Video Matters! When Communication Ability is Stressed, Video Helps

ABSTRACT

Keywords

© 1997 Copyright on this material is held by the authors.

INTRODUCTION

Most of the video-mediated communication research to date has focused on different tasks, but perhaps the effect of video appears when communication clarity is strained. People might use all forms of communication possible if verbal ability is impaired. Indeed, speaking to someone in a language that is not their native language could involve gestures and eye-contact to confirm when understanding is established. Research has shown that the moment there is any sort of processing difficulty, people supplement audition with vision [7]. In this study, we extend the research on video-mediated communication to look at its effect on non-native speaker interactions. If, as research has shown, people who easily communicate verbally find no advantage of seeing each other, would those who are stressed in communication use and benefit from video?

To test this notion, we compared the task performance of native speakers of English (NS) with non-native speakers of English (NNS), who were presumably more stressed in their verbal communication ability. These pairs collaborated over audio-only (audio) or audio-plus-video (video) communication channels. We were interested in how access to visual information affects the outcome of the communication. We hypothesized that there would be an interaction between language and communication media. If pairs share a native language and culture, then video connectivity may not provide any added benefit over audio. However, if native language and culture are not shared, then video might facilitate the communication by allowing for non-verbal communication and the ability to visually monitor each other's understanding.

Specifically, we do not expect to find a difference between NS pairs in the two media conditions. However, our central hypothesis is that the NNS pairs in the video condition will perform more effectively than their counterparts in the audio condition.

PROCEDURE

The task was modeled after the map task reported in [3]. In this task, one person was the instruction giver (SPEAKER) and the other was the instruction follower (RECEIVER). Each person had a copy of the map, but only the SPEAKER's map had the path on it. In order to introduce uncertainty and a need for clarification, some of the 12 landmarks on the two maps were different. The RECEIVER's task was to accurately draw the path on his or her own map based on the SPEAKER's instructions. The pairs worked on four maps of equal difficulty. They worked in different rooms and traded off the roles of SPEAKER and RECEIVER. In the video condition, only the speaker's face and hands were visible, not the object of conversation.

The study involved four conditions and each pair participated in only one condition (a between-Ss design). The conditions were: NNS-Audio, NNS-Video, NS-Audio, and NS- Video. Map order was counter-balanced within each condition.

Performance was a combination of standardized quality and time scores. Quality was measured as the amount by which the RECEIVER's map deviated from the SPEAKER's map. Task completion time was measured in seconds. Both the quality and time data were transformed using natural logarithms. In order to have one performance score for each pair, both the time and quality data were normalized and added together to produce a normalized performance score. This overall performance score was possible because there was no evidence of a speed/accuracy tradeoff. Pairs showed poor performance in either time, accuracy, or both.

RESULTS

Table 1. NNS and NS standardized performance scores

Audio	Audio+Video
NS pairs	-.74 (n=10)	-.34 (n=10)
NNS pairs	1.35 (n=7)	.15 (n=9)

Note: Higher scores indicate worse performance.

In addition, NNS pairs in the video condition had fewer miscommunications, as measured by the number of wrong turns, than NNS pairs in the audio condition (NNS-Audio = 3.71 vs. NNS-Video=1.75). This difference was not found for native speakers (NNS-Audio=.7 vs. NNS-Video=.9).

Post-experiment questionnaire data provided some additional support for the results. NNS pairs reported finding the video more useful than NS pairs (F(1,34)=6.20, p<.05).

CONCLUSIONS

REFERENCES

1. Finn, K., Sellen, A., & Wilbur, S. (Eds.) . (1997). Video-mediated communication. Hillsdale NJ: Lawrence Erlbaum Associates.
2. Chapanis, A. (1975). Interactive human communication. Scientific American, 232, 36-42.
3. Boyle, E. , Anderson, A. and Newlands, A. (1994). The effects of visibility on dialogue performance in a cooperative problem solving task, Language and Speech, 37,1, 1-20.
4. Green and Williges, R. C. (1995). Evaluation of alternative media used with a groupware editor in a simulated telecommunication environment. Human Factors, 37, 3, 283-289.
5. Olson, J., Olson, G. and Meader, D. (1995). What mix of video and audio is useful for remote real-time work. In Proceedings of CHI, 362-368.
6. Farmer, S. M. and Hyatt, C. W. (1994). Effects of task language demand and task complexity on computer-mediated work groups, Small Group Research, 25,3, 331-336.
7. Reisberg, D., McLean, J. and Goldfield, A. (1987). Easy to hear but hard to understand: A lip-reading advantage with intact auditory stimuli. In R. Campbell & B. Dodd (Eds.) Hearing by Eye: The psychology of lip-reading. Hillsdale, N. J.: Lawrence Erlbaum Associates