Language：English   Publishing type：Research paper (scientific journal)  

Prepulse inhibition (PPI) is a behavioural phenomenon in which a preceding weaker stimulus suppresses the startle response to a subsequent stimulus. The effect of PPI has been found to be reduced in psychiatric patients and is a promising neurophysiological indicator of psychiatric disorders. Because the neural circuit of the startle response has been identified at the cellular level, investigating the mechanism underlying PPI in Drosophila melanogaster larvae through experiment-based mathematical modelling can provide valuable insights. We recently identified PPI in Drosophila larvae and found that PPI was reduced in larvae mutated with the Centaurin gamma 1A (CenG1A) gene, which may be associated with autism. In this study, we used numerical simulations to investigate the neural mechanisms underlying PPI in Drosophila larvae. We adjusted the parameters of a previously developed Drosophila larvae computational model and demonstrated that the model could reproduce several behaviours, including PPI. An analysis of the temporal changes in neuronal activity when PPI occurs using our neural circuit model suggested that the activity of specific neurons triggered by prepulses has a considerable effect on PPI. Furthermore, we validated our speculations on PPI reduction in CenG1A mutants with simulations.

DOI： 10.1038/s41598-022-19210-8

PubMed

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.2337/db21-0644

Web of Science

researchmap

Publishing type：Research paper (scientific journal)   Publisher：{IOP} Publishing  

Abstract

L0-regularization-based compressed sensing (L0-RBCS) has the potential to outperform L1-regularization-based compressed sensing (L1-RBCS), but the optimization in L0-RBCS is difficult because it is a combinatorial optimization problem. To perform optimization in L0-RBCS, we propose a quantum–classical hybrid system consisting of a quantum machine and a classical digital processor. The coherent Ising machine (CIM) is a suitable quantum machine for this system because this optimization problem can only be solved with a densely connected network. To evaluate the performance of the CIM-classical hybrid system theoretically, a truncated Wigner stochastic differential equation (W-SDE) is introduced as a model for the network of degenerate optical parametric oscillators, and macroscopic equations are derived by applying statistical mechanics to the W-SDE. We show that the system performance in principle approaches the theoretical limit of compressed sensing and this hybrid system may exceed the estimation accuracy of L1-RBCS in actual situations, such as in magnetic resonance imaging data analysis.

DOI： 10.1088/2058-9565/ac6f19

researchmap

Other Link： https://iopscience.iop.org/article/10.1088/2058-9565/ac6f19/pdf

Publishing type：Research paper (scientific journal)   Publisher：Springer Science and Business Media LLC  

Abstract

A coherent Ising machine (CIM) is an open-dissipative Ising solver using optical pulses generated from a degenerate optical parametric oscillator as analog magnetizations. When solving real-world optimization problems with CIM, this solver has two difficulties: mutual coupling induced amplitude inhomogeneity and absence of natural way to implement Zeeman terms. For the approximate Gaussian formulation of CIMs with amplitude control feedback, we add artificial Zeemam terms using the target amplitude information. Here we show, for 16-spin CIM with Zeeman terms, the amplitude control increases the performance, particularly when Zeeman terms are competing against mutual coupling coefficients.

DOI： 10.1038/s42005-022-00927-x

researchmap

Other Link： https://www.nature.com/articles/s42005-022-00927-x

Authorship：Last author   Publishing type：Research paper (scientific journal)   Publisher：Elsevier BV  

DOI： 10.1016/j.neures.2022.02.008

researchmap

Language：Japanese   Publisher：Japanese Neural Network Society  

DOI： 10.3902/jnns.29.3

CiNii Research

researchmap

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)   Publisher：Elsevier BV  

The rapid progress of imaging devices such as two-photon microscopes has made it possible to measure the activity of thousands to tens of thousands of cells at single-cell resolution in a wide field of view (FOV) data. However, it is not possible to manually identify thousands of cells in such wide FOV data. Several research groups have developed machine learning methods for automatically detecting cells from wide FOV data. Many of the recently proposed methods using dynamic activity information rather than static morphological information are based on non-negative matrix factorization (NMF). In this review, we outline cell-detection methods related to NMF. For the purpose of raising issues on NMF cell detection, we introduce our current development of a non-NMF method that is capable of detecting about 17,000 cells in ultra-wide FOV data.

DOI： 10.1016/j.neures.2021.12.001

PubMed

researchmap

Publishing type：Research paper (scientific journal)   Publisher：IOP Publishing  

Abstract

Lithium-ion secondary batteries have been used in a wide variety of purposes, such as for powering mobile devices and electric vehicles, but their performance should be improved. One of the factors that limits their performance is the non-uniformity of the chemical reaction in the process of charging and discharging. Many attempts have been made to elucidate the mechanism behind this reaction non-uniformity. In this paper, to detect non-uniformity in various physical properties from Co K-edge two-dimensional x-ray absorption spectroscopy (2D-XAS) images of lithium ion batteries, we propose a method that consists of one-sided orthogonal non-negative matrix factorization in combination with removal of the reference signal. The difference between x-ray absorption spectra acquired at different positions in the battery is very small. However, even in such a situation, our method can decompose the 2D-XAS data into different spatial domains and their corresponding absorption spectra. From the spectral decomposition of the obtained absorption spectra, we confirmed a transition-energy shift of the main peak as evidence for a change in the state of charge and also found spectral changes due to orbital hybridization in the decomposed spectral components.

DOI： 10.1088/2399-6528/ac3268

researchmap

Other Link： https://iopscience.iop.org/article/10.1088/2399-6528/ac3268/pdf

Publishing type：Research paper (scientific journal)   Publisher：Physical Society of Japan  

DOI： 10.7566/JPSJ.90.094602

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Elsevier BV  

Fast and wide field-of-view imaging with single-cell resolution, high signal-to-noise ratio, and no optical aberrations have the potential to inspire new avenues of investigations in biology. However, such imaging is challenging because of the inevitable tradeoffs among these parameters. Here, we overcome these tradeoffs by combining a resonant scanning system, a large objective with low magnification and high numerical aperture, and highly sensitive large-aperture photodetectors. The result is a practically aberration-free, fast-scanning high optical invariant two-photon microscopy (FASHIO-2PM) that enables calcium imaging from a large network composed of ∼16,000 neurons at 7.5 Hz from a 9 mm2 contiguous image plane, including more than 10 sensory-motor and higher-order areas of the cerebral cortex in awake mice. Network analysis based on single-cell activities revealed that the brain exhibits small-world rather than scale-free behavior. The FASHIO-2PM is expected to enable studies on biological dynamics by simultaneously monitoring macroscopic activities and their compositional elements.

DOI： 10.1016/j.neuron.2021.03.032

PubMed

researchmap

Publishing type：Research paper (scientific journal)   Publisher：Physical Society of Japan  

DOI： 10.7566/jpsj.90.044003

researchmap

Authorship：Last author   Publishing type：Research paper (scientific journal)   Publisher：Japan Society of Applied Physics  

DOI： 10.7566/jpsj.90.014705

Scopus

researchmap

Authorship：Last author   Publishing type：Research paper (scientific journal)   Publisher：Elsevier BV  

DOI： 10.1016/j.brainresbull.2019.08.009

researchmap

Publishing type：Research paper (scientific journal)  

DOI： 10.1063/1.5041998

Web of Science

researchmap

Publishing type：Research paper (scientific journal)  

DOI： 10.1039/C7JA00334J

Web of Science

researchmap

Publishing type：Research paper (scientific journal)  

DOI： 10.1063/1.5041997

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1016/j.neunet.2018.02.014

Web of Science

Scopus

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.7566/JPSJ.86.104002

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：Springer Verlag  

DOI： 10.1007/978-3-319-70090-8_90

Scopus

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

CiNii Books

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.7566/JPSJ.84.094801

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1038/srep10253

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1371/journal.pone.0122263

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1016/j.neunet.2014.03.007

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1016/j.neulet.2014.04.011

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1371/journal.pone.0085790

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.2197/ipsjtrans.7.27

Scopus

researchmap

Language：English   Publisher：Information and Media Technologies Editorial Board  

Many associative memory models with synaptic decay such as the forgetting model and the zero-order decay model have been proposed and studied so far. The previous studies showed the relation between the storage capacity C and the synaptic decay coefficient α in each synaptic decay model. However, with the exceptions of a few studies, they did not compare the network retrieval performance between different synaptic decay models. We formulate the associative memory model with the β-th-order synaptic decay as an extension of the zero-order decay model. The parameter β denotes the synaptic decay order or the degree of the synaptic decay term, which enables us to compare the retrieval performance between different synaptic decay models. Using numerical simulations, we investigate the relation between the synaptic decay coefficient α and the storage capacity C of the network by varying the synaptic decay order β. The results show that the properties of the synaptic decay model are constant for a large decay order β. Moreover, we search the minimum β to avoid overloading and the optimal β to maximize the network retrieval performance. The minimum integer value of β to avoid overloading is -1. The optimal integer value of β to maximize the network retrieval performance is 1, i.e., the degree of the forgetting model, and the suboptimal integer β is 0, i.e., that of the zero-order synaptic decay model.

DOI： 10.11185/imt.9.67

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1371/journal.pone.0077395

Web of Science

Scopus

PubMed

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： 10.1007/978-3-642-42042-9_34

Scopus

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1103/PhysRevE.86.061911

Web of Science

Scopus

PubMed

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1103/PhysRevE.86.061911

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1371/journal.pone.0050232

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1007/s10015-012-0033-5

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1587/transinf.E95.D.2124

Web of Science

researchmap

Language：English   Publishing type：Part of collection (book)   Publisher：Springer New York  

DOI： 10.1007/978-1-4614-0739-3_8

Scopus

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Information Processing Society of Japan  

DOI： 10.2197/ipsjtrans.5.186

Scopus

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1159/000346266

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1103/PhysRevE.84.041902

Scopus

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1103/PhysRevE.84.041902

Web of Science

researchmap

Language：English  

DOI： 10.1016/j.neures.2011.07.893

Web of Science

researchmap

Language：English  

DOI： 10.1016/j.neures.2011.07.393

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

Web of Science

researchmap

Language：English  

DOI： 10.1016/j.neures.2011.07.1380

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1016/j.neunet.2010.06.006

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1143/JPSJ.79.124801

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1103/PhysRevE.81.021901

Scopus

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1103/PhysRevE.81.021901

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1016/j.bpj.2009.10.041

Web of Science

researchmap

Language：English  

DOI： 10.1016/j.neures.2010.07.1929

Web of Science

researchmap

Language：English  

DOI： 10.1016/j.neures.2010.07.288

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

researchmap

Language：English  

DOI： 10.1016/j.neures.2010.07.1006

Web of Science

researchmap

Language：English  

DOI： 10.1016/j.neures.2010.07.2058

Web of Science

researchmap

Language：English  

DOI： 10.1016/j.neures.2010.07.234

Web of Science

researchmap

Language：English  

DOI： 10.1016/j.neures.2010.07.448

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1016/j.neures.2009.01.012

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1007/s10827-008-0104-8

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English  

DOI： 10.1016/j.neures.2009.09.330

Web of Science

researchmap

Language：English  

DOI： 10.1016/j.neures.2009.09.262

Web of Science

researchmap

Language：English   Publishing type：Research paper (conference, symposium, etc.)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： 10.1088/1742-6596/197/1/012016

Web of Science

researchmap

Language：English  

DOI： 10.1016/j.neures.2009.09.668

Web of Science

researchmap

Language：English  

Web of Science

researchmap

Language：English  

Web of Science

researchmap

Language：English  

DOI： 10.1016/j.neures.2007.06.165

Web of Science

researchmap

Language：English  

DOI： 10.1016/j.neures.2007.06.810

Web of Science

researchmap

Language：English  

DOI： 10.1016/j.neures.2007.06.235

Web of Science

researchmap

Language：English  

DOI： 10.1016/j.neures.2007.06.164

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1007/s10827-008-0104-8

researchmap

Language：English  

Web of Science

researchmap

Language：English  

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1143/JPSJ.72.1334

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1103/PhysRevE.65.046223

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1103/PhysRevLett.88.024102

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1103/PhysRevE.65.046223

Scopus

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1109/72.846754

Web of Science

researchmap

Authorship：Lead author   Publishing type：Research paper (scientific journal)   Publisher：American Physical Society (APS)  

DOI： 10.1103/physrevlett.82.2800

researchmap

Other Link： http://harvest.aps.org/v2/journals/articles/10.1103/PhysRevLett.82.2800/fulltext

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1103/PhysRevE.58.4865

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1162/089976698300017692

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1007/s004220050431

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

Web of Science

researchmap

J-GLOBAL

researchmap

J-GLOBAL

researchmap

J-GLOBAL

researchmap

J-GLOBAL

researchmap

Language：Japanese  

Waterfall illusion is a motion aftereffect experienced when looking at a stationary stimulus after watching a stimulus moving in one direction for some time. The stationary stimulus appears to have movement in the opposite direction to the moving stimulus previously exposed. Srinivasan and Dvorak suggested the possibility that the fly could perceive the waterfall illusion, and its motion-sensitive neurons might induce rebound and inhibitory responses correlated with this perception. In this paper, we hypothesized that this perception is based on a network consisting of Lobula plate tangential cells (LPTCs), which were ones of motion-sensitive neurons. To test the hypothesis, we constructed a circuit model of LPTCs getting the newly introduced T-type calcium channel, and tried to reproduce this characteristic neural response after the stimulation was stopped. As a result of numerical simulation using the circuit model of LPTCs, we were able to reproduce the rebound and inhibitory responses. The result of simulation suggested that T-type calcium channel could act as a trigger for the rebound response, and LPTCs circuit might work as a reverberation circuit to promote the rebound response. This result also showed that the inhibitory response could be caused by suppression from the cells inducing the rebound response. In conclusion, this study showed one of the possible neural mechanisms underlying fly's waterfall illusion.

CiNii Books

researchmap

J-GLOBAL

researchmap

J-GLOBAL

researchmap

Language：Japanese   Publisher：電子情報通信学会  

researchmap

Language：Japanese   Publisher：電子情報通信学会  

researchmap

Language：Japanese   Publisher：電子情報通信学会  

researchmap

Language：Japanese   Publisher：電子情報通信学会  

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese   Publisher：電子情報通信学会  

CiNii Books

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese   Publisher：電子情報通信学会  

CiNii Books

researchmap

Language：Japanese  

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：電子情報通信学会  

CiNii Books

researchmap

Language：Japanese  

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：電子情報通信学会  

CiNii Books

researchmap

Language：Japanese   Publisher：電子情報通信学会  

CiNii Books

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人 日本生物物理学会  

DOI： 10.2142/biophys.57.036

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人日本地球化学会  

<p></p>

DOI： 10.14862/geochemproc.64.0_214

J-GLOBAL

researchmap

Language：Japanese   Publisher：電子情報通信学会  

researchmap

Language：Japanese   Publisher：電子情報通信学会  

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

<p>It has been known for sometime that brain tissue show marked low-frequency dielectric dispersion and large permittivity (>10<sup>6</sup>) in a frequency range of 10–100 Hz. We have shown that these properties can be due to electrically long neurites by solving equations that describe a model with a bundle of finite length cylindrical neurites subjected to oscillating extracellular electric fields. We also provide a notion of the apparent extracellular space and a generalized formulation for the EEG inverse problem that takes permittivity into account.</p>

DOI： 10.11316/butsuri.71.6_352

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：電子情報通信学会  

CiNii Books

researchmap

Language：Japanese   Publisher：電子情報通信学会  

CiNii Books

researchmap

Language：Japanese  

CiNii Books

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

Motion detection cell models of fly's early visual system have been proposed. Several models dealing with functions of LI and L2 pathway in fly's visual ganglia have also been proposed. Here we focused on following these three models, 4D model consisting of four correlation-type models and positive-and-negative-wave rectifiers, 2D model consisting of two correlation-type models and the rectifiers and 6D model consisting of six correlation-type models and the rectifiers. In this study, using electrophysiological data obtained from Lobula Prate Tangential Cells(LPTCs) in response to visual stimuli with various ambiguities, we estimated generalization errors of the above three models by cross-validation method. Consequently, we found that the generalization error of 2D model was smaller than the others.

CiNii Books

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人 人工知能学会  

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

It is well known that a correlation-type motion detection model can sub-quantitatively reproduce the response of the fly early visual system. A correlation-type motion detection model is classified as Reichardt and Eichner types. In this study, we compared the two models to determine which best describes electrophysiological data. In a previous work, the two models were compared by using data of motion-detection cells in response to flash visual stimuli. Here, we used data of cell responses to noisy motion stimuli in control of aperture. We compared the explanation capability of the two models by using leave-one-out cross-validation method. As a consequence, we found that Eichner-type model has higher generalization than that of Reichardt-type model independent of the individual difference of the flies.

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

Animals respond to various kinds of sensory stimuli presented in outside world. In nature, the sensory stimuli are often unapparent and noisy. It has been one of most important issues in neuroscience how the brain process unapparent stimuli and guarantee the robust behavioral reaction. Here, we investigate the precise properties of response to an unapparent wide-field motion stimulus with respect to behavior and neuronal activity in Drosophila melanogaster. The reactivity of behavioral response reduced proportionally to the increase of noise, on the other hand, however, the discriminative performance of motion direction showed robustness. Moreover, we found that an activity of motion-sensitive neurons is strongly correlated with these two distinct features of the behavioral response by in vivo whol cell patch-clamp recordings. Our results demonstrate that fly's discriminative capacity of motion direction has robusntess but the reactivity doesn't, and these two distinct features originate from motion-sensitive neuron's activity.

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

It is well known that a correlation-type motion detection model can sub-quantitatively reproduce the response of the fly early visual system. A correlation-type motion detection model is classified as Reichardt and Eichner types. In this study, we compared the two models to determine which best describes electrophysiological data. In a previous work, the two models were compared by using data of motion-detection cells in response to flash visual stimuli. Here, we used data of cell responses to noisy motion stimuli in control of aperture. We compared the explanation capability of the two models by using leave-one-out cross-validation method. As a consequence, we found that Eichner-type model has higher generalization than that of Reichardt-type model independent of the individual difference of the flies.

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

Animals respond to various kinds of sensory stimuli presented in outside world. In nature, the sensory stimuli are often unapparent and noisy. It has been one of most important issues in neuroscience how the brain process unapparent stimuli and guarantee the robust behavioral reaction. Here, we investigate the precise properties of response to an unapparent wide-field motion stimulus with respect to behavior and neuronal activity in Drosophila melanogaster. The reactivity of behavioral response reduced proportionally to the increase of noise, on the other hand, however, the discriminative performance of motion direction showed robustness. Moreover, we found that an activity of motion-sensitive neurons is strongly correlated with these two distinct features of the behavioral response by in vivo whol cell patch-clamp recordings. Our results demonstrate that fly's discriminative capacity of motion direction has robusntess but the reactivity doesn't, and these two distinct features originate from motion-sensitive neuron's activity.

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人情報処理学会  

It is well known that a correlation-type motion detection model can sub-quantitatively reproduce the response of the fly early visual system. A correlation-type motion detection model is classified as Reichardt and Eichner types. In this study, we compared the two models to determine which best describes electrophysiological data. In a previous work, the two models were compared by using data of motion-detection cells in response to flash visual stimuli. Here, we used data of cell responses to noisy motion stimuli in control of aperture. We compared the explanation capability of the two models by using leave-one-out cross-validation method. As a consequence, we found that Eichner-type model has higher generalization than that of Reichardttype model independent of the individual difference of the flies.

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人情報処理学会  

Animals respond to various kinds of sensory stimuli presented in outside world. In nature, the sensory stimuli are often unapparent and noisy. It has been one of most important issues in neuroscience how the brain process unapparent stimuli and guarantee the robust behavioral reaction. Here, we investigate the precise properties of response to an unapparent wide-field motion stimulus with respect to behavior and neuronal activity in Drosophila melanogaster. The reactivity of behavioral response reduced proportionally to the increase of noise, on the other hand, however, the discriminative performance of motion direction showed robustness. Moreover, we found that an activity of motion-sensitive neurons is strongly correlated with these two distinct features of the behavioral response by in vivo whol cell patch-clamp recordings. Our results demonstrate that fly's discriminative capacity of motion direction has robusntess but the reactivity doesn't, and these two distinct features originate from motion-sensitive neuron's activity.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人情報処理学会  

Animals respond to various kinds of sensory stimuli presented in outside world. In nature, the sensory stimuli are often unapparent and noisy. It has been one of most important issues in neuroscience how the brain process unapparent stimuli and guarantee the robust behavioral reaction. Here, we investigate the precise properties of response to an unapparent wide-field motion stimulus with respect to behavior and neuronal activity in Drosophila melanogaster. The reactivity of behavioral response reduced proportionally to the increase of noise, on the other hand, however, the discriminative performance of motion direction showed robustness. Moreover, we found that an activity of motion-sensitive neurons is strongly correlated with these two distinct features of the behavioral response by in vivo whol cell patch-clamp recordings. Our results demonstrate that fly's discriminative capacity of motion direction has robusntess but the reactivity doesn't, and these two distinct features originate from motion-sensitive neuron's activity.

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人情報処理学会  

It is well known that a correlation-type motion detection model can sub-quantitatively reproduce the response of the fly early visual system. A correlation-type motion detection model is classified as Reichardt and Eichner types. In this study, we compared the two models to determine which best describes electrophysiological data. In a previous work, the two models were compared by using data of motion-detection cells in response to flash visual stimuli. Here, we used data of cell responses to noisy motion stimuli in control of aperture. We compared the explanation capability of the two models by using leave-one-out cross-validation method. As a consequence, we found that Eichner-type model has higher generalization than that of Reichardttype model independent of the individual difference of the flies.

CiNii Books

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

It has been reported with multicellular calcium imaging methods that Ca^<2+> transients are entrained to the extracellular AC field in CA1 region of rat hippocampal slices. In this study, we apply statistical phase analysis method for the calcium imaging data and analyze the entrainment of Ca^<2+> transients to the extracellular AC field. We developed a statistical method consisting of the following process steps: automatic detection of the fluorescence intensity changes of individual cells, phase estimation from the fluorescence intensity changes and parameter estimation of phase equation. By numerical experiments on artificial data, we show that our method can approximately estimate natural frequencies and coupling coefficients of the phase equation well in the case of weakly coupling. We then apply the method to the imaging data of rat hippocampal slices to show that our method can estimate natural frequencies and coupling coefficients to the extracellular AC field.

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

Recent experimental findings suggest that the hippocampal network including CA1 and CA3 regions stores spatio-temporal spike patterns and retrieves temporally reversed and spread-out patterns. Under the assumption that the CA1 network functions as a hetero-associative memory, we derived pairs of spike-timing-dependent-plasticity (STDP) window function and phase response curve (PRO) that are optimal for retrieval of normal, reversal and spread-out spike patterns. In this research, we focus on the CA3 network with recurrent connections, which is suggested that it functions as an auto-associative memory. We numerically demonstratethat the recurrent network consisting of STDP window function and PRO can realize reversal and spread-outrecalls, and specify necessary conditions on pairs of STDP window function and PRO for realizing such recalls.

CiNii Books

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

Phase response curve (PRC) is an important measure to study neural ensemble synchrony. Recently, for this purpose several groups have tried to estimate the PRC from real neurons. Ermentrout et al. derived an equation that relates the PRC to the spike triggered average (STA) when applying white noise to oscillatory neurons. In this paper, we derive an equation that relates the PRC and STA when using colored noise. Here, the STA with colored noise is referred to as a colored spike triggered average (cSTA). On the basis of this relation, we propose a method for estimation of the PRC from the cSTA. Our numerical results suggest that the convergence speed of cSTA to its asymptotic value is faster than that of STA with white noise, and our estimation method can provide the more accurate PRC from the cSTA than from the STA with white noise under finite sample size situations.

CiNii Books

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：English  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

From the computational viewpoint, Lengyel et al. (2005) derived an optimal auto-associative memory algorithm to store and retrieve memory spike phase patterns. Moreover, they derived optimal relation between the coupling function and the spike-timing-dependent plasticity (STDP) window function for executing the algorithm. From the hardware viewpoint, we here derive an auto-associative memory algorithm, and compare optimal pairs of coupling function and STDP window function which are derived from our study with those from the previous study. First, we formulate an auto-associative memory algorithm as a phase reduction model consisting of a spike-timing-dependent plasticity (STDP) window function and a coupling function which is formalized in terms of a phase response curve (PRC). Next, we analytically derive the mutual information a stored phase pattern and the network retrieval pattern, and use it to evaluate memory retrieval performance. By maximizing the mutual information, we derive optimal pairs of STDP window functions and coupling functions. In line with the previous study, we search for a set of optimal coupling functions that are matched to the typical STDP window function. In the zero noise limit, we obtained the same relation between the coupling function and the STDP window function that Lengyel et al. derived.

CiNii Books

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

The Drosophila larva can orient its movements towards chemoattractants such as food sources using olfactory information. Almost all reports on chemotaxis of the Drosophila larva have focused on sharp turns, however as far as we know, there is no report suggesting the existence of successive slow orientations during locomotion to odorant sources. Firstly we sought to reveal from behaviors of Canton S (wild type)that the successive slow orientation, which will be named 'biased running' in this study, is used as an alternative behavioral strategy of the Drosophila larva during chemotaxis. Next, to verify the contribution of stereo olfaction to the biased running during chemotaxis, we studied behaviors of unilaterally and bilaterally rescued larvae that have unilaterally functional olfactory organs and bilaterally functional ones, respectively.

CiNii Books

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

既存の連想記憶モデルでは，記憶容量を超過してパターンを学習させると，ネットワークはオーバーロードを引き起こし，それまで学習したどのパターンも想起できなくなる．これまでオーバーロードを回避するモデルとして，指数的忘却モデルや 0 次減衰モデルといったシナプス減衰モデルが提案されてきた．本研究ではより一般的なシナプスの減衰を取り扱うため，0 次減衰モデルを β 次減衰するシナプスをもつ連想記憶モデルへと拡張した．減衰次数が 1 のときは指数的忘却モデルに相当する．シミュレーションにより，ネットワークの記憶想起性能を最大化する最適な減衰次数をもつシナプス減衰モデルを探索した．その結果，次数が整数値の場合，ネットワークの記憶容量を最大化する最適な減衰次数をもつモデルは指数的忘却モデルで，0 次減衰モデルは準最適であることがわかった．また，シナプス減衰次数が大きくなるにつれて，モデルの記憶特性がほぼ一定になることがわかった．

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

It has been reported that there are motion detection cells on both sides in fly visual system, which are mutually connected in a unique manner. However, little is known about the role of this mutual connection in visual information processing. Here, to investigate the role of the connection, we constructed a model of recurrent network with Morris-Lecar model .We conclude that the binocular motion vision by the LPTCs mutual network causes efficiency of population coding for the rotational motion stimuli to be higher, although neither of the rotational and translational motion have fundamentally dominance by nature.

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

It has been reported that there are motion detection cells on both sides in fly visual system, which are mutually connected in a unique manner. However, little is known about the role of this mutual connection in visual information processing. Here, to investigate the role of the connection, we constructed a model of recurrent network with Morris-Lecar model .We conclude that the binocular motion vision by the LPTCs mutual network causes efficiency of population coding for the rotational motion stimuli to be higher, although neither of the rotational and translational motion have fundamentally dominance by nature.

CiNii Books

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：English  

CiNii Books

researchmap

Other Link： http://id.nii.ac.jp/1001/00085820/

Language：Japanese   Publisher：The Physical Society of Japan (JPS)  

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

It has been reported that reverse replay and temporally-structured replay of ensemble spiking patterns occur during both awake behavior and sleep in the hippocampal CA1 region. In this paper, we explore optimal neural implementations for executing such memory replays, and interpret a computational role of spike-timing-dependent plasticity (STDP). First, we formulate a hetero-associative memory network model executing normal, reverse and temporally-structured replays as a phase reduction model consisting of a STDP window function and a phase response curve (PRC). Next, we analytically examine the network behavior in those replays, and derive the mutual information for measuring those replay qualities. By maximizing the mutual information, we search a set of optimal STDP window functions under the constraint of PRCs of the hippocampal CA1 pyramidal neurons recorded in vitro. The theoretically derived set of window functions qualitatively conforms to various types of STDP reported previously.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese  

ハエ視覚系の両側にはそれぞれ動き検知細胞があり，これらは両側の視野を繋ぐよう相互結合しており特徴的な神経回路を形成しているが報告されている．この神経回路は比較的少数の神経細胞で構成されているため，両眼視野統合のモデルとして有用であると考えられている．本研究では，動き検知細胞を形式ニューロンを用いてモデル化し数値的に解析することで，両眼視野統合とこの神経回路構造との関係を明らかにする．そのために，回転刺激と並進刺激にノイズを付加し，それぞれの刺激に対する LPTCs の応答性を相互情報量を用いて定量的に評価することで神経回路の符号器としての性能を調べた．シミュレーションの結果，神経回路を構成することで回転刺激に対する動き検知細胞の相互情報量が高くなり，反応性が向上することがわかった．一方で並進刺激に対しては逆に反応性が減少してしまうことがわかった．これらの結果から，動き検知細胞による神経回路は刺激依存的に情報処理特性が変化していることが示唆された．つまりこの神経回路は，並進刺激よりも回転刺激に対して反応性を上昇させ，回転刺激に対する応答性の確度を上昇させるような符号器として働くことがわかった．In the fly visual system, mutual recurrent connections among motion-sensitive neurons (Lobula Plate Tangential Cells (LPTCs)) on both hemispheres have been identified. The fly is an ideal model system to analyze the integration of visual information from both visual hemispheres. Here, to uncover a role for the mutual recurrent connection in the binocular integration, we construct a mathematical model of the LTPCs circuit consisting of for mal neurons. Using the mutual information measure, we numerically quantify the encoding capacity of the LTPCs circuit model in response to the following two kinds of motion stimuli. One is the rotation stimulus in which each eye receives an equal motion input. The other is the translation stimulus in which each eye receives an opposite motion input. We compare sensitivities of the LTPCs circuit model to these two stimuli with or without the mutual connection. We show that the mutual connection enhances the sensitivity of the neurons to the rotation stimulus whereas the mutual connection suppresses that to the translation stimulus. These results suggest that the LTCPs circuit is sensitive to congruent binocular stimuli and filters out conflicting stimuli due to the mutual recurrent connection between both hemispheres.

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

We employ a top-down approach to finding optimal neural implementations for executing a computation of the hippocampal CA1 network considered to function as a hetero-associative memory, and giving an implementation of computational roles of spike-timing-dependent plasticity (STDP). First, hypothesizing that the hippocampal CA1 network computationally works as the hetero-associative memory, we formulate a minimum model of memory functions as a phase reduction model consisting of a STDP window function and a phase response curve (PRC). Next, we analytically derive the mutual information between a stored phase memory pattern and a network output. By maximizing the mutual information, we search pairs of STDP window functions and PRCs optimally functioning as the hetero-associative memory. Here, we use PRCs of the hippocampal CA1 pyramidal neurons recorded in vitro, and under the constraint of measured PRCs, we search a set of optimal STDP window functions. A set of window functions we obtained qualitatively conform with various types of STDP reported previously.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

In the fly visual system, mutual recurrent connections among motion-sensitive neurons (Lobula Plate Tangential Cells (LPTCs)) on both hemispheres have been identified. The fly is an ideal model system to analyze the integration of visual information from both visual hemispheres. Here, to uncover a role for the mutual recurrent connection in the binocular integration, we construct a mathematical model of the LTPCs circuit consisting of formal neurons. Using the mutual information measure, we numerically quantify the encoding capacity of the LTPCs circuit model in response to the following two kinds of motion stimuli. One is the rotation stimulus in which each eye receives an equal motion input. The other is the translation stimulus in which each eye receives an opposite motion input. We compare sensitivities of the LTPCs circuit model to these two stimuli with or without the mutual connection. We show that the mutual connection enhances the sensitivity of the neurons to the rotation stimulus whereas the mutual connection suppresses that to the translation stimulus. These results suggest that the LTCPs circuit is sensitive to congruent binocular stimuli and filters out conflicting stimuli due to the mutual recurrent connection between both hemispheres.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese  

海馬 CA1 ネットワークで行われていると考えられている計算を最適に実現する神経実装をトップダウン的に求め，海馬 CA1 錐体細胞に存在するシナプス可塑性様式である spike-timing-dependent plasticity (STDP) の計算論的役割の可能性を示す．まず，海馬 CA1 での計算は相互想起型連想記憶であると仮定し，この計算を実現する最小記述モデルを，STDP 窓関数と位相応答曲線 (phase response curve，PRC) からなる位相縮約モデルで構成する．そして，STDP で記銘した位相パターンと回路の出力パターンの間の相互情報量を解析的に導出する．この相互情報量を最大化することにより，相互想起型連想記憶を実現する最適な STDＰ と PRC の組み合わせを探索する．我々は，電気生理実験で測定した海馬 CA1 錐体細胞の PRC を与え，この制約下で記憶想起に最適な STDP 窓関数を探索する．我々が求めた STDＰ 窓関数の集合は，先行研究で報告されている多様な窓関数と定性的に一致するものである．We employ a top-down approach to finding optimal neural implementations for executing a computation of the hippocampal CA1 network considered to function as a hetero-associative memory, and giving an implementa tion of computational roles of spike-timing-dependent plasticity (STDP). First, hypothesizing that the hippocampal CA1 network computationally works as the hetero-associative memory, we formulate a minimum model of memory functions as a phase reduction model consisting of a STDP window function and a phase response curve (PRC).Next, we analytically derive the mutual information between a stored phase memory pattern and a network output.By maximizing the mutual information, we search pairs of STDP window functions and PRCs optimally functioning as the hetero-associative memory. Here, we use PRCs of the hippocampal CA1 pyramidal neurons recorded in vitro,and under the constraint of measured PRCs, we search a set of optimal STDP window functions. A set of window functions we obtained qualitatively conform with various types of STDP reported previously.

CiNii Books

researchmap

Language：English  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

It has been reported that there are motion detection cells on both sides in fly visual system, which are mutually connected in a unique manner. However, little is known about the role of this mutual connection in visual information processing. Here, to elucidate the role of the connection, we constructed a model of recurrent network with formal neurons. By employing the stability analysis for the network model, we found that this model can form competitive network models with either bistable or triple-stable steady states depending on values of model parameters. Furhermore, by numerically calculating mutual information between neural responses and motion stimuli, under noisy conditions, we showed that neural sensitivity for characterizing stimulus coding is enhanced by the mutual connetion.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

The observation data of the calcium imaging techniques are indirect information about the intracellular calcium ion concentration. To quantitatively estimate the calcium ion concentration, we have to calibrate calcium fluorescence data. We applied the state space method to solve this calibration problem. We constructed the model of calcium dynamics containing discontinuous changes with the line process and the second-order trend model. We implemented our state-space model with the particle filter and performed Bayesian estimation for calcium ion concentration. We verified its effectiveness by applying it to artificial and real fluorescence data.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

Phase response curve (PRC) is one of important tools for better understanding interacting systems such as neural circuits. Recent electrophysiological experiments showed that PRCs of single neurons have different shapes depending on amplitudes of perturbation stimuli. A linear response element and higher-order nonlinear elements may exist together in observed the PRC due to the perturbation stimuli, whose amplitudes are finite, are used in the electrophysiological experiments. In order to clarify the influence of the perturbation stimulus amplitude on the PRCs, we analytically derive PRCs by using a spike response model. We compare the PRCs obtained numerically by using the spike response model with the PRCs derived analytically. We then show the relation among the perturbation stimulus amplitude, the linear element and the higher-order nonlinear elements of the PRCs, and discuss the influence of the perturbation stimulus amplitude on the PRCs.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese   Publisher：The Physical Society of Japan (JPS)  

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：The Physical Society of Japan (JPS)  

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：The Physical Society of Japan (JPS)  

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

It has been reported that synaptogenesis, formation of synaptic connection, continues to take place in certain regions of the postnatal brain including the hippocampal regions. According to the previous neurophysiological experiment, synapses with smaller strength tend to be eliminated with higher probability. We investigate the effect of synaptogenesis on memories in the neural circuit, proposing the abstract neural network model, the Hopfield model with the zero-order synaptic decay. Using the numerical simulations, we demonstrate the possibility that synaptogenesis plays a role in maintaining recent memories embedded in the network while avoiding overloading. We also show the storage capacity of the zero-order decay model as a function of the decay rate a which corresponds to the number of replacement synapses. Furthermore, we extend the zero-order decay model to β-th-order, and investigate its characteristics by varying the order of the decay term, β. The results show that the characteristics of the β-th-order decay model are constant for a large order β.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：The Physical Society of Japan  

DOI： 10.11316/jpsgaiyo.67.1.2.0_312_1

researchmap

Language：Japanese   Publisher：The Physical Society of Japan  

CiNii Books

researchmap

Language：Japanese   Publisher：The Physical Society of Japan  

CiNii Books

researchmap

Language：Japanese   Publisher：The Physical Society of Japan  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

Dielectric properties of the brain tissue are important in understanding how neural activities are related to LFP or EEG. It has been well-known that the permittivity and conductivity of biological tissue show strong frequency dependence (dispersion), however, little is known as to what causes the large permittivity at the low-frequency region. We postulated that the electrically long cell may partially explain the dielectric properties of brain tissues. To test this idea, we analytically calculated the membrane potential and extracellular potential in response to the extracellular current stimulation using a model consisting of purely resistive media and a passive long cable. Our results show that, the tissue with passive cable in purely resistive extracellular media may have large effective permittivity at low-frequency region.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

In this paper, the sparsely encoded Hopfield model with unit replacement is analyzed by a statistical mechanical approach, self-consistent signal-to-noise analysis (SCSNA). Using the SCSNA and the computer simulation, we show the relation between the stoage capacity and the number of replacement units at each sparseness. Moreover, we compare the unit replacement procedure with the forgetting process, which avoids overloading as well as unit replacement. Results show that the unit replacement model has a finite value of the optimal sparseness to maximize the storage capacity for the large number of replacement units, although the forgetting process model does not.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：The Physical Society of Japan (JPS)  

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

We propose a statistical algorithm for extracting non-uniform dendritic membrane properties from partially observable noisy data such as voltage imaging data. Recent experimental studies suggest that dendritic computation plays an important role in information processing in local neural circuits. The non-uniform distribution of membrane properties over dendrite remains unclear, even though it determines how different kinds of information are integrated in neural systems. We formulate a state-space model for a multi-compartment model and derive a distributed-constant type Kalman filter by means of belief propagation. By using the derived Kalman filter and the EM algorithm, we show that our proposed algorithm can simultaneously estimate the spatiotemporal distribution of membrane potentials and the non-uniform distribution of membrane properties from partially observable noisy data.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

Calcium imaging enables simultaneous recording from multiple neurons and glial cells. As the amount of data obtained by calcium imaging increases, efficient methods for data analysis is required. Finding cells from the movie data (cell sorting) is the first step in analyzing the calcium imaging data from multiple cells. In a recent study, a robust method for cell sorting using independent component analysis (ICA) was developed. Here, we take a different approach and propose a simple cell detection method by fitting data to generative model, making use of non-negative matrix factorization (NMF). Our understanding is that such simple model-based approaches have not been applied to calcium imaging data due to the difficulty in handling baseline intensity caused by background signals and artifacts. Thus we modify the NMF algorithm to handle baseline intensity explicitly. We tested our algorithm on artificial movie data and evaluated its performance. We then applied the algorithm to the imaging data of CA1 region of rat hippocampal slices obtained by confocal microscopy system to show that our method can indeed detect cells.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：Japanese   Publisher：The Physical Society of Japan (JPS)  

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：The Physical Society of Japan (JPS)  

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：The Physical Society of Japan (JPS)  

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

Rodent hippocampal CA1 place cells are known as the cells which encode the location of moving rats by theta phase precession in their place fields. Recently, it has been reported that there is a distortion of time corresponding to the rat's location encoded by a phase of the theta rhythm. Itskov et al. statistically derived the time-offset curve (TOC) representing time-distortion as a function of the theta phase. In this study, we try to reproduce the TOC by using a neural oscillator model based on the phase reduction theory. Finally, we show that the TOC theoretically obtained with phase response curves of hippocampal CA1 pyramidal neurons gives qualitative agreement with the observed TOC.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Statistical methods for estimating dendritic membrane properties, including membrane resistance as a representative example, from observed noisy signals have been proposed in the last decade. Most of these methods assume membrane properties to be uniform over a dendritic tree. However, it is known that these properties are actually non-uniformly distributed, and even change steeply along a dendrite. In this study, we propose a statistical method for estimating membrane resistance distributions from observed membrane potentials, applicable to any distribution form. Dynamics of membrane potential is expressed in the cable equation, and membrane resistance distributions are expressed in the line process model. The line process model assumes a piecewise smooth distribution. Hence, even in the case where there are steep changes, our method can accurately estimate the membrane resistance distribution.

CiNii Books

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Statistical methods for estimating dendritic membrane properties, including membrane resistance as a representative example, from observed noisy signals have been proposed in the last decade. Most of these methods assume membrane properties to be uniform over a dendritic tree. However, it is known that these properties are actually non-uniformly distributed, and even change steeply along a dendrite. In this study, we propose a statistical method for estimating membrane resistance distributions from observed membrane potentials, applicable to any distribution form. Dynamics of membrane potential is expressed in the cable equation, and membrane resistance distributions are expressed in the line process model. The line process model assumes a piecewise smooth distribution. Hence, even in the case where there are steep changes, our method can accurately estimate the membrane resistance distribution.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

Animals perform various behavior according to the integration of information from the outside world. Information from the world is often masked with noise, therefore the brain must process such unreliable information. The problem of how does the robust processing is carried out in the brain is the fundamental issues for the study of neuroscience. To approach such problem, we investigated the effects of noise in visual stimulus on motion detection in Drosophila melanogaster. Recently, we suggested that the feedback to visual system from the central brain exist in Drosophila. To investigate this hypothesis, we examined optomotor responses to the stimulus with noise in short-term memory mutant(rutabaga). We observed that the reduction in the noise robustness were found in short-term memory mutant. Enhancement of optomotor responses to the stimulus with noise in wilde type were observed, but in not short-term memory mutant. Taken together, these results strongly support the hypothesis we have proposed. And we found a stable optomotor response in rutabaga. This should reflect a defective in a mechanism witch designed to transit among competing percepts. It suggest that the noise robustness is closely related to this mechanism.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

Animals continuously collect and process information on the enviroment, e.g. luminance, temperature and chemical concentration, and translate them into proper behavioral actions. In particular, chemotaxis is the phenomenon in which animals direct their movements in response to concentration gradient of chemoattractants. In this study, we statistically analyzed the chemotaxis mechanisms in Drosophila larvae by composing hidden malkov models describing discreate transitions between forward and rotary motions in the chemotaxis. Here, this model was built under the assumption that a transition probability from forward to rotary motions can be formulated as a formal neuron model containing a moving average (MA) filter entered time sequence of chemical concentrations. Then, by using chemotaxis data measured from CantonS (wild type) and Or83b (olfactory mutant), we estimated its model parameters with Baum-Welch algorithm and we selected the lenght of the MA filter with Akaike information criteria. In the case of wild types, the length of estimated filters was one or two, and espcially in second-order filters, time differential filters were extracted. On the other hand, in the case of olfactory mutants, the length of estimated filters was zero, which suggests that the mutants use no information of odorant.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

Animals continuously collect and process information on the enviroment, e.g. luminance, temperature and chemical concentration, and translate them into proper behavioral actions. In particular, chemotaxis is the phenomenon in which animals direct their movements in response to concentration gradient of chemoattractants. In this study, we statistically analyzed the chemotaxis mechanisms in Drosophila larvae by composing hidden malkov models describing discreate transitions between forward and rotary motions in the chemotaxis. Here, this model was built under the assumption that a transition probability from forward to rotary motions can be formulated as a formal neuron model containing a moving average (MA) filter entered time sequence of chemical concentrations. Then, by using chemotaxis data measured from CantonS (wild type) and Or83b (olfactory mutant), we estimated its model parameters with Baum-Welch algorithm and we selected the lenght of the MA filter with Akaike information criteria. In the case of wild types, the length of estimated filters was one or two, and espcially in second-order filters, time differential filters were extracted. On the other hand, in the case of olfactory mutants, the length of estimated filters was zero, which suggests that the mutants use no information of odorant.

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

Animals perform various behavior according to the integration of information from the outside world. Information from the world is often masked with noise, therefore the brain must process such unreliable information. The problem of how does the robust processing is carried out in the brain is the fundamental issues for the study of neuroscience. To approach such problem, we investigate the effects of noise in visual stimulus on motion detection in Drosophila melanogaster. We observed the optomotor response to the stimulus including noise as an index of motion detection. In addition, we calculated a response of the classical elementary motion detector (EMD) model and compared fly responses to the EMD model. We showed that when the noise exceed a certain level the response reduce adruptly, while in the case of EMD model, the response gradualy reduce. And we found that the presence of the fly which showed the more stable response to the stimulus with noise than that without noise. Taken together, we suggest that the feedback to visual system from the central brain exist in Drosophila.

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

Animals perform various behavior according to the integration of information from the outside world. Information from the world is often masked with noise, therefore the brain must process such unreliable information. The problem of how does the robust processing is carried out in the brain is the fundamental issues for the study of neuroscience. To approach such problem, we investigate the effects of noise in visual stimulus on motion detection in Drosophila melanogaster. We observed the optomotor response to the stimulus including noise as an index of motion detection. In addition, we calculated a response of the classical elementary motion detector (EMD) model and compared fly responses to the EMD model. We showed that when the noise exceed a certain level the response reduce adruptly, while in the case of EMD model, the response gradualy reduce. And we found that the presence of the fly which showed the more stable response to the stimulus with noise than that without noise. Taken together, we suggest that the feedback to visual system from the central brain exist in Drosophila.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：English   Publisher：日本神経化学会  

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese   Publisher：情報処理学会  

位相応答曲線は素子の集団的挙動を理解するために有効な手段の 1 つであり，神経細胞の位相応答曲線が，近年，生理学実験において測定されている．しかし，神経細胞の位相応答曲線と神経細胞の膜特性との関係は明らかになっていない．本論文において我々は，スパイクレスポンスモデルを用いて神経細胞の位相応答曲線を解析的に導出し，位相応答曲線と膜特性の関係を解析的に明らかにする．我々は位相応答曲線がスパイクレスポンスモデルの膜特性を表す関数で記述できることを示し，数値実験により検証する．Phase response curve (PRC) is one of useful tools to understand of population dynamics and has been measured in recent physiological experiments. However, the relation between subthreshold membrane property of a single neuron and its PRC is unclear. In this paper, we derive the PRCs of single neurons by using spike response model and analytically clarify the relation between the subthreshold membrane property and the PRC. We analytically show that the PRC is described by the kernels of the spike response model and verify our theory by numerical simulations.

CiNii Books

researchmap

Publisher：一般社団法人情報処理学会  

樹状突起における電気的な応答特性は,神経細胞が受け取る多種多様な情報の統合様式を決定づける重要な因子である.しかしながら,樹状突起の計測には,空間分解能の限界やノイズ重畳の問題が存在しており,樹状突起上の電気的応答特性や膜電位の時空間分布を直接的に測定するのは困難である.本研究では,樹状突起の数理モデルに基づいて分布定数型カルマンフィルタを構成することにより,樹状突起上の一部の観測データのみから樹状突起全体の膜電位の時空間分布を推定するアルゴリズムを提案する.さらに,膜電位の推定と並行して EM アルゴリズムを用いて,直接測定が困難な樹状突起の電気的応答特性を推定する.提案アルゴリズムの有効性を検証するために,提案アルゴリズムを数値実験データに適用し,観測データ数と推定精度の関係を検討する.

researchmap

Publisher：一般社団法人情報処理学会  

カルシウムイメージングには原理的に2つの問題がある.1 つは,観測値は蛍光強度であるため Ca2+ 濃度を定量化する場合,較正の手続きが必要となる.もう 1 つは,カルシウムイメージングは微弱な蛍光信号を扱う光学計測であるため,ノイズの影響を強く受ける.我々は,これらの問題を解決する統計的な枠組みを提案する.カルシウムイメージングにより得られる蛍光信号を状態空間モデルで記述し,粒子フィルタの数値計算手法を用いて蛍光信号から Ca2+ 濃度と流量を推定するものである.数値実験によって提案手法の有効性を示し,従来の較正手法との比較を行う.

researchmap

Publisher：一般社団法人情報処理学会  

位相応答曲線は素子の集団的挙動を理解するために有効な手段の 1 つであり,神経細胞の位相応答曲線が,近年,生理学実験において測定されている.しかし,神経細胞の位相応答曲線と神経細胞の膜特性との関係は明らかになっていない.本論文において我々は,スパイクレスポンスモデルを用いて神経細胞の位相応答曲線を解析的に導出し,位相応答曲線と膜特性の関係を解析的に明らかにする.我々は位相応答曲線がスパイクレスポンスモデルの膜特性を表す関数で記述できることを示し,数値実験により検証する.

researchmap

Language：Japanese   Publisher：情報処理学会  

樹状突起における電気的な応答特性は，神経細胞が受け取る多種多様な情報の統合様式を決定づける重要な因子である．しかしながら，樹状突起の計測には，空間分解能の限界やノイズ重畳の問題が存在しており，樹状突起上の電気的応答特性や膜電位の時空間分布を直接的に測定するのは困難である．本研究では，樹状突起の数理モデルに基づいて分布定数型カルマンフィルタを構成することにより，樹状突起上の一部の観測データのみから樹状突起全体の膜電位の時空間分布を推定するアルゴリズムを提案する．さらに，膜電位の推定と並行して EM アルゴリズムを用いて，直接測定が困難な樹状突起の電気的応答特性を推定する．提案アルゴリズムの有効性を検証するために，提案アルゴリズムを数値実験データに適用し，観測データ数と推定精度の関係を検討する．The membrane electric property of dendrites are important determinants for the integration of information received by neurons. However, the spatial resolution in experimental measurements of dendrites is limited, and data obtained for dendrites by means of voltage imaging are noisy. It is thus difficult to directly measure the membrane properties and the spatiotemporal distribution of membrane potentials. In the present study, we have developed a statistical algorithm with a distributed constant type Kalman filter for estimating the spatiotemporal distribution of membrane potentials from noisy data observed at limited locations. In conjunction with estimating the membrane potentials, we estimate the membrane electric properties of dendrites by using the expectation maximization algorithm. To evaluate this method we apply it to the simulation data and examine the relation between the number of observable points and the estimated accuracy.

CiNii Books

researchmap

Language：Japanese   Publisher：情報処理学会  

カルシウムイメージングには原理的に2つの問題がある．1 つは，観測値は蛍光強度であるため Ca2+ 濃度を定量化する場合，較正の手続きが必要となる．もう 1 つは，カルシウムイメージングは微弱な蛍光信号を扱う光学計測であるため，ノイズの影響を強く受ける．我々は，これらの問題を解決する統計的な枠組みを提案する．カルシウムイメージングにより得られる蛍光信号を状態空間モデルで記述し，粒子フィルタの数値計算手法を用いて蛍光信号から Ca2+ 濃度と流量を推定するものである．数値実験によって提案手法の有効性を示し，従来の較正手法との比較を行う．Calcium imaging method has two major problems. First, we need to calibrate the observation data to quantify the Ca2+ concentration, because the observation data are the fluorescent signals of Ca2+ indicator dyes that are only indirect information about the Ca2+ concentration. Second, the fluorescent signals are too noisy. To address these problems, we propose a statistical method for the calcium imaging. We make a nonlinear state-space model describing the fluorescent signals obtained by the calcium imaging method, and estimate its state vector consisting of the Ca2+ concentration and Ca2+ flux by a particle filter based on the state-space model we make. By numerical experiments, we verify the effectiveness of the proposed method, and compare our method with the conventional calibration method.

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

The measurement of synaptic current at the distal dendrite with the somatic voltage clamp has an error due to a failure of space clamp in the whole neuron. Williams & Mitchell quantified space-clamp errors by using the simultaneous recording from the soma and apical dendrite of rat cortical layer V pyramidal neurons. It has been reported that the electrotonic length of cortical layer V pyramidal neuron is about 1. Contrary to this, Williams & Mitchell showed the space-clamp error is larger than that of the electric cable with electrotonic length 1. Recent experimental and theoretical studies have suggested that the membrane resistivity of the distal apical dendrites of cortical layer V pyramidal neurons may be significantly lower than that of the proximal dendrites and the soma. Therefore, it is possible that the large space-clamp error is due to the low membrane resistivity at distal apical dendrites. In this paper, we formulate an electric behavior of the apical dendrite with the leaky distal part when clamping the somatic membrane potential as a finite passive cable equation with the fixed and leaky boundary conditions, and by employing Green function method we elucidate the effect of the low membrane resistivity at distal apical dendrites on the space-clamp error. The solutions for a cable with a leak at one end qualitatively reproduce the space-clamp errors measured by Williams & Mitchell, whereas those without the leak cannot reproduce them even its electrotonic length is adjusted.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

Recently, some statistical methods for estimating Ca^<2+> concentration from fluorescent signals of Ca^<2+> indicator dyes have been proposed. These statistical models are characterized by the parametric indefiniteness in common. In this paper, we propose a state space model with a hierarchical Bayes structure. We introduce hyperpriors for generating parameters of the state space model, resulting in the parametric uniqueness. We implement this hierarchical Bayes model with the particle filter method, and by applying to artificial and real fluorescent signals, we verify the effectiveness of the proposed method.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

Previously we reported that, in response to an applied extracellular DC field, the membrane potential near the soma changes with slow time course, and the potential in the apical dendrites shows biphasic change with with fast and slow time course. For AC field, the potential in the apical dendrites shows frequency dependent changes. In our recent modeling studies, we have shown that a leaky cable model with low membrane resistivity on one end mimics these behaviors. As an alternative account for the experimental results, we tested the impact of the morphology of the cells for DC field induced responses, and found that the morphology for AC field by performing compartmental model simulations, to find that the morphology also can account for the frequency dependent behavior of the dendrites.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：The Physical Society of Japan (JPS)  

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：The Physical Society of Japan (JPS)  

CiNii Books

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese   Publisher：情報処理学会  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：Japanese  

樹状突起における電気的な応答特性は，神経細胞が受け取る多種多様な情報の統合様式を決定づける重要な因子である．しかしながら，樹状突起の計測には，空間分解能の限界やノイズ重畳の問題が存在しており，樹状突起上の電気的応答特性や膜電位の時空間分布を直接的に測定するのは困難である．本研究では，樹状突起の数理モデルに基づいて分布定数型カルマンフィルタを構成することにより，樹状突起上の一部の観測データのみから樹状突起全体の膜電位の時空間分布を推定するアルゴリズムを提案する．さらに，EM アルゴリズムを用いて直接測定が困難な樹状突起の電気的応答特性を推定する．提案アルゴリズムの有効性を検証するために，数値実験データを提案アルゴリズムに適用し，観測データ数と推定精度の関係を検討する．Membrane property in dendrite is one of important determinants for integration of information received by neurons. However, it is difficult to directly measure the membrane property as well as spatiotemporal distribution of membrane potential, since there exists limitation of spatial resolution in experimental measurements of dendrites and the observed data from the dendrites by means of voltage imaging are noisy. In this study, we propose a statistical estimation algorithm using distributed constant type Kalman filter which estimates spatiotemporal distribution from noisy data observed at limited locations. Membrane properties of dendrites are estimated by using the EM algorithm. We verify the effectiveness of the proposed method by applying simulation data to the proposed method and evaluate the relation between the spatial resolution of observation and estimate error.

CiNii Books

researchmap

Language：Japanese   Publisher：情報処理学会  

位相応答曲線は素子の集団的挙動を理解するために有効な手段の 1 つであり，神経細胞の位相応答曲線が，近年，生理学実験において測定されている．しかし，神経細胞の位相応答曲線と神経細胞の膜特性との関係は明らかになっていない．本論文において我々は，スパイクレスポンスモデルを用いて神経細胞の位相応答曲線を解析的に導出し，位相応答曲線と膜特性の関係を解析的に明らかにする．我々は位相応答曲線がスパイクレスポンスモデルの膜特性をあらわす関数で記述できることを示し，数値実験により検証する．Phase response curve (PRC) is one of useful tools to understand of population dynamics and has been measured in recent physiological experiments. However, the relation between subthreshold membrane property of a single neuron and its PRC is unclear. In this paper, we derive the PRCs of single neurons by using spike response model and analytically clarify the relation between the subthreshold membrane property and the PRC. We analytically show that the PRC is described by the kernels of the spike response model and verify our theory by numerical simulations.

CiNii Books

researchmap

Language：Japanese  

カルシウムイメージングには原理的に 2 つの問題がある．1 つは，観測値は蛍光強度であるため Ca2+ 濃度を定量化する場合，較正の手続きが必要となる．もう一つは，カルシウムイメージングは微弱な蛍光信号を扱う光学計測であるため，ノイズの影響を強く受ける．我々は，これらの問題を解決する統計的な枠組みを提案する．カルシウムイメージングにより得られる蛍光信号を状態空間モデルで記述し，粒子フィルタの数値計算手法を用いて蛍光信号から Ca2+ 濃度と流量を推定するものである．数値実験によって提案手法の有効性を示し，従来の較正手法との比較を行う．Calcium imaging method has two major problems. First, we need to calibrate the observation data to quantify the Ca2+ concentration, because the observation data are the fluorescent signals of Ca2+ indicator dyes that are only indirect information about the Ca2+ concentration. Second, the fluorescent signals are too noisy. To address these problems, we propose a statistical method for the calcium imaging. We make a nonlinear state-space model describing the fluorescent signals obtained by the calcium imaging method, and estimate its state vector consisting of the Ca2+ concentration and Ca2+ flux by a particle filter based on the state-space model we make. By numerical experiments, we verify the effectiveness of the proposed method, and compare our method with the conventional calibration method.

CiNii Books

researchmap

Language：Japanese   Publisher：情報処理学会  

樹状突起における電気的な応答特性は，神経細胞が受け取る多種多様な情報の統合様式を決定づける重要な因子である．しかしながら，樹状突起の計測には，空間分解能の限界やノイズ重畳の問題が存在しており，樹状突起上の電気的応答特性や膜電位の時空間分布を直接的に測定するのは困難である．本研究では，樹状突起の数理モデルに基づいて分布定数型カルマンフィルタを構成することにより，樹状突起上の一部の観測データのみから樹状突起全体の膜電位の時空間分布を推定するアルゴリズムを提案する．さらに，EM アルゴリズムを用いて直接測定が困難な樹状突起の電気的応答特性を推定する．提案アルゴリズムの有効性を検証するために，数値実験データを提案アルゴリズムに適用し，観測データ数と推定精度の関係を検討する．Membrane property in dendrite is one of important determinants for integration of information received by neurons. However, it is difficult to directly measure the membrane property as well as spatiotemporal distribution of membrane potential, since there exists limitation of spatial resolution in experimental measurements of dendrites and the observed data from the dendrites by means of voltage imaging are noisy. In this study, we propose a statistical estimation algorithm using distributed constant type Kalman filter which estimates spatiotemporal distribution from noisy data observed at limited locations. Membrane properties of dendrites are estimated by using the EM algorithm. We verify the effectiveness of the proposed method by applying simulation data to the proposed method and evaluate the relation between the spatial resolution of observation and estimate error.

CiNii Books

researchmap

Language：Japanese   Publisher：情報処理学会  

位相応答曲線は素子の集団的挙動を理解するために有効な手段の 1 つであり，神経細胞の位相応答曲線が，近年，生理学実験において測定されている．しかし，神経細胞の位相応答曲線と神経細胞の膜特性との関係は明らかになっていない．本論文において我々は，スパイクレスポンスモデルを用いて神経細胞の位相応答曲線を解析的に導出し，位相応答曲線と膜特性の関係を解析的に明らかにする．我々は位相応答曲線がスパイクレスポンスモデルの膜特性をあらわす関数で記述できることを示し，数値実験により検証する．Phase response curve (PRC) is one of useful tools to understand of population dynamics and has been measured in recent physiological experiments. However, the relation between subthreshold membrane property of a single neuron and its PRC is unclear. In this paper, we derive the PRCs of single neurons by using spike response model and analytically clarify the relation between the subthreshold membrane property and the PRC. We analytically show that the PRC is described by the kernels of the spike response model and verify our theory by numerical simulations.

CiNii Books

researchmap

Language：Japanese   Publisher：情報処理学会  

カルシウムイメージングには原理的に 2 つの問題がある．1 つは，観測値は蛍光強度であるため Ca2+ 濃度を定量化する場合，較正の手続きが必要となる．もう一つは，カルシウムイメージングは微弱な蛍光信号を扱う光学計測であるため，ノイズの影響を強く受ける．我々は，これらの問題を解決する統計的な枠組みを提案する．カルシウムイメージングにより得られる蛍光信号を状態空間モデルで記述し，粒子フィルタの数値計算手法を用いて蛍光信号から Ca2+ 濃度と流量を推定するものである．数値実験によって提案手法の有効性を示し，従来の較正手法との比較を行う．Calcium imaging method has two major problems. First, we need to calibrate the observation data to quantify the Ca2+ concentration, because the observation data are the fluorescent signals of Ca2+ indicator dyes that are only indirect information about the Ca2+ concentration. Second, the fluorescent signals are too noisy. To address these problems, we propose a statistical method for the calcium imaging. We make a nonlinear state-space model describing the fluorescent signals obtained by the calcium imaging method, and estimate its state vector consisting of the Ca2+ concentration and Ca2+ flux by a particle filter based on the state-space model we make. By numerical experiments, we verify the effectiveness of the proposed method, and compare our method with the conventional calibration method.

CiNii Books

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

It is necessary to elucidate the relation between neuronal dynamics and membrane response property to understand the dynamics of neural networks. Spike Response Model is an analytically tractable neuron model describing membrane property by kernel functions. The phase response curve (PRC), on the other hand, is one of useful tools to understand of neuronal dynamics. In our previous study, we analytically showed the relation between subthreshold membrane response property of a single neuron and the PRC. In recent experimental studies, it has been shown that the PRCs show the different shape depending on firing frequency. The firing frequency may influence the shape of a PRC due to the existence of refractory period regarding subthreshold membrane potential. In this study, we analytically derive the relation between the neuronal refractory period of a single neuron and the PRC of the SRM, and show the influence of neuronal refractory period on the PRC. Furthermore we estimate the kernel of the SRM from a conductance based model, and compare the PRCs estimated from the kernel with the PRCs obtained numerically.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

It is necessary to elucidate the relation between neuronal dynamics and membrane response property to understand the dynamics of neural networks. Spike Response Model is an analytically tractable neuron model describing membrane property by kernel functions. The phase response curve (PRC), on the other hand, is one of useful tools to understand of neuronal dynamics. In our previous study, we analytically showed the relation between subthreshold membrane response property of a single neuron and the PRC. In recent experimental studies, it has been shown that the PRCs show the different shape depending on firing frequency. The firing frequency may influence the shape of a PRC due to the existence of refractory period regarding subthreshold membrane potential. In this study, we analytically derive the relation between the neuronal refractory period of a single neuron and the PRC of the SRM, and show the influence of neuronal refractory period on the PRC. Furthermore we estimate the kernel of the SRM from a conductance based model, and compare the PRCs estimated from the kernel with the PRCs obtained numerically.

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

Neurons can be interpreted as an encoder transforming some specific external stimuli into spikes, and the mechanism of this neural encoder obey the neural dynamics. To relate the encoding process to the properties of ion channels, some researchers have investigated the relation between the structures of the stimuli associated with neural spikes (STE: Spike Triggered stimulus Ensemble) and the neural dynamics. In this report, we formulate the nth moment of STE (MSTE_n) of the neurons spiking periodically with the phase response curves (PRCs), and propose a new calculation technique to derive MSTE_n based on the higher-order perturbation expansion. The following three results were obtained for neural oscillators. 1. The approximated expression relating STC to PRC was derived. 2. The relation between STA and STC were disclosed analytically. 3. It became clear what factor mainly breaks the approximate expression holding between STA and PRC, which had derived in the previous work.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

In both physiological and artificial conditions, the dendrites of neurons can be exposed to electrical fields. Recent experimental studies suggested that the membrane resistivity of the distal apical dendrites of cortical and hippocampal pyramidal neurons may be significantly lower than that of the proximal dendrites and the soma. To understand the behavior of dendrites in time varying extracellular electrical fields, we obtained analytical solutions of cable equations for finite cylindrical cables with and without a leak conductance attached to one end employing the Green's function method. The solution of a cable with leak at one end for DC electrical fields shows reversal of polarization at the leaky end as has been shown previously employing the method of variable separation and Fourier expansion. The solution of a cable with leak at one end for AC electrical fields shows that the leaky end shows frequency preference in response amplitude. Our results suggest that a passive dendrite with low resistivity at the distal end would show frequency preference in response to sinusoidal extracellular local field potentials. The Green's function obtained in our study can be used to calculate response for any forms of extracellular electrical fields.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

The ionic dynamics depending on the nervous activity recently attracts attentions of many researchers due to the rapid progress of Ion-Imaging techniques. However, there are several problems to be addressed in the ion-imaging techniques. In this paper, we developed statistical estimation methods for calcium ion dynamics from noisy fluoresce signals. First, we composed a state-space model describing calcium ion inflow kinetics, binding kinetics of calcium buffers and calcium indicators, and optical measuring devices for detecting fluorescent signals. Then, we derived an algorithm based on the belief propagation method for estimating calcium ion concentrations, and the EM algorithm for estimating model parameters. We numerically verified the efficacy of our algorithms by applying to artificial data. Our algorithm proposed here makes it possible to estimate not only intracellular calcium ion concentration but also calcium ion inflow through the cell membrane and Endoplasmic Reticula (ER) membrane et al. It is expected that the understanding for the molecular mechanisms such as the calcium channels and Ca store can be promoted by our method, since variation of calcium inflows give the significant information these molecular mechanisms.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：The Physical Society of Japan (JPS)  

CiNii Books

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese   Publisher：The Physical Society of Japan (JPS)  

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：The Physical Society of Japan (JPS)  

CiNii Books

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

There is a critical memory capacity in the Hopfield model. If the number of embed patterns surpasses the crirical number, the network cannot retrieve all patterns due to the overloading. On the other hands, It has been reported that the network, in which a number of units are replaced by newborn ones when the networks learn, can keep up memory patterns studied recently without the overloading. We analyze the network with replacing units by utilizing statistical mechanics. We show that the critical number of replacing units is R_c=3.2 avoiding the overloading and independent of the network size. In addition, we demonstrate that there is the optimal number of replacing units to maximize the memory capacity. Finally, we compare a model with forgetting process with one with replacing process.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese   Publisher：The Physical Society of Japan (JPS)  

CiNii Books

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

We proposed a top-down framework based on the statistical mechanics to predict phase response curves (PRCs) of hippocampal CA3 pyramidal neurons. First, we hypothesized that the hippocampal CA3 network computationally works as an associative memory, and introduced a minimum model of the oscillator associative memory network consisting of time window functions of the spike timing-dependent synaptic plasticity (STDP) and PRCs, which fulfills physical limitations. We derived a free energy of this system by use of analytical approaches of the statistical mechanics. Then, by minimizing the free energy under a condition restrained by a given STDP time window function, we obtained an optimal PRC for the associative memory retrieving phase information embedded by the STDP. If a PRC predicted by the top-down framework had a resemblance to that of real pyramidal neurons, the hippocampal CA3 network might be optimized for working as the associative memory.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

Phase response curve (PRC) is important to clarify the interaction between neurons, which show oscillatory activities, such as hippocampal CA3 pyramidal cells. A linear element and a higher-order nonlinear element exist together in observed PRC because the amplitude of the perturbation stimulus is finite in physiological experiment. It is necessary to separate a linear element and a higher-order nonlinear element of PRC clearly in order to understand the behavior of dynamical systems of the networks, which have strong interaction between neurons. In this study, we analytically obtained the PRC of spike response model, in order to clarify the relation among the electrophysiological response property of single neuron, the linear element, and the higher-order nonlinear element of the PRC. Therefore, we can estimate the linear element and higher-order nonlinear element of the PRC based on linear response in subthreshold membrane potential, using the PRC of spike response model.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

To reveal mechanism of information processing in dendrite, we need to elucidate how membrane potential is distributed spatially and temporally over the dendrite. However there exists limitation of spatial resolution in experimental measurements of membrane potentials, such as multiple whole cell recordings and optical recording with voltage sensitive dyes. In this report, we propose a statistical estimation method using distributed constant type Kalman filter in order to estimate spatiotemporal distribution over the dendrite using experimental data of membrane potential observed at limited locations of dendrite. The application of the proposed method to simulation data of compartment models shows that spatiotemporal distribution can be estimated using simulation data of partially observable membrane potential in dendrite.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

We constructed a real-time measurement control system for the Dynamic Clamp method. The dynamic clamp enables us to realize real-time interaction between the living cell and the numerical model. For instance, we can create virtual local neuronal circuit of living cells connected by artificial synapses. Therefore we can expect the dynamic clamp mediates between theoretical and practical researches. In this paper, we built a Real-Time Linux based system, and carried out two kinds of electrophysiological experiments with the dynamic clamp system built here. First, by applying the Exponential Euler method, we realized real-time simulations of a Hodgkin-Huxley type model interacting with a real neuron. Next, we tried to measure phase response curves.In this experiment, it is successful to maintain a constant firing frequency. And our system contributed to progress of experimental data stock, and accuracy and efficiency of electrophysiological experiments.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

In hippocampal CA1 pyramidal cell,it has been shown experimentally that exteracellular electrical field applied parallel to the somato-dendritic axis induces unique changes in membrane potential. For example,extracellular DC field causes a monotonic change with slow time course near the cell body and biphasic changes in distal dendrite.A compartmental cable model with low membrane resistivity on one end shows similar profile of membrane potential change in response to DC field.To better understand the behavior of the model, here we tried to obtain solutions for the cable equation in which an extracellular electrical field is applied to a cable with a leaky end.We adapted Green's function method to solve the equation, and deviced boundary conditions to describe the leakiness on one end of the cable.The solution analytically shows the numerically calculated features of the cable in the DC field.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

Recently, phase response curves (PRCs) of various neurons have been estimated by perturbation-response experiments. We have proposed a Bayesian statistical method for estimating PRCs from physiological data fluctuated by noise. In this method, we can identify Langevin equations representing phase motion, because the probabilistic model assumed in our method is based on Langevin equations. Therefore, Fokker-Planck (FP) equations based on PRCs can predict the stochastic behavior of neuron perturbed by arbitrary inputs. Here we carried out the following experiments with the dynamic clamp. First, we estimated a PRC and intensity of noise by perturbing a hippocampal CA1 pyramidal neuron, and we obtained a FP equation based on the PRC. Next, by loading a periodic perturbation onto the same neuron, we measured a histogram of phase differences between the perturbation and the neural spike. The histogram was coincident with an equilibrium distribution of the FP equation.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese  

J-GLOBAL

researchmap

Language：Japanese   Publisher：The Physical Society of Japan (JPS)  

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：The Physical Society of Japan (JPS)  

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：The Physical Society of Japan (JPS)  

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：English   Publishing type：Research paper, summary (international conference)  

Web of Science

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：English   Publishing type：Research paper, summary (international conference)  

Web of Science

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

Non-uniform membrane property in dendrite is one of important determinants for manner of information integration in neurons, and therefore its distribution is important for information processings in brain. It is known that in the distal dendrite, many synaptic inputs are received and important information processings such as synaptic plasticity are performed. However, membrane property in distal dendrite is unclear since its direct measurements are difficult. To estimate membrane property in distal dendrite, we performed numerical simulation of membrane responses to perturbation to single neuron, using compartment model with non-uniformity of membrane property, and we compared numerical results with those observed in the hippocampal CA1 slices using optical recordings. Estimated membrane resistance at distal dendrite is lower than that at proximal dendrite and it was suggested that there is a steep decrease in the passive membrane property between distal and proximal dendrites.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

We propose an algorithm to estimate PRCs in the Bayesian framework. Firstly, we described analytically deterioration process for observed date by solving Fokker-Planck equation. This deterioration process enables us to discuss what perturbation dominantly retains linear response characteristic. Then, we derived reversal process for observation process by Bayes' theorem, and estimated a pure PRCs by a maximum a posteriori (MAP). Power spectrum constrain allows noise to operate multiplicatively to the system. Simulation by Morris-Lecar Neuron Model suggests our method estimates an optimal PRCs of In-Vitro neuron, even if the noise of neuron is unknown.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

Dendrites of hippocampal CA1 pyramidal neurons receive synaptic inputs from the entorhinal cortex (EC) and hippocampal CA3, and these two kinds of synaptic inputs are received separately in those dendrites; inputs from the EC are received at distal dendrite, whereas those from the CA3 are received at proximal dendrite. It has been suggested that it is difficult to induce somatic firing just with synaptic inputs to distal dendrite. In previous studies, Omori et al. estimated that distribution of specific membrane resistance, one of passive membrane properties, obeys step function; specific membrane resistance in distal dendrite is relatively small compared with that in proximal dendrite. By analysis of cable equation and numerical simulation of compartment model, we show that signal transfer from distal dendrite to soma is significantly improved due to non-uniformity of the specific membrane resistance obeying step function. Our results suggest that somatic firing can be induced even just with synaptic inputs to distal dendrite by the estimated non-uniform distribution of specific membrane resistance.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：English  

DOI： 10.1016/j.brainres.2006.09.095

Web of Science

researchmap

Language：English  

DOI： 10.1143/JPSJ.75.114802

Scopus

researchmap

Language：English  

DOI： 10.1143/JPSJ.75.114802

Web of Science

researchmap

Language：Japanese   Publisher：一般社団法人情報処理学会  

The phase response curves (PRCs) represennt an impulse response of oscillatory system to capture an essence of a nonequilibrium dynamics induced by a small disturbance. Wheareas measuring PRCs is important for bridging single neuron dynamics and network dynamics, estimating methods for PRCs have not yet been established. In this paper, we proposed a Bayesian approach to estimate PRCs from noisy data measured by perturbation-response experiments. First, we analyzed the stochastic process describing the observation process in perturbation-response experiments, and obtained a probability distribution of the deterioration process of PRCs. Then, by introducing a prior generating PRCs, we proposed the Maximum A Posteriori (MAP) estimation algorithm for PRCs.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

The phase response curves (PRCs) represennt an impulse response of oscillatory system to capture an essence of a nonequilibrium dynamics induced by a small disturbance. Wheareas measuring PRCs is important for bridging single neuron dynamics and network dynamics, estimating methods for PRCs have not yet been established. In this paper, we proposed a Bayesian approach to estimate PRCs from noisy data measured by perturbation-response experiments. First, we analyzed the stochastic process describing the observation process in perturbation-response experiments, and obtained a probability distribution of the deterioration process of PRCs. Then, by introducing a prior generating PRCs, we proposed the Maximum A Posteriori (MAP) estimation algorithm for PRCs.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人電子情報通信学会  

Dendritic membrane properties have been reported to be non-uniformly distributed in a single neuron. However, its distribution is still unclear since measurement of membrane properties in the whole single neuron is difficult to be performed. To reveal non-uniform membrane properties, we estimated distribution of specific membrane resistance by fitting multi-compartment model to voltage imaging data, which are supposed to be associated with the non-uniformity of the resistance, specifically, propagating EPSP induced by synaptic inputs, and also biphasic response to extracellular electric field. By numerical simulation, we found that these two optical recording data were consistently reproduced for sigmoid function. This implies that a steep decrease of specific membrane resistance exists in distal dendrite of hippocampal CA1 pyramidal neuron.

CiNii Books

researchmap

Language：Japanese   Publisher：日鉄技術情報センタ-  

We investigated the role of A-type K channels in the neuronal dendritic function. In the first half, the contribution of these channels to the induction of long-term potentiation (LTP) in hippocampal CA1 pyramidal neurons was examined experimentally. The results supported the hypothesis that A-type K channels contribute to the induction of LTP through controlling the dendritic intergration properties. In the second half, by using the NEURON simulator, we tried to elucidate mechanisms of boosting for EPSP amplitudes and inhibiting for ones. Here, it was theoretically demonstrated that the super-sensitivity, which can be controlled by the down-regulation of the A-type K channel, is the fundamental mechanism for regulation of such membrane responses.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：The Institute of Electronics, Information and Communication Engineers  

Dendritic membrane properties have been reported to be non-uniformly distributed in a single neuron. However, its distribution is still unclear since measurement of membrane properties in the whole single neuron is difficult to be performed. To reveal non-uniform membrane properties, we estimated distribution of specific membrane resistance by fitting multi-compartment model to voltage imaging data, which are supposed to be associated with the non-uniformity of the resistance, specifically, propagating EPSP induced by synaptic inputs, and also biphasic response to extracellular electric field. By numerical simulation, we found that these two optical recording data were consistently reproduced for sigmoid function. This implies that a steep decrease of specific membrane resistance exists in distal dendrite of hippocampal CA1 pyramidal neuron.

CiNii Books

J-GLOBAL

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

DOI： 10.1143/jpsj.75.114802

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人 日本物理学会  

DOI： 10.1143/jpsj.75.114802

CiNii Books

researchmap

Language：Japanese   Publisher：日本神経回路学会  

脳研究においては，実験家と理論家の間でモデルや理論の共有が出来ていない．脳はあまりにも複雑すぎて，研究者間で統一した記述のモデルが存在しない．また，脳を解析するために利用できる理論が存在する．しかし，モデルが共有できていない状況において，モデルに強く異存した理論を共有するのは難しい．この状況を打破してくれるのが，NEURONシミュレータをはじめとするプラットフォームシミュレータである．本稿では，NUERONシミュレータを介してモデルと理論的な自由度低減法である摂動法を共有して，理論家と実験家の協力で行っている研究を詳解する．今まで知られていない活動電位発生メカニズムや樹状突起の逆伝搬活動電位のシナプス情報統合における役割を述べる．

DOI： 10.3902/jnns.12.100

CiNii Books

J-GLOBAL

researchmap