Polymorphic Odor Information Representation Controlled by Sniffing-Coupled Rhythmic Input

2015 Fiscal Year Final Research Report

• Project/Area Number: 25330345
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Life / Health / Medical informatics
• Research Institution: Tohoku Gakuin University
• Principal Investigator: MAKINO Yoshinari 東北学院大学, 教養学部情報科学科, 准教授 (90250844)
• Project Period (FY): 2013-04-01 – 2016-03-31
• Keywords: 嗅覚システム / 情報表現 / リズム / 神経ネットワーク / 匂い / 時空間パターン

Outline of Final Research Achievements

Sniffing is an active sensing behavior strongly controlled depending on behavioral and stimulus context. Sniffing-coupled sub-threshold electrical activity is found in neurons of rodent olfactory bulb, and suggested to affect odor information representation. Here, we hypothesized that controlling the sniffing would be helpful in getting the odor information with different resolution, and examined it by network model simulation. Simulation results indicated following two points: in the level of single neuron, the sub-threshold rhythmic activity makes input-latency relationship to be linear in wide range of input strength, and by changing temporal features of the sub-threshold activity, the input-latency relationship vary in a regular way; on the level of neural network, controlling the sniffing-coupled input would play a key role in concentration invariance and speed-accuracy tradeoff in the odor information processing.processing.

Free Research Field

生物情報学