Biology subjects

Toyoizumi, T.

Publications and source records attributed to Toyoizumi, T..

14 recordsLinked to original sources

bioRxiv2026

A model of schema learning based on biological dimensionality reduction during sleep

Reorganizing learned knowledge into generalized representations and transferring it to future learning are essential aspects of cognition, often described as schema formation and use. However, the computational and circuit mechanisms underlying these processes remain unclear. Here, we propose a theoretical model in which schemas emerge through the formation and alignment of low-dimensional neural representations. In this model, high-dimensional input patterns are reorganized into low-dimensional manifolds through replay-driven Hebbian nonlinear dimensionality reduction. Manifold alignment simultaneously maps representations with shared task structure onto a common format, enabling downstream readout circuits to be reused across tasks. The model captures three core features of schema learning: rapid learning of similar tasks by reusing low-dimensional representations from prior experience, sleep-dependent generalization to unobserved relationships in transitive inference, and compositional recombination of schemas to solve novel tasks. Together, these results suggest a potential neural mechanism for forming, aligning, and recombining low-dimensional schemas to support future learning.

neuroscience↗

bioRxiv2026

A Dual-Pathway Prediction Error Model of Schizophrenia Spectrum Disorders:Bridging NMDA Hypofunction and Dopaminergic Hyperfunction

1.Schizophrenia spectrum disorders (SSDs) present a profound clinical enigma, manifesting as a heterogeneous continuum ranging from the chaotic volatility of acute psychosis to the impenetrable rigidity of systematized delusions. While neurobiological research has independently implicated NMDA receptor hypofunction or dopaminergic hyperfunction as cardinal pathophysiological distinct mechanisms, a computational framework capable of bridging these distinct cellular deficits to the spectrums vast phenomenological diversity remains elusive. Here, we propose a biologically plausible neural model using a dynamic Bayesian inference with separable positive and negative prediction-error pathways. We demonstrate that NMDA hypofunction selectively blunts negative prediction errors, fostering rigid, bias-dominated beliefs, while dopaminergic hyperfunction uniformly amplifies error signals, driving volatile, observation-dominated states. Their interaction reconstructs SSDs as a continuous bias-volatility spectrum, accounting for key neurophysiological markers and offering a theoretical foundation for mechanism-based patient stratification.

neuroscience↗

bioRxiv2024

A biological model of nonlinear dimensionality reduction

Obtaining appropriate low-dimensional representations from high-dimensional sensory inputs in an unsupervised manner is essential for straightforward downstream processing. Although nonlinear dimensionality reduction methods such as t-distributed stochastic neighbor embedding (t-SNE) have been developed, their implementation in simple biological circuits remains unclear. Here, we develop a biologically plausible dimensionality reduction algorithm compatible with t-SNE, which utilizes a simple three-layer feedforward network mimicking the Drosophila olfactory circuit. The proposed learning rule, described as three-factor Hebbian plasticity, is effective for datasets such as entangled rings and MNIST, comparable to t-SNE. We further show that the algorithm could be working in olfactory circuits in Drosophila by analyzing the multiple experimental data in previous studies. We finally suggest that the algorithm is also beneficial for association learning between inputs and rewards, allowing the generalization of these associations to other inputs not yet associated with rewards.

neuroscience↗

bioRxiv2024

Distributed encoding of hippocampal information in mossy cells

In neural information processing, the nervous system transmits neuronal activity across layers of neural circuits, occasionally passing through small layers composed only of sparse neurons. Hippocampal hilar mossy cells (MCs) constitute such a typical bottleneck layer. In vivo/vitro patch-clamp recordings revealed that MCs were reliably depolarized in response to sharp-wave ripples (SWRs), synchronous neuronal events transmitted from the CA3 region to the dentate gyrus via the MC layer. Machine-learning algorithms predicted the waveforms of SWRs in the CA3 region, based on the MC depolarization waveforms, suggesting that CA3 neural information is indeed transmitted to the MC layer. However, the prediction accuracy varied; i.e., a particular MC showed a more robust association with a particular SWR cluster, and the SWR cluster associated with one MC rarely overlapped with the SWR clusters associated with other MCs. Thus, CA3 network activity is distributed across MC ensembles with pseudo-orthogonal neural representations, allowing the small MC layer to effectively compress hippocampal information.

neuroscience↗

bioRxiv2023

Chaotic neural dynamics facilitate probabilistic computations through sampling

Cortical neurons exhibit highly variable responses over trials and time. Theoretical works posit that this variability arises potentially from chaotic network dynamics of recurrently connected neurons. Here we demonstrate that chaotic neural dynamics, formed through synaptic learning, allow networks to perform sensory cue integration in a sampling-based implementation. We show that the emergent chaotic dynamics provide neural substrates for generating samples not only of a static variable but also of a dynamical trajectory, where generic recurrent networks acquire these abilities with a biologically-plausible learning rule through trial and error. Furthermore, the networks generalize their experience in the stimulus-evoked samples to the inference without partial or all sensory information, which suggests a computational role of spontaneous activity as a representation of the priors as well as a tractable biological computation for marginal distributions. These findings suggest that chaotic neural dynamics may serve for the brain function as a Bayesian generative model.

neuroscience↗

bioRxiv2023

Distinguishing examples while building concepts in hippocampal and artificial networks

The hippocampal subfield CA3 is thought to function as an autoassociative network that stores experiences as memories. Information from these experiences arrives via the entorhinal cortex (EC), which projects to CA3 directly as well as indirectly through the dentate gyrus (DG). DG sparsifies and decorrelates the information before also projecting to CA3. The computational purpose for receiving two encodings of the same sensory information has not been firmly established. We model CA3 as a Hopfield-like network that stores both correlated and decorrelated encodings and retrieves them at low and high inhibitory tone, respectively. As more memories are stored, the dense, correlated encodings merge along shared features while the sparse, decorrelated encodings remain distinct. In this way, the model learns to transition between concept and example representations by controlling inhibitory tone. To experimentally test for the presence of these complementary encodings, we analyze the theta-modulated tuning of place cells in rat CA3. In accordance with our models prediction, these neurons exhibit more precise spatial tuning and encode more detailed task features during theta phases with sparser activity. Finally, we generalize the model beyond hippocampal architecture and find that feedforward neural networks trained in multitask learning benefit from a novel loss term that promotes hybrid encoding using correlated and decorrelated representations. Thus, the complementary encodings that we have found in CA3 can provide broad computational advantages for solving complex tasks.

neuroscience↗

bioRxiv2022

Information maximization explains state-dependent synaptic plasticity and memory reorganization during non-rapid eye movement sleep

Slow waves during the non-rapid eye movement (NREM) sleep reflect the alternating up and down states of cortical neurons; global and local slow waves promote memory consolidation and forgetting, respectively. Furthermore, distinct spike-timing-dependent plasticity (STDP) operates in these up and down states. The contribution of different plasticity rules to neural information coding and memory reorganization remains unknown. Here, we show that optimal synaptic plasticity for information maximization in a cortical neuron model provides a unified explanation for these phenomena. The model indicates that the optimal synaptic plasticity is biased towards depression as the baseline firing rate increases. This property explains the distinct STDP observed in the up and down states. Furthermore, it explains how global and local slow waves predominantly potentiate and depress synapses, respectively, if the background firing rate of excitatory neurons declines with the spatial scale of waves as the model predicts. The model provides a unifying account of the role of NREM sleep, bridging neural information coding, synaptic plasticity, and memory reorganization.

neuroscience↗

bioRxiv2021

Computational Roles of Intrinsic Synaptic Dynamics

Conventional theories assume that long-term information storage in the brain is implemented by modifying synaptic efficacy. Recent experimental findings challenge this view by demonstrating that dendritic spine sizes, or their corresponding synaptic weights, are highly volatile even in the absence of neural activity. Here we review previous computational works on the roles of these intrinsic synaptic dynamics. We first present the possibility for neuronal networks to sustain stable performance in their presence and we then hypothesize that intrinsic dynamics could be more than mere noise to withstand, but they may improve information processing in the brain. Highlights- Synapses exhibit changes due to intrinsic as well as extrinsic dynamics - Computational frameworks suggest stable network performance despite intrinsic changes - Intrinsic dynamics might be beneficial to information processing

neuroscience↗

bioRxiv2021

Detecting unreported recalls in memory experiments

In experiments on free recall from lists of items, not all memory retrievals are necessarily reported. Previous studies investigated unreported retrievals by attempting to induce their externalization. We show that, without any intervention, their statistics may be directly estimated through a model-free analysis of inter-response times - the silent intervals between recalls. A delay attributable to unreported recalls emerges in three situations: if the final item was already recalled ("silent recency effect"); if the item that, within the list, follows the latest recalled item was already recalled ("silent contiguity effect"); and in sequential recalls within highly performing trials ("sequential slowdown"). We then turn to reproducing all these effects by a minimal model where the discarding of memories ("bouncing") occurs either if they are repetitious or, in strategically organized trials, if they are not sequential. Based on our findings, we propose various approaches to further probing the submerged dynamics of memory retrieval.

animal behavior and cognition↗

bioRxiv2020

Predicting perturbation effects from resting state activity using functional causal flow

A crucial challenge in targeted manipulation of neural activity is to identify perturbation sites whose stimulation exerts significant effects downstream (high efficacy), a procedure currently achieved by labor-intensive trial-and-error. Targeted perturbations will be greatly facilitated by understanding causal interactions within neural ensembles and predicting the efficacy of perturbation sites before intervention. Here, we address this issue by developing a computational framework to predict how single-site micorstimulation alters the ensemble spiking activity in an alert monkeys prefrontal cortex. Our framework uses delay embedding techniques to infer the ensembles functional causal flow (FCF) based on the functional interactions inferred at rest. We validate FCF using ground truth data from models of cortical circuits, showing that FCF is robust to noise and can be inferred from brief recordings of even a small fraction of neurons in the circuit. A detailed comparison of FCF with several alternative methods, including Granger causality and transfer entropy, highlighted the advantages of FCF in predicting perturbation effects on empirical data. Our results provide the foundation for using targeted circuit manipulations to develop targeted interventions suitable for brain-machine interfaces and ameliorating cognitive dysfunctions in the human brain.

neuroscience↗

bioRxiv2019

Intrinsic spine dynamics are critical for recurrent network learning in models with and without autism spectrum disorder

It is often assumed that Hebbian synaptic plasticity forms a cell assembly, a mutually interacting group of neurons that encodes memory. However, in recurrently connected networks with pure Hebbian plasticity, cell assemblies typically diverge or fade under ongoing changes of synaptic strength. Previously assumed mechanisms that stabilize cell assemblies do not robustly reproduce the experimentally reported unimodal and long-tailed distribution of synaptic strengths. Here, we show that augmenting Hebbian plasticity with experimentally observed intrinsic spine dynamics can stabilize cell assemblies and reproduce the distribution of synaptic strengths. Moreover, we posit that strong intrinsic spine dynamics impair learning performance. Our theory explains how excessively strong spine dynamics, experimentally observed in several animal models of autism spectrum disorder, impair learning associations in the brain.

neuroscience↗

bioRxiv2018

Subcortical theta oscillations alternate with high amplitude neocortical population within synchronized states

Synchronized states are marked by large-amplitude low-frequency oscillations in the cortex. These states can be seen during quiet waking or slow-wave sleep. Within synchronized states, previous studies have noted a plethora of different types of activity, including delta oscillations (0.5-4 Hz) and slow oscillations (<1 Hz) in the cortex and large- and small-irregular activity in the hippocampus. However, it is not still fully characterized how neural populations contribute to the synchronized state. Here we apply independent component analysis (ICA) to parse which populations are involved in different kinds of cortical activity, and find two populations that alternate throughout synchronized states. One population broadly affects cortical deep layers, and is associated with larger amplitude slower cortical activity. The other population exhibits theta-frequency oscillations that are not easily observed in raw field potential recordings. These theta oscillations apparently come from below the cortex, suggesting hippocampal origin, and are associated with smaller amplitude faster cortical activity. Relative involvement of these two alternating populations may indicate different modes of operation within synchronized states.

neuroscience↗

bioRxiv2018

Multi-context blind source separation by error-gated Hebbian rule

Animals need to adjust their inferences according to the context they are in. This is required for the multi-context blind source separation (BSS) task, where an agent needs to infer hidden sources from their context-dependent mixtures. The agent is expected to invert this mixing process for all contexts. Here, we show that a neural network that implements the error-gated Hebbian rule (EGHR) with sufficiently redundant sensory inputs can successfully learn this task. After training, the network can perform the multi-context BSS without further updating synapses, by retaining memories of all experienced contexts. Finally, if there is a common feature shared across contexts, the EGHR can extract it and generalize the task to even inexperienced contexts. This demonstrates an attractive use of the EGHR for dimensionality reduction by extracting common sources across contexts. The results highlight the utility of the EGHR as a model for perceptual adaptation in animals.

neuroscience↗

bioRxiv2018

A Bayesian psychophysics model of sense of agency

Despite the increasing significance of sense of agency (SoA) research, the literature lacks a formal model: what computational principles underlie SoA, the registration that oneself initiated an action that caused something to happen? We theorize SoA in the framework of optimal Bayesian cue integration with mutually involved principles, namely, reliability of action and outcome sensory signals, their consistency with the causation of the outcome by the action, and the prior belief in causation. We used our Bayesian model to explain the intentional binding effect, hailed as reliable indicator of SoA. Our model explains temporal binding in both self-intended and unintentional actions suggesting that intentionality is not strictly necessary given high confidence in the action causing the outcome. Our Bayesian model also explains that if the sensory cues are reliable, SoA can emerge even for unintended actions. Our formal model therefore posits a precision-dependent causal agency.

neuroscience↗