The Dynamics of Bimodular Continuous Attractor Neural Networks with Moving Stimuli

Min Yan, Wen Hao Zhang, He Wang, K. Y.Michael Wong

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The single-layer continuous attractor neural network (CANN) model has been applied successfully to describe the tracking of moving stimuli of a single modality. Experimental evidence shows that stimuli of different modalities interact with each other in the neural system. To study these interaction effects, we generalize the single-module structure to a bimodular one. We found that when there is one static stimulus in one module and a moving one in the other, the network have very different behaviours depending on whether the inter-modular couplings are excitatory or inhibitory. We further compare the model with experimental observations that illustrate the interactions between two sensory modalities, such as the motion-bounce Illusion. Agreement between model and experimental results can be obtained for appropriate choice of parameters.

Original languageEnglish (US)
Title of host publicationNeural Information Processing - 24th International Conference, ICONIP 2017, Proceedings
EditorsDerong Liu, Shengli Xie, Yuanqing Li, El-Sayed M. El-Alfy, Dongbin Zhao
PublisherSpringer Verlag
Pages648-657
Number of pages10
ISBN (Print)9783319700922
DOIs
StatePublished - 2017
Externally publishedYes
Event24th International Conference on Neural Information Processing, ICONIP 2017 - Guangzhou, China
Duration: Nov 14 2017Nov 18 2017

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10637 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference24th International Conference on Neural Information Processing, ICONIP 2017
Country/TerritoryChina
CityGuangzhou
Period11/14/1711/18/17

Keywords

  • Continuous attractor neural networks
  • Motion-Bounce Illusion
  • Multisensory information processing

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint

Dive into the research topics of 'The Dynamics of Bimodular Continuous Attractor Neural Networks with Moving Stimuli'. Together they form a unique fingerprint.

Cite this