Abstract
Bibliobattle is a social book review game spreading in Japan recently. A presentation time limit for a Bibliobattle session is fixed at five minutes. This time limit was determined based on experiences of the founders who developed Bibliobattle. They said that the time limit is neither too short nor too long. However, the time limit has not been empirically investigated by comparing different time limits. In this paper, we investigate what kind of influence different time limits have on a presenter's speech in a Bibliobattle session. We conducted an experiment. For each experimental condition, we changed the time limit of the presenter's speech. We compared the book presenter's speech using a discourse analysis technique. It is proved that the ratio of utterance category is stable in different time limits. Additionally, we asked the presenters and audiences to answer several questionnaires to investigate the effects of different time limits. Most of all the questionnaries showed that the 3-, 5- and 7-minute time limits are not significantly different from each other. In contrast, there's a significant difference between 1 and 10 minutes. It was also revealed that a presenter and audiences have different feelings for the temporal differences in the presentation time limits of a Bibliobattle session. Audiences are more sensitive to the difference of time limit than presenters.
