Changes of utterances in the skill acquisition of collaborative conveyer task

Importance of developing human adaptive robots or systems is increasing these days. To accomplish it, we have to clarify features of human-human communication. In this study, we analyzed human speech while completing computerized collaborate task to clarify how humans speak in collaborative work. We extracted questioning speeches from conversations using CLAN, and classified them into four categories according to their content. We investigated whether the number and ratio of each type of questions changed over ten trials. The results indicate that leader role person is sensitive to task planning