BrainPy, a flexible, integrative, efficient, and extensible framework for general-purpose brain dynamics programming

A Heterogeneous Attractor Model for Neural Dynamical Mechanism of Movement Preparation

Preparatory activity is crucial for voluntary motor control, reducing reaction time and enhancing precision. To understand the neurodynamic mechanisms behind this, we construct a dynamical model within the motor cortex, which comprises coupled heterogeneous attractors to simulate delayed reaching tasks. This model replicates the neural activity patterns observed in the macaque motor cortex, within distinct attractor spaces for preparatory and executive activities. It can capture the transition from preparation to execution through shifts in an orthogonal subspace combined with a thresholding mechanism. Results show that the preparation duration modulates behavioral accuracy, with optimal preparation intervals enhancing performance. External inputs primarily shape the preparatory activity, while synaptic connections dominate execution. Our analysis of the network's multi-stable dynamics reveals that external inputs reshape the stable points of the heterogeneous attractor modules both before and after preparation, while synaptic strength affects dynamical stability and input sensitivity, allowing rapid and precise actions. Additionally, sensitivity to external perturbations decreases as preparatory time increases, emphasizing the importance of external inputs during preparation. Overall, this study provides insights into the neurodynamic mechanisms underlying the transition from motor preparation to execution and underscores the significance of preparatory activity for accurate motor control.

A systems model of alternating theta sweeps via firing rate adaptation

Place and grid cells provide a neural system for self-location and tend to fire in sequences within each cycle of the hippocampal theta rhythm when rodents run on a linear track. These sequences correspond to the decoded location of the animal sweeping forward from its current location ("theta sweeps"). However, recent findings in open-field environments show alternating left-right theta sweeps and propose a circuit for their generation. Here, we present a computational model of this circuit, comprising theta-modulated head-direction cells, conjunctive grid × direction cells, and pure grid cells, based on continuous attractor dynamics, firing rate adaptation, and modulation by the medial-septal theta rhythm. Due to firing rate adaptation, the head-direction ring attractor exhibits left-right sweeps coding for internal direction, providing an input to the grid cell attractor network shifted along the internal direction, via an intermediate layer of conjunctive grid × direction cells, producing left-right sweeps of position by grid cells. Our model explains the empirical findings, including the alignment of internal position and direction sweeps and the dependence of sweep length on grid spacing. It makes predictions for theta-modulated head-direction cells, including relationships between theta phase precession during turning, theta skipping, anticipatory firing, and directional tuning width, several of which we verify in experimental data from anteroventral thalamus. The model also predicts relationships between position and direction sweeps, running speed, and dorsal-ventral location within the entorhinal cortex. Overall, a simple intrinsic mechanism explains the complex theta dynamics of an internal direction signal within the hippocampal formation, with testable predictions.

Biophysical modeling and experimental analysis of the dynamics of C. elegans body-wall muscle cells

This study combines experimental techniques and mathematical modeling to investigate the dynamics of C. elegans body-wall muscle cells. Specifically, by conducting voltage clamp and mutant experiments, we identify key ion channels, particularly the L-type voltage-gated calcium channel (EGL-19) and potassium channels (SHK-1, SLO-2), which are crucial for generating action potentials. We develop Hodgkin-Huxley-based models for these channels and integrate them to capture the cells' electrical activity. To ensure the model accurately reflects cellular responses under depolarizing currents, we develop a parallel simulation-based inference method for determining the model's free parameters. This method performs rapid parallel sampling across high-dimensional parameter spaces, fitting the model to the responses of muscle cells to specific stimuli and yielding accurate parameter estimates. We validate our model by comparing its predictions against cellular responses to various current stimuli in experiments and show that our approach effectively determines suitable parameters for accurately modeling the dynamics in mutant cases. Additionally, we discover an optimal response frequency in body-wall muscle cells, which corresponds to a burst firing mode rather than regular firing mode. Our work provides the first experimentally constrained and biophysically detailed muscle cell model of C. elegans, and our analytical framework combined with robust and efficient parametric estimation method can be extended to model construction in other species.

Spindle oscillations emerge at the critical state of electrically coupled networks in the thalamic reticular nucleus

Spindle oscillation is a waxing-and-waning neural oscillation observed in the brain, initiated at the thalamic reticular nucleus (TRN) and typically occurring at 7-15 Hz. Experiments have shown that in the adult brain, electrical synapses, rather than chemical synapses, dominate between TRN neurons, suggesting that the traditional view of spindle generation via chemical synapses may need reconsideration. Based on known experimental data, we develop a computational model of the TRN network, where heterogeneous neurons are connected by electrical synapses. The model shows that the interplay between synchronizing electrical synapses and desynchronizing heterogeneity leads to multiple synchronized clusters with slightly different oscillation frequencies whose summed-up activity produces spindle oscillation as seen in local field potentials. Our results suggest that during spindle oscillation, the network operates at the critical state, which is known for facilitating efficient information processing. This study provides insights into the underlying mechanism of spindle oscillation and its functional significance.

Spike Neural Network of Motor Cortex Model for Arm Reaching Control

Motor cortex modeling is crucial for understanding movement planning and execution. While interconnected recurrent neural networks have successfully described the dynamics of neural population activity, most existing methods utilize continuous signal-based neural networks, which do not reflect the biological spike neural signal. To address this limitation, we propose a recurrent spike neural network to simulate motor cortical activity during an arm-reaching task. Specifically, our model is built upon integrate-and-fire spiking neurons with conductance-based synapses. We carefully designed the interconnections of neurons with two different firing time scales - "fast" and "slow" neurons. Experimental results demonstrate the effectiveness of our method, with the model's neuronal activity in good agreement with monkey's motor cortex data at both single-cell and population levels. Quantitative analysis reveals a correlation coefficient 0.89 between the model's and real data. These results suggest the possibility of multiple timescales in motor cortical control.

Unveiling serotonergic dysfunction of obsessive-compulsive disorder on prefrontal network dynamics: a computational perspective

Serotonin (5-HT) regulates working memory within the prefrontal cortex network, which is crucial for understanding obsessive-compulsive disorder. However, the mechanisms how network dynamics and serotonin interact in obsessive-compulsive disorder remain elusive. Here, we incorporate 5-HT receptors (5-HT1A, 5-HT2A) and dopamine receptors into a multistable prefrontal cortex network model, replicating the experimentally observed inverted U-curve phenomenon. We show how the two 5-HT receptors antagonize neuronal activity and modulate network multistability. Reduced binding of 5-HT1A receptors increases global firing, while reduced binding of 5-HT2A receptors deepens attractors. The obtained results suggest reward-dependent synaptic plasticity mechanisms may attenuate 5-HT related network impairments. Integrating serotonin-mediated dopamine release into circuit, we observe that decreased serotonin concentration triggers the network into a deep attractor state, expanding the domain of attraction of stable nodes with high firing rate, potentially causing aberrant reverse learning. This suggests a hypothesis wherein elevated dopamine concentrations in obsessive-compulsive disorder might result from primary deficits in serotonin levels. Findings of this work underscore the pivotal role of serotonergic dysregulation in modulating synaptic plasticity through dopamine pathways, potentially contributing to learned obsessions. Interestingly, serotonin reuptake inhibitors and antidopaminergic potentiators can counteract the over-stable state of high-firing stable points, providing new insights into obsessive-compulsive disorder treatment.

High-order sensory processing nanocircuit based on coupled VO2 oscillators

Conventional circuit elements are constrained by limitations in area and power efficiency at processing physical signals. Recently, researchers have delved into high-order dynamics and coupled oscillation dynamics utilizing Mott devices, revealing potent nonlinear computing capabilities. However, the intricate yet manageable population dynamics of multiple artificial sensory neurons with spatiotemporal coupling remain unexplored. Here, we present an experimental hardware demonstration featuring a capacitance-coupled VO2 phase-change oscillatory network. This network serves as a continuous-time dynamic system for sensory pre-processing and encodes information in phase differences. Besides, a decision-making module for special post-processing through software simulation is designed to complete a bio-inspired dynamic sensory system. Our experiments provide compelling evidence that this transistor-free coupling network excels in sensory processing tasks such as touch recognition and gesture recognition, achieving significant advantages of fewer devices and lower energy-delay-product compared to conventional methods. This work paves the way towards an efficient and compact neuromorphic sensory system based on nano-scale nonlinear dynamics.