Dinamical neuronal modeling of insect navigation based on polarization vision

2017 Fiscal Year Final Research Report

• Project/Area Number: 15KT0106
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 特設分野
• Research Field: Mathematical Sciences in Search of New Cooperation
• Research Institution: Kobe University
• Principal Investigator: SAKURA MIDORI 神戸大学, 理学研究科, 准教授 (60421989)
• Co-Investigator(Kenkyū-buntansha): 太田 泰広 神戸大学, 理学研究科, 教授 (10213745) ; 福山 克司 神戸大学, 理学研究科, 教授 (60218956) ; 岡田 龍一 兵庫県立大学, 環境人間学部, 研究員 (20423006)
• Research Collaborator: HENZE Miriam J. University of Queensland, QBI, 研究員
• Project Period (FY): 2015-07-10 – 2018-03-31
• Keywords: ニューラルネットワーク

Outline of Final Research Achievements

It is well known that many insects deduce their heading direction using polarized skylight. However, it is still unclear how the polarized skylight information, which forms complicate pattern in the sky and even heavily depending on solar azimuths and/or weather conditions, is coded by neural networks in the brain. In this study, we tried to clarify this by dynamical neural modeling using biological and astronomical data.
We first analyzed input manners of the polarized light information by a sensor map constructing from the morphological data of the polarized light-detecting area of the compound eye (DRA). Moreover, we tested the accuracy in orientation outputs of the neural network model constructed by the response properties of polarization-sensitive neurons in the brain.

Free Research Field

神経行動学