A dynamical model of visual motion processing for arbitrary stimuli including type II plaids.
Computational neuroscience
Dynamical model
Eye movement
Motion perception
Optimization
Journal
Neural networks : the official journal of the International Neural Network Society
ISSN: 1879-2782
Titre abrégé: Neural Netw
Pays: United States
ID NLM: 8805018
Informations de publication
Date de publication:
May 2023
May 2023
Historique:
received:
15
05
2022
revised:
23
02
2023
accepted:
25
02
2023
medline:
25
4
2023
pubmed:
7
3
2023
entrez:
6
3
2023
Statut:
ppublish
Résumé
To explore the operating principle of visual motion processing in the brain underlying perception and eye movements, we model the information processing of velocity estimate of the visual stimulus at the algorithmic level using the dynamical system approach. In this study, we formulate the model as an optimization process of an appropriately defined objective function. The model is applicable to arbitrary visual stimuli. We find that our theoretical predictions qualitatively agree with time evolution of eye movement reported by previous works across various types of stimulus. Our results suggest that the brain implements the present framework as the internal model of motion vision. We anticipate our model to be a promising building block for more profound understanding of visual motion processing as well as for the development of robotics.
Identifiants
pubmed: 36878170
pii: S0893-6080(23)00110-7
doi: 10.1016/j.neunet.2023.02.039
pii:
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
46-68Informations de copyright
Copyright © 2023 Elsevier Ltd. All rights reserved.
Déclaration de conflit d'intérêts
Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.