Referential Form and Word Duration in Video-Mediated and Face-to-Face Dialogues

It is widely believed that speakers adapt their speech to meet the comprehension needs of their listener. Yet recent work in face-to-face communication has shown that referential form is more sensitive to listener knowledge than word duration (Bard et al., 2000; Bard and Aylett, 2001). The nature of referring expressions in a collaborative problem solving task was explored in 2 experiments. Experiment 1 examined the duration of the names of landmarks on a map in video-mediated dialogues. It was found that second mentions of words were articulated less clearly than first mentions regardless of who said what. The same pattern of results has been observed in face-to- face communication which suggests that articulatory clarity is not influenced by video- mediation. Experiment 2 explored the impact of time pressure on referential form in face- to-face and computer-mediated dialogues. It was found that speakers responded to time pressure by shortening the names of the landmarks but only when the task was video- mediated. These results suggest that referential form is more sensitive to the impact of time pressure in video-mediated communication than in face-to-face communication. The implications of these findings will be discussed in terms of a dual process model of speech processing in spoken dialogue