Investigation of a method for extracting events that contribute to daily emotional experiences by applying natural language processing techniques

2021 Fiscal Year Final Research Report

• Project/Area Number: 20K22283
• Research Category: Grant-in-Aid for Research Activity Start-up
• Allocation Type: Multi-year Fund
• Review Section: 0110:Psychology and related fields
• Research Institution: Osaka University
• Principal Investigator: Muranaka Seiji 大阪大学, 人間科学研究科, 助教 (90878349)
• Project Period (FY): 2020-09-11 – 2022-03-31
• Keywords: 自然言語処理 / 情動制御 / トピックモデル

Outline of Final Research Achievements

This study aims to develop a methodology for identifying the relationship between the content of everyday event descriptions and the intensity of the emotional experiences generated by those events. We hypothesized that patterns that can be extracted from the descriptions and frequency of occurrence of certain topics would be related to the intensity of the emotional experiences. A total of 263 Japanese participated this study. Participants were asked to report the events of the day for 14 days using free descriptions and a Likert-type description of the emotions they experienced during the events. The free descriptions were converted into features using a topic model, and network analysis was conducted. The results suggested that those who were more likely to criticize others reported more about their leisure time on diaries and, thus, may be less likely to engage in positive reappraisal.

Free Research Field

臨床心理学

Academic Significance and Societal Importance of the Research Achievements

これまでに心理学研究で扱われてきた測定法で得られてきた人の体験を、表出行動の一つである言語的な情報から直接数値的な情報に変換でき、解析できる可能性を示した点では、学術的意義はあると考える。特に、患者等との対話の内容とその効果について深掘りしたい臨床心理学的な研究の発展に寄与するものと期待できる。
現在、コロナ禍により、心理支援を遠隔で実施するニーズとそれに応える動きが活発となっている。その中で、情報科学技術を活用した支援のあり方は注目が集まっている。本研究は、遠隔で支援する上で、患者の体験の記録を解析する一手法を提案するものとして、社会的意義も十分にあると考える。