Physiological Receptive Fields Research Articles

A computational study of the role of spatial receptive field structure in processing natural and non-natural scenes

The center-surround receptive field structure, ubiquitous in the visual system, is hypothesized to be evolutionarily advantageous in image processing tasks. We address the potential functional benefits and shortcomings of spatial localization and center-surround antagonism in the context of an integrate-and-fire neuronal network model with image-based forcing. Utilizing the sparsity of natural scenes, we derive a compressive-sensing framework for input image reconstruction utilizing evoked neuronal firing rates. We investigate how the accuracy of input encoding depends on the receptive field architecture, and demonstrate that spatial localization in visual stimulus sampling facilitates marked improvements in natural scene processing beyond uniformly-random excitatory connectivity. However, for specific classes of images, we show that spatial localization inherent in physiological receptive fields combined with information loss through nonlinear neuronal network dynamics may underlie common optical illusions, giving a novel explanation for their manifestation. In the context of signal processing, we expect this work may suggest new sampling protocols useful for extending conventional compressive sensing theory.

Improved Compressive Sensing of Natural Scenes Using Localized Random Sampling.

Compressive sensing (CS) theory demonstrates that by using uniformly-random sampling, rather than uniformly-spaced sampling, higher quality image reconstructions are often achievable. Considering that the structure of sampling protocols has such a profound impact on the quality of image reconstructions, we formulate a new sampling scheme motivated by physiological receptive field structure, localized random sampling, which yields significantly improved CS image reconstructions. For each set of localized image measurements, our sampling method first randomly selects an image pixel and then measures its nearby pixels with probability depending on their distance from the initially selected pixel. We compare the uniformly-random and localized random sampling methods over a large space of sampling parameters, and show that, for the optimal parameter choices, higher quality image reconstructions can be consistently obtained by using localized random sampling. In addition, we argue that the localized random CS optimal parameter choice is stable with respect to diverse natural images, and scales with the number of samples used for reconstruction. We expect that the localized random sampling protocol helps to explain the evolutionarily advantageous nature of receptive field structure in visual systems and suggests several future research areas in CS theory and its application to brain imaging.

Sparse Coding Models Can Exhibit Decreasing Sparseness while Learning Sparse Codes for Natural Images

The sparse coding hypothesis has enjoyed much success in predicting response properties of simple cells in primary visual cortex (V1) based solely on the statistics of natural scenes. In typical sparse coding models, model neuron activities and receptive fields are optimized to accurately represent input stimuli using the least amount of neural activity. As these networks develop to represent a given class of stimulus, the receptive fields are refined so that they capture the most important stimulus features. Intuitively, this is expected to result in sparser network activity over time. Recent experiments, however, show that stimulus-evoked activity in ferret V1 becomes less sparse during development, presenting an apparent challenge to the sparse coding hypothesis. Here we demonstrate that some sparse coding models, such as those employing homeostatic mechanisms on neural firing rates, can exhibit decreasing sparseness during learning, while still achieving good agreement with mature V1 receptive field shapes and a reasonably sparse mature network state. We conclude that observed developmental trends do not rule out sparseness as a principle of neural coding per se: a mature network can perform sparse coding even if sparseness decreases somewhat during development. To make comparisons between model and physiological receptive fields, we introduce a new nonparametric method for comparing receptive field shapes using image registration techniques.

Differential Targeting of Optical Neuromodulators to Ganglion Cell Soma and Dendrites Allows Dynamic Control of Center-Surround Antagonism

Retinal degenerative diseases cause photoreceptor loss and often result in remodeling and deafferentation of the inner retina. Fortunately, ganglion cell morphology appears to remain intact long after photoreceptors and distal retinal circuitry have degenerated. We have introduced the optical neuromodulators channelrhodopsin-2 (ChR2) and halorhodopsin (NpHR) differentially into the soma and dendrites of ganglion cells to recreate antagonistic center-surround receptive field interactions. We then reestablished the physiological receptive field dimensions of primate parafoveal ganglion cells by convolving Gaussian-blurred versions of the visual scene at the appropriate wavelength for each neuromodulator with the Gaussians inherent in the soma and dendrites. These Gaussian-modified ganglion cells responded with physiologically relevant antagonistic receptive field components and encoded edges with parafoveal resolution. This approach bypasses the degenerated areas of the distal retina and could provide a first step in restoring sight to individuals suffering from retinal disease.

Effects of acute ethyl alcohol consumption on a psychophysical measure of lateral inhibition in human vision

Acute consumption of ethyl alcohol affects a variety of visual functions. However, there have been few systematic attempts to investigate the neural mechanisms underlying these effects. Here, we employed the Westheimer paradigm to investigate the hypothesis that alcohol reduces lateral inhibition within human “perceptive fields”, the psychophysical analogue of physiological receptive fields. Westheimer functions obtained under alcohol and no-alcohol conditions at photopic, mesopic, and scotopic levels of adaptation showed changes consistent with an alcohol-induced decrease in lateral inhibition. We conclude that this decrease in lateral inhibition may be responsible for some of the changes in visual perception that result from alcohol consumption.

Sexual Dimorphism in the Hoverfly Motion Vision Pathway

Many insects perform high-speed aerial maneuvers in which they navigate through visually complex surrounds. Among insects, hoverflies stand out, with males switching from stationary hovering to high-speed pursuit at extreme angular velocities [1]. In dipterans, 50-60 large interneurons -- the lobula-plate tangential cells (LPTCs) -- detect changes in optic flow experienced during flight [2-5]. It has been predicted that large LPTC receptive fields are a requirement of accurate "matched filters" of optic flow [6]. Whereas many fly taxa have three horizontal system (HS) LPTC neurons in each hemisphere, hoverflies have four [7], possibly reflecting the more sophisticated flight behavior. We here show that the most dorsal hoverfly neuron (HS north [HSN]) is sexually dimorphic, with the male receptive field substantially smaller than in females or in either sex of blowflies. The (hoverfly-specific) HSN equatorial (HSNE) is, however, sexually isomorphic. Using complex optic flow, we show that HSN, despite its smaller receptive field, codes yaw velocity as well as HSNE. Responses to a target moving against a plain or textured background suggest that the male HSN could potentially play a role in target pursuit under some conditions.

On the Analysis and Interpretation of Inhomogeneous Quadratic Forms as Receptive Fields

In this letter, we introduce some mathematical and numerical tools to analyze and interpret inhomogeneous quadratic forms. The resulting characterization is in some aspects similar to that given by experimental studies of cortical cells, making it particularly suitable for application to second-order approximations and theoretical models of physiological receptive fields. We first discuss two ways of analyzing a quadratic form by visualizing the coefficients of its quadratic and linear term directly and by considering the eigenvectors of its quadratic term. We then present an algorithm to compute the optimal excitatory and inhibitory stimuli--those that maximize and minimize the considered quadratic form, respectively, given a fixed energy constraint. The analysis of the optimal stimuli is completed by considering their invariances, which are the transformations to which the quadratic form is most insensitive, and by introducing a test to determine which of these are statistically significant. Next we propose a way to measure the relative contribution of the quadratic and linear term to the total output of the quadratic form. Furthermore, we derive simpler versions of the above techniques in the special case of a quadratic form without linear term. In the final part of the letter, we show that for each quadratic form, it is possible to build an equivalent two-layer neural network, which is compatible with (but more general than) related networks used in some recent articles and with the energy model of complex cells. We show that the neural network is unique only up to an arbitrary orthogonal transformation of the excitatory and inhibitory subunits in the first layer.

The mapping of visual space by identified large second-order neurons in the dragonfly median ocellus

In adult dragonflies, the compound eyes are augmented by three simple eyes known as the dorsal ocelli. The outputs of ocellar photoreceptors converge on relatively few second-order neurons with large axonal diameters (L-neurons). We determine L-neuron morphology by iontophoretic dye injection combined with three-dimensional reconstructions. Using intracellular recording and white noise analysis, we also determine the physiological receptive fields of the L-neurons, in order to identify the extent to which they preserve spatial information. We find a total of 11 median ocellar L-neurons, consisting of five symmetrical pairs and one unpaired neuron. L-neurons are distinguishable by the extent and location of their terminations within the ocellar plexus and brain. In the horizontal dimension, L-neurons project to different regions of the ocellar plexus, in close correlation with their receptive fields. In the vertical dimension, dendritic arborizations overlap widely, paralleled by receptive fields that are narrow and do not differ between different neurons. These results provide the first evidence for the preservation of spatial information by the second-order neurons of any dorsal ocellus. The system essentially forms a one-dimensional image of the equator over a wide azimuthal area, possibly forming an internal representation of the horizon. Potential behavioural roles for the system are discussed.

Seeing with S cones.

The S cone is highly conserved across mammalian species, sampling the retinal image with less spatial frequency than other cone photoreceptors. In human and monkey retina, the S cone represents typically 5-10% of the cone mosaic and distributes in a quasi-regular fashion over most of the retina. In the fovea, the S cone mosaic recedes from a central "S-free" zone whose size depends on the optics of the eye for a particular primate species: the smaller the eye, the less extreme the blurring of short wavelengths, and the smaller the zone. In the human retina, the density of the S mosaic predicts well the spatial acuity for S-isolating targets across the retina. This acuity is likely supported by a bistratified retinal ganglion cell whose spatial density is about that of the S cone. The dendrites of this cell collect a depolarizing signal from S cones that opposes a summed signal from M and L cones. The source of this depolarizing signal is a specialized circuit that begins with expression of the L-AP4 or mGluR6 glutamate receptor at the S cone-->bipolar cell synapse. The pre-synaptic circuitry of this bistratified ganglion cell is consistent with its S-ON/(M+L)-OFF physiological receptive field and with a role for the ganglion cell in blue/yellow color discrimination. The S cone also provides synapses to other types of retinal circuit that may underlie a contribution to the cortical areas involved with motion discrimination.

Visual bridging of empty gaps in the optic flow.

This is a study of perception of bending motion and jointed rigid motions over large invisible segments of a bending line. In this project, we investigated the visual perception of changing form of lines, built up by a series of dots and presented under highly reduced pictorial conditions. The changing form was indicated by one or two moving and continuously changing visible fragments of the line. The most extreme condition studied was the perception of the bending of an initially vertical 24-dot line, visually represented only by the stationary base dot and the two moving dots at its top. In this experiment, nearly all subjects reported experiencing a smooth bending connection over the 21-dot empty gap. Three experiments are described and analyzed. The results suggest that the human visual system is astonishingly well adapted for derivation of relevant figural information from such severely reduced, continuously changing optical presentation. An explanation in terms of automatic sensory mechanisms related to the physiological receptive field effect is proposed.

Photoreceptor coupling in turtle retina.

Photoreceptors in the isolated turtle retina of two species of turtle, Chelydra serpentina and Pseudemus scripta elegans, were penetrated with double-barrel electrodes. Physiological responses were recorded through one barrel and Neurobiotin tracer was injected from the other. Intracellular injection of Neurobiotin revealed patterns of tracer-coupled photoreceptors. Both the patterns of tracer coupling and the electrophysiology suggest a high degree of specificity of connections. Rods seem to be coupled only to rods and green and red cones seem to be coupled to cones of the same spectral type. Receptive-field profiles, measured with a thin, sharply focused slit of light, often had well-defined peaks and troughs in sensitivity. We have taken advantage of this observation and used the position of a peak in sensitivity to locate the position on the retina of a coupled cell. In one rod, it was possible to correlate physiological and morphological data and to show that the peaks in the physiological receptive field occurred at positions on the retina where there were dye-coupled cells. This provides direct evidence that gap junctions produce the physiological coupling between rods.

A Mathematical Model of Perception with Reference to Peripheral Vision and its Application to the Hermann Grid Illusion

Central vision and peripheral vision are two of main components of visual perception. The functional differences between the two visions are considered with two points of view: the physiological receptive field and the psychological perceptive field.

Ratfish retina—intracellular recordings and HRP injections in an isolated, superfused all-rod retina

Hyperpolarizing responses to light were studied by intracellular recording in the isolated, superfused retina of the ratfish ( Hydrolagus colliei). Two of these hyperpolarizing units were identified as horizontal cells by iontophoretic injection of horseradish peroxidase (HRP). These cells had rather large cell bodies (15 × 30 μm, elliptical dendritic arborizations measuring 150 × 300 μm and no axons. Since their physiological receptive fields were found to be at least 2.15 mm in diameter, it appears likely that either the photoreceptors or the horizontal cells are electrically coupled. Electron microscopy of HRP-injected horizontal cells showed their dendrites to end laterally in ribbon synapses of rods only and revealed dendro-dendritic contacts resembling gap junctions between injected and uninjected horizontal cells. The spectral sensitivity function of a dark-adapted horizontal cell can be described by a Dartnall nomogram based on a retinene 1 pigment with λ max = 473 nm. These findings are consistent with the histological observation that the ratfish retina appears to contain only rod photoreceptors.

Sensory basis for directional wind detection in first instar cockroaches,Periplaneta americana

1. First instarPeriplaneta americana have only two wind-receptive filiform hair sensilla on each cercus, as compared to over 200 on the cercus of an adult cockroach. The two hairs have preferential planes of deflection which are approximately at right angles to each other (Figs. 1, 2). 2. The responses of single sensory neurons associated with the hairs were recorded extracellularly from the cereal nerve with tungsten electrodes. Movement of a hair in one direction within its preferential plane depolarized the underlying neuron and elicited an increased frequency of action potentials. Deflection in the opposite direction hyperpolarized the cell and caused the cessation of ongoing activity (Fig. 3). 3. Receptive fields for wind in the horizontal plane were plotted for all four hairs. Although the fields were broadly tuned and overlapped each other considerably, each hair sensillum dominated one quadrant of horizontal space, and all 360 ° were covered by the four hairs (Fig. 4). 4. The escape behavior of first instar nymphs, which depends on these sensilla, was oriented away from a source which delivered small puffs of wind. The nymph's accuracy of directed turning was no less than that of adult animals (Fig. 5). 5. When various combinations of hairs were removed from the cerci, the animals could no longer make correctly oriented turns to wind from all angles. The turns they did make reflected the physiological receptive fields of the remaining hairs and suggested that the behavior is based, to some extend, on comparisons of impulse frequency among the four sensory afferents (Fig. 6).