Information Processing In Neural Systems Research Articles

Quantum spin models for numerosity perception.

Humans share with animals, both vertebrates and invertebrates, the capacity to sense the number of items in their environment already at birth. The pervasiveness of this skill across the animal kingdom suggests that it should emerge in very simple populations of neurons. Current modelling literature, however, has struggled to provide a simple architecture carrying out this task, with most proposals suggesting the emergence of number sense in multi-layered complex neural networks, and typically requiring supervised learning; while simple accumulator models fail to predict Weber's Law, a common trait of human and animal numerosity processing. We present a simple quantum spin model with all-to-all connectivity, where numerosity is encoded in the spectrum after stimulation with a number of transient signals occurring in a random or orderly temporal sequence. We use a paradigmatic simulational approach borrowed from the theory and methods of open quantum systems out of equilibrium, as a possible way to describe information processing in neural systems. Our method is able to capture many of the perceptual characteristics of numerosity in such systems. The frequency components of the magnetization spectra at harmonics of the system's tunneling frequency increase with the number of stimuli presented. The amplitude decoding of each spectrum, performed with an ideal-observer model, reveals that the system follows Weber's law. This contrasts with the well-known failure to reproduce Weber's law with linear system or accumulators models.

A neural network model for timing control with reinforcement.

How do humans and animals perform trial-and-error learning when the space of possibilities is infinite? In a previous study, we used an interval timing production task and discovered an updating strategy in which the agent adjusted the behavioral and neuronal noise for exploration. In the experiment, human subjects proactively generated a series of timed motor outputs. Positive or negative feedback was provided after each response based on the timing accuracy. We found that the sequential motor timing varied at two temporal scales: long-term correlation around the target interval due to memory drifts and short-term adjustments of timing variability according to feedback. We have previously described these two key features of timing variability with an augmented Gaussian process, termed reward-sensitive Gaussian process (RSGP). In a nutshell, the temporal covariance of the timing variable was updated based on the feedback history to recreate the two behavioral characteristics mentioned above. However, the RSGP was mainly descriptive and lacked a neurobiological basis of how the reward feedback can be used by a neural circuit to adjust motor variability. Here we provide a mechanistic model and simulate the process by borrowing the architecture of recurrent neural networks (RNNs). While recurrent connection provided the long-term serial correlation in motor timing, to facilitate reward-driven short-term variations, we introduced reward-dependent variability in the network connectivity, inspired by the stochastic nature of synaptic transmission in the brain. Our model was able to recursively generate an output sequence incorporating internal variability and external reinforcement in a Bayesian framework. We show that the model can generate the temporal structure of the motor variability as a basis for exploration and exploitation trade-off. Unlike other neural network models that search for unique network connectivity for the best match between the model prediction and observation, this model can estimate the uncertainty associated with each outcome and thus did a better job in teasing apart adjustable task-relevant variability from unexplained variability. The proposed artificial neural network model parallels the mechanisms of information processing in neural systems and can extend the framework of brain-inspired reinforcement learning (RL) in continuous state control.

Information Closure Theory of Consciousness.

Information processing in neural systems can be described and analyzed at multiple spatiotemporal scales. Generally, information at lower levels is more fine-grained but can be coarse-grained at higher levels. However, only information processed at specific scales of coarse-graining appears to be available for conscious awareness. We do not have direct experience of information available at the scale of individual neurons, which is noisy and highly stochastic. Neither do we have experience of more macro-scale interactions, such as interpersonal communications. Neurophysiological evidence suggests that conscious experiences co-vary with information encoded in coarse-grained neural states such as the firing pattern of a population of neurons. In this article, we introduce a new informational theory of consciousness: Information Closure Theory of Consciousness (ICT). We hypothesize that conscious processes are processes which form non-trivial informational closure (NTIC) with respect to the environment at certain coarse-grained scales. This hypothesis implies that conscious experience is confined due to informational closure from conscious processing to other coarse-grained scales. Information Closure Theory of Consciousness (ICT) proposes new quantitative definitions of both conscious content and conscious level. With the parsimonious definitions and a hypothesize, ICT provides explanations and predictions of various phenomena associated with consciousness. The implications of ICT naturally reconcile issues in many existing theories of consciousness and provides explanations for many of our intuitions about consciousness. Most importantly, ICT demonstrates that information can be the common language between consciousness and physical reality.

Effect of Channel Block on Multiple Coherence Resonance Induced by Time Delay in Adaptive Neuronal Networks with Spike-Timing-Dependent Plasticity

In this paper, we study effect of channel block (CB) on multiple coherence resonance (MCR) in adaptive scale-free Hodgkin–Huxley neuronal networks with spike-timing-dependent plasticity (STDP). It is found that potassium CB suppresses MCR, but sodium CB can enhance MCR, and there is optimal sodium CB level by which MCR becomes most pronounced. In addition, STDP has a significant influence on the effect of CB on MCR. As adjusting rate [Formula: see text] of STDP increases, for potassium CB there is proper [Formula: see text] by which MCR is most pronounced; however, for sodium CB MCR is reduced. These findings could provide a new insight into effect of CB on information processing in neural systems.

Spiking patterns of a neuron model to stimulus: Rich dynamics and oxygen's role.

Neuronal spiking patterns, which are of fundamental importance for the understanding of information processing in neural systems, can be generated in response to different stimuli. We here investigate in detail the stimulus-induced spiking patterns in a biologically plausible neuron model in which the oxygen concentration and the dynamical concentrations of potassium, sodium, and chloride are considered. Various types of spiking patterns can be induced by the different external potassium accumulations in response to the stimulus, including two different types of epileptic seizure (SZ) and spreading depression (SD) states, two different mixed states of SD and SZ, SZ state with multi-burst, and tonic firing behaviors. Interestingly, we show that these rich spiking patterns can also be induced by the current stimulus with a low oxygen concentration. Furthermore, we reveal that the stimulus can induce two different phase transitions from the SD state to the SZ state according to the phase transition theory, which results in the different electrical activities. All these findings may provide insight into information processing in neural systems.

Digital configuration of astrocyte stimulation as a new technique to strengthen the impaired astrocytes in the tripartite synapse network

Recent findings have proven that glial cells and particularly astrocytes are responsible for some important roles in the central nervous system. Due to laboratory technical difficulties in the examination of astrocyte’s functions in information processing of neural systems, computational modeling is a suitable approach. In this paper, the process of synchronization in neuronal fluctuations in view of the role of astrocytes was investigated. Then, due to the destructive effects of hyper-synchronization, based on the dynamical model of the astrocyte biophysical model, a new bio-inspired stimulator was introduced. Many patients suffer from astrocyte-related epilepsy; however, drugs that affect astrocytes are not available currently. Therefore, here a new concept of “stimulating the impaired astrocytes to compensate for their malfunction” was introduced. This new mechanism was proposed to stimulate the impaired astrocytes in a population of tripartite synapses for restoration of the normal neural oscillations. Finally, a closed-loop interaction between the proposed stimulating method and a population of tripartite synapses was implemented on a Zynq system-on-chip (SoC) platform from Xilinx, because the Zynq SoC is an appropriate control platform in the implementation of the proposed architecture. Stimulator hardware design was optimized to attain a very low power consumption platform in real-time application. Results of hardware and software simulations confirmed that hyper-synchronization of a neuronal population caused by astrocytes dysfunction could be compensated by the bio-inspired stimulator.

Optimization of Intrinsic Stochastic Resonance in Adaptive Newman–Watts Network of Channel Blocked Hodgkin–Huxley Neurons

In this paper, we study stochastic resonance (SR) induced by channel noise in adaptive weighted Newman–Watts networks of Hodgkin–Huxley neurons with channel blocking (CB). It is found that the intrinsic SR is dependent on adaptive coupling and is strongly enhanced when the changing rate of adaptive coupling is optimal, and this phenomenon is independent of sodium and potassium CB levels. As CB increases, the channel noise for SR decreases, but the strength of intrinsic SR nearly does not change in the presence of adaptive coupling, which is different from the case for fixed coupling. These results show that intrinsic SR can be enhanced and optimized by adaptive coupling, and CB's effect on the intrinsic SR can be reduced by adaptive coupling. This implies that adaptive coupling could more efficiently improve the time precision of information processing in neural systems.

Efficient transfer entropy analysis of non-stationary neural time series.

Information theory allows us to investigate information processing in neural systems in terms of information transfer, storage and modification. Especially the measure of information transfer, transfer entropy, has seen a dramatic surge of interest in neuroscience. Estimating transfer entropy from two processes requires the observation of multiple realizations of these processes to estimate associated probability density functions. To obtain these necessary observations, available estimators typically assume stationarity of processes to allow pooling of observations over time. This assumption however, is a major obstacle to the application of these estimators in neuroscience as observed processes are often non-stationary. As a solution, Gomez-Herrero and colleagues theoretically showed that the stationarity assumption may be avoided by estimating transfer entropy from an ensemble of realizations. Such an ensemble of realizations is often readily available in neuroscience experiments in the form of experimental trials. Thus, in this work we combine the ensemble method with a recently proposed transfer entropy estimator to make transfer entropy estimation applicable to non-stationary time series. We present an efficient implementation of the approach that is suitable for the increased computational demand of the ensemble method's practical application. In particular, we use a massively parallel implementation for a graphics processing unit to handle the computationally most heavy aspects of the ensemble method for transfer entropy estimation. We test the performance and robustness of our implementation on data from numerical simulations of stochastic processes. We also demonstrate the applicability of the ensemble method to magnetoencephalographic data. While we mainly evaluate the proposed method for neuroscience data, we expect it to be applicable in a variety of fields that are concerned with the analysis of information transfer in complex biological, social, and artificial systems.

Manufacturing and Using Piggy-back Multibarrel Electrodes for <em>In vivo</em> Pharmacological Manipulations of Neural Responses

In vivo recordings from single neurons allow an investigator to examine the firing properties of neurons, for example in response to sensory stimuli. Neurons typically receive multiple excitatory and inhibitory afferent and/or efferent inputs that integrate with each other, and the ultimate measured response properties of the neuron are driven by the neural integrations of these inputs. To study information processing in neural systems, it is necessary to understand the various inputs to a neuron or neural system, and the specific properties of these inputs. A powerful and technically relatively simple method to assess the functional role of certain inputs that a given neuron is receiving is to dynamically and reversibly suppress or eliminate these inputs, and measure the changes in the neuron's output caused by this manipulation. This can be accomplished by pharmacologically altering the neuron's immediate environment with piggy-back multibarrel electrodes. These electrodes consist of a single barrel recording electrode and a multibarrel drug electrode that can carry up to 4 different synaptic agonists or antagonists. The pharmacological agents can be applied iontophoretically at desired times during the experiment, allowing for time-controlled delivery and reversible reconfiguration of synaptic inputs. As such, pharmacological manipulation of the microenvironment represents a powerful and unparalleled method to test specific hypotheses about neural circuit function. Here we describe how piggy-back electrodes are manufactured, and how they are used during in vivo experiments. The piggy-back system allows an investigator to combine a single barrel recording electrode of any arbitrary property (resistance, tip size, shape etc) with a multibarrel drug electrode. This is a major advantage over standard multi-electrodes, where all barrels have more or less similar shapes and properties. Multibarrel electrodes were first introduced over 40 years ago 1-3, and have undergone a number of design improvements 2,3 until the piggy-back type was introduced in the 1980s 4,5. Here we present a set of important improvements in the laboratory production of piggy-back electrodes that allow for deep brain penetration in intact in vivo animal preparations due to a relatively thin electrode shaft that causes minimal damage. Furthermore these electrodes are characterized by low noise recordings, and have low resistance drug barrels for very effective iontophoresis of the desired pharmacological agents.

Manufacturing and Using Piggy-back Multibarrel Electrodes for <em>In vivo</em> Pharmacological Manipulations of Neural Responses

In vivo recordings from single neurons allow an investigator to examine the firing properties of neurons, for example in response to sensory stimuli. Neurons typically receive multiple excitatory and inhibitory afferent and/or efferent inputs that integrate with each other, and the ultimate measured response properties of the neuron are driven by the neural integrations of these inputs. To study information processing in neural systems, it is necessary to understand the various inputs to a neuron or neural system, and the specific properties of these inputs. A powerful and technically relatively simple method to assess the functional role of certain inputs that a given neuron is receiving is to dynamically and reversibly suppress or eliminate these inputs, and measure the changes in the neuron's output caused by this manipulation. This can be accomplished by pharmacologically altering the neuron's immediate environment with piggy-back multibarrel electrodes. These electrodes consist of a single barrel recording electrode and a multibarrel drug electrode that can carry up to 4 different synaptic agonists or antagonists. The pharmacological agents can be applied iontophoretically at desired times during the experiment, allowing for time-controlled delivery and reversible reconfiguration of synaptic inputs. As such, pharmacological manipulation of the microenvironment represents a powerful and unparalleled method to test specific hypotheses about neural circuit function.Here we describe how piggy-back electrodes are manufactured, and how they are used during in vivo experiments. The piggy-back system allows an investigator to combine a single barrel recording electrode of any arbitrary property (resistance, tip size, shape etc) with a multibarrel drug electrode. This is a major advantage over standard multi-electrodes, where all barrels have more or less similar shapes and properties. Multibarrel electrodes were first introduced over 40 years ago 1-3, and have undergone a number of design improvements 2,3 until the piggy-back type was introduced in the 1980s 4,5. Here we present a set of important improvements in the laboratory production of piggy-back electrodes that allow for deep brain penetration in intact in vivo animal preparations due to a relatively thin electrode shaft that causes minimal damage. Furthermore these electrodes are characterized by low noise recordings, and have low resistance drug barrels for very effective iontophoresis of the desired pharmacological agents.

Dynamics about neural array with simple lateral inhibitory connections

Lateral inhibitory effect is a well-known feature of information processing in neural systems. This paper presents a neural array model with simple lateral inhibitory connections. After detailed examining into the dynamics of this kind of neural array, the author gives the sufficient conditions under which the outputs of the network will tend to a special stable pattern called spatial sparse pattern in which if the output of a neuron is 1, then the outputs of the neurons in its neighborhood are 0. This ability called spatial sparse coding plays an important role in self-coding, self-organization and associative memory for patterns and pattern sequences. The main conclusions about the dynamics of this kind of neural array which is related to spatial sparse coding are introduced.

Synchronization transitions on complex thermo-sensitive neuron networks with time delays

We have numerically studied the firing synchronization transitions on random thermo-sensitive neuron networks in dependence on information transmission delay tau, network randomness p, and coupling strength g. It is found that as tau is increased the neurons can exhibit transitions from burst synchronization (BS) to clustering anti-phase synchronization (APS), and further to spike synchronization (SS). It is also found that, with increasing p or g, there are transitions from spatiotemporal chaos to BS, then to APS, and finally to SS. However, the APS state with p or g exists only for intermediate tau values within a narrow range. For tau values outside this range, the APS state does not appear and the firings change directly from spatiotemporal chaos to BS or SS. These results show that, as time delay can do, network topology and coupling strength can also cause complex synchronization transitions in the neurons. In particular, the novel phenomenon of APS state with p or g shows that, with the help of appropriate random connections or coupling strength, the neurons may exhibit the APS behavior at a certain time delay for which the APS does not appear originally. These findings imply that time delay, network randomness, and coupling strength may have subtle effects on the firing behaviors on neuronal networks, and thus could play important roles in the information processing in neural systems.

Thermal impact on spiking properties in Hodgkin-Huxley neuron with synaptic stimulus

The effect of environmental temperature on neuronal spiking behaviors is investigated by numerically simulating the temperature dependence of spiking threshold of the Hodgkin-Huxley neuron subject to synaptic stimulus. We find that the spiking threshold exhibits a global minimum in a "comfortable temperature" range where spike initiation needs weakest synaptic strength, indicating the occurrence of optimal use of synaptic transmission in neural system. We further explore the biophysical origin of this phenomenon in ion channel gating kinetics and also discuss its possible biological relevance in information processing in neural systems.

Spatio-temporal patterns in a square-lattice Hodgkin-Huxley neural network

The collective behaviour of a square-lattice Hodgkin-Huxley neural network model with white noise is investigated by numerical methods. It is found that for an intermediate value of noise the Hodgkin-Huxley neurons in the square lattice exhibit an ordered circular structure. However, as the noise level increases, the ordered circular structures are distorted, and eventually totally destroyed. Thereby, the constructive role of appropriately pronounced random perturbations in the studied network is revealed. Furthermore, it is shown that as the diffusive coefficient increases, the typical width of the spatial waves also increases accordingly, which results in a decrease of the number of cycles by a given size of the spatial grid. More interestingly, it is observed that the spatio-temporal coherence resonance is enhanced as the diffusive coefficient is increased. Finally, the dependence of the typical width and the average period of the firing rate function on the diffusive coefficient is studied. Results presented in this paper should prove valuable for the understanding of information processing of neural systems in the presence of noise.

Progress in the neural sciences in the the century after Cajal (and the mysteries that remain).

One hundred years after Santiago Ramon y Cajal provided critical evidence for the "neuron doctrine," his cellular view of the brain remains the basis of modern neural science. This article begins with a review of how the early work of Ramon y Cajal, Charles Sherrington, and John Eccles and their contemporaries laid the groundwork for our current understanding of he information processing of neural systems and for understanding the task faced by studies of how the brain develops. The visual system is examined in some detail as a model for experimental investigation into the structure, operational mechanisms, and functions of large neural systems. Discussion of the phenomena of visual awareness and consciousness, links between the visual system and other brain systems, and disorders that disrupt voluntary control of cognition and emotion lead to a broader consideration of the problem of consciousness.

MIPAS: A neural network-based multimedia information processing and analysis system

This paper presents the structure and design criteria of a neural network-based multimedia information processing and analysis system (MIPAS) which can be used to deal with more-complicated intelligence issues. According to the structure and design criteria, a software environment (SEMIPAS), which supports the implementation of multimedia information (image + speech, image + characters, speech + characters, image + speech + characters) processing and analysis applications, is implemented and introduced. Under this software environment, a multimedia information processing and analysis system called “To Know the World” is constructed. Experiments show that the multimedia information processing and analysis is much more powerful and effective than single-medium information processing and analysis.

Neural information processing system for pig carcase grading in Danish slaughterhouses

This paper describes a full-sized integrated industrial application of neural networks of considerable extent and economic importance. Since May 1991, 16 million pig carcases have been individually graded every year in the Danish meat industry using a neural information processing system (NIPS) implemented at the 31 Classification Centres (CC). A CC is an 8 × 4 × 4 metre robot which fixs the pigs on a carousel and automatically positions and inserts nine probes into the carcases. A probe records a one-dimensional image, which is processed using multilayer perceptrons, yielding meat and fat thicknesses. The thicknesses from all the probes are subsequently fed to a recurrent neural network, which locates and predicts faulty or missing thicknesses. If there are too many predicted values, the robot repeats the measurement on the corresponding probes. The recurrent network employs a non-standard training method which alters the training data as well as the weights.