(Salle F104)

---

  Tanaka, Toyotomi (Nagoya University, Nagoya)
  Arakawa, Ayumu (Ritsumeikan University, Kyoto)
  Saito, Hirofumi (Nagoya University, Nagoya)
  Oi, Misato (Nagoya University, Nagoya)

Are Beat Gestures Contingent upon Disfluent Speech in the Animation-narration Task ?

Saito et al. (2004) indicated that beat gestures could be produced to recover disfl uent speech. Further, Tanaka et al. (2004) confi rmed that the number of fillers (e.g., « uh ») shows the same tendency as beat gestures. The purpose of this study is to investigate the functional relationship between beat gestures and fi llers in terms of the number of beat gestures produced before or after each fi ller. Eighteen university students participated in the experiment. They were required to watch all four parts of an animation, and then to narrate each part of the content to a video camera (half part are under Delayed Auditory Feedback / DAF condition, and the rest are under Natural Auditory Feedback / NAF). Half of them were instructed to produce gestures, and the remaining half was not. We compared the rate of two types of gesture, with and without the fi ller. The result suggests that beat gestures under the DAF condition are not contingent upon fi llers. The result showed that the gesture rate after the filler is higher than before it. In conclusion, the results were interpreted that fi llers production has an important role in speech planning, and accordingly it leads to gesture production intermediated by temporal suppression.