A 3D ray traced biological neural network learning model.

Training large neural networks on big datasets requires significant computational resources and time. Transfer learning reduces training time by pre-training a base model on one dataset and transferring the knowledge to a new model for another dataset. However, current choices of transfer learning algorithms are limited because the transferred models always have to adhere to the dimensions of the base model and can not easily modify the neural architecture to solve other datasets. On the other hand, biological neural networks (BNNs) are adept at rearranging themselves to tackle completely different problems using transfer learning. Taking advantage of BNNs, we design a dynamic neural network that is transferable to any other network architecture and can accommodate many datasets. Our approach uses raytracing to connect neurons in a three-dimensional space, allowing the network to grow into any shape or size. In the Alcala dataset, our transfer learning algorithm trains the fastest across changing environments and input sizes. In addition, we show that our algorithm also outperformance the state of the art in EEG dataset. In the future, this network may be considered for implementation on real biological neural networks to decrease power consumption.

Convolutional neural network models applied to neuronal responses in macaque V1 reveal limited nonlinear processing.

Computational models of the primary visual cortex (V1) have suggested that V1 neurons behave like Gabor filters followed by simple nonlinearities. However, recent work employing convolutional neural network (CNN) models has suggested that V1 relies on far more nonlinear computations than previously thought. Specifically, unit responses in an intermediate layer of VGG-19 were found to best predict macaque V1 responses to thousands of natural and synthetic images. Here, we evaluated the hypothesis that the poor performance of lower layer units in VGG-19 might be attributable to their small receptive field size rather than to their lack of complexity per se. We compared VGG-19 with AlexNet, which has much larger receptive fields in its lower layers. Whereas the best-performing layer of VGG-19 occurred after seven nonlinear steps, the first convolutional layer of AlexNet best predicted V1 responses. Although the predictive accuracy of VGG-19 was somewhat better than that of standard AlexNet, we found that a modified version of AlexNet could match the performance of VGG-19 after only a few nonlinear computations. Control analyses revealed that decreasing the size of the input images caused the best-performing layer of VGG-19 to shift to a lower layer, consistent with the hypothesis that the relationship between image size and receptive field size can strongly affect model performance. We conducted additional analyses using a Gabor pyramid model to test for nonlinear contributions of normalization and contrast saturation. Overall, our findings suggest that the feedforward responses of V1 neurons can be well explained by assuming only a few nonlinear processing stages.

Glutamatergic supramammillary nucleus neurons respond to threatening stressors and promote active coping.

Threat-response neural circuits are conserved across species and play roles in normal behavior and psychiatric diseases. Maladaptive changes in these neural circuits contribute to stress, mood, and anxiety disorders. Active coping in response to stressors is a psychosocial factor associated with resilience against stress-induced mood and anxiety disorders. The neural circuitry underlying active coping is poorly understood, but the functioning of these circuits could be key for overcoming anxiety and related disorders. The supramammillary nucleus (SuM) has been suggested to be engaged by threat. SuM has many projections and a poorly understood diversity of neural populations. In studies using mice, we identified a unique population of glutamatergic SuM neurons (SuMVGLUT2+::POA) based on projection to the preoptic area of the hypothalamus (POA) and found SuMVGLUT2+::POA neurons have extensive arborizations. SuMVGLUT2+::POA neurons project to brain areas that mediate features of the stress and threat responses including the paraventricular nucleus thalamus (PVT), periaqueductal gray (PAG), and habenula (Hb). Thus, SuMVGLUT2+::POA neurons are positioned as a hub, connecting to areas implicated in regulating stress responses. Here we report SuMVGLUT2+::POA neurons are recruited by diverse threatening stressors, and recruitment correlated with active coping behaviors. We found that selective photoactivation of the SuMVGLUT2+::POA population drove aversion but not anxiety like behaviors. Activation of SuMVGLUT2+::POA neurons in the absence of acute stressors evoked active coping like behaviors and drove instrumental behavior. Also, activation of SuMVGLUT2+::POA neurons was sufficient to convert passive coping strategies to active behaviors during acute stress. In contrast, we found activation of GABAergic (VGAT+) SuM neurons (SuMVGAT+) neurons did not alter drive aversion or active coping, but termination of photostimulation was followed by increased mobility in the forced swim test. These findings establish a new node in stress response circuitry that has projections to many brain areas and evokes flexible active coping behaviors.

Adult-born granule cells modulate CA2 network activity during retrieval of developmental memories of the mother.

Adult-born granule cells (abGCs) project to the CA2 region of the hippocampus, but it remains unknown how this circuit affects behavioral function. Here, we show that abGC input to the CA2 of adult mice is involved in the retrieval of remote developmental memories of the mother. Ablation of abGCs impaired the ability to discriminate between a caregiving mother and a novel mother, and this ability returned after abGCs were regenerated. Chemogenetic inhibition of projections from abGCs to the CA2 also temporarily prevented the retrieval of remote mother memories. These findings were observed when abGCs were inhibited at 4-6 weeks old, but not when they were inhibited at 10-12 weeks old. We also found that abGCs are necessary for differentiating features of CA2 network activity, including theta-gamma coupling and sharp wave ripples, in response to novel versus familiar social stimuli. Taken together, these findings suggest that abGCs are necessary for neuronal oscillations associated with discriminating between social stimuli, thus enabling retrieval of remote developmental memories of the mother by their adult offspring.

Keep calm and make neurons: The effects of glucocorticoids on human cortical neurogenesis.

Maternal well-being is important for the development of the fetus, with a key influence on its nervous system. In this issue of Neuron, Krontira et al.1 implicate glucocorticoids, the stress hormones, in the regulation of neural stem cell identity and proliferation, with long-lasting consequences on brain architecture and educational attainment.

Subthalamic nucleus synchronization between beta band local field potential and single-unit activity in Parkinson's disease.

Local field potential (LFP) oscillations in the beta band (13-30 Hz) in the subthalamic nucleus (STN) of Parkinson's disease patients have been implicated in disease severity and treatment response. The relationship between single-neuron activity in the STN and regional beta power changes remains unclear. We used spike-triggered average (STA) to assess beta synchronization in STN. Beta power and STA magnitude at the beta frequency range were compared in three conditions: STN versus other subcortical structures, dorsal versus ventral STN, and high versus low beta power STN recordings. Magnitude of STA-LFP was greater within the STN compared to extra-STN structures along the trajectory path, despite no difference in percentage of the total power. Within the STN, there was a higher percent beta power in dorsal compared to ventral STN but no difference in STA-LFP magnitude. Further refining the comparison to high versus low beta peak power recordings inside the STN to evaluate if single-unit activity synchronized more strongly with beta band activity in areas of high beta power resulted in a significantly higher STA magnitude for areas of high beta power. Overall, these results suggest that STN single units strongly synchronize to beta activity, particularly units in areas of high beta power.

Spiking Neural Membrane Systems with Adaptive Synaptic Time Delay.

Spiking neural membrane systems (or spiking neural P systems, SNP systems) are a new type of computation model which have attracted the attention of plentiful scholars for parallelism, time encoding, interpretability and extensibility. The original SNP systems only consider the time delay caused by the execution of rules within neurons, but not caused by the transmission of spikes via synapses between neurons and its adaptive adjustment. In view of the importance of time delay for SNP systems, which are a time encoding computation model, this study proposes SNP systems with adaptive synaptic time delay (ADSNP systems) based on the dynamic regulation mechanism of synaptic transmission delay in neural systems. In ADSNP systems, besides neurons, astrocytes that can generate adenosine triphosphate (ATP) are introduced. After receiving spikes, astrocytes convert spikes into ATP and send ATP to the synapses controlled by them to change the synaptic time delays. The Turing universality of ADSNP systems in number generating and accepting modes is proved. In addition, a small universal ADSNP system using 93 neurons and astrocytes is given. The superiority of the ADSNP system is demonstrated by comparison with the six variants. Finally, an ADSNP system is constructed for credit card fraud detection, which verifies the feasibility of the ADSNP system for solving real-world problems. By considering the adaptive synaptic delay, ADSNP systems better restore the process of information transmission in biological neural networks, and enhance the adaptability of SNP systems, making the control of time more accurate.

Representational drift as a result of implicit regularization.

Recent studies show that, even in constant environments, the tuning of single neurons changes over time in a variety of brain regions. This representational drift has been suggested to be a consequence of continuous learning under noise, but its properties are still not fully understood. To investigate the underlying mechanism, we trained an artificial network on a simplified navigational task. The network quickly reached a state of high performance, and many units exhibited spatial tuning. We then continued training the network and noticed that the activity became sparser with time. Initial learning was orders of magnitude faster than ensuing sparsification. This sparsification is consistent with recent results in machine learning, in which networks slowly move within their solution space until they reach a flat area of the loss function. We analyzed four datasets from different labs, all demonstrating that CA1 neurons become sparser and more spatially informative with exposure to the same environment. We conclude that learning is divided into three overlapping phases: (i) Fast familiarity with the environment; (ii) slow implicit regularization; and (iii) a steady state of null drift. The variability in drift dynamics opens the possibility of inferring learning algorithms from observations of drift statistics.

Drift of neural ensembles driven by slow fluctuations of intrinsic excitability.

Representational drift refers to the dynamic nature of neural representations in the brain despite the behavior being seemingly stable. Although drift has been observed in many different brain regions, the mechanisms underlying it are not known. Since intrinsic neural excitability is suggested to play a key role in regulating memory allocation, fluctuations of excitability could bias the reactivation of previously stored memory ensembles and therefore act as a motor for drift. Here, we propose a rate-based plastic recurrent neural network with slow fluctuations of intrinsic excitability. We first show that subsequent reactivations of a neural ensemble can lead to drift of this ensemble. The model predicts that drift is induced by co-activation of previously active neurons along with neurons with high excitability which leads to remodeling of the recurrent weights. Consistent with previous experimental works, the drifting ensemble is informative about its temporal history. Crucially, we show that the gradual nature of the drift is necessary for decoding temporal information from the activity of the ensemble. Finally, we show that the memory is preserved and can be decoded by an output neuron having plastic synapses with the main region.

Location analysis of presynaptically active and silent synapses in single-cultured hippocampal neurons.

A morphologically present but non-functioning synapse is termed a silent synapse. Silent synapses are categorized into "postsynaptically silent synapses," where AMPA receptors are either absent or non-functional, and "presynaptically silent synapses," where neurotransmitters cannot be released from nerve terminals. The presence of presynaptically silent synapses remains enigmatic, and their physiological significance is highly intriguing. In this study, we examined the distribution and developmental changes of presynaptically active and silent synapses in individual neurons. Our findings show a gradual increase in the number of excitatory synapses, along with a corresponding decrease in the percentage of presynaptically silent synapses during neuronal development. To pinpoint the distribution of presynaptically active and silent synapses, i.e., their positional information, we employed Sholl analysis. Our results indicate that the distribution of presynaptically silent synapses within a single neuron does not exhibit a distinct pattern during synapse development in different distance from the cell body. However, irrespective of neuronal development, the proportion of presynaptically silent synapses tends to rise as the projection site moves farther from the cell body, suggesting that synapses near the cell body may exhibit higher synaptic transmission efficiency. This study represents the first observation of changes in the distribution of presynaptically active and silent synapses within a single neuron.

Spontaneous persistent activity and inactivity in vivo reveals differential cortico-entorhinal functional connectivity.

Understanding the functional connectivity between brain regions and its emergent dynamics is a central challenge. Here we present a theory-experiment hybrid approach involving iteration between a minimal computational model and in vivo electrophysiological measurements. Our model not only predicted spontaneous persistent activity (SPA) during Up-Down-State oscillations, but also inactivity (SPI), which has never been reported. These were confirmed in vivo in the membrane potential of neurons, especially from layer 3 of the medial and lateral entorhinal cortices. The data was then used to constrain two free parameters, yielding a unique, experimentally determined model for each neuron. Analytic and computational analysis of the model generated a dozen quantitative predictions about network dynamics, which were all confirmed in vivo to high accuracy. Our technique predicted functional connectivity; e. g. the recurrent excitation is stronger in the medial than lateral entorhinal cortex. This too was confirmed with connectomics data. This technique uncovers how differential cortico-entorhinal dialogue generates SPA and SPI, which could form an energetically efficient working-memory substrate and influence the consolidation of memories during sleep. More broadly, our procedure can reveal the functional connectivity of large networks and a theory of their emergent dynamics.

Distributed and dynamical communication: a mechanism for flexible cortico-cortical interactions and its functional roles in visual attention.

Perceptual and cognitive processing relies on flexible communication among cortical areas; however, the underlying neural mechanism remains unclear. Here we report a mechanism based on the realistic spatiotemporal dynamics of propagating wave patterns in neural population activity. Using a biophysically plausible, multiarea spiking neural circuit model, we demonstrate that these wave patterns, characterized by their rich and complex dynamics, can account for a wide variety of empirically observed neural processes. The coordinated interactions of these wave patterns give rise to distributed and dynamic communication (DDC) that enables flexible and rapid routing of neural activity across cortical areas. We elucidate how DDC unifies the previously proposed oscillation synchronization-based and subspace-based views of interareal communication, offering experimentally testable predictions that we validate through the analysis of Allen Institute Neuropixels data. Furthermore, we demonstrate that DDC can be effectively modulated during attention tasks through the interplay of neuromodulators and cortical feedback loops. This modulation process explains many neural effects of attention, underscoring the fundamental functional role of DDC in cognition.

The impact of spike timing precision and spike emission reliability on decoding accuracy.

Precisely timed and reliably emitted spikes are hypothesized to serve multiple functions, including improving the accuracy and reproducibility of encoding stimuli, memories, or behaviours across trials. When these spikes occur as a repeating sequence, they can be used to encode and decode a potential time series. Here, we show both analytically and in simulations that the error incurred in approximating a time series with precisely timed and reliably emitted spikes decreases linearly with the number of neurons or spikes used in the decoding. This was verified numerically with synthetically generated patterns of spikes. Further, we found that if spikes were imprecise in their timing, or unreliable in their emission, the error incurred in decoding with these spikes would be sub-linear. However, if the spike precision or spike reliability increased with network size, the error incurred in decoding a time-series with sequences of spikes would maintain a linear decrease with network size. The spike precision had to increase linearly with network size, while the probability of spike failure had to decrease with the square-root of the network size. Finally, we identified a candidate circuit to test this scaling relationship: the repeating sequences of spikes with sub-millisecond precision in area HVC (proper name) of the zebra finch. This scaling relationship can be tested using both neural data and song-spectrogram-based recordings while taking advantage of the natural fluctuation in HVC network size due to neurogenesis.

Non-trivial relationship between behavioral avalanches and internal neuronal dynamics in a recurrent neural network.

Neuronal activity gives rise to behavior, and behavior influences neuronal dynamics, in a closed-loop control system. Is it possible then, to find a relationship between the statistical properties of behavior and neuronal dynamics? Measurements of neuronal activity and behavior have suggested a direct relationship between scale-free neuronal and behavioral dynamics. Yet, these studies captured only local dynamics in brain sub-networks. Here, we investigate the relationship between internal dynamics and output statistics in a mathematical model system where we have access to the dynamics of all network units. We train a recurrent neural network (RNN), initialized in a high-dimensional chaotic state, to sustain behavioral states for durations following a power-law distribution as observed experimentally. Changes in network connectivity due to training affect the internal dynamics of neuronal firings, leading to neuronal avalanche size distributions approximating power-laws over some ranges. Yet, randomizing the changes in network connectivity can leave these power-law features largely unaltered. Specifically, whereas neuronal avalanche duration distributions show some variations between RNNs with trained and randomized decoders, neuronal avalanche size distributions are invariant, in the total population and in output-correlated sub-populations. This is true independent of whether the randomized decoders preserve power-law distributed behavioral dynamics. This demonstrates that a one-to-one correspondence between the considered statistical features of behavior and neuronal dynamics cannot be established and their relationship is non-trivial. Our findings also indicate that statistical properties of the intrinsic dynamics may be preserved, even as the internal state responsible for generating the desired output dynamics is perturbed.

Neural correlates of tonal loudness, intensity discrimination, and duration discrimination.

A long-standing quest in audition concerns understanding relations between behavioral measures and neural representations of changes in sound intensity. Here, we examined relations between aspects of intensity perception and central neural responses within the inferior colliculus of unanesthetized rabbits (by averaging the population's spike count/level functions). We found parallels between the population's neural output and: (1) how loudness grows with intensity; (2) how loudness grows with duration; (3) how discrimination of intensity improves with increasing sound level; (4) findings that intensity discrimination does not depend on duration; and (5) findings that duration discrimination is a constant fraction of base duration.

A vasopressin circuit that modulates mouse social investigation and anxiety-like behavior in a sex-specific manner.

One of the largest sex differences in brain neurochemistry is the expression of the neuropeptide arginine vasopressin (AVP) within the vertebrate brain, with males having more AVP cells in the bed nucleus of the stria terminalis (BNST) than females. Despite the long-standing implication of AVP in social and anxiety-like behaviors, the circuitry underlying AVP's control of these behaviors is still not well defined. Using optogenetic approaches, we show that inhibiting AVP BNST cells reduces social investigation in males, but not in females, whereas stimulating these cells increases social investigation in both sexes, but more so in males. These cells may facilitate male social investigation through their projections to the lateral septum (LS), an area with the highest density of sexually differentiated AVP innervation in the brain, as optogenetic stimulation of BNST AVP → LS increased social investigation and anxiety-like behavior in males but not in females; the same stimulation also caused a biphasic response of LS cells ex vivo. Blocking the vasopressin 1a receptor (V1aR) in the LS eliminated all these responses. Together, these findings establish a sexually differentiated role for BNST AVP cells in the control of social investigation and anxiety-like behavior, likely mediated by their projections to the LS.

Neurometric amplitude modulation detection in the inferior colliculus of Young and Aged rats.

Amplitude modulation is an important acoustic cue for sound discrimination, and humans and animals are able to detect small modulation depths behaviorally. In the inferior colliculus (IC), both firing rate and phase-locking may be used to detect amplitude modulation. How neural representations that detect modulation change with age are poorly understood, including the extent to which age-related changes may be attributed to the inherited properties of ascending inputs to IC neurons. Here, simultaneous measures of local field potentials (LFPs) and single-unit responses were made from the inferior colliculus of Young and Aged rats using both noise and tone carriers in response to sinusoidally amplitude-modulated sounds of varying depths. We found that Young units had higher firing rates than Aged for noise carriers, whereas Aged units had higher phase-locking (vector strength), especially for tone carriers. Sustained LFPs were larger in Young animals for modulation frequencies 8-16 Hz and comparable at higher modulation frequencies. Onset LFP amplitudes were much larger in Young animals and were correlated with the evoked firing rates, while LFP onset latencies were shorter in Aged animals. Unit neurometric thresholds by synchrony or firing rate measures did not differ significantly across age and were comparable to behavioral thresholds in previous studies whereas LFP thresholds were lower than behavior.

Spike-based dynamic computing with asynchronous sensing-computing neuromorphic chip.

By mimicking the neurons and synapses of the human brain and employing spiking neural networks on neuromorphic chips, neuromorphic computing offers a promising energy-efficient machine intelligence. How to borrow high-level brain dynamic mechanisms to help neuromorphic computing achieve energy advantages is a fundamental issue. This work presents an application-oriented algorithm-software-hardware co-designed neuromorphic system for this issue. First, we design and fabricate an asynchronous chip called "Speck", a sensing-computing neuromorphic system on chip. With the low processor resting power of 0.42mW, Speck can satisfy the hardware requirements of dynamic computing: no-input consumes no energy. Second, we uncover the "dynamic imbalance" in spiking neural networks and develop an attention-based framework for achieving the algorithmic requirements of dynamic computing: varied inputs consume energy with large variance. Together, we demonstrate a neuromorphic system with real-time power as low as 0.70mW. This work exhibits the promising potentials of neuromorphic computing with its asynchronous event-driven, sparse, and dynamic nature.

Neuronal activation sequences in lateral prefrontal cortex encode visuospatial working memory during virtual navigation.

Working memory (WM) is the ability to maintain and manipulate information 'in mind'. The neural codes underlying WM have been a matter of debate. We simultaneously recorded the activity of hundreds of neurons in the lateral prefrontal cortex of male macaque monkeys during a visuospatial WM task that required navigation in a virtual 3D environment. Here, we demonstrate distinct neuronal activation sequences (NASs) that encode remembered target locations in the virtual environment. This NAS code outperformed the persistent firing code for remembered locations during the virtual reality task, but not during a classical WM task using stationary stimuli and constraining eye movements. Finally, blocking NMDA receptors using low doses of ketamine deteriorated the NAS code and behavioral performance selectively during the WM task. These results reveal the versatility and adaptability of neural codes supporting working memory function in the primate lateral prefrontal cortex.

Thalamic contributions to the state and contents of consciousness.

Consciousness can be conceptualized as varying along at least two dimensions: the global state of consciousness and the content of conscious experience. Here, we highlight the cellular and systems-level contributions of the thalamus to conscious state and then argue for thalamic contributions to conscious content, including the integrated, segregated, and continuous nature of our experience. We underscore vital, yet distinct roles for core- and matrix-type thalamic neurons. Through reciprocal interactions with deep-layer cortical neurons, matrix neurons support wakefulness and determine perceptual thresholds, whereas the cortical interactions of core neurons maintain content and enable perceptual constancy. We further propose that conscious integration, segregation, and continuity depend on the convergent nature of corticothalamic projections enabling dimensionality reduction, a thalamic reticular nucleus-mediated divisive normalization-like process, and sustained coherent activity in thalamocortical loops, respectively. Overall, we conclude that the thalamus plays a central topological role in brain structures controlling conscious experience.