October 2020
Volume 20, Issue 11
Open Access
Vision Sciences Society Annual Meeting Abstract  |   October 2020
Judging the emotion of natural groups
Author Affiliations
  • Susan Hao
    UC Berkeley
  • David Whitney
    UC Berkeley
  • Sonia Bishop
    UC Berkeley
Journal of Vision October 2020, Vol.20, 613. doi:https://doi.org/10.1167/jov.20.11.613
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Susan Hao, David Whitney, Sonia Bishop; Judging the emotion of natural groups. Journal of Vision 2020;20(11):613. https://doi.org/10.1167/jov.20.11.613.

      Download citation file:

      © ARVO (1962-2015); The Authors (2016-present)

  • Supplements

It has been argued that humans use summary statistics to rapidly judge group emotion. Studies have mainly used faces with posed expressions presented without context. Here, we used natural images and dimensional ratings of emotion (valence, arousal, and dominance) to investigate the influence of average emotion, maximal emotion, and context on judgments of group emotion. Stimuli contained three to five faces. Separate groups of participants viewed 1) the original images, 2) individual faces taken from the original images, and 3) context only (the image with faces removed). Stimulus ratings were averaged across participants. We conducted regression analyses with ratings of group emotion for the original image condition as the dependent measure. For valence, the arithmetic mean of individual face ratings explained the highest amount of unique variance in the ratings of group emotion with inclusion of context ratings increasing explained variance. For dominance, context explained the highest amount of unique variance. Including either the arithmetic mean or ratings of the most dominant face increased explained variance. For arousal, the maximally arousing face explained the highest proportion of unique variance in the group emotion rating; here context also made an additional significant contribution. We ran two additional conditions, one comprised the faces from the original image with context removed but position maintained; the second inserted new faces into these positions. For valence and dominance, the arithmetic mean, alone, explained significant variance in ratings of group emotion. For arousal, both the arithmetic mean and ratings of the maximally arousing face explained significant unique variance. These findings suggest that when we judge the emotion of a natural group, our relative reliance on mean versus maximal facial emotion and the extent of the role played by contextual information varies with the aspect of emotion concerned.


This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.