Origin of correlated activity between parasol retinal ganglion cells

Population decoding in rat barrel cortex: optimizing the linear readout of correlated population responses

Sensory information is encoded in the response of neuronal populations. How might this information be decoded by downstream neurons? Here we analyzed the responses of simultaneously recorded barrel cortex neurons to sinusoidal vibrations of varying amplitudes preceded by three adapting stimuli of 0, 6 and 12 µm in amplitude. Using the framework of signal detection theory, we quantified the performance of a linear decoder which sums the responses of neurons after applying an optimum set of weights. Optimum weights were found by the analytical solution that maximized the average signal-to-noise ratio based on Fisher linear discriminant analysis. This provided a biologically plausible decoder that took into account the neuronal variability, covariability, and signal correlations. The optimal decoder achieved consistent improvement in discrimination performance over simple pooling. Decorrelating neuronal responses by trial shuffling revealed that, unlike pooling, the performance of the optimal decoder was minimally affected by noise correlation. In the non-adapted state, noise correlation enhanced the performance of the optimal decoder for some populations. Under adaptation, however, noise correlation always degraded the performance of the optimal decoder. Nonetheless, sensory adaptation improved the performance of the optimal decoder mainly by increasing signal correlation more than noise correlation. Adaptation induced little systematic change in the relative direction of signal and noise. Thus, a decoder which was optimized under the non-adapted state generalized well across states of adaptation.

In the natural environment, animals are constantly exposed to sensory stimulation. A key question in systems neuroscience is how attributes of a sensory stimulus can be “read out” from the activity of a population of brain cells. We chose to investigate this question in the whisker-mediated touch system of rats because of its well-established anatomy and exquisite functionality. The whisker system is one of the major channels through which rodents acquire sensory information about their surrounding environment. The response properties of brain cells dynamically adjust to the prevailing diet of sensory stimulation, a process termed sensory adaptation. Here, we applied a biologically plausible scheme whereby different brain cells contribute to sensory readout with different weights. We established the set of weights that provide the optimal readout under different states of adaptation. The results yield an upper bound for the efficiency of coding sensory information. We found that the ability to decode sensory information improves with adaptation. However, a readout mechanism that does not adjust to the state of adaptation can still perform remarkably well.

Searching for collective behavior in a large network of sensory neurons

Maximum entropy models are the least structured probability distributions that exactly reproduce a chosen set of statistics measured in an interacting network. Here we use this principle to construct probabilistic models which describe the correlated spiking activity of populations of up to 120 neurons in the salamander retina as it responds to natural movies. Already in groups as small as 10 neurons, interactions between spikes can no longer be regarded as small perturbations in an otherwise independent system; for 40 or more neurons pairwise interactions need to be supplemented by a global interaction that controls the distribution of synchrony in the population. Here we show that such "K-pairwise" models--being systematic extensions of the previously used pairwise Ising models--provide an excellent account of the data. We explore the properties of the neural vocabulary by: 1) estimating its entropy, which constrains the population's capacity to represent visual information; 2) classifying activity patterns into a small set of metastable collective modes; 3) showing that the neural codeword ensembles are extremely inhomogenous; 4) demonstrating that the state of individual neurons is highly predictable from the rest of the population, allowing the capacity for error correction.

Transformation of stimulus correlations by the retina

Redundancies and correlations in the responses of sensory neurons may seem to waste neural resources, but they can also carry cues about structured stimuli and may help the brain to correct for response errors. To investigate the effect of stimulus structure on redundancy in retina, we measured simultaneous responses from populations of retinal ganglion cells presented with natural and artificial stimuli that varied greatly in correlation structure; these stimuli and recordings are publicly available online. Responding to spatio-temporally structured stimuli such as natural movies, pairs of ganglion cells were modestly more correlated than in response to white noise checkerboards, but they were much less correlated than predicted by a non-adapting functional model of retinal response. Meanwhile, responding to stimuli with purely spatial correlations, pairs of ganglion cells showed increased correlations consistent with a static, non-adapting receptive field and nonlinearity. We found that in response to spatio-temporally correlated stimuli, ganglion cells had faster temporal kernels and tended to have stronger surrounds. These properties of individual cells, along with gain changes that opposed changes in effective contrast at the ganglion cell input, largely explained the pattern of pairwise correlations across stimuli where receptive field measurements were possible.

An influential theory of early sensory processing argues that sensory circuits should conserve scarce resources in their outputs by reducing correlations present in their inputs. Measuring simultaneous responses from large numbers of retinal ganglion cells responding to widely different classes of visual stimuli, we find that output correlations increase when we present stimuli with spatial, but not temporal, correlations. On the other hand, we find evidence that retina adjusts to spatio-temporal structure so that retinal output correlations change less than input correlations would predict. Changes in the receptive field properties of individual cells, along with gain changes, largely explain this relative constancy of correlations over the population.

Synaptic noise is an information bottleneck in the inner retina during dynamic visual stimulation

In daylight, noise generated by cones determines the fidelity with which visual signals are initially encoded. Subsequent stages of visual processing require synapses from bipolar cells to ganglion cells, but whether these synapses generate a significant amount of noise was unknown. To characterize noise generated by these synapses, we recorded excitatory postsynaptic currents from mammalian retinal ganglion cells and subjected them to a computational noise analysis. The release of transmitter quanta at bipolar cell synapses contributed substantially to the noise variance found in the ganglion cell, causing a significant loss of fidelity from bipolar cell array to postsynaptic ganglion cell. Virtually all the remaining noise variance originated in the presynaptic circuit. Circuit noise had a frequency content similar to noise shared by ganglion cells but a very different frequency content from noise from bipolar cell synapses, indicating that these synapses constitute a source of independent noise not shared by ganglion cells. These findings contribute a picture of daylight retinal circuits where noise from cones and noise generated by synaptic transmission of cone signals significantly limit visual fidelity.

NaV1.1 channels in axon initial segments of bipolar cells augment input to magnocellular visual pathways in the primate retina

In the primate visual system, the ganglion cells of the magnocellular pathway underlie motion and flicker detection and are relatively transient, while the more sustained ganglion cells of the parvocellular pathway have comparatively lower temporal resolution, but encode higher spatial frequencies. Although it is presumed that functional differences in bipolar cells contribute to the tuning of the two pathways, the properties of the relevant bipolar cells have not yet been examined in detail. Here, by making patch-clamp recordings in acute slices of macaque retina, we show that the bipolar cells within the magnocellular pathway, but not the parvocellular pathway, exhibit voltage-gated sodium (NaV), T-type calcium (CaV), and hyperpolarization-activated, cyclic nucleotide-gated (HCN) currents, and can generate action potentials. Using immunohistochemistry in macaque and human retinae, we show that NaV1.1 is concentrated in an axon initial segment (AIS)-like region of magnocellular pathway bipolar cells, a specialization not seen in transient bipolar cells of other vertebrates. In contrast, CaV3.1 channels were localized to the somatodendritic compartment and proximal axon, but were excluded from the AIS, while HCN1 channels were concentrated in the axon terminal boutons. Simulations using a compartmental model reproduced physiological results and indicate that magnocellular pathway bipolar cells initiate spikes in the AIS. Finally, we demonstrate that NaV channels in bipolar cells augment excitatory input to parasol ganglion cells of the magnocellular pathway. Overall, the results demonstrate that selective expression of voltage-gated channels contributes to the establishment of parallel processing in the major visual pathways of the primate retina.

Two-photon imaging of nonlinear glutamate release dynamics at bipolar cell synapses in the mouse retina

Alpha/Y-type retinal ganglion cells encode visual information with a receptive field composed of nonlinear subunits. This nonlinear subunit structure enhances sensitivity to patterns composed of high spatial frequencies. The Y-cell's subunits are the presynaptic bipolar cells, but the mechanism for the nonlinearity remains incompletely understood. We investigated the synaptic basis of the subunit nonlinearity by combining whole-cell recording of mouse Y-type ganglion cells with two-photon fluorescence imaging of a glutamate sensor (iGluSnFR) expressed on their dendrites and throughout the inner plexiform layer. A control experiment designed to assess iGluSnFR's dynamic range showed that fluorescence responses from Y-cell dendrites increased proportionally with simultaneously recorded excitatory current. Spatial resolution was sufficient to readily resolve independent release at intermingled ON and OFF bipolar terminals. iGluSnFR responses at Y-cell dendrites showed strong surround inhibition, reflecting receptive field properties of presynaptic release sites. Responses to spatial patterns located the origin of the Y-cell nonlinearity to the bipolar cell output, after the stage of spatial integration. The underlying mechanism differed between OFF and ON pathways: OFF synapses showed transient release and strong rectification, whereas ON synapses showed relatively sustained release and weak rectification. At ON synapses, the combination of fast release onset with slower release offset explained the nonlinear response of the postsynaptic ganglion cell. Imaging throughout the inner plexiform layer, we found transient, rectified release at the central-most levels, with increasingly sustained release near the borders. By visualizing glutamate release in real time, iGluSnFR provides a powerful tool for characterizing glutamate synapses in intact neural circuits.

Gap junctions are essential for generating the correlated spike activity of neighboring retinal ganglion cells

Neurons throughout the brain show spike activity that is temporally correlated to that expressed by their neighbors, yet the generating mechanism(s) remains unclear. In the retina, ganglion cells (GCs) show robust, concerted spiking that shapes the information transmitted to central targets. Here we report the synaptic circuits responsible for generating the different types of concerted spiking of GC neighbors in the mouse retina. The most precise concerted spiking was generated by reciprocal electrical coupling of GC neighbors via gap junctions, whereas indirect electrical coupling to a common cohort of amacrine cells generated the correlated activity with medium precision. In contrast, the correlated spiking with the lowest temporal precision was produced by shared synaptic inputs carrying photoreceptor noise. Overall, our results demonstrate that different synaptic circuits generate the discrete types of GC correlated activity. Moreover, our findings expand our understanding of the roles of gap junctions in the retina, showing that they are essential for generating all forms of concerted GC activity transmitted to central brain targets.

Method for stationarity-segmentation of spike train data with application to the Pearson cross-correlation

Correlations among neurons are supposed to play an important role in computation and information coding in the nervous system. Empirically, functional interactions between neurons are most commonly assessed by cross-correlation functions. Recent studies have suggested that pairwise correlations may indeed be sufficient to capture most of the information present in neural interactions. Many applications of correlation functions, however, implicitly tend to assume that the underlying processes are stationary. This assumption will usually fail for real neurons recorded in vivo since their activity during behavioral tasks is heavily influenced by stimulus-, movement-, or cognition-related processes as well as by more general processes like slow oscillations or changes in state of alertness. To address the problem of nonstationarity, we introduce a method for assessing stationarity empirically and then “slicing” spike trains into stationary segments according to the statistical definition of weak-sense stationarity. We examine pairwise Pearson cross-correlations (PCCs) under both stationary and nonstationary conditions and identify another source of covariance that can be differentiated from the covariance of the spike times and emerges as a consequence of residual nonstationarities after the slicing process: the covariance of the firing rates defined on each segment. Based on this, a correction of the PCC is introduced that accounts for the effect of segmentation. We probe these methods both on simulated data sets and on in vivo recordings from the prefrontal cortex of behaving rats. Rather than for removing nonstationarities, the present method may also be used for detecting significant events in spike trains.

Regulation of spatial selectivity by crossover inhibition

Signals throughout the nervous system diverge into parallel excitatory and inhibitory pathways that later converge on downstream neurons to control their spike output. Converging excitatory and inhibitory synaptic inputs can exhibit a variety of temporal relationships. A common motif is feedforward inhibition, in which an increase (decrease) in excitatory input precedes a corresponding increase (decrease) in inhibitory input. The delay of inhibitory input relative to excitatory input originates from an extra synapse in the circuit shaping inhibitory input. Another common motif is push-pull or "crossover" inhibition, in which increases (decreases) in excitatory input occur together with decreases (increases) in inhibitory input. Primate On midget ganglion cells receive primarily feedforward inhibition and On parasol cells receive primarily crossover inhibition; this difference provides an opportunity to study how each motif shapes the light responses of cell types that play a key role in visual perception. For full-field stimuli, feedforward inhibition abbreviated and attenuated responses of On midget cells, while crossover inhibition, though plentiful, had surprisingly little impact on the responses of On parasol cells. Spatially structured stimuli, however, could cause excitatory and inhibitory inputs to On parasol cells to increase together, adopting a temporal relation very much like that for feedforward inhibition. In this case, inhibitory inputs substantially abbreviated a cell's spike output. Thus inhibitory input shapes the temporal stimulus selectivity of both midget and parasol ganglion cells, but its impact on responses of parasol cells depends strongly on the spatial structure of the light inputs.

Independent control of reciprocal and lateral inhibition at the axon terminal of retinal bipolar cells

Bipolar cells (BCs), the second order neurons in the vertebrate retina, receive two types of GABAergic feedback inhibition at their axon terminal: reciprocal and lateral inhibition. It has been suggested that two types of inhibition may be mediated by different pathways. However, how each inhibition is controlled by excitatory BC output remains to be clarified. Here, we applied single/dual whole cell recording techniques to the axon terminal of electrically coupled BCs in slice preparation of the goldfish retina, and found that each inhibition was regulated independently. Activation voltage of each inhibition was different: strong output from a single BC activated reciprocal inhibition, but could not activate lateral inhibition. Outputs from multiple BCs were essential for activation of lateral inhibition. Pharmacological examinations revealed that composition of transmitter receptors and localization of Na(+) channels were different between two inhibitory pathways, suggesting that different amacrine cells may mediate each inhibition. Depending on visual inputs, each inhibition could be driven independently. Model simulation showed that reciprocal and lateral inhibition cooperatively reduced BC outputs as well as background noise, thereby preserving high signal-to-noise ratio. Therefore, we conclude that excitatory BC output is efficiently regulated by the dual operating mechanisms of feedback inhibition without deteriorating the quality of visual signals.

Controlling gain one photon at a time

Adaptation is a salient property of sensory processing. All adaptational or gain control mechanisms face the challenge of obtaining a reliable estimate of the property of the input to be adapted to and obtaining this estimate sufficiently rapidly to be useful. Here, we explore how the primate retina balances the need to change gain rapidly and reliably when photons arrive rarely at individual rod photoreceptors. We find that the weakest backgrounds that decrease the gain of the retinal output signals are similar to those that increase human behavioral threshold, and identify a novel site of gain control in the retinal circuitry. Thus, surprisingly, the gain of retinal signals begins to decrease essentially as soon as background lights are detectable; under these conditions, gain control does not rely on a highly averaged estimate of the photon count, but instead signals from individual photon absorptions trigger changes in gain.

DOI:http://dx.doi.org/10.7554/eLife.00467.001

To process the sights and sounds around us, our senses must be attuned to a huge range of signals: from barely audible whispers to deafening rock concerts, and from dim glimmers of light to bright spotlights. Sensory neurons face the challenge of encoding this huge range of inputs within their much more restricted response range. Thus, neurons in our eyes and ears must continually adjust their gain or sensitivity to match changes in the light and sound inputs. These gain control processes must operate rapidly to keep up with the ever-changing input signals, but must also operate accurately so as not to distort the inputs.

The trade-off between rapid and accurate gain control can be illustrated by considering how the retina processes information at low light levels. There are two main types of light-sensitive cells in the retina: rods and cones. Vision at night relies on the ability of the rods to detect single photons—the smallest unit of light. In starlight, an individual rod will register photons only rarely, and most of the time, the majority of the rods will not register any photons. Neurons in the retinal circuits that read out the rod signals receive input from hundreds or thousands of rods, and those rod inputs are highly amplified to allow detection of the responses produced when a tiny fraction of the rods absorbs a photon. But this amplification is dangerous, as it could easily saturate retinal signals when light levels increase. Gain control mechanisms are needed to avoid such saturation.

Schwartz and Rieke now add to our understanding of this process by examining how the retinas of non-human primates behave in low light. They reveal that levels of background light that can only just be detected behaviorally trigger retinal gain controls; these gain controls operate when less than 1% of rods absorb a photon. Under these conditions, the physics of light itself will cause considerable variability in the stream of photons arriving at the retina, leading to high variability in the gain of retinal responses. Nonetheless, changes in gain occurred rapidly following changes in background, indicating that the underlying mechanisms spend little time averaging incident photons. Taken together, these findings will require revisiting our ideas about how adaptational mechanisms balance the competing demands of speed and reliability to help us see the world around us.

DOI:http://dx.doi.org/10.7554/eLife.00467.002

A model-based spike sorting algorithm for removing correlation artifacts in multi-neuron recordings

We examine the problem of estimating the spike trains of multiple neurons from voltage traces recorded on one or more extracellular electrodes. Traditional spike-sorting methods rely on thresholding or clustering of recorded signals to identify spikes. While these methods can detect a large fraction of the spikes from a recording, they generally fail to identify synchronous or near-synchronous spikes: cases in which multiple spikes overlap. Here we investigate the geometry of failures in traditional sorting algorithms, and document the prevalence of such errors in multi-electrode recordings from primate retina. We then develop a method for multi-neuron spike sorting using a model that explicitly accounts for the superposition of spike waveforms. We model the recorded voltage traces as a linear combination of spike waveforms plus a stochastic background component of correlated Gaussian noise. Combining this measurement model with a Bernoulli prior over binary spike trains yields a posterior distribution for spikes given the recorded data. We introduce a greedy algorithm to maximize this posterior that we call "binary pursuit". The algorithm allows modest variability in spike waveforms and recovers spike times with higher precision than the voltage sampling rate. This method substantially corrects cross-correlation artifacts that arise with conventional methods, and substantially outperforms clustering methods on both real and simulated data. Finally, we develop diagnostic tools that can be used to assess errors in spike sorting in the absence of ground truth.

Gap junctional coupling in the vertebrate retina: variations on one theme?

Gap junctions connect cells in the bodies of all multicellular organisms, forming either homologous or heterologous (i.e. established between identical or different cell types, respectively) cell-to-cell contacts by utilizing identical (homotypic) or different (heterotypic) connexin protein subunits. Gap junctions in the nervous system serve electrical signaling between neurons, thus they are also called electrical synapses. Such electrical synapses are particularly abundant in the vertebrate retina where they are specialized to form links between neurons as well as glial cells. In this article, we summarize recent findings on retinal cell-to-cell coupling in different vertebrates and identify general features in the light of the evergrowing body of data. In particular, we describe and discuss tracer coupling patterns, connexin proteins, junctional conductances and modulatory processes. This multispecies comparison serves to point out that most features are remarkably conserved across the vertebrate classes, including (i) the cell types connected via electrical synapses; (ii) the connexin makeup and the conductance of each cell-to-cell contact; (iii) the probable function of each gap junction in retinal circuitry; (iv) the fact that gap junctions underlie both electrical and/or tracer coupling between glial cells. These pan-vertebrate features thus demonstrate that retinal gap junctions have changed little during the over 500 million years of vertebrate evolution. Therefore, the fundamental architecture of electrically coupled retinal circuits seems as old as the retina itself, indicating that gap junctions deeply incorporated in retinal wiring from the very beginning of the eye formation of vertebrates. In addition to hard wiring provided by fast synaptic transmitter-releasing neurons and soft wiring contributed by peptidergic, aminergic and purinergic systems, electrical coupling may serve as the 'skeleton' of lateral processing, enabling important functions such as signal averaging and synchronization.

Dynamic range of vertebrate retina ganglion cells: importance of active dendrites and coupling by electrical synapses

The vertebrate retina has a very high dynamic range. This is due to the concerted action of its diverse cell types. Ganglion cells, which are the output cells of the retina, have to preserve this high dynamic range to convey it to higher brain areas. Experimental evidence shows that the firing response of ganglion cells is strongly correlated with their total dendritic area and only weakly correlated with their dendritic branching complexity. On the other hand, theoretical studies with simple neuron models claim that active and large dendritic trees enhance the dynamic range of single neurons. Theoretical models also claim that electrical coupling between ganglion cells via gap junctions enhances their collective dynamic range. In this work we use morphologically reconstructed multi-compartmental ganglion cell models to perform two studies. In the first study we investigate the relationship between single ganglion cell dynamic range and number of dendritic branches/total dendritic area for both active and passive dendrites. Our results support the claim that large and active dendrites enhance the dynamic range of a single ganglion cell and show that total dendritic area has stronger correlation with dynamic range than with number of dendritic branches. In the second study we investigate the dynamic range of a square array of ganglion cells with passive or active dendritic trees coupled with each other via dendrodendritic gap junctions. Our results suggest that electrical coupling between active dendritic trees enhances the dynamic range of the ganglion cell array in comparison with both the uncoupled case and the coupled case with cells with passive dendrites. The results from our detailed computational modeling studies suggest that the key properties of the ganglion cells that endow them with a large dynamic range are large and active dendritic trees and electrical coupling via gap junctions.

Recording from defined populations of retinal ganglion cells using a high-density CMOS-integrated microelectrode array with real-time switchable electrode selection

In order to understand how retinal circuits encode visual scenes, the neural activity of defined populations of retinal ganglion cells (RGCs) has to be investigated. Here we report on a method for stimulating, detecting, and subsequently targeting defined populations of RGCs. The possibility to select a distinct population of RGCs for extracellular recording enables the design of experiments that can increase our understanding of how these neurons extract precise spatio-temporal features from the visual scene, and how the brain interprets retinal signals. We used light stimulation to elicit a response from physiologically distinct types of RGCs and then utilized the dynamic-configurability capabilities of a microelectronics-based high-density microelectrode array (MEA) to record their synchronous action potentials. The layout characteristics of the MEA made it possible to stimulate and record from multiple, highly overlapping RGCs simultaneously without light-induced artifacts. The high-density of electrodes and the high signal-to-noise ratio of the MEA circuitry allowed for recording of the activity of each RGC on 14±7 electrodes. The spatial features of the electrical activity of each RGC greatly facilitated spike sorting. We were thus able to localize, identify and record from defined RGCs within a region of mouse retina. In addition, we stimulated and recorded from genetically modified RGCs to demonstrate the applicability of optogenetic methods, which introduces an additional feature to target a defined cell type. The developed methodologies can likewise be applied to other neuronal preparations including brain slices or cultured neurons.

A-current and type I/type II transition determine collective spiking from common input

The mechanisms and impact of correlated, or synchronous, firing among pairs and groups of neurons are under intense investigation throughout the nervous system. A ubiquitous circuit feature that can give rise to such correlations consists of overlapping, or common, inputs to pairs and populations of cells, leading to common spike train responses. Here, we use computational tools to study how the transfer of common input currents into common spike outputs is modulated by the physiology of the recipient cells. We focus on a key conductance, g(A), for the A-type potassium current, which drives neurons between "type II" excitability (low g(A)), and "type I" excitability (high g(A)). Regardless of g(A), cells transform common input fluctuations into a tendency to spike nearly simultaneously. However, this process is more pronounced at low g(A) values. Thus, for a given level of common input, type II neurons produce spikes that are relatively more correlated over short time scales. Over long time scales, the trend reverses, with type II neurons producing relatively less correlated spike trains. This is because these cells' increased tendency for simultaneous spiking is balanced by an anticorrelation of spikes at larger time lags. These findings extend and interpret prior findings for phase oscillators to conductance-based neuron models that cover both oscillatory (superthreshold) and subthreshold firing regimes. We demonstrate a novel implication for neural signal processing: downstream cells with long time constants are selectively driven by type I cell populations upstream and those with short time constants by type II cell populations. Our results are established via high-throughput numerical simulations and explained via the cells' filtering properties and nonlinear dynamics.

Amacrine and bipolar inputs to midget and parasol ganglion cells in marmoset retina

Retinal ganglion cells receive excitatory synapses from bipolar cells and inhibitory synapses from amacrine cells. Previous studies in primate suggest that the strength of inhibitory amacrine input is greater to cells in peripheral retina than to foveal (central) cells. A comprehensive study of a large number of ganglion cells at different eccentricities, however, is still lacking. Here, we compared the amacrine and bipolar input to midget and parasol ganglion cells in central and peripheral retina of marmosets (Callithrix jacchus). Ganglion cells were labeled by retrograde filling from the lateral geniculate nucleus or by intracellular injection. Presumed amacrine input was identified with antibodies against gephyrin; presumed bipolar input was identified with antibodies against the GluR4 subunit of the AMPA receptor. In vertical sections, about 40% of gephyrin immunoreactive (IR) puncta were colocalized with GABAA receptor subunits, whereas immunoreactivity for gephyrin and GluR4 was found at distinct sets of puncta. The density of gephyrin IR puncta associated with ganglion cell dendrites was comparable for midget and parasol cells at all eccentricities studied (up to 2 mm or about 16 degrees of visual angle for midget cells and up to 10 mm or >80 degrees of visual angle for parasol cells). In central retina, the densities of gephyrin IR and GluR4 IR puncta associated with the dendrites of midget and parasol cells are comparable, but the average density of GluR4 IR puncta decreased slightly in peripheral parasol cells. These anatomical results indicate that the ratio of amacrine to bipolar input does not account for the distinct functional properties of parasol and midget cells or for functional differences between cells of the same type in central and peripheral retina.

Adaptation-dependent synchronous activity contributes to receptive field size change of bullfrog retinal ganglion cell

Nearby retinal ganglion cells of similar functional subtype have a tendency to discharge spikes in synchrony. The synchronized activity is involved in encoding some aspects of visual input. On the other hand, neurons always continuously adjust their activities in adaptation to some features of visual stimulation, including mean ambient light, contrast level, etc. Previous studies on adaptation were primarily focused on single neuronal activity, however, it is also intriguing to investigate the adaptation process in population neuronal activities. In the present study, by using multi-electrode recording system, we simultaneously recorded spike discharges from a group of dimming detectors (OFF-sustained type ganglion cells) in bullfrog retina. The changes in receptive field properties and synchronization strength during contrast adaptation were analyzed. It was found that, when perfused using normal Ringer's solution, single neuronal receptive field size was reduced during contrast adaptation, which was accompanied by weakening in synchronization strength between adjacent neurons' activities. When dopamine (1 µM) was applied, the adaptation-related receptive field area shrinkage and synchronization weakening were both eliminated. The activation of D1 receptor was involved in the adaptation-related modulation of synchronization and receptive field. Our results thus suggest that the size of single neuron's receptive field is positively related to the strength of its synchronized activity with its neighboring neurons, and the dopaminergic pathway is responsible for the modulation of receptive field property and synchronous activity of the ganglion cells during the adaptation process.

The effect of mean luminance change and grating pedestals on contrast perception: model simulations suggest a common, retinal, origin

The percept of a contrast target is substantially affected by co-occurring changes in mean luminance or underlying ("pedestal") contrast elements. These two types of modulatory effects have traditionally been studied as separate phenomena. However, regardless of different higher-level mechanisms, both classes of phenomena will necessarily also depend on shared mechanisms in the first stages of vision, starting with the primary responses of photoreceptors. Here we present model simulations showing that important aspects of both classes may be explained by the temporal dynamics of photoreceptor responses read by integrate-and-fire operators. The model is physiologically justified and all its parameters are constrained by experimental evidence. Although there remains plenty of room for additional mechanisms to shape the exact quantitative realization of the perceptual functions in different situations, we suggest that signature features may be inherited from primary retinal signaling.

Amacrine cell-mediated input to bipolar cells: Variations on a common mechanistic theme

Feedback is a ubiquitous feature of neural circuits in the mammalian central nervous system (CNS). Analogous to pure electronic circuits, neuronal feedback provides either a positive or negative influence on the output of upstream components/neurons. Although the particulars (i.e., connectivity, physiological encoding/processing/signaling) of circuits in higher areas of the brain are often unclear, the inner retina proves an excellent model for studying both the anatomy and physiology of feedback circuits within the functional context of visual processing. Inner retinal feedback to bipolar cells is almost entirely mediated by a single class of interneurons, the amacrine cells. Although this might sound like a simple circuit arrangement with an equally simple function, anatomical, molecular, and functional evidence suggest that amacrine cells represent an extremely diverse class of CNS interneurons that contribute to a variety of retinal processes. In this review, I classify the amacrine cells according to their anatomical output synapses and target cell(s) (i.e., bipolar cells, ganglion cells, and/or amacrine cells) and discuss specifically our current understandings of amacrine cell-mediated feedback and output to bipolar cells on the synaptic, cellular, and circuit levels, while drawing connections to visual processing.

Modeling the impact of common noise inputs on the network activity of retinal ganglion cells

Synchronized spontaneous firing among retinal ganglion cells (RGCs), on timescales faster than visual responses, has been reported in many studies. Two candidate mechanisms of synchronized firing include direct coupling and shared noisy inputs. In neighboring parasol cells of primate retina, which exhibit rapid synchronized firing that has been studied extensively, recent experimental work indicates that direct electrical or synaptic coupling is weak, but shared synaptic input in the absence of modulated stimuli is strong. However, previous modeling efforts have not accounted for this aspect of firing in the parasol cell population. Here we develop a new model that incorporates the effects of common noise, and apply it to analyze the light responses and synchronized firing of a large, densely-sampled network of over 250 simultaneously recorded parasol cells. We use a generalized linear model in which the spike rate in each cell is determined by the linear combination of the spatio-temporally filtered visual input, the temporally filtered prior spikes of that cell, and unobserved sources representing common noise. The model accurately captures the statistical structure of the spike trains and the encoding of the visual stimulus, without the direct coupling assumption present in previous modeling work. Finally, we examined the problem of decoding the visual stimulus from the spike train given the estimated parameters. The common-noise model produces Bayesian decoding performance as accurate as that of a model with direct coupling, but with significantly more robustness to spike timing perturbations.

Balanced synaptic input shapes the correlation between neural spike trains

Stimulus properties, attention, and behavioral context influence correlations between the spike times produced by a pair of neurons. However, the biophysical mechanisms that modulate these correlations are poorly understood. With a combined theoretical and experimental approach, we show that the rate of balanced excitatory and inhibitory synaptic input modulates the magnitude and timescale of pairwise spike train correlation. High rate synaptic inputs promote spike time synchrony rather than long timescale spike rate correlations, while low rate synaptic inputs produce opposite results. This correlation shaping is due to a combination of enhanced high frequency input transfer and reduced firing rate gain in the high input rate state compared to the low state. Our study extends neural modulation from single neuron responses to population activity, a necessary step in understanding how the dynamics and processing of neural activity change across distinct brain states.

A processing platform for optoelectronic/optogenetic retinal prosthesis

The field of retinal prosthesis has been steadily developing over the last two decades. Despite the many obstacles, clinical trials for electronic approaches are in progress and already demonstrating some success. Optogenetic/optoelectronic retinal prosthesis may prove to have even greater capabilities. Although resolutions are now moving beyond recognition of simple shapes, it will nevertheless be poor compared to normal vision. If we define the aim to be to return mobility and natural scene recognition to the patient, it is important to maximize the useful visual information we attempt to transfer. In this paper, we highlight a method to simplify the scene, perform spatial image compression, and then apply spike coding. We then show the potential for translation on standard consumer processors. The algorithms are applicable to all forms of visual prosthesis, but we particularly focus on optogenetic approaches.

Cone photoreceptor contributions to noise and correlations in the retinal output

Transduction and synaptic noise generated in retinal cone photoreceptors determine the fidelity with which light inputs are encoded, and the readout of cone signals by downstream circuits determines whether this fidelity is used for vision. We examined the effect of cone noise on visual signals by measuring its contribution to correlated noise in primate retinal ganglion cells. Correlated noise was strong in the responses of dissimilar cell types with shared cone inputs. The dynamics of cone noise could account for rapid correlations in ganglion cell activity, and the extent of shared cone input could explain correlation strength. Furthermore, correlated noise limited the fidelity with which visual signals were encoded by populations of ganglion cells. Thus, a simple picture emerges: cone noise, traversing the retina through diverse pathways, accounts for most of the noise and correlations in the retinal output and constrains how higher centers exploit signals carried by parallel visual pathways.

Cell types, circuits, computation

How does the connectivity of a neuronal circuit, together with the individual properties of the cell types that take part in it, result in a given computation? We examine this question in the context of retinal circuits. We suggest that the retina can be viewed as a parallel assemblage of many small computational devices, highly stereotypical and task-specific circuits afferent to a given ganglion cell type, and we discuss some rules that govern computation in these devices. Multi-device processing in retina poses conceptual problems when it is contrasted with cortical processing. We lay out open questions both on processing in retinal circuits and on implications for cortical processing of retinal inputs.

Encoding visual information in retinal ganglion cells with prosthetic stimulation

Retinal prostheses aim to restore functional vision to those blinded by outer retinal diseases using electric stimulation of surviving retinal neurons. The ability to replicate the spatiotemporal pattern of ganglion cell spike trains present under normal viewing conditions is presumably an important factor for restoring high-quality vision. In order to replicate such activity with a retinal prosthesis, it is important to consider both how visual information is encoded in ganglion cell spike trains, and how retinal neurons respond to electric stimulation. The goal of the current review is to bring together these two concepts in order to guide the development of more effective stimulation strategies. We review the experiments to date that have studied how retinal neurons respond to electric stimulation and discuss these findings in the context of known retinal signaling strategies. The results from such in vitro studies reveal the advantages and disadvantages of activating the ganglion cell directly with the electric stimulus (direct activation) as compared to activation of neurons that are presynaptic to the ganglion cell (indirect activation). While direct activation allows high temporal but low spatial resolution, indirect activation yields improved spatial resolution but poor temporal resolution. Finally, we use knowledge gained from in vitro experiments to infer the patterns of elicited activity in ongoing human trials, providing insights into some of the factors limiting the quality of prosthetic vision.

Light increases the gap junctional coupling of retinal ganglion cells

We examined the effect of light adaptation on the gap junctional coupling of α-ganglion cells (α-GCs) in rabbit and mouse retinas. We assayed changes in coupling by measuring parameters of tracer coupling following injection of α-GCs with Neurobiotin and the concerted spike activity of α-GC neighbours under dark- and light-adapted conditions. We found that light adaptation using mesopic or photopic background lights resulted in a dramatic increase in the labelling intensity, number, and spatial extent of ganglion and amacrine cells coupled to OFF α-GCs when compared to levels seen under dark adaptation. While this augmentation of coupling by light did not produce an increase in the concerted spontaneous activity of OFF α-GC neighbours, it did significantly increase correlated light-evoked spiking. This was seen as an increase in the number of correlated spikes for α-GC neighbours and an extension of correlations to second-tier neighbours that was not seen under dark-adapted conditions. Pharmacological studies in the rabbit retina indicated that dopamine mediates the observed changes in coupling by differentially activating D1 and D2 receptors under different adaptation states. In this scheme, activation of dopamine D1 receptors following light exposure triggers cAMP-mediated intracellular pathways resulting in an increase in gap junctional conductance. Overall, our results indicate that as we move from night to day there is an enhanced electrical coupling between α-GCs, thereby increasing the concerted activity believed to strengthen the capacity and efficiency of information flow across the optic nerve.

Fast inference of interactions in assemblies of stochastic integrate-and-fire neurons from spike recordings

We present two Bayesian procedures to infer the interactions and external currents in an assembly of stochastic integrate-and-fire neurons from the recording of their spiking activity. The first procedure is based on the exact calculation of the most likely time courses of the neuron membrane potentials conditioned by the recorded spikes, and is exact for a vanishing noise variance and for an instantaneous synaptic integration. The second procedure takes into account the presence of fluctuations around the most likely time courses of the potentials, and can deal with moderate noise levels. The running time of both procedures is proportional to the number S of spikes multiplied by the squared number N of neurons. The algorithms are validated on synthetic data generated by networks with known couplings and currents. We also reanalyze previously published recordings of the activity of the salamander retina (including from 32 to 40 neurons, and from 65,000 to 170,000 spikes). We study the dependence of the inferred interactions on the membrane leaking time; the differences and similarities with the classical cross-correlation analysis are discussed.

Noise correlations improve response fidelity and stimulus encoding

Computation in the nervous system often relies on the integration of signals from parallel circuits with different functional properties. Correlated noise in these inputs can, in principle, have diverse and dramatic effects on the reliability of the resulting computations. Such theoretical predictions have rarely been tested experimentally because of a scarcity of preparations that permit measurement of both the covariation of a neuron's input signals and the effect on a cell's output of manipulating such covariation. Here we introduce a method to measure covariation of the excitatory and inhibitory inputs a cell receives. This method revealed strong correlated noise in the inputs to two types of retinal ganglion cell. Eliminating correlated noise without changing other input properties substantially decreased the accuracy with which a cell's spike outputs encoded light inputs. Thus, covariation of excitatory and inhibitory inputs can be a critical determinant of the reliability of neural coding and computation.

Modulation of temporal precision in thalamic population responses to natural visual stimuli

Natural visual stimuli have highly structured spatial and temporal properties which influence the way visual information is encoded in the visual pathway. In response to natural scene stimuli, neurons in the lateral geniculate nucleus (LGN) are temporally precise - on a time scale of 10-25 ms - both within single cells and across cells within a population. This time scale, established by non stimulus-driven elements of neuronal firing, is significantly shorter than that of natural scenes, yet is critical for the neural representation of the spatial and temporal structure of the scene. Here, a generalized linear model (GLM) that combines stimulus-driven elements with spike-history dependence associated with intrinsic cellular dynamics is shown to predict the fine timing precision of LGN responses to natural scene stimuli, the corresponding correlation structure across nearby neurons in the population, and the continuous modulation of spike timing precision and latency across neurons. A single model captured the experimentally observed neural response, across different levels of contrasts and different classes of visual stimuli, through interactions between the stimulus correlation structure and the nonlinearity in spike generation and spike history dependence. Given the sensitivity of the thalamocortical synapse to closely timed spikes and the importance of fine timing precision for the faithful representation of natural scenes, the modulation of thalamic population timing over these time scales is likely important for cortical representations of the dynamic natural visual environment.

Correlated firing among major ganglion cell types in primate retina

Retinal ganglion cells exhibit substantial correlated firing: a tendency to fire nearly synchronously at rates different from those expected by chance. These correlations suggest that network interactions significantly shape the visual signal transmitted from the eye to the brain. This study describes the degree and structure of correlated firing among the major ganglion cell types in primate retina. Correlated firing among ON and OFF parasol, ON and OFF midget, and small bistratified cells, which together constitute roughly 75% of the input to higher visual areas, was studied using large-scale multi-electrode recordings. Correlated firing in the presence of constant, spatially uniform illumination exhibited characteristic strength, time course and polarity within and across cell types. Pairs of nearby cells with the same light response polarity were positively correlated; cells with the opposite polarity were negatively correlated. The strength of correlated firing declined systematically with distance for each cell type, in proportion to the degree of receptive field overlap. The pattern of correlated firing across cell types was similar at photopic and scotopic light levels, although additional slow correlations were present at scotopic light levels. Similar results were also observed in two other retinal ganglion cell types. Most of these observations are consistent with the hypothesis that shared noise from photoreceptors is the dominant cause of correlated firing. Surprisingly, small bistratified cells, which receive ON input from S cones, fired synchronously with ON parasol and midget cells, which receive ON input primarily from L and M cones. Collectively, these results provide an overview of correlated firing across cell types in the primate retina, and constraints on the underlying mechanisms.

A new look at state-space models for neural data

State space methods have proven indispensable in neural data analysis. However, common methods for performing inference in state-space models with non-Gaussian observations rely on certain approximations which are not always accurate. Here we review direct optimization methods that avoid these approximations, but that nonetheless retain the computational efficiency of the approximate methods. We discuss a variety of examples, applying these direct optimization techniques to problems in spike train smoothing, stimulus decoding, parameter estimation, and inference of synaptic properties. Along the way, we point out connections to some related standard statistical methods, including spline smoothing and isotonic regression. Finally, we note that the computational methods reviewed here do not in fact depend on the state-space setting at all; instead, the key property we are exploiting involves the bandedness of certain matrices. We close by discussing some applications of this more general point of view, including Markov chain Monte Carlo methods for neural decoding and efficient estimation of spatially-varying firing rates.

NMDA receptor contributions to visual contrast coding

In the retina, it is not well understood how visual processing depends on AMPA- and NMDA-type glutamate receptors. Here we investigated how these receptors contribute to contrast coding in identified guinea pig ganglion cell types in vitro. NMDA-mediated responses were negligible in ON alpha cells but substantial in OFF alpha and delta cells. OFF delta cell NMDA receptors were composed of GluN2B subunits. Using a novel deconvolution method, we determined the individual contributions of AMPA, NMDA, and inhibitory currents to light responses of each cell type. OFF alpha and delta cells used NMDA receptors for encoding either the full contrast range (alpha), including near-threshold responses, or only a high range (delta). However, contrast sensitivity depended substantially on NMDA receptors only in OFF alpha cells. NMDA receptors contribute to visual contrast coding in a cell-type-specific manner. Certain cell types generate excitatory responses using primarily AMPA receptors or disinhibition.

Rebound spiking as a neural mechanism for surface filling-in

Perceptual filling-in is the phenomenon where visual information is perceived although information is not physically present. For instance, the blind spot, which corresponds to the retinal location where there are no photoreceptor cells to capture the visual signals, is filled-in by the surrounding visual signals. The neural mechanism for such immediate filling-in of surfaces is unclear. By means of computational modeling, we show that surround inhibition produces rebound or after-discharge spiking in neurons that otherwise do not receive sensory information. The behavior of rebound spiking mimics the immediate surface filling-in illusion observed at the blind spot and also reproduces the filling-in of an empty object after a background flash, like in the color dove illusion. In conclusion, we propose rebound spiking as a possible neural mechanism for surface filling-in.

How connectivity, background activity, and synaptic properties shape the cross-correlation between spike trains

Functional interactions between neurons in vivo are often quantified by cross-correlation functions (CCFs) between their spike trains. It is therefore essential to understand quantitatively how CCFs are shaped by different factors, such as connectivity, synaptic parameters, and background activity. Here, we study the CCF between two neurons using analytical calculations and numerical simulations. We quantify the role of synaptic parameters, such as peak conductance, decay time, and reversal potential, and analyze how various patterns of connectivity influence CCF shapes. In particular, we find that the symmetry of the CCF distinguishes in general, but not always, the case of shared inputs between two neurons from the case in which they are directly synaptically connected. We systematically examine the influence of background synaptic inputs from the surrounding network that set the baseline firing statistics of the neurons and modulate their response properties. We find that variations in the background noise modify the amplitude of the cross-correlation function as strongly as variations of synaptic strength. In particular, we show that the postsynaptic neuron spiking regularity has a pronounced influence on CCF amplitude. This suggests an efficient and flexible mechanism for modulating functional interactions.

Origins of correlated activity in an olfactory circuit

Multineuronal recordings often reveal synchronized spikes in different neurons. How correlated spike timing affects neural codes depends on the statistics of correlations, which in turn reflects the connectivity that gives rise to correlations. However, determining the connectivity of neurons recorded in vivo can be difficult. Here, we investigate the origins of correlated activity in genetically-labeled neurons of the Drosophila antennal lobe. Dual recordings show synchronized spontaneous spikes in projection neurons (PNs) postsynaptic to the same type of olfactory receptor neuron (ORN). Odors increase these correlations. The primary origin of correlations lies in the divergence of each ORN onto every PN in its glomerulus. Reciprocal PN-PN connections make a smaller contribution to correlations, and PN spike trains in different glomeruli are only weakly correlated. PN axons from the same glomerulus reconverge in the lateral horn, where pooling redundant signals may allow lateral horn neurons to average out noise that arises independently in these PNs.

Ideal observer analysis of signal quality in retinal circuits

The function of the retina is crucial, for it must encode visual signals so the brain can detect objects in the visual world. However, the biological mechanisms of the retina add noise to the visual signal and therefore reduce its quality and capacity to inform about the world. Because an organism's survival depends on its ability to unambiguously detect visual stimuli in the presence of noise, its retinal circuits must have evolved to maximize signal quality, suggesting that each retinal circuit has a specific functional role. Here we explain how an ideal observer can measure signal quality to determine the functional roles of retinal circuits. In a visual discrimination task the ideal observer can measure from a neural response the increment threshold, the number of distinguishable response levels, and the neural code, which are fundamental measures of signal quality relevant to behavior. It can compare the signal quality in stimulus and response to determine the optimal stimulus, and can measure the specific loss of signal quality by a neuron's receptive field for non-optimal stimuli. Taking into account noise correlations, the ideal observer can track the signal-to-noise ratio available from one stage to the next, allowing one to determine each stage's role in preserving signal quality. A comparison between the ideal performance of the photon flux absorbed from the stimulus and actual performance of a retinal ganglion cell shows that in daylight a ganglion cell and its presynaptic circuit loses a factor of approximately 10-fold in contrast sensitivity, suggesting specific signal-processing roles for synaptic connections and other neural circuit elements. The ideal observer is a powerful tool for characterizing signal processing in single neurons and arrays along a neural pathway.

Relationship between spontaneous and evoked spike-time correlations in primate visual cortex

Coincident spikes have been implicated in vision-related processes such as feature binding, gain modulation, and long-distance communication. The source of these spike-time correlations is unknown. Although several studies have proposed that cortical spikes are correlated based on stimulus structure, others have suggested that spike-time correlations reflect ongoing cortical activity present even in the absence of a coherent visual stimulus. To examine this issue, we collected single-unit recordings from primary visual cortex (V1) of the anesthetized and paralyzed prosimian bush baby using a 100-electrode array. Spike-time correlations for pairs of cells were compared under three conditions: a moving grating at the cells' preferred orientation, an equiluminant blank screen, and a dark condition with eyes covered. The amplitudes, lags, and widths of cross-correlation histograms (CCHs) were strongly correlated between these conditions although for the blank stimulus and dark condition, the CCHs were broader with peaks lower in amplitude. In both preferred stimulus and blank conditions, the CCH amplitudes were greater when the cells within the pair had overlapping receptive fields and preferred similar orientations rather than nonoverlapping receptive fields and different orientations. These data suggest that spike-time correlations present in evoked activity are generated by mechanisms common to those operating in spontaneous conditions.

The structure of large-scale synchronized firing in primate retina

Synchronized firing among neurons has been proposed to constitute an elementary aspect of the neural code in sensory and motor systems. However, it remains unclear how synchronized firing affects the large-scale patterns of activity and redundancy of visual signals in a complete population of neurons. We recorded simultaneously from hundreds of retinal ganglion cells in primate retina, and examined synchronized firing in completely sampled populations of approximately 50-100 ON-parasol cells, which form a major projection to the magnocellular layers of the lateral geniculate nucleus. Synchronized firing in pairs of cells was a subset of a much larger pattern of activity that exhibited local, isotropic spatial properties. However, a simple model based solely on interactions between adjacent cells reproduced 99% of the spatial structure and scale of synchronized firing. No more than 20% of the variability in firing of an individual cell was predictable from the activity of its neighbors. These results held both for spontaneous firing and in the presence of independent visual modulation of the firing of each cell. In sum, large-scale synchronized firing in the entire population of ON-parasol cells appears to reflect simple neighbor interactions, rather than a unique visual signal or a highly redundant coding scheme.

Synchronized firing in the retina

Synchronized firing in neural populations has been proposed to constitute an elementary aspect of the neural code, but a complete understanding of its origins and significance has been elusive. Synchronized firing has been extensively documented in retinal ganglion cells, the output neurons of the retina. However, differences in synchronized firing across species and cell types have led to varied conclusions about its mechanisms and role in visual signaling. Recent work on two identified cell populations in the primate retina, the ON-parasol and OFF-parasol cells, permits a more unified understanding. Intracellular recordings reveal that synchronized firing in these cell types arises primarily from common synaptic input to adjacent pairs of cells. Statistical analysis indicates that local pairwise interactions can explain the pattern of synchronized firing in the entire parasol cell population. Computational analysis reveals that the aggregate impact of synchronized firing on the visual signal is substantial. Thus, in the parasol cells, the origin and impact of synchronized firing on the neural code may be understood as locally shared input which influences the visual signals transmitted from eye to brain.