Partial lesions of striate cortex were made in newborn and adolescent or young adult macaque monkeys, one newborn squirrel monkey, and adult squirrel and owl monkeys. After survival times ranging from 3 1/2 weeks to 8 years, the retinas were examined for transneuronal retrograde ganglion cell loss and retinal projections to the dorsal lateral geniculate nucleus, and other targets were examined for changes. After lesions in infant macaque monkeys and long postoperative survivals, nearly 80% of the ganglion cells were lost in the altered portions of the retinas. The degeneration appeared to be exclusively of ganglion cells projecting to the parvocellular layers of the lateral geniculate nucleus, and the loss of this class of cell appeared to be complete or nearly complete for the affected portions of the retina. Cases with shorter survivals showed that nine-tenths of the potential loss occurred within 6 months, and about half of the potential loss took place within one month. In cases where lesions were placed in adolescent and young adult macaque monkeys, the loss also was of ganglion cells projecting to the parvocellular layers. However, the rate of cell loss was slower so that little or no cell loss was apparent after six months, and only one-third to three-fourths of the potential loss occurred within 12–14 months. A cell loss of 22% was measured in the altered portions of the retina of a squirrel monkey lesioned as an infant and surviving for 6 months, but no regions of ganglion cell loss were apparent in the retinas of owl and squirrel monkeys lesioned as adults and surviving as long as two or more years.

We conclude that nearly 80% of the ganglion cells project to the parvocellular layers in macaque monkeys, and that the ultimate survival of these ganglion cells depends on the presence of target neurons in the parvocellular layers. Age is important in that the loss of ganglion cells proceeds rapidly in infant macaque monkeys, but slowly in older animals. Infant New World monkeys, judging from one squirrel monkey, are also susceptible to ganglion cell loss, although apparently at a rate comparable to older macaque monkeys. Finally, adult New World monkeys do not appear to be susceptible to ganglion cell loss. These age and species differences in rates of loss and susceptibility to loss challenge a “sustaining collateral” hypothesis proposed earlier (Weller et al., 1979), and suggest alternatives and modifications.

Visual information encoded by the middle-wavelength-sensitive (MWS) and long-wavelength-sensitive (LWS) cones in the primate retina are processed by both depolarizing (ON) and hyperpolarizing (OFF) bipolar cells. In contrast, signals from the short-wavelength-sensitive (SWS) cones and dark-adapted rod photoreceptors are thought to be carried almost exclusively by ON bipolar cells (Gouras & Evers, 1985). Consequently, it would be expected that functional inactivation of the retinal ON channels at the bipolar cell level would produce selective deficits in visual functions mediated by rods and SWS cones. We have examined this hypothesis by injecting rhesus monkeys with 2-amino-4-phosphonobutyric acid (APB), a pharmacological agent that reduces the responsiveness of retinal ON neurons, and psychophysically measuring the changes in spectral sensitivities. Under adaptation conditions that isolated rod function, APB caused, as expected, a substantial loss in rod-mediated spectral sensitivity. However, under photopic conditions, cone-mediated spectral sensitivity, including that associated with the SWS cones, was relatively unaffected. These results demonstrate distinct organizational differences between the rod and cone systems; specifically, they indicate that the rod system is more dependent upon retinal ON channels than the cone system. Our failure to find a selective visual deficit related to SWS cone function under photopic viewing conditions suggests that the OFF system can mediate stimulus detection throughout the visible spectrum and that the ability of the OFF system to process signals from the SWS cones has been underestimated.

The color vision of three domestic dogs was examined in a series of behavioral discrimination experiments. Measurements of increment-threshold spectral sensitivity functions and direct tests of color matching indicate that the dog retina contains two classes of cone photopigment. These two pigments are computed to have spectral peaks of about 429 nm and 555 nm. The results of the color vision tests are all consistent with the conclusion that dogs have dichromatic color vision.

It is generally accepted that the purine nucleoside, adenosine, plays a neuromodulatory role in the central nervous system (CNS) (Daly et al., 1981; Phillis ' Wu, 1983; Williams, 1986; Williams, 1987; Snyder, 1985). Adenosine is thought to exert its primary effects presynaptically, by inhibiting the release of neurotransmitters including ³-aminobutyric acid (GABA) and acetylcholine (ACh) (Phillis ' Barraco, 1985; Proctor ' Dunwiddie, 1987). In mammalian retina, cell bodies that are strongly labeled for adenosine-like immunoreactivity (ALIR) have been localized to the ganglion cell layer (GCL) (Braas et al., 1987; Blazynski et al., 1989). Rabbit retinal cells that are labeled by markers for both ACh and GABA are located in the GCL and inner nuclear layer (INL) (Tauchi ' Masland, 1984; Vaney ' Young, 1988b; Brecha et al., 1988). It is now demonstrated in the rabbit retina that approximately 50% of the cells labeled for ALIR within the GCL represent true ganglion cells, with the remainder presumed to be displaced cholinergic amacrine cells (DAPI accumulating). In addition, some of these same cells also demonstrate immunoreactivity to glutamate decarboxylase (GAD), involved in the biosynthesis of the neurotransmitter GABA. Thus, in a particular class of retinal neurons, two fast-acting neurotransmitters as well as a putative neuromodulator have been co-localized.

Cat-301 is a monoclonal antibody which recognizes a cell surface associated antigen of selected neurons in the central nervous system (CNS). In the visual system, cat-301 selectively labels Y-like cells in several visual structures, including portions of the lateral geniculate nucleus complex and visual cortex. The cat superior colliculus (SC) also receives Y input and contains cells driven by Y input which are selectively distributed in the deep superficial gray and deeper laminae. If cat-301 is selective to the Y-cell system in SC, labeled cells should be restricted to those laminae. To test this hypothesis, we have examined quantitatively the laminar distribution, percentage, size, and morphology of cells in SC labeled by the cat-301 antibody. Cat-301 labeled a variety of cells in the cat SC. Labeled cells were found within the deep portion of the superficial gray layer (6.6%), optic layer (27.6%), intermediate gray layer (26.9%), and the deep gray and white layers (38.5%). By contrast, only 2 of 667 labeled cells (0.3%) were found within that part of the upper superficial gray layer innervated exclusively by W input and thought to contain only W-driven cells. When considered as a percentage of the total cell population, cat-301 labeled cells represented less than 3% of cells in the superficial gray layer and approximately 15% in the deeper layers. Neurons labeled by cat-301 were all of medium to large size (mean average diameter = 33.3μm; range = 15–84μm) and included vertical fusiform and stellate cells in the upper layers and the very large neurons found in the intermediate gray and deeper layers. These results provide further evidence that the cat-301 antibody selectively recognizes the Y channel of the cat visual system.

The number, dendritic morphology, and retinal distribution of displaced ganglion cells were studied in two anuran species, Xenopus laevis and Bufo marinus. Horseradish peroxidase or cobaltic lysine complex was applied to the cut end of the optic nerve, and the size, shape, and retinal position of retrogradely filled ganglion cells displaced into the inner nuclear layer were determined in retinal wholemount and sectioned material. Approximately 1% of ganglion cells in Xenopus and 0.1% in Bufo were found to be displaced. In both species, many of the previously described orthotopic ganglion cell types (Straznicky & Straznicky, 1988; Straznicky et al., 1990) were present among displaced ganglion cells. In Xenopus more displaced ganglion cells were found in the retinal periphery than in the retinal center, and they formed 3 or 4 distinct bands around the optic nerve head. In Bufo the incidence of displaced ganglion cells was higher along the visual streak than in the dorsal and ventral peripheral retina. These results indicate that the distribution of displaced ganglion cells approximates the retinal distribution of orthotopic ganglion cells. One of the likely mechanisms to account for this developmental paradox may be that the formation of the inner plexiform layer, adjacent to the ciliary margin, acts as a mechanical barrier by preventing the entry of some of the late developing ganglion cells into the ganglion cell layer.

Rod and cone photoresponses in a variety of species have been accurately described with linear multistage filter models. In this study, the response latency and initial coding of intensity at two higher levels of visual processing are related to such photoreceptor responses. One level is the retinal output (spiking discharges from frog ganglion cells, based on experimental data reported here), the other is the perceptual level in humans (psychophysical latency and brightness functions, based on data from the literature). Photoreceptor responses are described with the “independent activation” model of Baylor et al. (1974). The intensity dependence of the early ganglion cell discharge, its latency and initial impulse frequency, is shown to follow from such a waveform, assuming that 1) latency L = l + D, where l is the time it takes for the rod response linearly summed over the ganglion cell's receptive field to reach a criterion amplitude, and D is a constant delay; and 2) the initial frequency (below saturation) is proportional to the steepness of rise of the summed rod response at time l. It is shown that the intensity dependences of 1) human visual latency and 2) brightness sensation, including effects of stimulus area and duration, are accounted for by the same model. The predicted functions are not power functions of intensity, but approximate such over wide ranges. Thus, a large body of psychophysical data is explained simply by the waveform of photoreceptor responses.

The number of subregions in the activity profiles of simple cells varies in different cells from 2–8; that is, the number of cycles in the weighting function varies from 1–4. The distribution of receptive-field (RF) sizes at eccentricities of 0-6 deg are clustered at half-octave intervals and form a discrete distribution with maxima at 0.62, 0.9, 1.24, 1.8, 2.48, and 3.4 deg. The spatial frequencies to which the cells are tuned are also clustered at half-octave intervals, forming a discrete distribution peaking at 0.45, 0.69, 0.9, 1.35, 1.88, 2.7, 3.8, and 5.6 cycles/deg. If we divide the RF sizes by the size of the period of the subregions, then the average indices of complexity (really existing) or the number of cycles in the weighting function form (after normalization) the sequences: 1, 1.41, 2.0, 2.9, 4.15.

The relation between the bandwidth of the spatial-frequency characteristic and the optimal spatial frequency is in accordance with predictions of the Fourier hypothesis. The absolute bandwidth does not change with the number of cycles/module. This means that inside the module the absolute bandwidth does not change with the number of the harmonic. The results allow us to suggest the following. A module of the striate cortex, which is a group of cells with RFs of equal size projected onto the same area of central visual field, accounts for the Fourier description of the image. The basis functions of the module are composed of four harmonics only, irrespective of size and position of the module.

Besides linear cells (sinusoidal and cosinusoidal elements), the module contains nonlinear cells, performing a nonlinear summation of the responses of sinusoidal and cosinusoidal elements. Such cells are characterized by an index of complexity which is more than the number of cycles in the weighting function and by marked overlap of ON and OFF zones. The analysis of organization suggests that the cells can measure the amplitude and phase of the stimulus.

The peripheral retina of the sea lamprey develops in a 5-year-long process in which only certain neurons differentiate each year. The growth of cell layers, the differentiation of the neurons, and the morphology of their dendrites and axons were studied with normal, HRP, and Golgi preparations. Ganglion cells are differentiated in 3-year-old larvae, amacrine and horizontal cells in 4-year-old larvae, photoreceptor cells in stage I transformers, and bipolar cells in stage III transformers. Each new development is expressed as a radial gradient of differentiation. As a result of this protracted and stepped process, lamprey retinal neurons, particularly ganglion cells, differentiate in the absence of other cells to which they will ultimately be connected and may express their individual genetic programs more fully than in other vertebrate retinas. This could account for the unusual relationship of the ganglion cell, inner plexiform, and optic nerve layers and for the very high ratio of displaced to orthotopic ganglion cells.

Neurons in the nucleus praetectalis olivaris (NPO) were antidromically identified by electrical stimulation of the nucleus of Edinger-Westphal (EW), the location of preganglionic pupilloconstrictor motoneurons. Electrical stimulation within the NPO leads to bilateral pupil constriction. Single neurons recorded in the NPO respond tonically to light stimuli, and their discharge frequency increases linearly with logarithmic increase in light intensity. This characteristic identifies NPO neurons as luminance detectors. They have large receptive fields mostly lying in the upper and contralateral quadrant of the visual field.

Cats with impaired binocular vision show a significantly reduced binocular summation of the pupillary light reflex (BSP), i.e. the increase of pupil constriction during binocular illumination when compared to monocular illumination is less than in normal animals. The investigation of ocular dominance and subthreshold binocular interactions in the NPO of normal and innate microstrabismic cats revealed two possible mechanisms for BSP and its reduction in strabismic subjects. First, the percentage of neurons increasing their discharge rate by illuminating either eye is significantly reduced in the NPO of innate microstrabismic cats (6.6%) when compared to normal cats (22% of all neurons tested). Second, in most NPO neurons of normal cats the subthreshold influence of the ipsilateral eye leads to an increase in neuronal activity during binocular stimulation when compared to monocular stimulation of the contralateral eye (binocular summation). The subthreshold influence of the ipsilateral eye in most NPO neurons of microstrabismic cats, however, is inhibitory, i.e. the neuronal discharge rate during binocular stimulation is decreased when compared to monocular stimulation of the contralateral eye (binocular inhibition). However, there is no significant correlation between BSP and binocularity in the NPO in individual animals. This suggests that BSP may be additionally influenced by visual structures other than NPO.

The existence of multiple areas of extrastriate visual cortex raises the question of how the response properties of each area are derived from its visual input. This question was investigated for one such area in the cat, referred to here as the Clare-Bishop area (Hubel & Wiesel, 1969); it is the region of lateral suprasylvian cortex that receives input from area 17. A novel approach was used, in which kainic acid was injected locally into the Clare-Bishop area, making it possible to record directly from afferent inputs.

The response properties of the great majority of a sample of 424 presumed afferents resembled cells in areas 17 and 18. Thus, a systematic comparison was made with cells from area 17's upper layers, the source of its projection to the Clare-Bishop area (Gilbert & Kelly, 1975), to see whether these afferents had distinctive properties that might distinguish them from cells projecting to areas 18 or 19. Some differences did emerge: (1) The smallest receptive fields typical of area 17 were relatively scarce among afferents. (2) Direction-selective afferents were more abundant than were such cells in area 17. (3) End-stopped afferents were extremely rare, although end-stopped cells were common in area 17's upper layers.

Despite these differences, afferents were far more similar in their properties to cells in areas 17 and 18 than to cells in the Clare-Bishop area. Compared to the latter, afferents showed major discrepancies in receptive-field size, in direction selectivity, in end-stopping, and in ocular dominance distribution. These differences seem most likely to stem from circuitry intrinsic to the Clare-Bishop area.

We examined the role of the lateral suprasylvian (LS) cortex in motion perception by testing the ability of three cats to detect moving targets and to discriminate differences in stimulus direction and speed before and after making bilateral ibotenic acid lesions in LS. The lesions had little or no effect on contrast sensitivity for detecting moving sinusoidal gratings. Moreover, we found no deficits in discriminating opposite directions of motion: the cats discriminated grating directions at threshold contrasts. All three cats, however, showed permanent deficits in discriminating differences in speed and in flicker rate. The deficits were most pronounced at higher temporal and spatial frequencies and at lower contrasts. This result suggests that LS plays an important role in the analysis of stimulus speed. It appears that information needed for discriminating opposite directions of motion may be signalled by visual areas outside LS.

The anatomical substrate of the pupillary light reflex was investigated in normal and innate microstrabismic cats using anatomical methods as well as electrical stimulation. The bilateral retinal input to the nucleus praetectalis olivaris (NPO), the pretectal relay station in the subcortical pupilloconstrictor pathway, was identified to come from the ventral retina where the upper visual field is represented. Orthodromic electrical stimulation revealed that retinal information is transmitted to on-tonic neurons in the NPO mainly via slowly conducting axons probably originating from W- and X-type retinal ganglion cells.

For the first time, a direct cortical input to on-tonic neurons in the NPO could be demonstrated. This cortical input originates from caudolateral parts of the occipital cortex. Putative input structures are those subdivisions of areas 19 and 20a where the upper part of the visual field is represented.

A direct, predominantly contralateral projection with a weak ipsilateral component from NPO to the nucleus of Edinger-Westphal, and an interhemispheric connection between the NPOs could be demonstrated. With respect to the anatomical connections as described in this study, no differences between normal and innate microstrabismic cats could be found.

The results are discussed with respect to the binocular summation of the pupillary light reflex and its reduction in subjects with impaired binocular vision.

It is believed that spatial summation in most simple cells is a linear process. If this were so, then the Fourier transform of a simple cell's line weighting function should predict the cell's spatial frequency tuning curve. We have compared such predictions with experimental measurements and have found a consistent discrepancy: the predicted tuning curve is much too broad. We show qualitatively that this kind of discrepancy is consistent with the well-known threshold nonlinearity shown by most cortical cells. We have tested quantitatively whether a response threshold could explain the observed disagreements between predictions and measurements: a least-squares minimization routine was used to fit the inverse Fourier Transform of the measured frequency tuning curve to the measured line weighting function. The fitting procedure permitted us to introduce a threshold to the reconstructed line weighting function. The results of the analysis show that, for all of the cells tested, the Fourier transforms produced better predictions when a response threshold was included in the model. For some cells, the actual magnitude of the response threshold was measured independently and found to be compatible with that suggested by the model. The effects of nonlinearities of spatial summation are considered.

Averaged grating-evoked cortical potentials were recorded from area 17 of awake cats. Peak latency of early components of the visual-evoked potential (VEP) response to stimulus onset increased as a function of spatial frequency, while amplitude tended to be largest at intermediate spatial frequencies. Latency increased and amplitude generally decreased to lower spatial-frequency stimuli (<0.25 cycle/deg) in the presence of a uniform flickering field (UFF). The UFF had a relatively small or opposite effect on peak latency and amplitude for higher spatial-frequency stimuli (>0.50 cycle/deg). The VEP response to stimulus offset was present only at low spatial frequencies and was virtually eliminated by the presence of the UFF. The effects were similar whether the target and UFF background were simultaneously presented or briefly separated; however, the UFF had no effect when the two were spatially separated. The effects of the UFF background on VEP onset response increased with increasing temporal frequency from 2–8 Hz; offset responses were affected similarly at all temporal frequencies. These effects are similar to those observed in humans and suggest that two spatio-temporally tuned mechanisms contribute to the early VEP response. In the cat, the mechanisms seem to correspond to X and Y cells in the dorsal lateral geniculate nucleus.

The projections of the nucleus of the optic tract were studied in the cat by simultaneous use of two distinct retrograde tracers (Fast Blue and Diamidino Yellow) injected in the inferior olive and the prepositus hypoglossi nucleus. Following injections of diamidino yellow in one structure and fast blue in the other, a significant number of retrogradely labeled neurons projecting to either target were observed dispersed in the nucleus of the optic tract. Three populations of labeled cells were found: one which projected to the inferior olive, a second to the nucleus prepositus hypoglossi, and a third which projected by means of a bifurcating axon to both of these structures. Quantification of these results reveals that 72% of the total number of labeled neurons are labeled by the IO injection, the remaining cells being labeled by the NPH injection. Double-labeled neurons represent more than 7% of the total number of the labeled cells. Tentative inferences as to the electrophysiological properties of the nucleus of the optic tract are discussed in the context of the optokinetic system.

Dopamine D1 antagonists have been shown to alter drastically the spontaneous and light-evoked activity of ganglion cells in the rabbit retina (Jensen & Daw, 1984, 1986). A major target of dopaminergic neurons in mammalian retinas appears to be rod All amacrine cells (Pourcho, 1982; Voigt & Wässle, 1987). In the present study, the following questions were addressed: (1) Do dopamine D1 antagonists alter the activity of ganglion cells through actions primarily on rod All amacrine cells? (2) Are the effects of dopamine D1 antagonists on ganglion cells due to an inhibition of dopamine-stimulated adenylate cyclase activity?

Using an isolated, superfused retinal preparation, the ability of several pharmacological agents to counteract the physiological effects of the dopamine D1 antagonist (+)-SCH 23390 on rabbit ganglion cells was examined. The glycine antagonist strychnine abolished the effects of (+)-SCH 23390 on the spontaneous and light-evoked activity of OFF-center ganglion cells, whereas the excitatory amino-acid antagonist kynurenic acid abolished the effects of (+)-SCH 23390 on the spontaneous and light-evoked activity of ON-center ganglion cells. The findings obtained with these antagonists can be explained in terms of the known synaptic connections of All amacrine cells.

Both 8-(4-chlorophenylthio) cyclic AMP, a membrane-permeable cAMP analog, and forskolin, an activator of adenylate cyclase, reversed the effects of (+)-SCH 23390 on the spontaneous and light-evoked activity of OFF-center ganglion cells but not ON-center ganglion cells. These findings suggest that the effects of dopamine D1 antagonists on OFF-center ganglion cells are due to an inhibition of dopamine-stimulated adenylate cyclase, with the ensuing lowering of cellular cAMP levels. The effects of dopamine D1 antagonists on ON-center ganglion cells appear, however, to be independent of intracellular cAMP levels.

It has been suggested that development of central connections in the mammalian visual system is governed by a simple Hebbian rule of synaptic modifiability. Under such a rule, simultaneity of presynaptic and postsynaptic action potentials is a prerequisite for enhanced synaptic efficacy. The present paper reports the results of a study designed to test whether this hypothesis is applicable to the development of the thalamo-cortical visual pathway.

In four-week-old kittens, exposure to a 2-d period of monocular deprivation was used to render the vast majority of cortical cells capable of being activated only by the nondeprived eye. During a subsequent 3–5 month recovery period, both eyes were open but surgically misaligned. This combination of conditions was chosen so that during the recovery period presynaptic activity originating from the initially deprived eye would be decorrelated from postsynaptic action potentials in cortical neurons. If synaptic modification is regulated by a simple Hebbian mechanism, then in this situation the deprived eye should be unable to recover control of cortical cells.

In fact, the present results indicate that during the recovery period the proportion of cortical neurons dominated by the deprived eye rose to a level equal to that of the nondeprived eye – a result contrary to that predicted by a simple Hebbian rule of development. Histological analysis indicated that a similar level of recovery was present both within and outside of cortical layer IV, the main thalamo-recipient layer. As expected, the induced strabismus resulted in a failure of cortical binocularity to recover in these kittens.

Although these results argue against a simple Hebbian mechanism of development, they are compatible with the hypothesis that synaptic modifiability is dependent upon correlations between presynaptic activity and local, subthreshold, postsynaptic changes. This alternative hypothesis has the advantage of allowing modification of local synaptic circuits within the dendritic arbors of a single neuron.

One subpopulation of amacrine in the turtle retina was shown to contain met-enkephalin by means of immunocytochemistry, and another was demonstrated to have a high-affinity uptake system for [3H]-dopamine by means of tutoradiography. Although the amacrine soma size, density, and distribution of their neurites in IPL substrata was similar in retinas in which met-enkephalin and dopamine were localized, combined light microscope immuncytochemistry-autoradiography demonstrated that these two neurotransmitter systems did not coexist in the same cells. Because the two amacrine cell subtypes ramify in the same IPL substrata, neuronal interaction between them is possible. Release experiments showed that the potassium-induced release of [3H]-dopamine from the superfused turtle retina was reduced by 40% when enkephlian was added to the superfusate. The inhibition of [3H]-dopamine release could be blocked by the addition of naloxone. The addition of enkephalin had no effect of the potassium-induced release of [3H]-GABA from the superfused retina. These findings suggest that an enkephalinergic modulation of the dopaminergic amacrine cell system exits in the turtle retina.

The trajectory of retinal projections and the location of retinorecipient nuclei in the quail brain was examined after application of horseradish peroxidase (HRP) either to the cut end of the optic nerve or following intraocular injection of HRP. Retinal projections to the hypothalamus, dorsalateral anterior thalamus (rostralateral part, magnocellular part, and lateral part), lateral anterior thalamus, lateroventral geniculate nucleus, lateral geniculate intercalated nuclei (rostral and caudal parts), ventrolateral thalamus, superficial synencephalic nucleus, external nucleus, tectal gray, diffuse pretectal area, pretectal optic area, ectomammillary nucleus, and optic tectum were revealed. Retinal projections observed in quail were compared with results obtained in other avian species and considered in relation to possible anatomic pathways underlying photoperiodism and circadian rhythms.