirium tremens. Psychopathology 1995; 28: 247- 55. 53. Vaphiades MS, Celesia GG, Brigell MG. Positive spontaneous phenomena limited to the hemianopic field in lesions of central visual pathways. Neurology 1996; 47: 408- 17. 54. Nagaratnam N, Virk S, Brdarevic O. Musical hallucinations asso­ciated with recurrence of a right occipital meningioma. Br J Clin Pract 1996; 50: 56- 7. 55. Liu GT, Schatz NJ, Galetta SL, Volpe NJ, Skobieranda F, Kos-morsky GS. Persistent positive visual phenomena in migraine. Neurology 1995; 45: 664- 8. 56. Rothrock JF. Successful treatment of persistant migraine aura with divalproex sodium. Neurology 1997; 48: 261- 2. 57. Kawasaki A, Purvin V. Persistent palinopsia following ingestion of lysergic acid diethylamide ( LSD). Arch Ophthalmol 1996; 114: 47- 50. 58. Muller T, Buttner T, Kuhn W, Heinz A, Przuntek H. Palinopsia as sensory epileptic phenomenon. Acta Neurol Scand 1995; 91: 433- 6. 59. Robert J, Miro O, Pedrol E, Cardellach F. Palinopsia as a mani­festation of migraine associated with other symptoms. Med Clin 1995; 105: 76- 7. 60. Takahashi N, Kawamura M, Hirayama K, Shiota J, Isono O. Prosopagnosia: a clinical and anatomical study of four patients. Cortex 1995; 31: 317- 29. 61. Carlesimo GA, Caltagirone C. Components in the visual process­ing of known and unknown faces. J Clin Exp Neuropsychol 1995; 17: 691- 705. 62. Kosslyn SM, Hamilton SE, Bernstein JH. The perception of cur­vature can be selectively disrupted in prosopagnosia. Brain Cognit 1995; 27: 36- 58. 63. Farah MJ, Wilson KD, Drain HM, Tanaka JR. The inverted face inversion effect in prosopagnosia: evidence for mandatory, face-specific perceptual mechanisms. Vision Res 1995; 35: 2089- 93. 64. Bub DN. Arguin M. Visual word activation in pure alexia. Brain Lang 1995; 49: 77- 103. 65. Feinberg TE, Dyckes- Berke D, Miner CR, Roane DM. Knowl­edge, implicit knowledge and metaknowledge in visual agnosia and pure alexia. Brain 1995; 118: 789- 800. 66. Price CJ, Humphreys GW. Contrasting effects of letter- spacing in alexia: further evidence that different strategies generate word length effects in reading. Q J Exp Psychol [ A] 1995; 3: 573- 97. 67. Livingstone M. Segregation of form, color, movement, and depth processing in the visual system: Anatomy, Physiology, Art and Illusion. In: Cohen B, Bodis- Wollner I, eds. Vision and the Brain. The organization of the central visual system. New York: Raven Press, 1990: 119- 38. J Neuro- Ophthalmol, Vol. 17, No. 3, 1997