Experimental econophysics of social learning and collective intelligence

2015 Fiscal Year Final Research Report

• Project/Area Number: 25610109
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Mathematical physics/Fundamental condensed matter physics
• Research Institution: Kitasato University
• Principal Investigator: Mori Shintaro 北里大学, 理学部, 講師 (70296424)
• Co-Investigator(Kenkyū-buntansha): TAKAHASHI TAIKI 北海道大学, 文学研究科, 准教授 (60374170)
• Co-Investigator(Renkei-kenkyūsha): HISAKADO MASATO 金融庁, 検査局, 主任専門検査官 (60623923)
• Research Collaborator: HINO MASAFUMI 株式会社NEC
• Project Period (FY): 2013-04-01 – 2016-03-31
• Keywords: 相転移 / 情報カスケード / 普遍クラス / 集団実験 / 集合知 / 多腕バンディット

Outline of Final Research Achievements

It has been considered that the social learning where agents get information from the behaviors and choices of other agents is the most important factor in the description and the understanding of social and economical phenomena. We have controlled the uncertainty of the information of human collectives in an information cascade experiment and showed that there occurs a phase transition of
non-linear Polya urn. Theoretically, we clarify when the phase transition becomes continuous or discontinuous and the universality class of the former continuous one. Furthermore, we study several ideas how to increase the accuracy of collective intelligence in information cascade.

Free Research Field

統計物理学