The effects of intraocular injections of ethylcholine mustard aziridinium ion (AF64A), an irreversible inhibitor of choline uptake, on the rabbit retina were assessed electrophysiologically, pharmacologically, anatomically, and behaviorally. Survival times from 1 day to 30 days were investigated. After 24 h, the shortest time tested, the directional selectivity of On–Off responding ganglion cells having the characteristic morphology of On–Off directionally selective directionally selective (DS) ganglion cells, as revealed by intracellular dye injection, was significantly reduced, both by an apparent decrease of preferred direction responses and an increase in responses to null-direction movement. No toxin-mediated changes in the dendritic trees of these cells were noted. Cells in AF64A-affected retinas having the DS morphology did not respond significantly to GABAergic or cholinergic agents such as picrotoxin and eserine, but did respond to nicotine. Recordings from a small random sample of other ganglion cell classes in the same retinas yielded no obvious changes in response properties. The direct effects on starburst (cholinergic) amacrine cells, which were identified by intraocular injection of the fluorescent dye DAPI with the AF64A, were investigated by intracellular injections of Lucifer yellow, and by immunohistochemical staining with antibodies to choline acetyltransferase (ChAT). Although starburst amacrine cell somas survived the AF64A treatment for at least several days, the dendrites could not be visualized by fluorescent dye injection in affected retinas due to dye leakage of the injected fluorescent dye from either the soma or proximal dendritic region. ChAT staining revealed a sequence in which ChAT-positive cells were undetectable first in the inner nuclear layer, and then in the ganglion cell layer. Cholinergic amacrine cells in the central retina were also affected before those in the periphery. The electrophysiological changes observed typically preceded the loss of ChAT activity. Behavioral tests for optokinetic nystagmus responses also revealed a lack of such responses in the affected eyes.

A survey of the shapes of dopaminergic (DA) neurons in the retinas of representative vertebrates reveals that they are divisible into three groups. In teleosts and Cebus monkey, DA cells are interplexiform (IPC) neurons with an ascending process that ramifies to create an extensive arbor in the outer plexiform layer (OPL). All other vertebrates studied, including several primate species, have either DA amacrine cells or IPCs with an ascending process that either does not branch within the OPL or does so to a very limited degree. DA neurons of non-teleosts exhibit a dense plexus of fine caliber fibers which extends in the distal most sublamina of the inner plexiform layer (IPL). Teleosts lack this plexus. In all vertebrates, DA cells are distributed more or less evenly and at a low density (10–60 cells/mm2) over the retinal surface. Dendritic fields of adjacent DA neurons overlap. Most of the membrane area of the DA cell is contained within the plexus of fine fibers, which we postulate to be the major source of dopamine release. Thus, dopamine release can be modeled as occurring uniformly from a thin sheet located either in the OPL (teleosts) or in the distal IPL (most other vertebrates) or both (Cebus monkey). Assuming that net lateral spread of dopamine is zero, the fall of dopamine concentration with distance at right angles to the sheet (i.e. in the scleral-vitreal axis) will be exponential. The factors that influence the rate of fall – diffusion in extracellular space, uptake, and transport – are not yet quantified for dopamine, hence the dopamine concentration around its target cells cannot yet be assessed. This point is important in relation to the thresholds for activation of D1 and D2 dopamine receptors that are found on a variety of retinal cells.

Bipolar cells (BCs) are critical relay neurons in the retina that are organized into parallel signaling pathways. The three main signaling pathways in the mammalian retina are the rod, ON cone, and OFF cone BCs. Rod BCs mediate incrementing dim light signals from rods, and ON cone and OFF cone BCs mediate incrementing and decrementing brighter light signals from cones, respectively. The outputs of BCs are shaped by inhibitory inputs from GABAergic and glycinergic amacrine cells in the inner plexiform layer, mediated by three distinct types of inhibitory receptors: GABAA, GABAC, and glycine receptors. The three main BC pathways receive distinct forms of inhibition from these three receptors that shape their light-evoked inhibitory signals. Rod BC inhibition is dominated by slow GABAC receptor inhibition, while OFF cone BCs are dominated by glycinergic inhibition. The inhibitory inputs to BCs are also shaped by serial inhibitory connections between GABAergic amacrine cells that limit the spatial profile of BC inhibition. We discuss our recent studies on how inhibitory inputs to BCs are shaped by receptor expression, receptor properties, and neurotransmitter release properties and how these affect the output of BCs.

In macaque, it has long been known since the late nineties that the medial parieto-occipital sulcus (POS) contains two regions, V6 and V6A, important for visual motion and action. While V6 is a retinotopically organized extrastriate area, V6A is a broadly retinotopically organized visuomotor area constituted by a ventral and dorsal subdivision (V6Av and V6Ad), both containing arm movement-related cells active during spatially directed reaching movements. In humans, these areas have been mapped only in recent years thanks to neuroimaging methods. In a series of brain mapping studies, by using a combination of functional magnetic resonance imaging methods such as wide-field retinotopy and task-evoked activity, we mapped human areas V6 (Pitzalis et al., 2006) and V6Av (Pitzalis et al., 2013d) retinotopically and defined human V6Ad functionally as a pointing-selective region situated anteriorly in the close proximity of V6Av (Tosoni et al., 2014). Like in macaque, human V6 is a motion area (e.g., Pitzalis et al., 2010, 2012, 2013a,b,c), while V6Av and V6Ad respond to pointing movements (Tosoni et al., 2014). The retinotopic organization (when present), anatomical position, neighbor relations, and functional properties of these three areas closely resemble those reported for macaque V6 (Galletti et al., 1996, 1999a), V6Av, and V6Ad (Galletti et al., 1999b; Gamberini et al., 2011). We suggest that information on objects in depth which are translating in space, because of the self-motion, is processed in V6 and conveyed to V6A for evaluating object distance in a dynamic condition such as that created by self-motion, so to orchestrate the eye and arm movements necessary to reach or avoid static and moving objects in the environment.

We have reinvestigated receptive-field structure of ganglion cells of the macaque parafovea using counterphase modulation of a bipartite field. Receptive fields were mapped with luminance, chromatic, and cone-isolating stimuli. Center sizes of middle (M) and long (L) wavelength cone opponent cells of the parvocellular (PC) pathway were consistent with previous estimates (Gaussian radii of 2–4 min of arc, corresponding to center diameters of 6–12 min of arc). We calculate that a large factor of the enlargement relative to cone radius could be blur due to the eye's natural optics. Maps were consistent with cone selectivity in surround mechanisms, which had radii of 5–8 min of arc. For magnocellular (MC) cells, center size estimates were also consistent with grating measurements from the literature (also Gaussian radii of 2–4 min of arc). The surround mechanism contributing the MC-cell frequency-doubled response to chromatic modulation appears to possess a subunit structure, and we speculate it derives from nonlinear summation of signals from M,L-cone opponent subunits, such as midget bipolar cells.

Most primate retinas have an area dedicated for high visual acuity called the fovea centralis. Little is known about specific mechanisms that drive development of this complex central retinal specialization. The primate area of high acuity (AHA) is characterized by the presence of a pit that displaces the inner retinal layers. Virtual engineering models were analyzed with finite element analysis (FEA) to identify mechanical mechanisms potentially critical for pit formation. Our hypothesis is that the pit emerges within the AHA because it contains an avascular zone (AZ). The absence of blood vessels makes the tissue within the AZ more elastic and malleable than the surrounding vascularized retina. Models evaluated the contribution to pit formation of varying elasticity ratios between the AZ and surrounding retina, AZ shape, and width. The separate and interactive effects of two mechanical variables, intraocular pressure (IOP) and ocular growth-induced retinal stretch, on pit formation were also evaluated. Either stretch or IOP alone produced a pit when applied to a FEA model having a highly elastic AZ surrounded by a less elastic region. Pit depth and width increased when the elasticity ratio increased, but a pit could not be generated in models lacking differential elasticity. IOP alone produced a deeper pit than did stretch alone and the deepest pit resulted from the combined effects of IOP and stretch. These models predict that the pit in the AHA is formed because an absence of vasculature makes the inner retinal tissue of the AZ very deformable. Once a differential elasticity gradient is established, pit formation can be driven by either IOP or ocular growth-induced retinal stretch.

Vertebrate vision starts with photoisomerization of the 11-cis-retinal chromophore to all-trans-retinal. Biosynthesis of 11-cis-retinal is required to maintain vision. A key enzyme catalyzing the oxidation of 11-cis-retinol is 11-cis-retinol dehydrogenase (11-cis-RDH), which is encoded by the RDH5 gene. 11-cis-RDH is expressed in the RPE and not in the neural retina. The consequences of a lack of 11-cis-RDH were studied in a family with fundus albipunctatus. We identified the causative novel RDH5 mutation, Arg157Trp, that replaces an amino acid residue conserved among short-chain alcohol dehydrogenases. Three-dimensional structure modeling and in vitro experiments suggested that this mutation destabilizes proper folding and inactivates the enzyme. Studies using RPE membranes indicated the existence of an alternative oxidizing system for the production of 11-cis-retinal. In vivo visual consequences of this null mutation showed complex kinetics of dark adaptation. Rod and cone resensitization was extremely delayed following full bleaches; unexpectedly, the rate of cone recovery was slower than rods. Cones showed a biphasic recovery with an initial rapid component and an elevated final threshold. Other unanticipated results included normal rod recovery following 0.5% bleach and abnormal recovery following bleaches in the 2–12% range. These intermediate bleaches showed rapid partial recovery of rods with transitory plateaux. Pathways in addition to 11-cis-RDH likely provide 11-cis-retinal for rods and cones and can maintain normal kinetics of visual recovery but only under certain constraints and less efficiently for cone than rod function.

The responses of neurons in areas V1 (17) and V2 (18) of anesthetized and paralyzed rhesus monkeys and cats were recorded while presenting a set of computer-generated visual stimuli that varied in pattern, texture, luminance, and contrast. We find that a class of extrastriate cortical cells in cats and monkeys can signal the presence of boundaries regardless of the cue or cues that define the boundaries. These cue-invariant (CI) cells were rare in area V1 but easily found in V2. CI cortical cells responded more strongly to more salient boundaries regardless of the cue defining the boundaries. Many CI cortical cells responded to illusory contours and exhibited the same degree of orientation and direction selectivity when tested with boundaries defined by different cues. These cells have significant computational power inherent in their receptive fields since they were able to generalize across stimuli and integrate multiple cues simultaneously in order to signal boundaries. Cells in higher order cortical areas such as MT (Albright, 1992), MST (Geesaman & Anderson, 1996), and IT (Sary et al., 1993) have previously been reported to respond in a cue invariant fashion. The present results suggest that the ability to respond to boundaries in a cue-invariant manner originates at relatively early stages of cortical processing.

Antibodies directed against the calcium-binding proteins, parvalbumin and calbindin, can be used to label distinct neuronal subgroups in the primate visual pathway. We analyzed parvalbumin immunoreactivity (P-IR) and calbindin immunoreactivity (C-IR) in the lateral geniculate nucleus (LGN) and visual cortex of the marmoset, Callithrix jacchus. We compared marmosets which were identified as having dichromatic or trichromatic color vision. Within the LGN, the density of P-IR neurones is highest in the parvocellular and magnocellular laminae, but C-IR neurones are found mainly in the koniocellular division of the LGN, that is, the interlaminar zones and S laminae. Not all interlaminar zone cells are C-IR. In the visual cortex, P-IR neurones are present in all laminae except lamina 1, in areas V1 and V2. Neurones which are strongly C-IR are mainly located in laminae 2 and 3 in V1 and V2. Lightly C-IR neurones are concentrated in lamina 4, and are more numerous in V1 than in V2. Quantitative analysis showed no differences in the density or distribution of IR neurones in either LGN or visual cortex when dichromat and trichromat animals were compared. We conclude that this functional difference is not associated with differences in the neurochemistry of calcium-binding proteins in the primary visual pathways.

To further characterize the H1 and H2 horizontal cell populations in macaque monkey retinae, cells were injected with the tracer Neurobiotin following intracellular recordings. Tracer coupling between cells of the same type revealed all H1 or H2 cells in small patches around the injected cell. The mosaics of their cell bodies and the tiling of the retina with their dendrites were analyzed. Morphological differences between the H1 and H2 cells observable in Neurobiotin-labeled patches made it possible to recognize H1 and H2 cells in retinae immunolabeled for the calcium-binding proteins parvalbumin and calbindin, and thus to study their relative spatial densities across the retina. These data, together with the intracellularly stained patches, show that H1 cells outnumber H2 cells at all eccentricities. There is, however, a change in the relative proportions of H1 and H2 cells with eccentricity: close to the fovea the ratio of H1 to H2 cells is ∼4 to 1, in midperipheral retina ∼3 to 1, and in peripheral retina ∼2 to 1. In both the Neurobiotin-stained and the immunostained retinae, about 3–5% of the H2 cells were obviously misplaced into the ganglion cell layer. Several features of the morphology of the misplaced H2 cells suggest that they represent the so-called “biplexiform ganglion cells” previously described in Golgi studies of primate retina.

It is now well established that environmental signals mediated via neurotransmitters and hormones can induce responses in cells which involve a cascade of receptors, G proteins, and second messengers. These in turn can induce transcription factors which regulate long-term changes in gene expression. It has been proposed that the stimulus-transcription coupling properties of these DNA-binding proteins can be exploited to visualize activated neurons by way of immunostaining. We have used standard immunohistochemical techniques to detect the expression of one specific transcription factor, Zif268, in the visual cortex (area 17, V1) of vervet monkeys (Cercopithecus aethiops). Immunopositive neurons were present in large numbers throughout the visual cortex of the normal animal, being concentrated in layers 2/3 and 6 and at moderate levels in 4Cβ and 5. To determine if Zif268 expression was affected by visual stimulation in the monkey, we restricted light input to one eye with the aim of revealing ocular-dominance columns in striate cortex. We found that short-term monocular deprivation induced either by enucleation, intravitreal TTX injection, or eyelid suturing resulted in dramatic changes in Zif268 levels, revealing vertically oriented columns of reduced Zif268 staining interdigitated with columns of normal expression. Furthermore, these columns were discernible after just 2 h of monocular blockade. A comparison of the ocular-dominance pattern obtained with Zif268 immunostaining and cytochrome oxidase histochemistry in long-term monocularly deprived animals showed a coincident reduction of both markers along columns that were precisely aligned in adjacent sections, indicating that Zif268 expression is restricted to cortical regions of high metabolic activity. Simultaneous immunostaining for Zif268 and the calcium-binding proteins calbindin and parvalbumin showed a negative correlation, suggesting that the Zif268 protein may be expressed selectively within excitatory neurons. A similar approach with immunostaining for neurofilament and microtubule-associated proteins (SMI-32 and MAP2) revealed pyramidal neurons which were regularly found to contain a Zif268-positive nucleus. Furthermore, confocal images of lucifer yellow filled neurons possessing Zif268-positive nuclei all showed pyramidal morphology. Taken together, these results point to activity-dependent expression of Zif268 within a subset of excitatory neurons.

In 1789, when neither the physical basis of hue nor the retinal basis of color perception was established, the mathematician Gaspard Monte stated firmly that our color perceptions do not depend on the absolute value of the physical variable, but are influenced by the context and in particular by our estimate of the illuminant. He used this insight to explain color contrast effects and the Paradox of Monge (the desaturation of red objects seen through a red filter). He proposed that we can estimate the chromaticity of the illuminant in any scene because all surfaces reflect to us varying mixtures of (i) the body color and (ii) a specular component that represents the illuminant. He also realized that white objects have a special property: Provided that they are illuminated by a single illuminant, such objects exhibit no variation in chromaticity across their surface. Thus at least one of the unique hues exists as an external reference on which observers can agree. It is suggested that other unique hues may also have a basis in the external world.

The intergeniculate leaflet (IGL) is an integral part of the circadian visual system. It receives direct retinal input and relays photic information to the circadian clock in the suprachiasmatic nucleus (SCN) through a geniculohypothalamic tract (GHT). In both rat and hamster, neuropeptide Y immunoreactive (NPY-IR) IGL cells project through the GHT to the SCN. However, the hamster GHT also contains enkephalin-IR (ENK-IR) fibers, presumably of IGL origin. In the present investigations, the IGL was examined for NPY-, ENK-, or dual-IR cells. Their projections to the SCN, contralateral IGL and pretectum were also studied. The results show that the hamster IGL contains both NPY- and ENK-IR neurons and that about 50% of these are immunoreactive to both peptides. Double-label retrograde analysis indicates that cells of each peptide class project to the SCN. Similarly, IGL neurons, many of which are NPY- and ENK-IR, project to the pretectum, particularly the posterior limitans nucleus. While numerous IGL neurons project contralaterally, very few are NPY- or ENK-IR.The distribution of SCN- and pretectum-projecting cells, in conjunction with the distribution of peptide-IR neurons, allows expansion of the IGL definition to include the region medial to the ventral lateral geniculate nucleus (VLG). The VLG is ventrolateral to the IGL and does not contain either neurons projecting to the SCN nor NPY- or ENK-IR cells, but does have numerous neurons projecting to the pretectum. The results substantiate and expand the previous definition of the hamster IGL, elaborate the species difference in IGL organization, and demonstrate the increased breadth of the circadian visual system.

The electroretinogram (ERG) oscillatory potential (OP) is a high-frequency, low-amplitude potential that is superimposed on the rising phase of the b-wave. It provides noninvasive evaluation of inner retina function in vivo and is a useful tool in basic research as well as in the clinic. While the OP is widely believed to be generated mainly by activity of the inner retina, the exact underlying neural mechanisms are not well understood. We have investigated the retinal mechanisms that underlie OP generation in Dutch-belted rabbits. The OP was isolated by band-filtering (100–1000 Hz) ERG signals. We used pharmacological agents that block specific transmitter receptors or voltage-gated channels in order to examine contributions of various retinal mechanisms to OP generation. Our results show that the OP elicited by a bright brief flash can be classified into early, intermediate, and late subgroups that are likely generated mainly by photoreceptors, action-potential-independent, and action-potential-dependent mechanisms in the ON pathway of the inner retina, respectively. ON bipolar cells themselves make only a small direct contribution to OP generation, as do horizontal cells and neurons in the OFF pathway.

This paper reviews evidence that the superior colliculus (SC) of the midbrain represents visual direction and certain aspects of saccadic eye movements in the distribution of activity across a population of cells. Accurate and precise eye movements appear to be mediated, in part at least, by cells of the SC that have large sensory receptive fields and/or discharge in association with a range of saccades. This implies that visual points or saccade targets are represented by patches rather than points of activity in the SC. Perturbation of the pattern of collicular discharge by focal inactivation modifies saccade amplitude and direction in a way consistent with distributed coding. Several models have been advanced to explain how such a code might be implemented in the colliculus. Evidence related to these hypotheses is examined and continuing uncertainties are identified.

Microiontophoresis of γ-aminobutyric acid (GABA) was used to reversibly inactivate small sites of defined orientation/direction specificity in layers II-IV of cat area 17 while single cells were recorded in the same area at a horizontal distance of ~350–700 jam. We compared the effect of inactivating iso-orientation sites (where orientation preference was within 22.5 deg) and cross-orientation sites (where it differed by 45–90 deg) on orientation tuning and directionality. The influence of iso-orientation inactivation was tested in 33 cells, seven of which were subjected to alternate inactivation of two iso-orientation sites with opposite direction preference. Of the resulting 40 inactivations, only two (5%) caused significant changes in orientation tuning, whereas 26 (65%) elicited effects on directionality: namely, an increase or a decrease in response to a cell's preferred direction when its direction preference was the same as that at an inactivation site, and an increase in response to a cell's nonpreferred direction when its direction preference was opposite that at an inactivation site. It is argued that the decreases in response to the preferred direction reflected a reduction in the strength of intracortical iso-orientation excitatory connections, while the increases in response were due to the loss of iso-orientation inhibition. Of 35 cells subjected to cross-orientation inactivation, only six (17%) showed an effect on directionality, whereas 21 (60%) showed significant broadening of orientation tuning, with an increase in mean tuning width at half-height of 126%. The effects on orientation tuning were due to increases in response to nonoptimal orientations. Changes in directionality also resulted from increased responses (to preferred or nonpreferred directions) and were always accompanied by broadening of tuning. Thus, the effects of cross-orientation inactivation were presumably due to the loss of a cross-orientation inhibitory input that contributes mainly to orientation tuning by suppressing responses to nonoptimal orientations. Differential effects of iso-orientation and cross-orientation inactivation could be elicited in the same cell or in different cells from the same inactivation site. The results suggest the involvement of three different intracortical processes in the generation of orientation tuning and direction selectivity in area 17: (1) suppression of responses to nonoptimal orientations and directions as a result of cross-orientation inhibition and iso-orientation inhibition between cells with opposite direction preferences; (2) amplification of responses to optimal stimuli via iso-orientation excitatory connections; and (3) regulation of cortical amplification via iso-orientation inhibition.

The ganglion cells of the primate retina include two major anatomical and functional classes: P cells which project to the four parvocellular layers of the lateral geniculate nucleus (LGN), and M cells which project to the two magnocellular layers. The characteristics of the P-cell receptive field are central to understanding early form and color vision processing (Kaplan et al., 1990; Schiller & Logothetis, 1990). In this and in the following paper, P-cell dynamics are systematically analyzed in terms of linear and nonlinear response properties. Stimuli that favor either the center or the surround of the receptive field were produced on a CRT and modulated with a broadband signal composed of multiple m-sequences (Benardete et al., 1992b; Benardete & Victor, 1994). The first-order responses were calculated and analyzed in this paper (part I). The findings are: (1) The first-order responses of the center and surround depend linearly on contrast. (2) The dynamics of the center and surround are well described by a bandpass filter model. The most significant difference between center and surround dynamics is a delay of approximately 8 ms in the surround response. (3) In the LGN, these responses are attenuated and delayed by an additional 1–5 ms. (4) The spatial transfer function of the P cell in response to drifting sine gratings at three temporal frequencies was measured. This independent method confirmed the delay between the (first-order) responses of the center and surround. This delay accounts for the dependence of the spatial transfer function on the frequency of stimulation.

Recently, we introduced a phototransduction model that was able to account for the reproducibility of vertebrate rod single-photon responses (SPRs) (Hamer et al., 2003). The model was able to reproduce SPR statistics by means of stochastic activation and inactivation of rhodopsin (R*), transducin (Gα), and phosphodiesterase (PDE). The features needed to capture the SPR statistics were (1) multiple steps of R* inactivation by means of multiple phosphorylations (followed by arrestin capping) and (2) phosphorylation dependence of the affinity between R* and the three molecules competing to bind with R* (Gα, arrestin, and rhodopsin kinase). The model was also able to account for several other rod response features in the dim-flash regime, including SPRs obtained from rods in which various elements of the cascade have been genetically disabled or disrupted. However, the model was not tested under high light-level conditions. We sought to evaluate the extent to which the multiple phosphorylation model could simultaneously account for single-photon response behavior, as well as responses to high light levels causing complete response saturation and/or significant light adaptation (LA). To date no single model, with one set of parameters, has been able to do this. Dim-flash responses and statistics were simulated using a hybrid stochastic/deterministic model and Monte-Carlo methods as in Hamer et al. (2003). A dark-adapted flash series, and stimulus paradigms from the literature eliciting various degrees of light adaptation (LA), were simulated using a full differential equation version of the model that included the addition of Ca2+-feedback onto rhodopsin kinase via recoverin. With this model, using a single set of parameters, we attempted to account for (1) SPR waveforms and statistics (as in Hamer et al., 2003); (2) a full dark-adapted flash-response series, from dim flash to saturating, bright flash levels, from a toad rod; (3) steady-state LA responses, including LA circulating current (as in Koutalos et al., 1995) and LA flash sensitivity measured in rods from four species; (4) step responses from newt rods (Forti et al., 1989) over a large dynamic range; (5) dynamic LA responses, such as the step-flash paradigm of Fain et al. (1989), and the two-flash paradigm of Murnick and Lamb (1996); and (6) the salient response features from four knockout rod preparations. The model was able to meet this stringent test, accounting for almost all the salient qualitative, and many quantitative features, of the responses across this broad array of stimulus conditions, including SPR reproducibility. The model promises to be useful in testing hypotheses regarding both normal and abnormal photoreceptor function, and is a good starting point for development of a full-range model of cone phototransduction. Informative limitations of the model are also discussed.

Psychophysical studies suggest that different colors have different perceptual status: red and blue for example are thought of as elementary sensations whereas yellowish green is not. The dominant account for such perceptual asymmetries attributes them to specificities of the neuronal representation of colors. Alternative accounts involve cultural or linguistic arguments. What these accounts have in common is the idea that there are no asymmetries in the physics of light and surfaces that could underlie the perceptual structure of colors, and this is why neuronal or cultural processes must be invoked as the essential underlying mechanisms that structure color perception. Here, we suggest a biological approach for surface reflection properties that takes into account only the information about light that is accessible to an organism given the photopigments it possesses, and we show that now asymmetries appear in the behavior of surfaces with respect to light. These asymmetries provide a classification of surface properties that turns out to be identical to the one observed in linguistic color categorization across numerous cultures, as pinned down by cross cultural studies. Further, we show that data from psychophysical studies about unique hues and hue cancellation are consistent with the viewpoint that stimuli reported by observers as special are those associated with this singularity-based categorization of surfaces under a standard illuminant. The approach predicts that unique blue and unique yellow should be aligned in chromatic space while unique red and unique green should not, a fact usually conjectured to result from nonlinearities in chromatic pathways.

The Aii amacrine cell of mammalian retina collects signals from several hundred rods and is hypothesized to transmit quantal “single-photon” signals at scotopic (starlight) intensities. One problem for this theory is that the quantal signal from one rod when summed with noise from neighboring rods would be lost if some mechanism did not exist for removing the noise. Several features of the Aii might together accomplish such a noise removal operation: The Aii is interconnected into a syncytial network by gap junctions, suggesting a noise-averaging function, and a quantal signal from one rod appears in five Aii cells due to anatomical divergence. Furthermore, the Aii contains voltage-gated Na+ and K+ channels and fires slow action potentials in vitro, suggesting that it could selectively amplify quantal photon signals embedded in uncorrelated noise. To test this hypothesis, we simulated a square array of AII somas (Rm = 25,000 Ohm-cm2) interconnected by gap junctions using a compartmental model. Simulated noisy inputs to the Aii produced noise (3.5 mV) uncorrelated between adjacent cells, and a gap junction conductance of 200 pS reduced the noise by a factor of 2.5, consistent with theory. Voltage-gated Na+ and K+ channels (Na+: 4 nS, K+: 0.4 nS) produced slow action potentials similar to those found in vitro in the presence of noise. For a narrow range of Na+ and coupling conductance, quantal photon events (-5–10 mV) were amplified nonlinearly by subthreshold regenerative events in the presence of noise. A lower coupling conductance produced spurious action potentials, and a greater conductance reduced amplification. Since the presence of noise in the weakly coupled circuit readily initiates action potentials that tend to spread throughout the AII network, we speculate that this tendency might be controlled in a negative feedback loop by up-modulating coupling or other synaptic conductances in response to spiking activity.