Circular Receptive Fields Research Articles

A Radar-Nearest-Neighbor based data-driven approach for crowd simulation

In this work, a learnable data-driven motion model namely Multi-Feature Fusion Recursive Neural Network (MFF-RNN) is proposed. The model yields pedestrians’ velocities by learning from the designed motion states consisting of the relative distances and velocities with neighbors, as well as individuals’ previous velocity sequences. A novel Radar-Nearest-Neighbor (Radar-NN) method is developed to determine the nearest neighbors of a pedestrian by treating him/her as a radar and detecting the surrounding environment within a limited circular receptive field. Bidirectional flow scenarios are adopted to evaluate the performance of the proposed model and the lane formation phenomenon can be successfully reproduced. The simulation results coincide with that of experiments and are superior to the work of Ma et al. in pedestrian trajectories, distributions, as well as fundamental diagrams. By calculating five evaluation metrics, it shows that the errors of our model are reduced by 34.1–79.0% compared with their work.

Population Receptive Field Shapes in Early Visual Cortex Are Nearly Circular.

The visual field region where a stimulus evokes a neural response is called the receptive field (RF). Analytical tools combined with functional MRI (fMRI) can estimate the RF of the population of neurons within a voxel. Circular population RF (pRF) methods accurately specify the central position of the pRF and provide some information about the spatial extent (diameter) of the RF. A number of investigators developed methods to further estimate the shape of the pRF, for example, whether the shape is more circular or elliptical. There is a report that there are many pRFs with highly elliptical pRFs in early visual cortex (V1-V3; Silson et al., 2018). Large aspect ratios (>2) are difficult to reconcile with the spatial scale of orientation columns or visual field map properties in early visual cortex. We started to replicate the experiments and found that the software used in the publication does not accurately estimate RF shape: it produces elliptical fits to circular ground-truth data. We analyzed an independent data set with a different software package that was validated over a specific range of measurement conditions, to show that in early visual cortex the aspect ratios are <2. Furthermore, current empirical and theoretical methods do not have enough precision to discriminate ellipses with aspect ratios of 1.5 from circles. Through simulation we identify methods for improving sensitivity that may estimate ellipses with smaller aspect ratios. The results we present are quantitatively consistent with prior assessments using other methodologies.SIGNIFICANCE STATEMENT We evaluated whether the shape of many population receptive fields (RFs) in early visual cortex is elliptical and differs substantially from circular. We evaluated two tools for estimating elliptical models of the pRF; one tool was valid over the measured compliance range. Using the validated tool, we found no evidence that confidently rejects circular fits to the pRF in visual field maps V1, V2, and V3. The new measurements and analyses are consistent with prior theoretical and experimental assessments in the literature.

Global oscillation regime change by gated inhibition

The role of sensory inputs in the modelling of synchrony regimes is exhibited by means of networks of spiking cells where the relative strength of the inhibitory interaction is controlled by the activation of a linear unit working as a gating variable. Adaptation to stimulus size is determined by the value of a changing length scale, modelled by the time-varying radius of a circular receptive field. In this set-up, ‘consolidation’ time intervals relevant to attentional effects are shown to depend on the dynamics governing the evolution of the introduced length scale.

A new angle on the role of feedfoward inputs in the generation of orientation selectivity in primary visual cortex

A new angle on the role of feedfoward inputs in the generation of orientation selectivity in primary visual cortex

A simple neural autism model based on sensory invariants

Autistic individuals show a variety of curious behavioral and perceptual patterns which generally seem to lack a common underlying cause. We propose a theory of learning and perception which suggests that the general impairment of the autistic brain may be a restriction on the class of connectivity patterns (or `features') that can be utilized for learning perceptual and cognitive tasks. In particular, we argue that the autistic brain does not make proper use of features which pool information over larger areas of the input space, which would allow them to make use of symmetries and develop invariants to permutations. We present preliminary findings with a simple neuronal model which provides useful analogies to psycho-physical experimental findings in support of this hypothesis. Our model consists of a two layer perceptron model. 'Neurons' in the model are connectionist neurons, thus are not meant to represent real biological neurons but rather functional entities that serve our task. In the first layer we have a set of model neurons with fixed random receptive fields. The second neuron layer learns by quadratic optimization to discriminate a noisy “smiley face” from a non-smiling face placed at random locations within the visual field. The only difference between the autistic and non-autistic version of the model is the shape of the receptive fields – autistic neurons have circular receptive fields while non-autistic neurons have horizontal receptive fields. As shown in Figure 2, the non-autistic version is able to solve the discrimination task with a much smaller number of neurons in the first layer. However if the “smiley face” is presented only partly to the model, the autistic version is actually better at the discrimination task then the non-autistic version of the model. This coincides –among other parallels to psycho-physical experiments– with findings in which autistic children show better performance in recognizing faces from parts [1]. Figure 1 Example receptive fields. Left: autistic, Right: non-autistic. Figure 2 Left: Discrimination error (y-axis) versus number of neurons (x-axis) for both non-autistic (full line) and autistic model (dashed). In both types the error reduces with the number of neurons, the non-autistic type shows a better performance. Right: In ...

Field experiments on the effectiveness of ‘eyespots’ as predator deterrents

Many potential prey species possess features that falsely convey a risk to the predator. One form of visual signal hypothesized to have this function are the circular features commonly found on the wings of diurnal lepidopterans: ‘eyespots’. Eyespots have been found to reduce predation by intimidating predators before an attack or startling them once an attack has been initiated, and have been proposed to work because predators falsely classify the markings as the eyes of the predators' own enemies. However, this argument lacks unambiguous experimental support because the data are consistent with the hypothesis that eyespots are effective merely because they are highly conspicuous. High contrast concentric circles are liable to be highly effective in stimulating the centre-surround arrangement of receptive fields typical of the vertebrate retina. Here, we use a field technique involving artificial moth-like stimuli, with or without different forms of grey-scale eyespots, to test these two opposing hypotheses. Across five experiments, the consistent effect is that predation risk is reduced by patterns that have high internal pattern contrast and also contrast highly with the target background, irrespective of the specific pattern arrangement. Circles are more effective than less eye-like shapes, but this constitutes only limited support for the eye-mimicry hypothesis because such patterns will also be highly conspicuous to animals with circular receptive fields in their retinae. We conclude that, contrary to popular belief, ‘eyespots’ can be effective predator deterrents without mimicking eyes.

Feedforward Construction of the Receptive Field and Orientation Selectivity of Visual Neurons in the Pigeon

How the receptive field (RF) of visual cells is formed and how to explain the orientation selectivity have been intensely studied and debated. Here we provided direct electrophysiological evidence by single-unit recording and electrophysiological mapping that the elongated excitatory RF of a visual cell in the pigeon nucleus isthmi is constructed from aligned circular excitatory RFs of tectal cells, whereas its inhibitory RF originates from intranuclear inhibitory circuits. The orientation selectivity of an isthmic cell is mainly determined by its excitatory RF and sharply tuned by its inhibitory RF. Retrograde tracing showed that the tectal cells converging onto an isthmic cell are arranged in a narrow dorsoventral column in the tectum. According to the retinotopic map on the tectum, the excitatory RFs of these tectal cells are aligned in a line orthogonal to the horizontal meridian of the visual field in agreement with the result obtained by electrophysiological mapping.

Synaptic physiology and receptive field structure in the early visual pathway of the cat.

How does the cortical circuitry analyze the visual scene? Here we explore the earliest levels of striate cortical processing: the first stage, where orientation sensitivity emerges, and the second stage, where stimulus selectivity is further refined. The approach is whole cell recording from cat in vivo. Neurons in the lateral geniculate nucleus of the thalamus have circular receptive fields whose subregions, center and surround are concentrically arranged and have the reverse sign, on or off. These neurons supply cortical simple cells, whose receptive fields have on and off subregions that are elongated and lie side by side. Feedforward models hold that orientation sensitivity depends on this thalamocortical change in receptive field structure and an arrangement within subregions such that stimuli of the reverse contrast evoke synaptic responses of the opposite polarity-push-pull. Our work provides support for feedforward models and emphasizes that push-pull is key in the geniculostriate pathway, preserved from retina by thalamic relay cells and reiterated, point by point, by cortical simple cells. Also, we help define the cortical push-pull circuit by identifying inhibitory simple cells. Lastly, separate experiments that compare the first and second levels of cortical processing suggest that differences in the synaptic physiology of connections at the two (thalamocortical versus intracortical) stages underlie differential selectivity for properties such as motion.

Spatial properties of neurones in the lateral geniculate nucleus of the pigmented ferret.

We used quantitative electrophysiological techniques to study the spatial properties of single units recorded extracellularly in the lateral geniculate and perigeniculate nuclei of the adult pigmented ferret. All neurones examined had approximately circular receptive fields, whose central regions gave responses antagonistic to those elicited from the surrounds. We presented sinusoidally modulated grating patterns, either drifting or counterphased, to obtain spatial frequency tuning curves, contrast sensitivity functions and to assess linearity or non-linearity of each neurone's response. In the ferret, as in other species, two types of lateral geniculate neurone could be distinguished, and we have termed these X-cells and Y-cells; both groups responded briskly to visual stimulation but X-cells gave sustained and linear responses whereas Y-cells responded transiently and non-linearly. Perigeniculate cells gave linear responses. For neurones in the lateral geniculate and perigeniculate nuclei, both the limit of spatial resolution ('acuity') and optimum spatial frequency were inversely related to receptive field eccentricity and the diameter of the receptive field centre. We recorded geniculate neurones in the ferret with acuities up to 3 cycles deg-1 and contrast sensitivities up to 114, values that are lower than those found previously for many geniculate cells in the cat.

Spatial distribution and interaction of responses in receptive fields of cat lateral geniculate neurons

Responses of lateral geniculate neurons to local photic stimulation and to adaptation of the central, antagonistic, and disinhibiting zones of their receptive fields were compared in unanesthetized cats immobilized with D-tubocurarine. Under most conditions of local adaptation, activation of on- and off-responses of neurons occurred after stimulation of the peripheral zones and inhibition of responses after stimulation of the central zone of the receptive field. As a result most neurons acquired the ability to generate a considerable on- and off-signal in response to stimulation. Comparison of this fact with the properties of on-off neurons [7] supports the view that under light-adaptation conditions the processing of large volumes of visual information and the more sophisticated performance of visual functions are connected with activation of responses from peripheral zones of circular receptive fields. It is concluded that local adaptation to light can extend the functional capacity of circular receptive fields.

Recursive features of circular receptive fields.

The distribution of excitability in retinal receptive fields may be well approximated by functions with recursive features. Physiological data do not exclude an implementation of recursive structures in the visual system. It is the most remarkable advantage of a recursive visual system, that cortical receptive fields tuned to different spatial frequencies will have an identical neuronal circuitry. Structural consequences for retina, LGN and visual cortex are discussed.

Color vision mechanisms in monkey striate cortex: dual-opponent cells with concentric receptive fields.

1. I have recorded with tungsten microelectrodes from single cells in the monkey's visual cortex and have specifically studied those neurons which were sensitive to the color of the stimulus. In the primate striate cortex there are four classes of color-coded cells. The cells described in this paper have concentric receptive fields with one red-green opponent-color system in the field center and the opposite organization in the surround. These dual-opponent cells were nost sensitive to the simultaneous presentation of two different colors, one covering the field center and the other illuminating the surround. They are probable involved in the perception of simultaneous color-contrast phenomena. 2. Spectral sensitivity curves revealed that both the field centers and the surrounds received opposite types of inputs from red-sensitive and green-sensitive cones. None of the cells tested had inputs from rods. 3. Area-sensitivity curves showed that peripheral suppression was present for both phases of the center opponent-color system. The boundary between the center and the surround was the same for both sets of opponent systems. Some cells had "silent" surrounds, which did not respond to annular stimuli. 4. Multiple-unit recordings from a concentric cell and one of its presumed afferents yielded information regarding its possible synaptic inputs. In some cases the cells appeared to receive contacts from red/green opponent-color geniculated fibers with circular receptive fields that lacked an antagonistic surround (similar to Wiesel and Hubel's (37) type II class). In other instances the afferents had on-center, off surround receptive fields or the reverse, but received inputs from only one cone type, either red or green (similar to Wiesel and Hubel's type III class). 5. Concentric cells were always driven by only one eye. 6. The laminar distribution of these cells was limited almost entirely to layer IV and its subdivisions. 7. The cumulative evidence presented in this paper indicates that the concentric cells probably received direct geniculate inputs and, therefore, they are the first cortical stage in the integration of color-contrast information.

Orientation detectors in the primary somatosensory neocortex of the raccoon

An analysis of response properties of single neurons in the forepaw region of the primary somatosensory neocortex (SI) of the raccoon was undertaken to gain a better understanding of the neural bases of tactile form discrimination. In the course of this investigation a new type of feature detector neuron which responded preferentially to tactile stimulus orientation was discovered. Within a sample of 110 neurons responding to light touch, two categories of cells were noted in terms of the type of stimulation best able to excite them. One category (linear units) was preferentially responsive to indentations of the skin using a narrow elongated stimulus probe applied in a particular orientation. The other category (round field units) gave a maximum response to stimulation with a round stimulus probe applied anywhere within a roughly circular receptive field. The response pattern of some SI neurons to a maintained indentation of the skin consisted of early and late excitatory responses, separated by a depression in responding. The second excitatory response appeared to be more influenced by variations in stimulus parameters than was the first.

Receptive fields of single cells in the visual cortex of the hooded rat

The receptive fields of 107 single cells in area 17 of the hooded rat were examined. About half the cells responded to stationary as well as moving stimuli and about half only to movement. A variety of receptive field types were observed. Some of the cells responding to stationary stimuli had circular receptive fields, some with and some without annuli, some had elongated receptive fields, some had irregular receptive fields. Of the cells that responded only to movement, some were orientation or direction specific and some were not. Only two cells were found that responded to stimulation of the ipsilateral eye. Columnar organization of the cortex was not observed.

Two-dimensional analysis of retinal information transfer.

A two-dimensional, time-dependent analysis of the mammalian retina is presented. A computer simulation of experimental data considers ganglion-cell activity as a two-dimensional process. The properties of the simulation model were examined with several input stimuli. Circular receptive fields enhance not only borders, but other spatial complexities, such as ends of lines and angles. The enhancement is remarkably insensitive to the shape of receptive fields. The processing of the input picture by on-center and off-center receptive fields is complementary but the division of labor varies with the task.

Visual receptive fields of striate cortex neurons in awake monkeys.

THE EXPERIMENTS of Lettvin and colleagues (25) and Hubel and Wiesel (19) have indicated that a neuron in the vertebrate visual system is sensitive to particular characteristics of the visual stimulus. These characteristics have been referred to as the “trigger features” of the stimulus (3) and include such factors as size, shape, orientation, color, and rate and direction of movement; the area of the visual field over which the stimulus is effective is referred to as the receptive field of the neuron (17). Receptive-field characteristics of visual system neurons have subsequently been determined in a wide range of animals (for references see 7, 16). In addition, Hubel and Wiesel (19-24, 30) have shown that there are changes in the characteristics of the stimulus needed to activate a neuron at successively higher levels of the visual system. In both cats and monkeys they observed a progressive change in the predominant type of receptive field from the circular receptive fields of retinal ganglion and lateral geniculate cells, through the elongated receptive fields of the simplest cortical cells, to the more complex fields of other cortical neurons. This understanding of the visual system has been derived entirely from experiments in which the animal was paralyzed and frequently also anesthetized. Such procedures have been designed to eliminate all eye movements, particularly the small eye movements of physiological nystagmus that are characteristic of normal vision (1). The purpose of the present experiments was to determine whether receptive fields of striate cortex neurons in the awake, behaving animal are similar to those in the anesthetized,