Biology subjects

Asabuki, T.

Publications and source records attributed to Asabuki, T..

14 recordsLinked to original sources

bioRxiv2026

Spike-history gating of plateau potentials enables closed-loop rewriting of CA1 representations

Behavioral time scale synaptic plasticity (BTSP) allows hippocampal CA1 neurons to form or shift place fields after a single plasticity-inducing plateau potential, but what determines when individual neurons become eligible for another plasticity event remains unclear. Here we propose that recent somatic spike history provides a cell-specific gate for plateau initiation, making neurons sensitive to rising activity after relative silence while suppressing repeated plasticity during sustained firing. In a two-compartment spiking model of CA1, this gate interacts with entorhinal input to create a feedback loop in which existing representations bias subsequent plateau locations. The resulting dynamics produce gradual place-field drift, reward-associated overrepresentation, recovery of environment-specific representations after remapping, and cross-day reinstatement. Our model further predicts that stronger residual activity biases subsequent plateaus toward previous field locations, a relationship we identify by reanalyzing experimental data. Thus, the expression of neural representations can help determine when and where they are subsequently rewritten.

neuroscience↗

bioRxiv2026

Intrinsic representational drift from predictive excitatory-inhibitory plasticity

Neural representations drift gradually over time even under stable environmental conditions, but the synaptic mechanisms driving this drift remain unclear. Here we show that representational drift can arise intrinsically from predictive synaptic plasticity in spiking excitatory-inhibitory networks. During repeated exposure to unchanged input patterns, individual neurons gradually changed their preferred pattern while ensemble-level coding remained stable. These changes in preference were preceded by a weakening of the net excitatory-inhibitory drive supporting the current preference relative to competing patterns, and this relative drive predicted changes on the following trial. Extending the model to hippocampal place coding reproduced experience-dependent tuning-curve drift in CA1, including the dissociation between elapsed time and intervening exposure. At the population level, drift was expressed as coordinated rotation and translation of neural state space. Thus, representational drift can emerge as an intrinsic consequence of predictive E/I plasticity that maintains balanced, selective population codes.

neuroscience↗

bioRxiv2026

Parallel processing of orthogonal manifolds enables zero-shot composition in recurrent networks

Animals flexibly combine learned behaviors into novel actions without practicing their combinations, yet the computational mechanisms that enable independently acquired computations to be expressed in parallel remain unclear. Here we show that feedback geometry during learning determines whether recurrent dynamics can be recombined through zero-shot parallel composition. Using recurrent networks trained by a local predictive plasticity rule, we found that distinct feedback vectors embed independently learned computations in separable dynamical subspaces, allowing novel input combinations to co-activate these components and generate composite outputs without joint training. In contrast, aligned feedback vectors, as well as networks trained by backpropagation through time, exhibited accurate single-task performance but failed to support parallel composition, demonstrating that task acquisition and future reusability are dissociable properties of learning. A combined input evoked a single composite population trajectory, whose projections onto feedback-shaped task subspaces recovered the independently learned component dynamics. The same principle reproduced additive reach-posture geometry observed in motor cortex and generalized to higher-dimensional movement primitives. These results identify feedback geometry as a computational principle by which learning systems structure recurrent dynamics for future compositional reuse.

neuroscience↗

bioRxiv2026

Predictability of evoked activity governs the geometry of spontaneous neural manifolds

Spontaneous neural activity reflects sensory experience yet occupies a lower-dimensional subspace than evoked responses. However, the principle governing which components of experience are incorporated into this intrinsic activity has remained unknown. Here, we propose that temporal predictability is one of the organizing principles that selects which evoked dimensions enter spontaneous neural manifolds. We show that local predictive synaptic plasticity provides a circuit mechanism that selectively embeds components of evoked activity that are predictable on the intrinsic timescale of recurrent dynamics, while excluding unpredictable fluctuations. As a consequence, the dimensionality of spontaneous activity is not fixed but depends on environmental timescales: rapidly fluctuating inputs are excluded, whereas slowly varying components are retained as contextual dimensions. This framework reproduces on- and off-manifold coding observed in visual cortex, and reconciles apparently conflicting developmental observations that spontaneous and evoked activity become both more similar and more geometrically distinct over maturation. Together, these results identify temporal predictability as a key principle linking environmental statistics to the geometry of intrinsic population dynamics.

neuroscience↗

bioRxiv2026

Learning sculpts orthogonal task manifolds for continual skill learning in recurrent networks

Humans and animals can learn and seamlessly perform a vast repertoire of behaviors. However, how neural populations incorporate new skills without disrupting previously learned ones remains poorly understood. This challenge is known as catastrophic forgetting in artificial neural networks and is especially severe in recurrent networks, where computation relies on stable internal dynamics. While current machine learning approaches often rely on explicit weight-protection strategies, such approaches do not directly address this challenge. Here we show that orthogonal task manifolds can emerge in recurrent neural networks from a local predictive, error-driven learning rule. Our model preserves task-specific latent dynamics that remain resilient to interference from subsequent learning. Using low-rank connectivity ablations, we causally isolate these latent dynamics, selectively impairing individual tasks while leaving others intact. We further show that the proposed principle generalizes beyond low-dimensional to high-dimensional naturalistic movie replay, suggesting a scalable mechanism for continual learning. Our results identify a solution to catastrophic forgetting by preserving previously learned dynamics in recurrent connectivity, thereby providing a mechanistic bridge between artificial recurrent networks and biological neural circuits.

neuroscience↗

bioRxiv2025

Human-specific fast synaptic kinetics enable rapid detection of predictive features

Humans can discern the underlying principles in complex environments, but the cellular basis of this ability is unclear. By comparing layer 2/3 cortical pyramidal neurons in five mammals (mouse, rat, marmoset, macaque and human), we found that human neurons exhibited distinctively faster kinetics of excitatory synaptic transmissions. In a computational model, neurons with human-type synapses detected hidden patterns more quickly in noise, and this superiority was only evident in unsupervised learning. Tracking synaptic weights revealed that human synapses were potentiated sharply and persistently by the patterns, while macaque-type synapses were broadly potentiated, which were easily eroded. These data suggest that the human-specific acceleration of synaptic kinetics enables the rapid extraction of latent structures in complex environments, providing a cellular basis for our efficient cognitive abilities.

neuroscience↗

bioRxiv2024

Taming the chaos gently: a Predictive Alignment learning rule in recurrent neural networks

Recurrent neural circuits often face inherent complexities in learning and generating their desired outputs, especially when they initially exhibit chaotic spontaneous activity. While the celebrated FORCE learning rule can train chaotic recurrent networks to produce coherent patterns by suppressing chaos, it requires non-local plasticity rules and extremely quick plasticity, raising the question of how synapses adapt on local, biologically plausible timescales to handle potential chaotic dynamics. We propose a novel framework called "Predictive Alignment", which tames the chaotic recurrent dynamics to generate a variety of patterned activities via a biologically plausible plasticity rule. Unlike most recurrent learning rules, predictive alignment does not aim to directly minimize output error to train recurrent connections, but rather it tries to efficiently suppress chaos by aligning recurrent prediction with chaotic activity. We show that the proposed learning rule can perform supervised learning of multiple target signals, including complex low-dimensional attractors, delay matching tasks that require short-term temporal memory, and finally even dynamic movie clips with high-dimensional pixels. Our findings shed light on how predictions in recurrent circuits can support learning.

neuroscience↗

bioRxiv2023

Learning predictive signals within a local recurrent circuit

The predictive coding hypothesis proposes that top-down predictions are compared with incoming bottom-up sensory information, with prediction errors signaling the discrepancies between these inputs. While this hypothesis explains the presence of prediction errors, recent experimental studies suggest that prediction error signals can emerge within a local circuit, that is, from bottom-up sensory input alone. In this paper, we test whether local circuits alone can generate predictive signals by training a recurrent spiking network using local plasticity rules. Our network model replicates prediction errors resembling various experimental results, such as a biphasic pattern of prediction errors and context-specific representation of error signals. Our findings shed light on how synaptic plasticity can shape prediction errors and enables the acquisition and updating of an internal model of sensory input within a recurrent neural network.

neuroscience↗

bioRxiv2023

Embedding stochastic dynamics of the environment in spontaneous activity by prediction-based plasticity

The brain learns an internal model of the environment through sensory experiences, which is essential for high-level cognitive processes. Recent studies show that spontaneous activity reflects such learned internal model. Although computational studies have proposed that Hebbian plasticity can learn the switching dynamics of replayed activities, it is still challenging to learn dynamic spontaneous activity that obeys the statistical properties of sensory experience. Here, we propose a pair of biologically plausible plasticity rules for excitatory and inhibitory synapses in a recurrent spiking neural network model to embed stochastic dynamics in spontaneous activity. The proposed synaptic plasticity rule for excitatory synapses seeks to minimize the discrepancy between stimulus-evoked and internally predicted activity, while inhibitory plasticity maintains the excitatory-inhibitory balance. We show that the spontaneous reactivation of cell assemblies follows the transition statistics of the models evoked dynamics. We also demonstrate that simulations of our model can replicate recent experimental results of spontaneous activity in songbirds, suggesting that the proposed plasticity rule might underlie the mechanism by which animals learn internal models of the environment.

neuroscience↗

bioRxiv2023

Learning rules for cortical-like spontaneous replay of an internal model

The brain is thought to construct an optimal internal model representing the probabilistic structure of the environment accurately. Evidence suggests that spontaneous brain activity gives such a model by cycling through activity patterns evoked by previous sensory experiences with the experienced probabilities. The brains spontaneous activity emerges from internally-driven neural population dynamics. However, how cortical neural networks encode internal models into spontaneous activity is poorly understood. Recent computational and experimental studies suggest that a cortical neuron can implement complex computations, including predictive responses, through soma-dendrite interactions. Here, we show that a recurrent network of spiking neurons subject to the same predictive learning principle provides a novel mechanism to learn the spontaneous replay of probabilistic sensory experiences. In this network, the learning rules minimize probability mismatches between stimulus-evoked and internally driven activities in all excitatory and inhibitory neurons. This learning paradigm generates stimulus-specific cell assemblies that internally remember their activation probabilities using within-assembly recurrent connections. Our model contrasts previous models that encode the statistical structure of sensory experiences into Markovian transition patterns among cell assemblies. We demonstrate that the spontaneous activity of our model well replicates the behavioral biases of monkeys performing perceptual decision making. Our results suggest that interactions between intracellular processes and recurrent network dynamics are more crucial for learning cognitive behaviors than previously thought.

neuroscience↗

bioRxiv2021

Intrinsic bursts facilitate learning of Levy flight movements in recurrent neural network models

Isolated spikes and bursts of spikes are thought to provide the two major modes of information coding by neurons. Bursts are known to be crucial for fundamental processes between neuron pairs, such as neuronal communications and synaptic plasticity. Deficits in neuronal bursting can also impair higher cognitive functions and cause mental disorders. Despite these findings on the roles of bursts, whether and how bursts have an advantage over isolated spikes in the network-level computation remains elusive. Here, we demonstrate in a computational model that not isolated spikes but intrinsic bursts can greatly facilitate learning of Levy flight random walk trajectories by synchronizing burst onsets across neural population. Levy flight is a hallmark of optimal search strategies and appears in cognitive behaviors such as saccadic eye movements and memory retrieval. Our results suggest that bursting is a crucial component of sequence learning by recurrent neural networks in the brain.

neuroscience↗

bioRxiv2021

Neural mechanisms of context-dependent segmentation tested on large-scale recording data

The brain performs various cognitive functions by learning the spatiotemporal salient features of the environment. This learning likely requires unsupervised segmentation of hierarchically organized spike sequences, but the underlying neural mechanism is only poorly understood. Here, we show that a recurrent gated network of neurons with dendrites can context-dependently solve difficult segmentation tasks. Dendrites in this model learn to predict somatic responses in a self-supervising manner while recurrent connections learn a context-dependent gating of dendro-somatic current flows to minimize a prediction error. These connections select particular information suitable for the given context from input features redundantly learned by the dendrites. The model selectively learned salient segments in complex synthetic sequences. Furthermore, the model was also effective for detecting multiple cell assemblies repeating in large-scale calcium imaging data of more than 6,500 cortical neurons. Our results suggest that recurrent gating and dendrites are crucial for cortical learning of context-dependent segmentation tasks.

neuroscience↗

bioRxiv2019

Artificial Dendritic Neurons Enable Self-Supervised Temporal Feature Extraction

The brain identifies potentially salient features within continuous information streams to appropriately process external and internal temporal events. This requires the compression or abstraction of information streams, for which no effective information principles are known. Here, we propose conditional entropy minimization learning as the fundamental principle of such temporal processing. We show that this learning rule resembles Hebbian learning with backpropagating action potentials in dendritic neuron models. Moreover, networks of the dendritic neurons can perform a surprisingly wide variety of complex unsupervised learning tasks. Our model not only accounts for the mechanisms of chunking of temporal inputs in the human brain but also accomplishes blind source separation of correlated mixed signals, which cannot be solved by conventional machine learning methods, such as independent-component analysis.\n\nOne Sentence SummaryNeurons use soma-dendrite interactions to self-supervise the learning of characteristic features of various temporal inputs.

neuroscience↗

bioRxiv2017

Chunking sequence information by mutually predicting recurrent neural networks

Interpretation and execution of complex sequences is crucial for various cognitive tasks such as language processing and motor control. The brain solves this problem arguably by dividing a sequence into discrete chunks of contiguous items. While chunking has been accounted for by predictive uncertainty, alternative mechanisms have also been suggested, and the mechanism underlying chunking is poorly understood. Here, we propose a class of unsupervised neural networks for learning and identifying repeated patterns in sequence input with various degrees of complexity. In this model, a pair of reservoir computing modules, each of which comprises a recurrent neural network and readout units, supervise each other to consistently predict others responses to frequently recurring segments. Interestingly, this system generates neural responses similar to those formed in the basal ganglia during habit formation. Our model extends reservoir computing to higher cognitive function and demonstrates its resemblance to sequence processing by cortico-basal ganglia loops.

neuroscience↗