Year: 2025
Type: dissertation
Abstract: note learning is rewriting notes while reviewing the lecture's content postclass.Learners can deepen their understanding and remember the content through writing.However, there are problems such as a decline in motivation and a decrease in the quality of notes because learners usually take notes alone.Many studies have pointed out that interventions during writing summary notes are essential, but previous methods required other's labor and were less accessible for interventions.In this research, I implemented a note-taking system with writing assistants using generative AI and conducted a user study to survey whether supporting summary note-taking by assistants' interventions is effective.The writing assistants generate explanations or questions based on the content in learners' summary notes and of the lecture.The user study was a between-subjects experiment with three groups -a group with the assistant that generates explanations (the explaining group), a group with the assistant that generates questions (the questioning group), and a group without assistants (the control group).They reviewed their notes and took a test one week later.I did not find significant differences in almost all the quantitative results, such as the number of characters, the quality of summary notes, and the test scores across the three groups.However, through interviews with the participants, I found positive and negative points about the writing assistants and the format of explanations and questions.Thanks to the assistants, participants noticed missing points in their notes and felt they understood the lecture content well.On the other hand, they could not concentrate on note-taking and remember the content the assistants did not men-v tion.Finally, I described the design implications of writing assistants for summary note-taking.概要
Author: Katori, Hiroki
Institution: The University of Tokyo
Language: English
Subfield: Education
Field: Social Sciences
Domain: Social Sciences
SDG: Quality education
Open Access status: green
OpenAlex ID: W7146407578
