Spike Timing Research Articles

Correlated signatures of social behavior in cerebellum and anterior cingulate cortex.

The cerebellum has been implicated in the regulation of social behavior. Its influence is thought to arise from communication, via the thalamus, to forebrain regions integral in the expression of social interactions, including the anterior cingulate cortex (ACC). However, the signals encoded or the nature of the communication between the cerebellum and these brain regions is poorly understood. Here, we describe an approach that overcomes technical challenges in exploring the coordination of distant brain regions at high temporal and spatial resolution during social behavior. We developed the E-Scope, an electrophysiology-integrated miniature microscope, to synchronously measure extracellular electrical activity in the cerebellum along with calcium imaging of the ACC. This single coaxial cable device combined these data streams to provide a powerful tool to monitor the activity of distant brain regions in freely behaving animals. During social behavior, we recorded the spike timing of multiple single units in cerebellar right Crus I (RCrus I) Purkinje cells (PCs) or dentate nucleus (DN) neurons while synchronously imaging calcium transients in contralateral ACC neurons. We found that during social interactions a significant subpopulation of cerebellar PCs were robustly inhibited, while most modulated neurons in the DN were activated, and their activity was correlated with positively modulated ACC neurons. These distinctions largely disappeared when only non-social epochs were analyzed suggesting that cerebellar-cortical interactions were behaviorally specific. Our work provides new insights into the complexity of cerebellar activation and co-modulation of the ACC during social behavior and a valuable open-source tool for simultaneous, multimodal recordings in freely behaving mice.

Correlated signatures of social behavior in cerebellum and anterior cingulate cortex

The cerebellum has been implicated in the regulation of social behavior. Its influence is thought to arise from communication, via the thalamus, to forebrain regions integral in the expression of social interactions, including the anterior cingulate cortex (ACC). However, the signals encoded or the nature of the communication between the cerebellum and these brain regions is poorly understood. Here, we describe an approach that overcomes technical challenges in exploring the coordination of distant brain regions at high temporal and spatial resolution during social behavior. We developed the E-Scope, an electrophysiology-integrated miniature microscope, to synchronously measure extracellular electrical activity in the cerebellum along with calcium imaging of the ACC. This single coaxial cable device combined these data streams to provide a powerful tool to monitor the activity of distant brain regions in freely behaving animals. During social behavior, we recorded the spike timing of multiple single units in cerebellar right Crus I (RCrus I) Purkinje cells (PCs) or dentate nucleus (DN) neurons while synchronously imaging calcium transients in contralateral ACC neurons. We found that during social interactions a significant subpopulation of cerebellar PCs were robustly inhibited, while most modulated neurons in the DN were activated, and their activity was correlated with positively modulated ACC neurons. These distinctions largely disappeared when only non-social epochs were analyzed suggesting that cerebellar-cortical interactions were behaviorally specific. Our work provides new insights into the complexity of cerebellar activation and co-modulation of the ACC during social behavior and a valuable open-source tool for simultaneous, multimodal recordings in freely behaving mice.

Machine learning-based high-frequency neuronal spike reconstruction from low-frequency and low-sampling-rate recordings

Recording neuronal activity using multiple electrodes has been widely used to understand the functional mechanisms of the brain. Increasing the number of electrodes allows us to decode more variety of functionalities. However, handling massive amounts of multichannel electrophysiological data is still challenging due to the limited hardware resources and unavoidable thermal tissue damage. Here, we present machine learning (ML)-based reconstruction of high-frequency neuronal spikes from subsampled low-frequency band signals. Inspired by the equivalence between high-frequency restoration and super-resolution in image processing, we applied a transformer ML model to neuronal data recorded from both in vitro cultures and in vivo male mouse brains. Even with the x8 downsampled datasets, our trained model reasonably estimated high-frequency information of spiking activity, including spike timing, waveform, and network connectivity. With our ML-based data reduction applicable to existing multichannel recording hardware while achieving neuronal signals of broad bandwidths, we expect to enable more comprehensive analysis and control of brain functions.

Wired together, change together: Spike timing modifies transmission in converging assemblies.

The precise timing of neuronal spikes may lead to changes in synaptic connectivity and is thought to be crucial for learning and memory. However, the effect of spike timing on neuronal connectivity in the intact brain remains unknown. Using closed-loop optogenetic stimulation in CA1 of freely moving mice, we generated unique spike patterns between presynaptic pyramidal cells (PYRs) and postsynaptic parvalbumin (PV)-immunoreactive cells. The stimulation led to spike transmission changes that occurred together across all presynaptic PYRs connected to the same postsynaptic PV cell. The precise timing of all presynaptic and postsynaptic cell spikes affected transmission changes. These findings reveal an unexpected plasticity mechanism, in which the spike timing of an entire cell assembly has a more substantial impact on effective connectivity than that of individual cell pairs.

Dataset of human-single neuron activity during a Sternberg working memory task

We present a dataset of 1809 single neurons recorded from the human medial temporal lobe (amygdala and hippocampus) and medial frontal lobe (anterior cingulate cortex, pre-supplementary motor area, ventral medial prefrontal cortex) across 41 sessions from 21 patients that underwent seizure monitoring with depth electrodes. Subjects performed a screening task (907 neurons) to identify images for which highly selective cells were present. Subjects then performed a working memory task (902 neurons), in which they were sequentially presented with 1–3 images for which highly selective cells were present and, following a maintenance period, were asked if the probe was identical to one of the maintained images. This Neurodata Without Borders formatted dataset includes spike times, extracellular spike waveforms, stimuli presented, behavior, electrode locations, and subject demographics. As validation, we replicate previous findings on the selectivity of concept cells and their persistent activity during working memory maintenance. This large dataset of rare human single-neuron recordings and behavior enables the investigation of the neural mechanisms of working memory in humans.

Nerve Wrap for Local Delivery of FK506/Tacrolimus Accelerates Nerve Regeneration.

Peripheral nerve injuries (PNIs) occur frequently and can lead to devastating and permanent sensory and motor function disabilities. Systemic tacrolimus (FK506) administration has been shown to hasten recovery and improve functional outcomes after PNI repair. Unfortunately, high systemic levels of FK506 can result in adverse side effects. The localized administration of FK506 could provide the neuroregenerative benefits of FK506 while avoiding systemic, off-target side effects. This study investigates the utility of a novel FK506-impregnated polyester urethane urea (PEUU) nerve wrap to treat PNI in a previously validated rat infraorbital nerve (ION) transection and repair model. ION function was assessed by microelectrode recordings of trigeminal ganglion cells responding to controlled vibrissae deflections in ION-transected and -repaired animals, with and without the nerve wrap. Peristimulus time histograms (PSTHs) having 1 ms bins were constructed from spike times of individual single units. Responses to stimulus onsets (ON responses) were calculated during a 20 ms period beginning 1 ms after deflection onset; this epoch captures the initial, transient phase of the whisker-evoked response. Compared to no-wrap controls, rats with PEUU-FK506 wraps functionally recovered earlier, displaying larger response magnitudes. With nerve wrap treatment, FK506 blood levels up to six weeks were measured nearly at the limit of quantification (LOQ ≥ 2.0 ng/mL); whereas the drug concentrations within the ION and muscle were much higher, demonstrating the local delivery of FK506 to treat PNI. An immunohistological assessment of ION showed increased myelin expression for animals assigned to neurorrhaphy with PEUU-FK506 treatment compared to untreated or systemic-FK506-treated animals, suggesting that improved PNI outcomes using PEUU-FK506 is mediated by the modulation of Schwann cell activity.

Disruption of TLE epileptiform activity retarded the seizure and reduced pathological HFOs

In temporal lobe epilepsy (TLE), the epileptogenic zones, such as the temporal lobe structure, could generate pathological high-frequency oscillations (pHFOs, 250–500 Hz) before the ictal period. These pHFOs have also been observed during the process of seizures in both TLE patients and animals, exhibiting a critical role as promising biomarkers for TLE seizures. TLE seizures could be modulated via regulating the neural excitability in epileptogenic zones, for that TLE is primarily associated with the excitation-inhibition imbalance. However, whether these kinds of modulations could also impact the pHFOs characteristics during TLE seizures is still unclear. For this purpose, we pharmaco-genetically inhibited the principal cells (PCs) in the mouse CA3 region and tracked the difference in the behavioral and electrophysiological features during LiCl-pilocarpine-induced TLE seizure between the hM4Di+CNO (experimental) mice and mCherry+CNO (control) mice. Delayed latency, decreased averaged duration, and reduced counts of the generalized seizure were observed in the experimental mice. Besides, the electrophysiological characteristics, such as the firing rate of PCs and the count of pHFO, exhibited significant decline in the CA3 and CA1 regions. During TLE seizure, there existed strong phase-coupling between pHFO and PCs spike timing in the control mice, while it was abolished in the experimental mice. In addition, we also found that the counts of pHFO were significantly associated with the behavioral features, indicating the close relationships within them. Collectively, our findings suggested that alterations in pHFO and the retardation of seizures may be attributed to disruptions in neuronal excitability, and the variations of electrophysiological features were related to seizure severity during TLE seizures. These results provide valuable insights into the role of pHFOs in TLE and shed light on the underlying mechanisms involved.

Leaky Integrate‐and‐Fire Neuron Based on Organic Electrochemical Transistor for Spiking Neural Networks with Temporal‐Coding

AbstractSpiking neural networks (SNNs) employ discrete spikes that mimic the firing of neurons in biological systems to process and transmit information. This characteristic enables SNNs to effectively capture temporal dynamics and capitalize on the time information inherent in time‐varying inputs, such as motion, audio/video streams, and other sequential data. Currently, most hardware implementations of SNNs are designed to use rate‐coding, where information is encoded in the rate of spikes. However, it still remains challenging for the hardware implementation of temporal coding in SNNs, which allows for higher input sparsity and exploits additional dimensions such as precise spike timing and relative spike timings. This study presents hardware implementations of SNNs constructed by organic electrochemical transistors (OECTs), processing temporal‐coded information. The protic dynamics in response to electrical stimuli enable the emulation of temporal integration, reset, and leaking of membrane potential in a simple leaky integrate‐and‐fire (LIF) neuron circuit. By utilizing these features, the emulated LIF neuron can be employed to construct SNNs capable of processing temporal‐coded information in complex tasks including coincidence detection and dynamic handwriting recognition, exhibiting high performance and good tolerance even when dealing with noisy datasets.

Bifurcation and negative self-feedback mechanisms for enhanced spike-timing precision of inhibitory interneurons

<abstract> <p>A high spike-timing precision characterized by a small variation in interspike intervals of neurons is important for information processing in various brain functions. An experimental study on fast-spiking interneurons has shown that inhibitory autapses functioning as negative self-feedback can enhance spike-timing precision. In the present paper, bifurcation and negative self-feedback mechanisms for the enhanced spike-timing precision to stochastic modulations are obtained in two theoretical models, presenting theoretical explanations to the experimental finding. For stochastic spikes near both the saddle-node bifurcation on an invariant cycle (SNIC) and the subcritical Hopf (SubH) bifurcation with classes 1 and 2 excitabilities, respectively, enhanced spike-timing precision appears in large ranges of the conductance and the decaying rate of inhibitory autapses, closely matching the experimental observation. The inhibitory autaptic current reduces the membrane potential after a spike to a level lower than that in the absence of inhibitory autapses and the threshold to evoke the next spike, making it more difficult for stochastic modulations to affect spike timings, and thereby enhancing spike-timing precision. In addition, firing frequency near the SubH bifurcation is more robust than that near the SNIC bifurcation, resulting in a higher spike-timing precision for the SubH bifurcation. The bifurcation and negative self-feedback mechanisms for the enhanced spike-timing precision present potential measures to modulate the neuronal dynamics or the autaptic parameters to adjust the spike-timing precision.</p> </abstract>

Optimizing skyrmionium movement and stability via stray magnetic fields in trilayer nanowire constructs.

Skyrmioniums, known for their unique transport and regulatory properties, are emerging as potential cornerstones for future data storage systems. However, the stability of skyrmionium movement faces considerable challenges due to the skyrmion Hall effect, which is induced by deformation. In response, our research introduces an innovative solution: we utilized micro-magnetic simulations to create a sandwiched trilayer nanowire structure augmented with a stray magnetic field. This combination effectively guides the skyrmionium within the ferromagnetic (FM) layer. Our empirical investigations reveal that the use of a stray magnetic field not only reduces the size of the skyrmionium but also amplifies its stability. This dual-effect proficiently mitigates the deformation of skyrmionium movement and boosts their thermal stability. We find these positive outcomes are most pronounced at a particular intensity of the stray magnetic field. Importantly, the required stray magnetic field can be generated using a heavy metal (HM1) layer of suitable thickness, rendering the practical application of this approach plausible in real-world experiments. Additionally, we analyze the functioning mechanism based on the Landau-Lifshitz-Gilbert (LLG) equation and energy variation. We also develop a deep spiking neural network (DSNN), which achieves a remarkable recognition accuracy of 97%. This achievement is realized through supervised learning via the spike timing dependent plasticity rule (STDP), considering the nanostructure as an artificial synapse device that corresponds to the electrical properties of the nanostructure. In conclusion, our study provides invaluable insights for the design of innovative information storage devices utilizing skyrmionium technology. By tackling the issues presented by the skyrmion Hall effect, we outline a feasible route for the practical application of this advanced technology. Our research, therefore, serves as a robust platform for continued investigations in this field.

(Invited) Nanoscale Devices for Accelerating AI Algorithms

The energy consumption of training and deploying state-of-the-art artificial intelligence (AI) models has experienced exponential growth, driven by increasing model parameters and the necessary training data. Recent models demand over 1 billion PetaFLOPs of total computation for training, potentially taking weeks to complete. With current GPU performance at approximately 1-3 TOPS/W, GPU training energy consumption alone can exceed 100,000 kWh, equivalent to the monthly energy expenditure of 100 US households. Over the next decade, these models are expected to scale up further, driving total computing energy to constitute a significant portion of global consumption. In this talk, we describe both synaptic and neuronal devices that can accelerate AI algorithms with potentially multiple orders of magnitude improvement in power efficiency.First, we describe an oscillatory retinal neuron (ORN) that directly converts incident DC light into voltage spikes. Coupled arrays of this device result in an imager that carries out in-sensor processing while capturing an image. Uniquely, the conversion from input light to output voltage spikes occurs without external power. When coupled in arrays, the neighboring neurons interact with each other to influence the spiking frequency spectrum. This allows the arrays to carry out frequency multiplexed computation on an input image. It is shown that this approach can achieve>20,000 TOPS/W, multiple orders of magnitude greater than the current approaches. Theory and simulation is used to elucidate how coupled ORNs carry out computation on an input image. When coupled, each output frequency band encodes a unique computation on the input image. By tuning the coupling impedances and the frequency bands, user defined computations can be carried out on the input. We experimentally show that this can be carried out with a 3x3 array that demonstrates simultaneous edge detection, intensity filtering, image segmentation and other functions. This hardware is then used to demonstrate improvement in MNIST handwritten digit classification accuracy over a traditional imager connected to a fully connected network.Next, we describe spiking synaptic devices that can be directly fabricated in the back-end of line of CMOS devices. These devices consist of an InP transistor channel with an engineered gate stack. Using a uniform gate insulator, we can demonstrate behaviors of biological synapses, such as potentiation, depression, spike number dependent plasticity, and spike timing dependent plasticity. By introducing a heterostructured gate insulator, it is shown that short-term to long-term memory transitions can be designed into the device. Finally, by using a transparent gate, an in-sensor synaptic phototransistor is demonstrated and the performance of these devices at a system level is demonstrated.

Supervised Learning in Multilayer Spiking Neural Networks With Spike Temporal Error Backpropagation.

The brain-inspired spiking neural networks (SNNs) hold the advantages of lower power consumption and powerful computing capability. However, the lack of effective learning algorithms has obstructed the theoretical advance and applications of SNNs. The majority of the existing learning algorithms for SNNs are based on the synaptic weight adjustment. However, neuroscience findings confirm that synaptic delays can also be modulated to play an important role in the learning process. Here, we propose a gradient descent-based learning algorithm for synaptic delays to enhance the sequential learning performance of single spiking neuron. Moreover, we extend the proposed method to multilayer SNNs with spike temporal-based error backpropagation. In the proposed multilayer learning algorithm, information is encoded in the relative timing of individual neuronal spikes, and learning is performed based on the exact derivatives of the postsynaptic spike times with respect to presynaptic spike times. Experimental results on both synthetic and realistic datasets show significant improvements in learning efficiency and accuracy over the existing spike temporal-based learning algorithms. We also evaluate the proposed learning method in an SNN-based multimodal computational model for audiovisual pattern recognition, and it achieves better performance compared with its counterparts.

A novel role for APP in neural circuit function

AbstractBackgroundDespite the importance of the amyloid‐precursor protein (APP) to Alzheimer’s Disease pathophysiology and therapy, and notable pan‐cellular and synaptic abundance, the physiological function of APP and its homologues remains to be fully elucidated in‐vivo. Here, we interrogated how the APP family regulates neuronal cell function and long‐range neural circuit dynamics in the intact brain.MethodAPPflox/flox/APLP2flox /flox/APLP1−/− mice were crossed with NexCre mice expressing Cre prenatally from ∼E12.5 in postmitotic neuronal precursor cells of the cortex and hippocampus, producing a conditional triple knockout model (cTKO) of the APP family in excitatory forebrain neurons that is associated with pronounced behavioral deficits (Steubler et al., 2021, EMBO Journal). Neuronal function at the single cell/unit and circuit‐level was characterised using in‐vivo two‐photon and mesoscopic calcium imaging, and bilateral Neuropixels recordings in awake and anesthetised mice.ResultcTKO mice exhibited pronounced suppression in spike timing variability and dynamic excitability in visual cortex and hippocampus CA1 relative to controls, with a marked inter‐hemispheric decoherence of oscillatory activity (including low‐frequency and sleep‐associated slow waves) and dysfunctional intrahemispheric propagation of neuronal activity. Further, cTKO mice also displayed behavioural abnormalities in open field tests relative to controls which was linked to functional deficits. High‐resolution synaptic imaging revealed a significant reduction in the number of synapses containing the NMDA receptor subunit GluN1 in cTKO mice versus controls, with partial NMDA receptor agonism (d‐cycloserine, 30mg/kg, i.p.) ameliorating cellular and circuit‐level impairments as well as behavioral impairments.ConclusionThese findings provide novel in‐vivo evidence for an essential role of APP in regulating synaptic function via an effect on NMDA receptors, that elucidate the physiological role of this preeminent protein family and have important implications for disease and therapy.

Spike Timing Dependent Gradient for Direct Training of Fast and Efficient Binarized Spiking Neural Networks

Spike Timing Dependent Gradient for Direct Training of Fast and Efficient Binarized Spiking Neural Networks

Emulation of Pavlovian conditioning and pattern recognition through fully connected neural networks using Holmium oxide (Ho2O3) based synaptic RRAM device.

In this manuscript, we report on the paramagnetic Ho2O3 -based synaptic resistive random-access memory (RRAM) device for the implementation of neuronal functionalities such as long-term potentiation (LTP), long-term depression (LTD) and spike timing dependent plasticity (STDP) respectively. The plasticity of the artificial synapse is also studied by varying pulse amplitude, pulse width, and pulse interval. In addition, we could classify handwritten Modified National Institute of Standards and Technology (MNIST) data set using a fully connected neural network (FCN). The device-based FCN records a high classification accuracy of 93.47% which is comparable to the software-based test accuracy of 97.97%. This indicates the highly optimized behavior of our synaptic device for hardware neuromorphic applications. Successful emulation of Pavlovian classical conditioning for associative learning of the biological brain is achieved. We believe that the present device consists the potential to utilize in neuromorphic applications.

Lightweight spiking neural network training based on spike timing dependent backpropagation

Spiking neural networks are energy efficient and biological interpretability, communicating through sparse, asynchronous spikes, which makes them suitable for neuromorphic hardware. However, due to the nature of binary weights and spike trains in time-coded binarized spiking neural networks, their forward propagation may cause neurons to not fire spikes, and their backward propagation has non-differentiable problems. Moreover, the current use of deep and complex network structures generates a large number of redundant parameters. Therefore, we need effective methods to improve energy efficiency without reducing accuracy. We propose a dynamic threshold model that can reduce the number of dead neurons. We combine the backpropagation algorithm and the spike timing dependent plasticity algorithm to avoid the non-differentiable problem. We propose a neuron pruning strategy based on adaptive firing time threshold. This pruning strategy prunes 267 neurons in a network of 600 neurons, reducing the network size and obtaining a more compact network structure. The energy efficiency is improved by 0.55×, while the classification accuracy is lost by 1.1%.

Granule cells perform frequency-dependent pattern separation in a computational model of the dentate gyrus.

Mnemonic discrimination (MD) may be dependent on oscillatory perforant path input frequencies to the hippocampus in a "U"-shaped fashion, where some studies show that slow and fast input frequencies support MD, while other studies show that intermediate frequencies disrupt MD. We hypothesize that pattern separation (PS) underlies frequency-dependent MD performance. We aim to study, in a computational model of the hippocampal dentate gyrus (DG), the network and cellular mechanisms governing this putative "U"-shaped PS relationship. We implemented a biophysical model of the DG that produces the hypothesized "U"-shaped input frequency-PS relationship, and its associated oscillatory electrophysiological signatures. We subsequently evaluated the network's PS ability using an adapted spatiotemporal task. We undertook systematic lesion studies to identify the network-level mechanisms driving the "U"-shaped input frequency-PS relationship. A minimal circuit of a single granule cell (GC) stimulated with oscillatory inputs was also used to study potential cellular-level mechanisms. Lesioning synapses onto GCs did not impact the "U"-shaped input frequency-PS relationship. Furthermore, GC inhibition limits PS performance for fast frequency inputs, while enhancing PS for slow frequency inputs. GC interspike interval was found to be input frequency dependent in a "U"-shaped fashion, paralleling frequency-dependent PS observed at the network level. Additionally, GCs showed an attenuated firing response for fast frequency inputs. We conclude that independent of network-level inhibition, GCs may intrinsically be capable of producing a "U"-shaped input frequency-PS relationship. GCs may preferentially decorrelate slow and fast inputs via spike timing reorganization and high frequency filtering.

Model-based analysis of subthreshold mechanisms of spinal cord stimulation for pain

Objective. Spinal cord stimulation (SCS) is a common treatment for chronic pain. For decades, SCS maximized overlap between stimulation-induced paresthesias and the patient’s painful areas. Recently developed SCS paradigms relieve pain at sub-perceptible amplitudes, yet little is known about the neural response to these new waveforms or their analgesic mechanisms of action. Therefore, in this study, we investigated the neural response to multiple forms of paresthesia-free SCS. Approach. We used computational modeling to investigate the neurophysiological effects and the plausibility of commonly proposed mechanisms of three paresthesia-free SCS paradigms: burst, 1 kHz, and 10 kHz SCS. Specifically, in C- and Aβ-fibers, we investigated the effects of different SCS waveforms on spike timing and activation thresholds, as well as how stochastic ion channel gating affects the response of dorsal column axons. Finally, we characterized membrane polarization of superficial dorsal horn neurons. Main results. We found that none of the SCS waveforms activate nor modulate spike timing in C-fibers. Spike timing was modulated in Aβ-fibers only at suprathreshold amplitudes. Ion channel stochasticity had little effect on Aβ-fiber activation thresholds but produced heterogeneous spike timings at suprathreshold amplitudes. Finally, local cells were preferentially polarized in their axon terminals, and the magnitude of this polarization was dependent on cellular morphology and position relative to the stimulation electrodes. Significance. Overall, the mechanisms of action of subparesthetic SCS remain unclear. Our results suggest that no SCS waveforms directly activate C-fibers, and modulation of spike timing is unlikely at subthreshold amplitudes. We conclude that potential subthreshold neuromodulatory effects of SCS on local cells are likely to be presynaptic in nature, as axons are preferentially depolarized during SCS.

Algebraic approach to spike-time neural codes in the hippocampus.

Although temporal coding through spike-time patterns has long been of interest in neuroscience, the specific structures that could be useful for spike-time codes remain highly unclear. Here, we introduce an analytical approach, using techniques from discrete mathematics, to study spike-time codes. As an initial example, we focus on the phenomenon of "phase precession" in the rodent hippocampus. During navigation and learning on a physical track, specific cells in a rodent's brain form a highly structured pattern relative to the oscillation of population activity in this region. Studies of phase precession largely focus on its role in precisely ordering spike times for synaptic plasticity, as the role of phase precession in memory formation is well established. Comparatively less attention has been paid to the fact that phase precession represents one of the best candidates for a spike-time neural code. The precise nature of this code remains an open question. Here, we derive an analytical expression for a function mapping points in physical space to complex-valued spikes by representing individual spike times as complex numbers. The properties of this function make explicit a specific relationship between past and future in spike patterns of the hippocampus. Importantly, this mathematical approach generalizes beyond the specific phenomenon studied here, providing a technique to study the neural codes within precise spike-time sequences found during sensory coding and motor behavior. We then introduce a spike-based decoding algorithm, based on this function, that successfully decodes a simulated animal's trajectory using only the animal's initial position and a pattern of spike times. This decoder is robust to noise in spike times and works on a timescale almost an order of magnitude shorter than typically used with decoders that work on average firing rate. These results illustrate the utility of a discrete approach, based on the structure and symmetries in spike patterns across finite sets of cells, to provide insight into the structure and function of neural systems.

Spiking Neural Network with plasticity in the time domain recovers temporal information from a noisy pattern using reference spikes

Neurons in the brain communicate with each other by sending trains of spikes that can encode information using the timings of the spikes. Spiking Neural Networks (SNNs) are biologically plausible neural networks that can model this transfer of information and that, by incorporating plasticity, can detect repeating patterns that may be embedded in a train of spikes with a noisy background. Although existing model networks are capable of learning to detect the occurrence of repeated patterns, they are not tailored to recover the entirety of the time sequence of the spikes in the pattern. Here we present a network that, in addition to typical parameters of plasticity such as weights and time delays, uses a new plasticity on the time domain that can recover most of the spikes in a pattern. This new plasticity acts by modifying the timings of reference spikes that come from a hypothesized upstream network that could potentially encode preexisting memories. The model is shown to be robust under several noise perturbation scenarios, and its overall performance demonstrates the benefits of using reference spikes to improve the temporal information processing ability of SNNs.