Dynamical Reorganization of Attractor Structure in Neural Networks with Dynamic Synapses

  • Katori Yuichi
    FIRST, Aihara Innovative Mathematical Modelling Project, Japan Science and Technology Agency
  • Aihara Kazuyuki
    Institute of Industrial Science, The University of Tokyo

書誌事項

公開日
2014-03-17
DOI
  • 10.15248/proc.1.602
公開者
The Institute of Electronics, Information and Communication Engineers

説明

We investigate the dynamical properties of a neural network with dynamic synapses, whose transmission efficacy is modulated by short-term plasticity, and we use a mean field model that approximates the population dynamics of spiking neurons. In particular, we consider a neural network with recurrent connections via depression and facilitation synapses, and we analyze the influence of synaptic modulation on the dynamics of synaptic activity via slow-fast analysis with time-parameterized bifurcation parameters. The model is described by three variables: one represents synaptic activity, whereas the other two represent modulation in synaptic transmission efficacy. The variables that represent the synaptic modulation can be considered as slow variables that affect the properties of synaptic activity, which can be regarded as the fast variable. The analysis indicates that an attractor in the fast system corresponding to an active state of the neural network appears or disappears according to the activities of the neural network. The concept of dynamical reorganization of the attractor structure can potentially uncover the mechanism of flexible brain functions.

収録刊行物

  • IEICE Proceeding Series

    IEICE Proceeding Series 1 602-605, 2014-03-17

    The Institute of Electronics, Information and Communication Engineers

参考文献 (8)*注記

もっと見る

キーワード

詳細情報 詳細情報について

  • CRID
    1390283687150831744
  • DOI
    10.15248/proc.1.602
  • ISSN
    21885079
  • 本文言語コード
    en
  • データソース種別
    • JaLC
    • Crossref
    • OpenAIRE
  • 抄録ライセンスフラグ
    使用不可

問題の指摘

ページトップへ