Demonstration and extension of Mizunami-Unoki learning model.

2014 Fiscal Year Final Research Report

• Project/Area Number: 24370030
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Partial Multi-year Fund
• Section: 一般
• Research Field: Animal physiology/Animal behavior
• Research Institution: Hokkaido University
• Principal Investigator: MIZUNAMI Makoto 北海道大学, 理学(系)研究科(研究院), 教授 (30174030)
• Project Period (FY): 2012-04-01 – 2015-03-31
• Keywords: 学習 / 記憶 / 予測誤差 / ドーパミン / オクトパミン / 昆虫 / 視覚 / 嗅覚

Outline of Final Research Achievements

I obtained three major results in this study. At first, I demonstrated the validity of the prediction error theory. In associative learning in mammals, it is accepted that the error between actual and predicted reward determines whether learning occurs: This theory stems from the finding of a blocking phenomenon, but blocking can also be accounted for by other theories. I demonstrated blocking in classical conditioning in crickets and obtained evidence to support the prediction error theory. Second, I found evidence of “habit formation”, which make the conditioned response insensitive to devaluation of the US. The results demonstrate, for the first time, habit formation in classical conditioning in any animals. Thirdly, I performed intracellular recordings from many of output neurons of the lobes of the mushroom body in cockroaches. Some of these neurons exhibited changes in their responses to an odor presentation after pairing of the odor with sucrose reward.

Free Research Field

行動生理学