Deconstructing Mere Presence and Audience Effect During Videoconferencing: Video versus Screen-Sharing Mediated Performance Changes

The study aims to understand how the impact of social presence on task performance (social facilitation effect), usually measured in face-to-face settings, can be generalised towards remote videoconferencing. The social facilitation effect is expressed in the improvement of task performance on easy tasks, and detriment on difficult tasks, during a social situation versus when performing alone. We tested which videoconferencing channels are responsible for this performance change. The interaction occurred within an experimentally controlled naturalistic videoconferencing setting. The participants performed visual-reasoning tasks as quickly and accurately as possible under several conditions: when screen-sharing their task performance, having their video on, seeing the video of the researchers’ interactive avatar, and with all these channels on or all off. Based on two social facilitation effect phenomena, we predicted that participants' performance might change when it is watched (audience effect) by the researcher through screen-sharing and when participants or the researcher co-share their videos during videoconferencing (mere presence effect). We found that having participant video visible to the companion improved participants' performance accuracy on difficult tasks, whilst task screen-sharing improved speed on correct easy tasks, with no significant effect from the researchers' visual presence. We entertain the notion of soft-presence and propose ways forward.