This paper proposes the use of crossword puzzles to promote serendipitous learning, defined here as learning that emerges when system design implicitly encourages users to voluntarily engage in additional information seeking. We prepared a game environment in which crossword puzzles are presented on demand from newspaper article data from the past ten years, and players can browse the corresponding articles while solving the puzzles. Thirty participants were asked to freely play crossword puzzles, and a few days later, they individually answered personalized 4-choice quizzes generated according to their actual puzzle play. Interestingly, for quiz items related to words that participants failed to input correctly even though they had viewed the corresponding newspaper articles during puzzle play, the correct answer rate was higher than for quiz items generated under other conditions. In the questionnaire conducted together with the 4-choice quizzes, several participants reported that they had voluntarily looked up the words they could not solve in the puzzles, which is consistent with this result. Furthermore, focusing on quizzes related to newspaper articles viewed during puzzle play, we confirmed that participants who reported engaging in self-directed learning achieved higher accuracy than those who did not.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Serendipitous Learning Triggered by Crossword Puzzles with Newspaper Articles

  • Yasuyuki Sumi,
  • Ryusuke Hioki

摘要

This paper proposes the use of crossword puzzles to promote serendipitous learning, defined here as learning that emerges when system design implicitly encourages users to voluntarily engage in additional information seeking. We prepared a game environment in which crossword puzzles are presented on demand from newspaper article data from the past ten years, and players can browse the corresponding articles while solving the puzzles. Thirty participants were asked to freely play crossword puzzles, and a few days later, they individually answered personalized 4-choice quizzes generated according to their actual puzzle play. Interestingly, for quiz items related to words that participants failed to input correctly even though they had viewed the corresponding newspaper articles during puzzle play, the correct answer rate was higher than for quiz items generated under other conditions. In the questionnaire conducted together with the 4-choice quizzes, several participants reported that they had voluntarily looked up the words they could not solve in the puzzles, which is consistent with this result. Furthermore, focusing on quizzes related to newspaper articles viewed during puzzle play, we confirmed that participants who reported engaging in self-directed learning achieved higher accuracy than those who did not.