When we attend live performances, one important part of the experience is the ‘collective engagement’1 with others attending the same performance, and feeling and sensing others’ silent reactions and overt responses. Most of us have experienced the thrill of being amongst a wildly appreciative audience - whether that appreciation is expressed through cheers and whistles, screaming and dancing along, or through the magical hush at the end where no one wants to spoil the moment by applauding. Most of us have also experienced being amidst a restless audience, whose coughs and fidgets feel like a referendum on the quality of the performance. And there can be the surprising feeling of having reactions that are out of sync with the majority - not finding funny what the rest of the audience is laughing at enthusiastically, or being moved to tears by something that seems to leave others entirely unmoved.
While such anecdotes feel familiar, systematic understanding of the dynamics of audience engagement and reaction is surprisingly limited. When and to what extent are audience members affected by each other’s reactions? Are some people affected more than others? How does audience influence vary across performance genres, venues, and kinds of audiences? Which overt behaviours are contagious (whether they were intended as communicative signals or not), under which circumstances and for whom? How much do audience members agree in their assessment of what has happened in the performance, and in their sense of what others’ reactions are? How accurately do performers assess audience response?
Answering these kinds of questions is tricky. Directly addressing them has often required intrusive measures that may well disrupt the audience experience: hooking people up to unfamiliar equipment like heart rate sensors or EEG caps, or asking audience members to continuously turn a dial during the performance to report their evaluative reaction. Retrospective reports by audience members have their own problems: they can be far removed from the original audience experience, and the kinds of report that people provide can be vague and not specific to particular performances.
New technologies and methods are opening new possibilities for investigating audience reactions to live performances in real-time and in non-intrusive ways. Advances in computer vision have provided new tools for tracking visible audience responses in live performance situations: automated processing of facially displayed emotion, gaze direction, and body movements, based on external HD video. We have exploited such tools in our studies of the social dynamics of live audiences ranging from stand-up comedy2 to contemporary dance3. We have also developed new tools for combining and simultaneously visualizing multiple fine-grained data streams - e.g., facial expression, breathing and movement - from the perspective of any audience member or the performer4,5.
It is also now possible to unobtrusively collect audience members’ self-reports about their experience - their characterizations of the performance, their judgments of peak moments - immediately afterwards using their own mobile devices, before they have even left their seats. This is in principle no different than passing out paper and pencil questionnaires, but the speed and ease of deployment is greater, and people may feel particularly willing to disclose private thoughts and reactions on devices they regularly use to do just that6. Further, it is possible to measure, immediately or later, the extent to which they agree with others’ characterizations. With such tools, researchers can develop new kinds of profiles of audience members’ connectedness in particular performances: the extent to which audience members with different backgrounds and genre-specific expertise, or with different spatial proximity and prior connection with other audience members, agree with each other’s patterns of interpretation and evaluation7.
As we see it, the next methodological opportunity and challenge will be to integrate these different streams of data—audience members’ visible behaviours, self-generated characterizations of the performance, and extent of agreement with each other’s characterizations—so as to create new ways of understanding audience interaction and influence.
Symposium that focuses on these methodological challenges and opportunities will be held on Friday July 14th, 1pm-4pm at the Digital Catapult, 101 Euston Road, London.
Resources
1 Radbourne, J., Glow, H., and Johanson, K. (2013b) “Knowing and measuring the audience experience,” in The Audience Experience: A Critical Analysis of Audiences in the Performing Arts, eds J. Radbourne, H. Glow, and K. Johanson (Chicago, IL: Intellect Ltd., University of Chicago Press), 1–13.
2 Katevas, K., Healey, P. G., & Harris, M. T. (2015) Robot Comedy Lab: Experimenting with the social dynamics of live performance. Frontiers in Psychology, 6. http://journal.frontiersin.org/ article/10.3389/fpsyg.2015.01253/full
3 Theodoru, L., Healey, P.G.T. and Smeraldi, F. (forthcoming) Overt audience responses to contemporary dance: Is hand and body movement a signal of engagement? International Symposium on Performance Science, Reykjavik, Iceland, 30 August - 02 September 2017.
4 Harris, M.T. and Healey, P.G.T. (forthcoming) Visualising performer–audience dynamics. International Symposium on Performance Science, Reykjavik, Iceland, 30 August - 02 September 2017.
5 http://tobyz.net/projects/visualising-performer-audience-dynamics
6 Schober, M.F., Conrad, F.G., Antoun, C., Ehlen, P., Fail, S., Hupp, A.L., Johnston, M., Vickers, L., Yan, H., & Zhang, C. (2015) Precision and disclosure in text and voice interviews on smartphones. PLOS ONE 10(6): e0128337. http:// journals.plos.org/plosone/article?id=10.1371/ journal.pone.0128337
7 Schober, M.F., & Spiro, N. (2016) Listeners’ and performers’ shared understanding of jazz improvisations. Frontiers in Psychology 7:1629. http://journal.frontiersin.org/article/10.3389/ fpsyg.2016.01629/full