Cooperative Work and Shared Visual Context: An Empirical Study of Comprehension Problems in Side-by-Side and Remote Help Dialogues

Abstract

If the sharing of context is now widely acknowledged as a condition for successful communication, existing studies do not allow us to determine whether it is necessary to restore the maximum of shared visual information to obtain the best communicative performance. To address this issue, three help dialogue conditions distinguished by the range of shared visual information are compared. The analyses are focused on the comprehension problems raised by each condition. The results highlight that comprehension efficiency in help dialogues is not necessarily linked to the quantity of shared visual information. This study suggests two reasons for this observation: (a) Help requesters in remote help dialogues adapt the content of their requests to the effective amount of shared visual information, and (b) helpers adapt their interpretive strategies to the available shared resources. On the other hand, it is shown that the inabil-ity to visually share some specific task-related information strongly affects communication efficiency. Implications for the design of computer-mediated communication systems are drawn from these results.

Seeing the Face and Observing the Actions: The Effects of Nonverbal Cues on Mediated Tutoring Dialogue

“…A possible interpretation is that kinesic cues, increasing the level of social presence, led tutor to monitor more carefully learners' ongoing task and to encourage them, helping both students and tutors in mutual understanding process. Outcomes from experiment 2 partially corroborated the first ones, and also confirmed some previous researches (e.g., kinesic cues improve proactive behavior) [9]. Moreover, ostensive-inferential cues let the tutor and the student focus on intrinsic tutoring contents, and improved student's performance.…”

“…Based on studies concerning the grounding processes in communication [6], affordances of visual information in mediated communication [1,9] and tutoring dialogues [19], we created the following coding scheme: a) the tutor's proactive behavior: we distinguished the tutor's spontaneous interventions towards the student as proactive interventions from the tutor's reactive interventions when s(he) replied to a student's call; b) the mutual understanding: we categorized all verbal markers that students and tutors had used during their dialogue to ground their mutual understanding (Table 1); c) the tutoring intrinsic speech acts: we categorized the tutors' and students' speech acts related to the intrinsic nature of the tutoring dialogue ( …”

Gestures Over Video Streams to Support Remote Collaboration on Physical Tasks

“…Examples of comparable instructions from the Scene Camera and DOVE + auto-erase conditions are shown in Figure 15. As can be seen in the examples on the right side of the figure, messages using the drawing tool incorporated deictic pronouns such as "this" and "that" in ways undistinguishable from the use of deixis in the side-by-side conditions of our and others' previous studies (e.g., Bauer et al, 1999;Fussell et al, 2000;Karsenty, 1999;Kraut et al, 2003). Furthermore, as shown in the left side of this figure, and likewise consistent with our earlier studies, Workers in the scene camera condition could use deictic pronouns because they knew the Helper could view their activities on the video feed, but Helpers had to use lengthier verbal expressions to denote objects and locations.…”

“…For example, a speaker might say, "take that piece and put it there," while using a pointed index finger to indicate the intended piece and target location. Pointing gestures provide a quick and efficient way to indicate objects and locations that would otherwise require lengthy verbal descriptions (e.g., Bauer, Kortuem, & Segall, 1999;Fussell et al, 2000;Karsenty, 1999). In a study of collaborative bicycle repair (Fussell et al, 2000), for example, we found that when participants worked side-by-side, and thus both gestures and task objects were visually shared, participants used more pointing gestures and deictic expressions to refer to task objects and this use of deictic expressions was associated with shorter, more efficient referring expressions and faster task performance than when pairs were linked by audio-only or audio-video connections.…”

Supporting the Changing Roles of Maintenance Operators in Mining: A Human Factors Perspective

“…To meet this need, many systems have been proposed or developed in the literature (e.g., [1]). Technologies used to support remote collaboration include email exchanges, telephone calls, video conferencing and video-mediated gesturing.…”