2018 Fiscal Year Final Research Report

Analysing risk perception by latent semantic analysis model and application to risk communication

Research Project

PDF

Project/Area Number	16K17290
Research Category	Grant-in-Aid for Young Scientists (B)
Allocation Type	Multi-year Fund
Research Field	Social psychology
Research Institution	University of Shizuoka (2017-2018) The University of Tokyo (2016)
Principal Investigator	Tamari Yuki 静岡県立大学, 経営情報学部, 講師 (60737360)
Research Collaborator	TAKEMURA kazuhisa IDENO takashi
Project Period (FY)	2016-04-01 – 2019-03-31
Keywords	リスク認知 / 意思決定 / 過程追跡法 / 言語プロトコル法 / 決定方略
Outline of Final Research Achievements	Structures of risk perception in the decision-making process were considered from the perspective of decision strategies. Verbal protocol data were analyzed by using probabilistic latent semantic indexing and computer simulation of multi-attribute decision making. Topics assumed behind verbal protocol data were extracted, then decision strategies were estimated in each extracted topic. A combination of decision strategies, which is the most similar to the observed term frequency, was explored by using Jensen Shanon divergence. The results suggested that the proposed technique would reveal the structure of risk perception from the viewpoint of decision strategies. This method would facilitate risk communication and consensus building.
Free Research Field	社会心理学
Academic Significance and Societal Importance of the Research Achievements	これまで，意思決定の状況依存性を記述する様々な決定方略が見いだされてきたが，決定方略の推定方法は確立されておらず，定性的記述が主であると言える。本分析枠組みを用いることで、プロセスデータなどの背後にある情報処理の類型を、定量的に検討できる可能性が示唆された。また、リスクコミュニケーションにおいては、自分自身と相手のリスク認知の構造を把握することが肝要であると言える。本分析枠組みを用いることで、自分自身の情報処理の類型を、複数のトピック毎に把握できる可能性がある。これらの情報を呈示することで、リスクコミュニケーションにおける相互理解および合意形成の一助となることが期待される。