Optimal compensation for neuron loss

Parametrization of the dying brain: A case report from ICU bed-side EEG monitoring

BACKGROUND

Cortical high-frequency activation immediately before death has been reported, raising questions about an enhanced conscious state at this critical time. Here, we analyzed an electroencephalogram (EEG) from a comatose patient during the dying process with a standard bedside monitor and spectral parameterization techniques.

METHODS

We report neurophysiologic features of a dying patient without major cortical injury. Sixty minutes of frontal EEG activity was recorded using the Sedline™ monitor. Quantitative metrics of the frequency spectrum, the non-oscillatory 1/f characteristic, and signal complexity with Lemple-Ziv-Welch and permutation entropy were calculated. In addition to comparing the EEG trajectories over time, we provide a comparison to EEG records obtained from other studies with well-known vigilance states (sleep, anesthesia, and wake).

RESULTS

Although we observed changes in high-frequency activation during the dying process, larger alterations of the aperiodic EEG components were also noted. These changes differed dramatically when compared to EEG records representative of wake, slow-wave sleep, or anesthesia. Although still fundamentally unique, the neuronal activity present in the dying brain is more similar to REM sleep than any other state we tested.

CONCLUSION

Even in patients with coma, temporal dynamics in quantitative EEG features (including the aperiodic components) can be observed in the final hour before death.

Efficient coding in biophysically realistic excitatory-inhibitory spiking networks

The principle of efficient coding posits that sensory cortical networks are designed to encode maximal sensory information with minimal metabolic cost. Despite the major influence of efficient coding in neuroscience, it has remained unclear whether fundamental empirical properties of neural network activity can be explained solely based on this normative principle. Here, we derive the structural, coding, and biophysical properties of excitatory-inhibitory recurrent networks of spiking neurons that emerge directly from imposing that the network minimizes an instantaneous loss function and a time-averaged performance measure enacting efficient coding. We assumed that the network encodes a number of independent stimulus features varying with a time scale equal to the membrane time constant of excitatory and inhibitory neurons. The optimal network has biologically-plausible biophysical features, including realistic integrate-and-fire spiking dynamics, spike-triggered adaptation, and a non-specific excitatory external input. The excitatory-inhibitory recurrent connectivity between neurons with similar stimulus tuning implements feature-specific competition, similar to that recently found in visual cortex. Networks with unstructured connectivity cannot reach comparable levels of coding efficiency. The optimal ratio of excitatory vs inhibitory neurons and the ratio of mean inhibitory-to-inhibitory vs excitatory-to-inhibitory connectivity are comparable to those of cortical sensory networks. The efficient network solution exhibits an instantaneous balance between excitation and inhibition. The network can perform efficient coding even when external stimuli vary over multiple time scales. Together, these results suggest that key properties of biological neural networks may be accounted for by efficient coding.

Enteric nervous system regeneration and functional cure of experimental digestive Chagas disease with trypanocidal chemotherapy

Digestive Chagas disease (DCD) is an enteric neuropathy caused by Trypanosoma cruzi infection. There is a lack of evidence on the mechanism of pathogenesis and rationales for treatment. We used a female C3H/HeN mouse model that recapitulates key clinical manifestations to study how infection dynamics shape DCD pathology and the impact of treatment with the front-line, anti-parasitic drug benznidazole. Curative treatment 6 weeks post-infection resulted in sustained recovery of gastrointestinal transit function, whereas treatment failure led to infection relapse and gradual return of DCD symptoms. Neuro/immune gene expression patterns shifted from chronic inflammation to a tissue repair profile after cure, accompanied by increased cellular proliferation, glial cell marker expression and recovery of neuronal density in the myenteric plexus. Delaying treatment until 24 weeks post-infection led to partial reversal of DCD, suggesting the accumulation of permanent tissue damage over the course of chronic infection. Our study shows that murine DCD pathogenesis is sustained by chronic T. cruzi infection and is not an inevitable consequence of acute stage denervation. The risk of irreversible enteric neuromuscular tissue damage and dysfunction developing highlights the importance of prompt diagnosis and treatment. These findings support the concept of treating asymptomatic, T. cruzi-infected individuals with benznidazole to prevent DCD development.

Approximating Nonlinear Functions With Latent Boundaries in Low-Rank Excitatory-Inhibitory Spiking Networks

Deep feedforward and recurrent neural networks have become successful functional models of the brain, but they neglect obvious biological details such as spikes and Dale's law. Here we argue that these details are crucial in order to understand how real neural circuits operate. Towards this aim, we put forth a new framework for spike-based computation in low-rank excitatory-inhibitory spiking networks. By considering populations with rank-1 connectivity, we cast each neuron's spiking threshold as a boundary in a low-dimensional input-output space. We then show how the combined thresholds of a population of inhibitory neurons form a stable boundary in this space, and those of a population of excitatory neurons form an unstable boundary. Combining the two boundaries results in a rank-2 excitatory-inhibitory (EI) network with inhibition-stabilized dynamics at the intersection of the two boundaries. The computation of the resulting networks can be understood as the difference of two convex functions and is thereby capable of approximating arbitrary non-linear input-output mappings. We demonstrate several properties of these networks, including noise suppression and amplification, irregular activity and synaptic balance, as well as how they relate to rate network dynamics in the limit that the boundary becomes soft. Finally, while our work focuses on small networks (5-50 neurons), we discuss potential avenues for scaling up to much larger networks. Overall, our work proposes a new perspective on spiking networks that may serve as a starting point for a mechanistic understanding of biological spike-based computation.

Distinctive properties of biological neural networks and recent advances in bottom-up approaches toward a better biologically plausible neural network

Although it may appear infeasible and impractical, building artificial intelligence (AI) using a bottom-up approach based on the understanding of neuroscience is straightforward. The lack of a generalized governing principle for biological neural networks (BNNs) forces us to address this problem by converting piecemeal information on the diverse features of neurons, synapses, and neural circuits into AI. In this review, we described recent attempts to build a biologically plausible neural network by following neuroscientifically similar strategies of neural network optimization or by implanting the outcome of the optimization, such as the properties of single computational units and the characteristics of the network architecture. In addition, we proposed a formalism of the relationship between the set of objectives that neural networks attempt to achieve, and neural network classes categorized by how closely their architectural features resemble those of BNN. This formalism is expected to define the potential roles of top-down and bottom-up approaches for building a biologically plausible neural network and offer a map helping the navigation of the gap between neuroscience and AI engineering.

Clinical and neuroimaging markers of neurodegeneration in first-degree relatives of patients with REM sleep behavior disorder with and without isolated rapid eye movement sleep without atonia: A case-control clinical and dopamine PET study

OBJECTIVES

The current study aimed to examine the neurodegenerative implication of isolated REM sleep without atonia (RSWA) among first-degree relatives of patients with REM sleep behaviour disorder (RBD).

METHODS

This cross-sectional case-control study recruited three groups of subjects: First-degree relatives of RBD patients with isolated RSWA (n = 17), first-degree relatives of RBD patients without isolated RSWA (n = 18), and normal controls who did not have any RWSA and family history of RBD (n = 15). Prodromal Parkinson's Disease likelihood ratio by the updated MDS Research Criteria and striatal dopaminergic transmission function of the subjects as assessed by triple-tracer (18F-DOPA, 11C-Raclopride, and 18F-FDG) PET/CT scan were used as proxy markers of neurodegeneration.

RESULTS

In contrary to our hypothesis, the three groups did not differ in their pre- or post-striatal dopaminergic transmission function, and their Prodromal Parkinson's Disease likelihood ratio. However, they differed significantly in their frequency of a having first-degree relatives with Parkinson's disease or dementia of Lewy body (first-degree relativess with RSWA vs first degree relatives without RSWA vs normal controls = 58.8% vs 22.2% vs 0%, p = 0.001).

CONCLUSION

FDRs of RBD patients with isolated RSWA did not have increased neurodegenerative markers compared to FDRs of RBD patients without isolated RSWA and normal control, despite an paradoxical increase in frequency of Parkinson's disease or dementia of Lewy body among their family compared to FDRs of RBD patients without isolated RSWA. Further longitudinal follow-up study will be needed to ascertain their long-term prognosis.

Computational methods to study information processing in neural circuits

The brain is an information processing machine and thus naturally lends itself to be studied using computational tools based on the principles of information theory. For this reason, computational methods based on or inspired by information theory have been a cornerstone of practical and conceptual progress in neuroscience. In this Review, we address how concepts and computational tools related to information theory are spurring the development of principled theories of information processing in neural circuits and the development of influential mathematical methods for the analyses of neural population recordings. We review how these computational approaches reveal mechanisms of essential functions performed by neural circuits. These functions include efficiently encoding sensory information and facilitating the transmission of information to downstream brain areas to inform and guide behavior. Finally, we discuss how further progress and insights can be achieved, in particular by studying how competing requirements of neural encoding and readout may be optimally traded off to optimize neural information processing.

Disrupted visual input unveils the computational details of artificial neural networks for face perception

Background

Convolutional Neural Network (DCNN), with its great performance, has attracted attention of researchers from many disciplines. The studies of the DCNN and that of biological neural systems have inspired each other reciprocally. The brain-inspired neural networks not only achieve great performance but also serve as a computational model of biological neural systems.

Methods

Here in this study, we trained and tested several typical DCNNs (AlexNet, VGG11, VGG13, VGG16, DenseNet, MobileNet, and EfficientNet) with a face ethnicity categorization task for experiment 1, and an emotion categorization task for experiment 2. We measured the performance of DCNNs by testing them with original and lossy visual inputs (various kinds of image occlusion) and compared their performance with human participants. Moreover, the class activation map (CAM) method allowed us to visualize the foci of the "attention" of these DCNNs.

Results

The results suggested that the VGG13 performed the best: Its performance closely resembled human participants in terms of psychophysics measurements, it utilized similar areas of visual inputs as humans, and it had the most consistent performance with inputs having various kinds of impairments.

Discussion

In general, we examined the processing mechanism of DCNNs using a new paradigm and found that VGG13 might be the most human-like DCNN in this task. This study also highlighted a possible paradigm to study and develop DCNNs using human perception as a benchmark.

Altered Functional Connectivity of Basal Ganglia in Mild Cognitive Impairment and Alzheimer's Disease

(1) Background: Alzheimer's disease (AD), an age-progressive neurodegenerative disease that affects cognitive function, causes changes in the functional connectivity of the default-mode network (DMN). However, the question of whether AD-related changes occur in the functional connectivity of the basal ganglia has rarely been specifically analyzed. This study aimed to measure the changes in basal ganglia functional connectivity among patients with AD and mild cognitive impairment (MCI) in their resting state using the functional connectivity density (FCD) value, the functional connectivity (FC) intensity, and the graph theory index, and to confirm their influence on clinical manifestations. (2) Methods: Resting-state functional MRI (rs-fMRI) and neuropsychological data from 48 participants in the Alzheimer's Disease Neuroimaging Initiative (ADNI) were used for analyses. The 48 ADNI participants comprised 16 patients with AD, 16 patients with MCI, and 16 normal controls (NCs). The functional connectivity of basal ganglia was evaluated by FCDs, FC strength, and graph theory index. We compared voxel-based FCD values between groups to show specific regions with significant variation and significant connectivity from ROI conduction to ROI analysis. Pearson's correlation analyses between functional connectivity and several simultaneous clinical variables were also conducted. Additionally, receiver operating characteristic (ROC) analyses associated with classification were conducted for both FCD values and graph theory indices. (3) Results: The level of FCD in patients with cognitive impairment showed obvious abnormalities (including short-range and long-range FCD). In addition to DMN-related regions, aberrant functional connectivity was also found to be present in the basal ganglia, especially in the caudate and amygdala. The FCD values of the basal ganglia (involving the caudate and amygdala) were closely related to scores from the Mini-Mental State Examination (MMSE) and the Functional Activities Questionnaire (FAQ); meanwhile, the graph theory indices (involving global efficiency and degree) of the basal ganglia (involving the caudate, amygdala, and putamen) were also found to be closely correlated with MMSE scores. In ROC analyses of both FCD and graph theory, the amygdala was of the utmost importance in the early-stage detection of MCI; additionally, the caudate nucleus was found to be crucial in the progression of cognitive decline and AD diagnosis. (4) Conclusions: It was systematically confirmed that there is a phenomenon of change in the functional connections in the basal ganglia during cognitive decline. The findings of this study could improve our understanding of AD and MCI pathology in the basal ganglia and make it possible to propose new targets for AD treatment in further studies.

Drifting neuronal representations: Bug or feature?

The brain displays a remarkable ability to sustain stable memories, allowing animals to execute precise behaviors or recall stimulus associations years after they were first learned. Yet, recent long-term recording experiments have revealed that single-neuron representations continuously change over time, contravening the classical assumption that learned features remain static. How do unstable neural codes support robust perception, memories, and actions? Here, we review recent experimental evidence for such representational drift across brain areas, as well as dissections of its functional characteristics and underlying mechanisms. We emphasize theoretical proposals for how drift need not only be a form of noise for which the brain must compensate. Rather, it can emerge from computationally beneficial mechanisms in hierarchical networks performing robust probabilistic computations.

The geometry of robustness in spiking neural networks

Neural systems are remarkably robust against various perturbations, a phenomenon that still requires a clear explanation. Here, we graphically illustrate how neural networks can become robust. We study spiking networks that generate low-dimensional representations, and we show that the neurons’ subthreshold voltages are confined to a convex region in a lower-dimensional voltage subspace, which we call a 'bounding box'. Any changes in network parameters (such as number of neurons, dimensionality of inputs, firing thresholds, synaptic weights, or transmission delays) can all be understood as deformations of this bounding box. Using these insights, we show that functionality is preserved as long as perturbations do not destroy the integrity of the bounding box. We suggest that the principles underlying robustness in these networks — low-dimensional representations, heterogeneity of tuning, and precise negative feedback — may be key to understanding the robustness of neural systems at the circuit level.

Neuron Circuit Failure and Pattern Learning in Electronic Spiking Neural Networks

Biological neural networks demonstrate remarkable resilience and the ability to compensate for neuron losses over time. Thus, the effects of neural/synaptic losses in the brain go mostly unnoticed until the loss becomes profound. This study analyses the capacity of electronic spiking networks to compensate for the sudden, random neuron failure (“death”) due to reliability degradation or other external factors such as exposure to ionizing radiation. Electronic spiking neural networks with memristive synapses are designed to learn spatio-temporal patterns representing 25 or 100-pixel characters. The change in the pattern learning ability of the neural networks is observed as the afferents (input layer neurons) in the network fail/die during network training. Spike-timing-dependent plasticity (STDP) learning behavior is implemented using shaped action potentials with a realistic, non-linear memristor model. This work focuses on three cases: (1) when only neurons participating in the pattern are affected, (2) when non-participating neurons (those that never present spatio-temporal patterns) are disabled, and (3) when random/non-selective neuron death occurs in the network (the most realistic scenario). Case 3 is further analyzed to compare what happens when neuron death occurs over time versus when multiple afferents fail simultaneously. Simulation results emphasize the importance of non-participating neurons during the learning process, concluding that non-participating afferents contribute to improving the learning ability and stability of the neural network. Instantaneous neuron death proves to be more detrimental for the network compared to when afferents fail over time. To a surprising degree, the electronic spiking neural networks can sometimes retain their pattern recognition capability even in the case of significant neuron death.

Supervised training of spiking neural networks for robust deployment on mixed-signal neuromorphic processors

Mixed-signal analog/digital circuits emulate spiking neurons and synapses with extremely high energy efficiency, an approach known as "neuromorphic engineering". However, analog circuits are sensitive to process-induced variation among transistors in a chip ("device mismatch"). For neuromorphic implementation of Spiking Neural Networks (SNNs), mismatch causes parameter variation between identically-configured neurons and synapses. Each chip exhibits a different distribution of neural parameters, causing deployed networks to respond differently between chips. Current solutions to mitigate mismatch based on per-chip calibration or on-chip learning entail increased design complexity, area and cost, making deployment of neuromorphic devices expensive and difficult. Here we present a supervised learning approach that produces SNNs with high robustness to mismatch and other common sources of noise. Our method trains SNNs to perform temporal classification tasks by mimicking a pre-trained dynamical system, using a local learning rule from non-linear control theory. We demonstrate our method on two tasks requiring temporal memory, and measure the robustness of our approach to several forms of noise and mismatch. We show that our approach is more robust than common alternatives for training SNNs. Our method provides robust deployment of pre-trained networks on mixed-signal neuromorphic hardware, without requiring per-device training or calibration.

Modularity and robustness of frontal cortical networks

Neural activity underlying short-term memory is maintained by interconnected networks of brain regions. It remains unknown how brain regions interact to maintain persistent activity while exhibiting robustness to corrupt information in parts of the network. We simultaneously measured activity in large neuronal populations across mouse frontal hemispheres to probe interactions between brain regions. Activity across hemispheres was coordinated to maintain coherent short-term memory. Across mice, we uncovered individual variability in the organization of frontal cortical networks. A modular organization was required for the robustness of persistent activity to perturbations: each hemisphere retained persistent activity during perturbations of the other hemisphere, thus preventing local perturbations from spreading. A dynamic gating mechanism allowed hemispheres to coordinate coherent information while gating out corrupt information. Our results show that robust short-term memory is mediated by redundant modular representations across brain regions. Redundant modular representations naturally emerge in neural network models that learned robust dynamics.

Cochlear neural degeneration disrupts hearing in background noise by increasing auditory cortex internal noise

Correlational evidence in humans suggests that selective difficulties hearing in noisy, social settings may reflect premature auditory nerve degeneration. Here, we induced primary cochlear neural degeneration (CND) in adult mice and found direct behavioral evidence for selective detection deficits in background noise. To identify central determinants for this perceptual disorder, we tracked daily changes in ensembles of layer 2/3 auditory cortex parvalbumin-expressing inhibitory neurons and excitatory pyramidal neurons with chronic two-photon calcium imaging. CND induced distinct forms of plasticity in cortical excitatory and inhibitory neurons that culminated in net hyperactivity, increased neural gain, and reduced adaptation to background noise. Ensemble activity measured while mice detected targets in noise could accurately decode whether individual behavioral trials were hits or misses. After CND, random surges of hypercorrelated cortical activity occurring just before target onset reliably predicted impending detection failures, revealing a source of internal cortical noise underlying perceptual difficulties in external noise.

Going in circles is the way forward: the role of recurrence in visual inference

Biological visual systems exhibit abundant recurrent connectivity. State-of-the-art neural network models for visual recognition, by contrast, rely heavily or exclusively on feedforward computation. Any finite-time recurrent neural network (RNN) can be unrolled along time to yield an equivalent feedforward neural network (FNN). This important insight suggests that computational neuroscientists may not need to engage recurrent computation, and that computer-vision engineers may be limiting themselves to a special case of FNN if they build recurrent models. Here we argue, to the contrary, that FNNs are a special case of RNNs and that computational neuroscientists and engineers should engage recurrence to understand how brains and machines can (1) achieve greater and more flexible computational depth (2) compress complex computations into limited hardware (3) integrate priors and priorities into visual inference through expectation and attention (4) exploit sequential dependencies in their data for better inference and prediction and (5) leverage the power of iterative computation.

Neural compensation in presymptomatic hAPP mouse models of Alzheimer's disease

Largely inspired from clinical concepts like brain reserve, cognitive reserve, and neural compensation, here we review data showing how neural circuits reorganize in presymptomatic and early symptomatic hAPP mice to maintain memory intact. By informing on molecular alterations and compensatory adaptations which take place in the brain before mice show cognitive impairments, these data can help to identify ultra-early disease markers that could be targeted in a therapeutic perspective aimed at preventing rather than treating cognitive deterioration.

Firing rate homeostasis counteracts changes in stability of recurrent neural networks caused by synapse loss in Alzheimer's disease

The impairment of cognitive function in Alzheimer's disease is clearly correlated to synapse loss. However, the mechanisms underlying this correlation are only poorly understood. Here, we investigate how the loss of excitatory synapses in sparsely connected random networks of spiking excitatory and inhibitory neurons alters their dynamical characteristics. Beyond the effects on the activity statistics, we find that the loss of excitatory synapses on excitatory neurons reduces the network's sensitivity to small perturbations. This decrease in sensitivity can be considered as an indication of a reduction of computational capacity. A full recovery of the network's dynamical characteristics and sensitivity can be achieved by firing rate homeostasis, here implemented by an up-scaling of the remaining excitatory-excitatory synapses. Mean-field analysis reveals that the stability of the linearised network dynamics is, in good approximation, uniquely determined by the firing rate, and thereby explains why firing rate homeostasis preserves not only the firing rate but also the network's sensitivity to small perturbations.

Learning to represent signals spike by spike

Networks based on coordinated spike coding can encode information with high efficiency in the spike trains of individual neurons. These networks exhibit single-neuron variability and tuning curves as typically observed in cortex, but paradoxically coincide with a precise, non-redundant spike-based population code. However, it has remained unclear whether the specific synaptic connectivities required in these networks can be learnt with local learning rules. Here, we show how to learn the required architecture. Using coding efficiency as an objective, we derive spike-timing-dependent learning rules for a recurrent neural network, and we provide exact solutions for the networks’ convergence to an optimal state. As a result, we deduce an entire network from its input distribution and a firing cost. After learning, basic biophysical quantities such as voltages, firing thresholds, excitation, inhibition, or spikes acquire precise functional interpretations.

Spiking neural networks can encode information with high efficiency in the spike trains of individual neurons if the synaptic weights between neurons are set to specific, optimal values. In this regime, the networks exhibit irregular spike trains, high trial-to-trial variability, and stimulus tuning, as typically observed in cortex. The strong variability on the level of single neurons paradoxically coincides with a precise, non-redundant, and spike-based population code. However, it has remained unclear whether the specific synaptic connectivities required in these spiking networks can be learnt with local learning rules. In this study, we show how the required architecture can be learnt. We derive local and biophysically plausible learning rules for recurrent neural networks from first principles. We show both mathematically and using numerical simulations that these learning rules drive the networks into the optimal state, and we show that the optimal state is governed by the statistics of the input signals. After learning, the voltages of individual neurons can be interpreted as measuring the instantaneous error of the code, given by the error between the desired output signal and the actual output signal.

Recurrent interactions in local cortical circuits

Most cortical synapses are local and excitatory. Local recurrent circuits could implement amplification, allowing pattern completion and other computations^1-4. Cortical circuits contain subnetworks that consist of neurons with similar receptive fields and increased connectivity relative to the network average^5,6. Cortical neurons that encode different types of information are spatially intermingled and distributed over large brain volumes^5-7, and this complexity has hindered attempts to probe the function of these subnetworks by perturbing them individually⁸. Here we use computational modelling, optical recordings and manipulations to probe the function of recurrent coupling in layer 2/3 of the mouse vibrissal somatosensory cortex during active tactile discrimination. A neural circuit model of layer 2/3 revealed that recurrent excitation enhances sensory signals by amplification, but only for subnetworks with increased connectivity. Model networks with high amplification were sensitive to damage: loss of a few members of the subnetwork degraded stimulus encoding. We tested this prediction by mapping neuronal selectivity⁷ and photoablating^9,10 neurons with specific selectivity. Ablation of a small proportion of layer 2/3 neurons (10-20, less than 5% of the total) representing touch markedly reduced responses in the spared touch representation, but not in other representations. Ablations most strongly affected neurons with stimulus responses that were similar to those of the ablated population, which is also consistent with network models. Recurrence among cortical neurons with similar selectivity therefore drives input-specific amplification during behaviour.

Monitor-Based Spiking Recurrent Network for the Representation of Complex Dynamic Patterns

Neural networks are powerful computation tools for mimicking the human brain to solve realistic problems. Since spiking neural networks are a type of brain-inspired network, called the novel spiking system, Monitor-based Spiking Recurrent network (MbSRN), is derived to learn and represent patterns in this paper. This network provides a computational framework for memorizing the targets using a simple dynamic model that maintains biological plasticity. Based on a recurrent reservoir, the MbSRN presents a mechanism called a ‘monitor’ to track the components of the state space in the training stage online and to self-sustain the complex dynamics in the testing stage. The network firing spikes are optimized to represent the target dynamics according to the accumulation of the membrane potentials of the units. Stability analysis of the monitor conducted by limiting the coefficient penalty in the loss function verifies that our network has good anti-interference performance under neuron loss and noise. The results of solving some realistic tasks show that the MbSRN not only achieves a high goodness-of-fit of the target patterns but also maintains good spiking efficiency and storage capacity.

Emerging links between cell competition and Alzheimer's disease

Alzheimer's disease (AD) causes a progressive loss of memory and other cognitive functions, which inexorably debilitates patients. There is still no cure for AD and effective treatments to delay or revert AD are urgently needed. On a molecular level, the excessive accumulation of amyloid-β (Aβ) peptides triggers a complex cascade of pathological events underlying neuronal death, whose details are not yet completely understood. Our laboratory recently discovered that cell competition may play a protective role against AD by eliminating less fit neurons from the brain of Aβ-transgenic flies. Loss of Aβ-damaged neurons through fitness comparison with healthy counterparts is beneficial for the organism, delaying cognitive decline and motor disability. In this Review, we introduce the molecular mechanisms of cell competition, including seminal works on the field and latest advances regarding genetic triggers and effectors of cell elimination. We then describe the biological relevance of competition in the nervous system and discuss how competitive interactions between neurons may arise and be exacerbated in the context of AD. Selection of neurons through fitness comparison is a promising, but still emerging, research field that may open new avenues for the treatment of neurological disorders.

Analyzing biological and artificial neural networks: challenges with opportunities for synergy?

Deep neural networks (DNNs) transform stimuli across multiple processing stages to produce representations that can be used to solve complex tasks, such as object recognition in images. However, a full understanding of how they achieve this remains elusive. The complexity of biological neural networks substantially exceeds the complexity of DNNs, making it even more challenging to understand the representations they learn. Thus, both machine learning and computational neuroscience are faced with a shared challenge: how can we analyze their representations in order to understand how they solve complex tasks? We review how data-analysis concepts and techniques developed by computational neuroscientists can be useful for analyzing representations in DNNs, and in turn, how recently developed techniques for analysis of DNNs can be useful for understanding representations in biological neural networks. We explore opportunities for synergy between the two fields, such as the use of DNNs as in silico model systems for neuroscience, and how this synergy can lead to new hypotheses about the operating principles of biological neural networks.

Fast-spiking GABA circuit dynamics in the auditory cortex predict recovery of sensory processing following peripheral nerve damage

Cortical neurons remap their receptive fields and rescale sensitivity to spared peripheral inputs following sensory nerve damage. To address how these plasticity processes are coordinated over the course of functional recovery, we tracked receptive field reorganization, spontaneous activity, and response gain from individual principal neurons in the adult mouse auditory cortex over a 50-day period surrounding either moderate or massive auditory nerve damage. We related the day-by-day recovery of sound processing to dynamic changes in the strength of intracortical inhibition from parvalbumin-expressing (PV) inhibitory neurons. Whereas the status of brainstem-evoked potentials did not predict the recovery of sensory responses to surviving nerve fibers, homeostatic adjustments in PV-mediated inhibition during the first days following injury could predict the eventual recovery of cortical sound processing weeks later. These findings underscore the potential importance of self-regulated inhibitory dynamics for the restoration of sensory processing in excitatory neurons following peripheral nerve injuries.

DOI:http://dx.doi.org/10.7554/eLife.21452.001