Neural Firing Activity Research Articles

CaMKIIα-NpHR-Mediated Optogenetic Inhibition of DRG Glutamatergic Neurons by Flexible Optic Fiber Alleviates Chronic Neuropathic Pain.

Glutamatergic neurons of the dorsal root ganglion (DRGg) exert a significant effect on peripheral nociceptive signal transmission. However, assessing the explicit modulatory effect of DRGg during chronic neuropathic pain (CNP) with neuromodulation techniques remains largely unexplored. Therefore, we inhibited DRGg by optogenetic stimulation and examined whether it could alleviate CNP and associated anxiety-related behaviors in a chronic compressed DRG (CCD) rat model. The CCD pain model was established by inserting an L-shaped rod into the lumbar 5 (L5) intervertebral foramen, and either AAV2-CaMKIIα-eNpHR3.0-mCherry or AAV2-CaMKIIα-mCherry was injected into the L5 DRG. Flexible optic fibers were implanted to direct yellow light into the L5 DRG. Pain and anxiety-related behavioral responses were assessed using mechanical threshold, mechanical latency, thermal latency, and open field tests. In vivo single-unit extracellular recording from the DRG and ventral posterolateral (VPL) thalamus was performed. CNP and anxiety-related behavioral responses along with increased neural firing activity of the DRG and VPL thalamus were observed in CCD animals. Enhanced expression of nociception-influencing molecules was found in the DRG and spinal dorsal horn (SDH). In contrast during optogenetic stimulation, specific DRGg inhibition markedly alleviated the CNP responses and reduced the DRG and VPL thalamic neural hyperactivity in CCD animals. Inhibition of DRGg also reduced the active expression of nociceptive signal mediators in the DRG and SDH. Taken together, our findings suggest that CaMKIIα-NpHR-mediated optogenetic inhibition of DRGg can produce antinociceptive effects in CCD rats during peripheral nerve injury-induced CNP condition by altering peripheral nociceptive signal input in the spinothalamic tract.

BrainForest: Neuromorphic Multiplier-Less Bit-Serial Weight-Memory-Optimized 1024-Tree Brain-State Classification Processor.

Personalized brain implants have the potential to revolutionize the treatment of neurological disorders and augment cognition. Medical implants that deliver therapeutic stimulation in response to detected seizures have already been deployed for the treatment of epilepsy. These devices require low-power integrated circuits for life-long operation. This constraint impedes the integration of machine-learning driven classifiers that could improve treatment outcomes. This paper introduces BrainForest, a neuromorphic multiplier-less bit-serial weight-memory-optimized brain-state classification processor. Thearchitecture achieves state-of-the-art energy efficiency using two layers of neuron models to implement the spectral and temporal functions needed for classification: 1) resonate-and-fire neurons are used to extract physiological signal band energy EEG biomarkers 2) leaky integrator neurons are used to build multi-timescale representations for classification. Sparse neural model firing activity is used to clock-gate device logic, thereby decreasing power consumption by 93%. Anenergy-optimized 1024-tree boosted decision forest performs the classification used to trigger stimulation in response to detected pathological brain states. TheIC is implemented in 65nm CMOS with state-of-the-art power consumption (best case: 9.6µW, typical: 118µW), achieving a seizure sensitivity of 97.5% with a false detection rate of 2.08 per hour.

Low-intensity transcranial ultrasound stimulation inhibits epileptic seizures in motor cortex by modulating hippocampus neural activity.

Prior studies indicate that applying low-intensity transcranial ultrasound stimulation (TUS) to the hippocampus can suppress epileptic seizures. Nevertheless, it is unclear how TUS regulates hippocampal neural activity, and whether and how epileptic discharges in the motor cortex are suppressed by modulating hippocampal neural activity. To explore the answers of above questions, ultrasound was utilized to investigate the responses to the aforementioned inquiries by stimulating the hippocampus of mice with penicillin-induced epilepsy, while simultaneously recording the local field potentials (LFPs) in the hippocampus and the motor cortex (M1) throughout the experiment. The results showed that TUS (1) reduced the amplitude and the strength of the θ frequency band in LFPs in the hippocampus and M1, (2) decreased the coupling strength of the δ-γ, θ-γ and α-γ frequency bands in the hippocampus and M1, (3) weakened the correlation of neural activity between the hippocampus and M1. The above results indicated that TUS effectively suppressed abnormal slow neural oscillations in the hippocampus, had a significant decoupling effect on slow-fast neural oscillations, and reduced the correlation of hippocampus-cortical neural activity. TUS of the hippocampus may be through the hippocampus-cortical circuits to suppress abnormal neural firing activity in M1.

62 Neural Correlates of Cognitive Function in the Basal Ganglia

Objective:Timing, or the decision of when to act, is essential to mammalian behaviors from escaping predators to driving a car. It requires cognitive functions such as working memory for time-based rules and attention to the passing of time. Thus, it can be used as a proxy for higher order executive functions that are difficult to measure but are impaired in many neurological disorders. Therefore, insights from studies of interval timing, tasks which require estimating time intervals of several seconds, have great value for our understanding of human disease. Crucial to timing is the basal ganglia, which integrates cortical activity with midbrain dopamine signals and sends out signals to the spinal cord that regulate movement, motivation, and other behaviors. We have previously found that within the basal ganglia, medium spiny neurons of the striatum exhibit ramping activity in time-related tasks. In other words, they gradually increase or decrease firing frequency across a timed interval, and this is thought to encode time. Yet it is still unknown how the encoding of time is translated into time-based motor responses. To answer this question, we turned to the external globus pallidus (GPe) because it is a regulatory hub within the basal ganglia and is thus well positioned to regulate timing behavior. We sought to examine how the GPe functions in response to time-based demands.Participants and Methods:We recorded from neuronal ensembles using 16 channel electrode arrays implanted in the GPe of five mice while they performed an interval timing task called the switch interval timing task. Spike sorting was then used to identify signal from individual neurons.Results:Data were compiled from 43 neurons over several trials. Principal component analysis of neural firing activity was then conducted and revealed a downward ramping pattern in GPe neurons during interval timing trials. Data were then separated based on trials in which mice made correct decisions and those in which mice made a mistake. We found that when mice make correct timing decisions, there is downward ramping activity in the GPe, yet when mice make timing mistakes, this ramping pattern is lost.Conclusions:Our findings suggest that the GPe processes timing signals through ramping activity, before projecting to the output nuclei of the basal ganglia. This is a novel finding and contributes to a growing understanding of the temporal code of the basal ganglia. The full extent of this code is still unknown, but this insight contributes to a better understanding of how the globus pallidus represents cognition. If we can better explain the neural correlates of timing, we can use this knowledge to inform therapeutic interventions for basal ganglia dysfunction, which could have profound implications for diseases like Parkinson’s disease, which affects millions around the world.

Nonlinear mechanism for the enhanced bursting activities induced by fast inhibitory autapse and reduced activities by fast excitatory autapse.

The paradoxical phenomena that excitatory modulation does not enhance but reduces or inhibitory modulation not suppresses but promotes neural firing activities have attracted increasing attention. In the present study, paradoxical phenomena induced by both fast excitatory and inhibitory autapses in a "Fold/Big Homoclinic" bursting are simulated, and the corresponding nonlinear and biophysical mechanisms are presented. Firstly, the enhanced conductance of excitatory autapse induces the number of spikes per burst and firing rate reduced, while the enhanced inhibitory autapse cause both indicators increased. Secondly, with fast-slow variable dissection, the burst of bursting is identified to locate between a fold bifurcation and a big saddle-homoclinic orbit bifurcation of the fast subsystem. Enhanced excitatory or inhibitory autapses cannot induce changes of both bifurcation points, i.e., burst width. However, width of slow variable between two successive spikes within a burst becomes wider for the excitatory autapse and narrower for the inhibitory autapse, resulting in the less and more spikes per burst, respectively. Last, the autaptic current of fast autapse mainly plays a role during the peak of action potential, differing from the slow autaptic current with exponential decay, which can play roles following the peak of action potential. The fast excitatory autaptic current enhances the amplitude of the action potential and reduces the repolarization of the action potential to lengthen the interspike interval (ISI) of the spiking of the fast subsystem, resulting in the wide width of slow variable between successive spikes. The fast inhibitory autaptic current reduces the amplitude of action potential and ISI of spiking, resulting in narrow width of slow variable. The novel example of the paradoxical responses for both fast modulations and nonlinear mechanism extend the contents of neurodynamics, which presents potential functions of the fast autapse.

The Influence of Neural Activity and Neural Cytoarchitecture on Cerebrovascular Arborization: A Computational Model.

Normal functioning of the brain relies on a continual and efficient delivery of energy by a vast network of cerebral blood vessels. The bidirectional coupling between neurons and blood vessels consists of vasodilatory energy demand signals from neurons to blood vessels, and the retrograde flow of energy substrates from the vessels to neurons, which fuel neural firing, growth and other housekeeping activities in the neurons. Recent works indicate that, in addition to the functional coupling observed in the adult brain, the interdependence between the neural and vascular networks begins at the embryonic stage, and continues into subsequent developmental stages. The proposed Vascular Arborization Model (VAM) captures the effect of neural cytoarchitecture and neural activity on vascular arborization. The VAM describes three important stages of vascular tree growth: (i) The prenatal growth phase, where the vascular arborization depends on the cytoarchitecture of neurons and non-neural cells, (ii) the post-natal growth phase during which the further arborization of the vasculature depends on neural activity in addition to neural cytoarchitecture, and (iii) the settling phase, where the fully grown vascular tree repositions its vascular branch points or nodes to ensure minimum path length and wire length. The vasculature growth depicted by VAM captures structural characteristics like vascular volume density, radii, mean distance to proximal neurons in the cortex. VAM-grown vasculature agrees with the experimental observation that the neural densities do not covary with the vascular density along the depth of the cortex but predicts a high correlation between neural areal density and microvascular density when compared over a global scale (across animals and regions). To explore the influence of neural activity on vascular arborization, the VAM was used to grow the vasculature in neonatal rat whisker barrel cortex under two conditions: (i) Control, where the whiskers were intact and (ii) Lesioned, where one row of whiskers was cauterized. The model captures a significant reduction in vascular branch density in lesioned animals compared to control animals, concurring with experimental observation.

Differential Effects of the G-Protein-Coupled Estrogen Receptor (GPER) on Rat Embryonic (E18) Hippocampal and Cortical Neurons.

Estrogen plays fundamental roles in nervous system development and function. Traditional studies examining the effect of estrogen in the brain have focused on the nuclear estrogen receptors (ERs), ERα and ERβ. Studies related to the extranuclear, membrane-bound G-protein-coupled ER (GPER/GPR30) have revealed a neuroprotective role for GPER in mature neurons. In this study, we investigated the differential effects of GPER activation in primary rat embryonic day 18 (E18) hippocampal and cortical neurons. Microscopy imaging, multielectrode array (MEA), and Ca2+ imaging experiments revealed that GPER activation with selective agonist, G-1, and nonselective agonist, 17β-estradiol (E2), increased neural growth, neural firing activity, and intracellular Ca2+ more profoundly in hippocampal neurons than in cortical neurons. The GPER-mediated Ca2+ rise in hippocampal neurons involves internal Ca2+ store release via activation of phospholipase C (PLC) and extracellular entry via Ca2+ channels. Immunocytochemistry results revealed no observable difference in GPER expression/localization in neurons, yet real-time qPCR (RT-qPCR) and Western blotting showed a higher GPER expression in the cortex than hippocampus, implying that GPER expression level may not fully account for its robust physiological effects in hippocampal neurons. We used RNA sequencing data to identify distinctly enriched pathways and significantly expressed genes in response to G-1 or E2 in cultured rat E18 hippocampal and cortical neurons. In summary, the identification of differential effects of GPER activation on hippocampal and cortical neurons in the brain and the determination of key genes and molecular pathways are instrumental toward an understanding of estrogen’s action in early neuronal development.

Deciphering functional roles of synaptic plasticity and intrinsic neural firing in developing mouse visual cortex layer IV microcircuit.

Between the onset of the critical period of mouse primary visual cortex and eye opening at postnatal day 14 is a complex process and that is vital for the cognitive function of vision. The onset of the critical period of mouse primary visual cortex involves changes of the intrinsic firing property of each neuron and short term plasticity of synapses. In order to investigate the functional role of each factor in regulating the circuit firing activity during the critical period plasticity, we adopted the Markram's model for short term plasticity and Wilson's model for intrinsic neuron firing activity, and construct a microcircuit for mouse visual cortex layer IV based on the connection probabilities from experimental results. Our results indicate that, during CP development, the most critical factors that regulate the firing pattern of microcircuit is the short term plasticity of the synapse from PC to PV and SST interneurons, which upregulates the PV interneuron firing and produces new balance between excitation and inhibition; the intrinsic firing activity of PC and PV during development downregulates the firing frequency of the circuits. In addition, we have investigated the function of feedforward excitatory thalamic-cortical projection to PC and PV interneuron during CP, and found that neural firing activity largely depends on the TC input and the results are similar to the local circuit with minor differences. We conclude that the short term plasticity development during critical period plays a crucial role in regulating the circuit behavior.

Multiple and complex codimension-2 bifurcations underlying post-inhibitory rebound spike and excitability transition

Neuron exhibits nonlinear dynamics such as excitability transition and post-inhibitory rebound (PIR) spike related to bifurcations, which are associated with information processing, locomotor modulation, or brain disease. PIR spike is evoked by inhibitory stimulation instead of excitatory stimulation, which presents a challenge to the threshold concept. In this paper, 7 codimension-2 or degenerate bifurcations related to 10 codimension-1 bifurcations are acquired in a neuronal model, which presents the bifurcations underlying the excitability transition and PIR spike. Type I excitability corresponds to saddle-node bifurcation on an invariant cycle (SNIC) bifurcation, and type II excitability to saddle-node (SN) bifurcation or sub-critical Hopf (SubH) bifurcation or sup-critical Hopf (SupH) bifurcation. The excitability transition from type I to II corresponds to the codimension-2 bifurcation, Saddle-Node Homoclinic orbit (SNHO) bifurcation, via which SNIC bifurcation terminates and meanwhile big homoclinic orbit (BHom) bifurcation and SN bifurcation emerge. A degenerate bifurcation via which BHom bifurcation terminates and limit point of cycle (LPC) bifurcation emerges is responsible for spiking transition from type I to II, and the roles of other codimension-2 bifurcations (Cusp, Bogdanov-Takens, and Bautin) are discussed. In addition, different from the widely accepted viewpoint that PIR spike is mainly evoked near Hopf bifurcation rather than SNIC bifurcation, PIR spike is identified to be induced near SNIC or BHom or LPC bifurcations, and threshold curves resemble that of Hopf bifurcation. The complex bifurcations present comprehensive and deep understandings of excitability transition and PIR spike, which are helpful for the modulation to neural firing activities and physiological functions.

192-OR: T-Type Calcium Channel Cav3.2 in GABAergic Neurons Regulates Rhythmic Glucose Metabolism

Systemic insulin sensitivity shows a diurnal rhythm with a peak upon waking. The GABAergic (γ-aminobutyric acid-producing) neurons in the suprachiasmatic nucleus (SCN) (SCNGABA neurons) were reported to control the diurnal rhythm of insulin-mediated suppression of hepatic glucose production in mice in earlier work1. SCNGABA neural firing activity displays a diurnal rhythm, with the intrinsic ionic mechanism underlying this temporal pattern not completely understood. Here we provided evidence supporting a role of the T type voltage-gated calcium ion channel subunit 3.2 (Cav3.2) in SCNGABA neurons in regulating rhythmic neural firing and glucose homeostasis. We found that Cav3.2 global knockout (Cav3.2KO) mice displayed disrupted diurnal rhythm of blood glucose levels. Selective deletion of Cav3.2 in the SCNGABA neurons abolished the oscillatory firing activity of SCNGABA neurons and led to glucose intolerance. Furthermore, our real-time calcium imaging data in freely moving mice showed that SCN GABA neuronal activity displayed a circadian rhythm pattern in WT control mice but not Cav3.2KO mice or mice with Cav3.2 selectively deleted in the SCNGABA neurons. These findings provide insights into how the calcium ion channel regulates the central circadian clock and glucose metabolism, with implications in the extended dawn phenomenon in type 2 diabetes. Keywords: Type 2 diabetes, calcium channel, Cav3.2, GABA neurons, circadian rhythm Reference: 1 Ding, G. et al. REV-ERB in GABAergic neurons controls diurnal hepatic insulin sensitivity. Nature 592, 763-767, doi:10.1038/s41586-021-03358-w (2021) . Disclosure P. Xu: None. W. Zhou: None. Z. Sun: None. Y. He: None. Funding NIH (P20 GM135002)

Geometric characterization of dynamical structure for neural firing activities induced by inhibitory pulse.

In general, inhibitory stimuli are thought to inhibit neuronal firing, but they may actually enhance firing sometimes, such as post-inhibitory rebound spike (PIR spike) and post-inhibitory facilitation (PIF) phenomena, which play an important role in human neuronal activities. We study responses to inhibitory pulse in a classical neuron model (Quartic adaptive Integrate-and-fire model) well known to reproduce a number of biologically realistic behaviors. The three phenomena that we study are PIR, in which a neuron fires after an inhibitory pulse, and PIF, in which a subthreshold excitatory input can induce a spike if it is applied with proper timing after an inhibitory pulse, as well as period firing after inhibitory pulse. When the system features focus and saddle two equilibriums, the three phenomena will be occurred under the inhibitory pulse, while all three phenomena will not be induced when the system features node and saddle two equilibriums. Using dynamical systems theory, we explain the threshold mechanism of enhancement of neural firing response induced by inhibitory pulse and analyze the origin of these phenomena from several factors. We also describe the geometric characterization of dynamical structures of these three phenomena. This study therefore enrich the paradoxical phenomena that induced by inhibitory input and advance our understanding of its role.

Distribution of equal states for amplitude fluctuations in epileptic EEG

Distribution of equal states for amplitude fluctuations in epileptic EEG

A Measure of Concurrent Neural Firing Activity Based on Mutual Information.

Multiple methods have been developed in an attempt to quantify stimulus-induced neural coordination and to understand internal coordination of neuronal responses by examining the synchronization phenomena in neural discharge patterns. In this work we propose a novel approach to estimate the degree of concomitant firing between two neural units, based on a modified form of mutual information (MI) applied to a two-state representation of the firing activity. The binary profile of each single unit unfolds its discharge activity in time by decomposition into the state of neural quiescence/low activity and state of moderate firing/bursting. Then, the MI computed between the two binary streams is normalized by their minimum entropy and is taken as positive or negative depending on the prevalence of identical or opposite concomitant states. The resulting measure, denoted as Concurrent Firing Index based on MI (CFIMI), relies on a single input parameter and is otherwise assumption-free and symmetric. Exhaustive validation was carried out through controlled experiments in three simulation scenarios, showing that CFIMI is independent on firing rate and recording duration, and is sensitive to correlated and anti-correlated firing patterns. Its ability to detect non-correlated activity was assessed using ad-hoc surrogate data. Moreover, the evaluation of CFIMI on experimental recordings of spiking activity in retinal ganglion cells brought insights into the changes of neural synchrony over time. The proposed measure offers a novel perspective on the estimation of neural synchrony, providing information on the co-occurrence of firing states in the two analyzed trains over longer temporal scales compared to existing measures.

Light/Clock Influences Membrane Potential Dynamics to Regulate Sleep States.

The circadian rhythm is a fundamental process that regulates the sleep–wake cycle. This rhythm is regulated by core clock genes that oscillate to create a physiological rhythm of circadian neuronal activity. However, we do not know much about the mechanism by which circadian inputs influence neurons involved in sleep–wake architecture. One possible mechanism involves the photoreceptor cryptochrome (CRY). In Drosophila, CRY is receptive to blue light and resets the circadian rhythm. CRY also influences membrane potential dynamics that regulate neural activity of circadian clock neurons in Drosophila, including the temporal structure in sequences of spikes, by interacting with subunits of the voltage-dependent potassium channel. Moreover, several core clock molecules interact with voltage-dependent/independent channels, channel-binding protein, and subunits of the electrogenic ion pump. These components cooperatively regulate mechanisms that translate circadian photoreception and the timing of clock genes into changes in membrane excitability, such as neural firing activity and polarization sensitivity. In clock neurons expressing CRY, these mechanisms also influence synaptic plasticity. In this review, we propose that membrane potential dynamics created by circadian photoreception and core clock molecules are critical for generating the set point of synaptic plasticity that depend on neural coding. In this way, membrane potential dynamics drive formation of baseline sleep architecture, light-driven arousal, and memory processing. We also discuss the machinery that coordinates membrane excitability in circadian networks found in Drosophila, and we compare this machinery to that found in mammalian systems. Based on this body of work, we propose future studies that can better delineate how neural codes impact molecular/cellular signaling and contribute to sleep, memory processing, and neurological disorders.

Memristor-based oscillatory behavior in the FitzHugh–Nagumo and Hindmarsh–Rose models

The neural firing activities related to information coding maintaining the information transmission vary qualitatively considering the electromagnetic induction. The firing of a single neuron can be investigated by Hopf bifurcation analysis. In this paper, with the help of the center manifold theory and algebraic invariants method, general parameter conditions are obtained for the existence and stability of Hopf bifurcation in the memristive FitzHugh–Nagumo (FHN) and Hindmarsh–Rose (HR) models. By studying the roots of the characteristic polynomial, the parameter ranges that cause the systems to undergo oscillations have been studied. With the help of the Lyapunov functions, the general form of the first Lyapunov coefficient is obtained for the memristive FHN model. By using the center manifold theory and algebraic invariants method, the memristive HR model is reduced to a more manageable model which inherits the local dynamics of the original model. The effects of electromagnetic induction on neural oscillations are studied for general parameter conditions. The oscillatory bursting firing regimes for the models are illustrated by numerical simulations.

Delayed excitatory self-feedback-induced negative responses of complex neuronal bursting patterns* *Project supported by the National Natural Science Foundation of China (Grant Nos. 11872276 and 11762001) and the Young Talents of Science and Technology in Universities of Inner Mongolia Autonomous Region, China (Grant No. NJYT-20-A09).

In traditional viewpoint, excitatory modulation always promotes neural firing activities. On contrary, the negative responses of complex bursting behaviors to excitatory self-feedback mediated by autapse with time delay are acquired in the present paper. Two representative bursting patterns which are identified respectively to be “Fold/Big Homoclinic” bursting and “Circle/Fold cycle” bursting with bifurcations are studied. For both burstings, excitatory modulation can induce less spikes per burst for suitable time delay and strength of the self-feedback/autapse, because the modulation can change the initial or termination phases of the burst. For the former bursting composed of quiescent state and burst, the mean firing frequency exhibits increase, due to that the quiescent state becomes much shorter than the burst. However, for the latter bursting pattern with more complex behavior which is depolarization block lying between burst and quiescent state, the firing frequency manifests decrease in a wide range of time delay and strength, because the duration of both depolarization block and quiescent state becomes long. Therefore, the decrease degree of spike number per burst is larger than that of the bursting period, which is the cause for the decrease of firing frequency. Such reduced bursting activity is explained with the relations between the bifurcation points of the fast subsystem and the bursting trajectory. The present paper provides novel examples of paradoxical phenomenon that the excitatory effect induces negative responses, which presents possible novel modulation measures and potential functions of excitatory self-feedback/autapse to reduce bursting activities.

Unsupervised Adaptive Weight Pruning for Energy-Efficient Neuromorphic Systems.

To tackle real-world challenges, deep and complex neural networks are generally used with a massive number of parameters, which require large memory size, extensive computational operations, and high energy consumption in neuromorphic hardware systems. In this work, we propose an unsupervised online adaptive weight pruning method that dynamically removes non-critical weights from a spiking neural network (SNN) to reduce network complexity and improve energy efficiency. The adaptive pruning method explores neural dynamics and firing activity of SNNs and adapts the pruning threshold over time and neurons during training. The proposed adaptation scheme allows the network to effectively identify critical weights associated with each neuron by changing the pruning threshold dynamically over time and neurons. It balances the connection strength of neurons with the previous layer with adaptive thresholds and prevents weak neurons from failure after pruning. We also evaluated improvement in the energy efficiency of SNNs with our method by computing synaptic operations (SOPs). Simulation results and detailed analyses have revealed that applying adaptation in the pruning threshold can significantly improve network performance and reduce the number of SOPs. The pruned SNN with 800 excitatory neurons can achieve a 30% reduction in SOPs during training and a 55% reduction during inference, with only 0.44% accuracy loss on MNIST dataset. Compared with a previously reported online soft pruning method, the proposed adaptive pruning method shows 3.33% higher classification accuracy and 67% more reduction in SOPs. The effectiveness of our method was confirmed on different datasets and for different network sizes. Our evaluation showed that the implementation overhead of the adaptive method regarding speed, area, and energy is negligible in the network. Therefore, this work offers a promising solution for effective network compression and building highly energy-efficient neuromorphic systems in real-time applications.

Dysfunction of serotonergic activity and emotional responses across the light-dark cycle in mice lackingmelatonin MT2 receptors.

Melatonin (MLT) levels fluctuate according to the external light/dark cycle in both diurnal and nocturnal mammals. We previously demonstrated that melatoninMT2 receptor knockout (MT2 -/- ) mice show a decreased nonrapid eyemovement sleep over 24hours and increased wakefulness during the inactive (light) phase. Here, we investigated the role of MT2 receptors in physiological light/dark cycle fluctuations in the activity of dorsal raphe nucleus (DRN) serotonin (5-HT) neurons and anxiety- and depression-like behavior. We found that the 5-HT burst-firing activity was tonically reduced across the whole 24hours in MT2 -/- mice compared with MT2 +/+ mice. Importantly, the physiological changes in the spontaneous firing activity of DRN 5-HT neurons during the light/dark cycle were nullified in MT2 -/- mice, with a higher DRN 5-HT neural firing activity during the light phase in MT2 -/- than in MT2 +/+ mice. The role of MT2 receptors over DRN 5-HT neurons was confirmed by acute pharmacological studies in which the selective MT2 receptors agonist UCM1014 dose dependently inhibited DRN 5-HT activity, mostly during the dark phase. Compared withMT2 +/+ , MT2 -/- mice displayed an anxiety-like phenotype in the novelty-suppressed feeding and in the light/dark box tests; while anxiety levels in the light/dark box test were lower during the dark thanduring the light phase inMT2 +/+ mice, the opposite was seen in MT2 -/- mice. No differences between MT2 +/+ and MT2 -/- mice were observed for depression-like behavior in the forced swim and in the sucrose preference tests. These results suggest that MT2 receptor genetic inactivation impacts 5-HT neurotransmission and interferes with anxiety levels by perturbing thephysiologic light/dark pattern.

Bifurcations of Negative Responses to Positive Feedback Current Mediated by Memristor in a Neuron Model with Bursting Patterns

In contrast to traditional viewpoint that positive feedback current always enhances neural firing activities, in the present paper, we identify that the excitatory feedback current mediated by memristor can induce negative responses of bursting patterns, which can be well interpreted with bifurcations. For the Hindmarsh–Rose neuron model without memristor, the period-adding bifurcations of bursting patterns and increase of firing frequency can be induced by increasing the excitatory effect of the background current. After introducing a memristor to simulate the biological synapse or electromagnetic induction effect, inverse period-adding or complex bifurcations of bursting patterns are induced by the excitatory feedback current mediated by the memristor. The number of spikes per burst becomes smaller and the firing frequency becomes lower when increasing the positive feedback gain. Such negative responses of bursting patterns to the positive feedback current are demonstrated in a circuit designed with Digital Signal Processor systems of the MatLab software. Furthermore, the underlying bifurcation mechanism of the negative responses to the positive feedback is acquired with fast–slow variable dissection method. With increasing feedback gain, the initial phase of the burst, which corresponds to a saddle-node bifurcation point of the fast subsystem, delays, while the termination phase of the burst, which corresponds to a saddle-homoclinic bifurcation point, remains unchanged. Therefore, the burst becomes narrower with increasing feedback gain, which leads to decrease in the number of spikes within a burst and decrease in firing frequency. The results present a paradoxical nonlinear phenomenon and the dynamical mechanism, which is helpful for understanding the functions of memristor and roles of the electromagnetic induction current.

Self-sustained activity of low firing rate in balanced networks

Self-sustained activity of low firing rate in balanced networks