Background complexity and emotion recognition in video conferencing

Undergraduate Just-In-Time Abstract

Poster Presentation 33.347: Sunday, May 17, 2026, 8:30 am – 12:30 pm, Banyan Breezeway
Session: Undergraduate Just-In-Time 2

Jane Kalla1, Yelda Semizer1; 1New Jersey Institute of Technology

Virtual communication has become a part of our daily life, allowing us to connect with others across locations around the world. While video conferencing tools facilitate remote communication, they may also present perceptual challenges, especially in cases where visual features are complex or degraded in quality. For example, allowing users to control their virtual backgrounds may increase user satisfaction; however, the resulting perceptually rich displays may introduce additional demands on the perceptual system. In this study, we examined the extent to which the visual complexity of attendees’ backgrounds influences the ability to accurately judge the average emotion of a group. Using a controlled set of faces with happy and neutral expressions, we generated a series of composite images in which the mean and standard deviation of emotional expressions of face groups were systematically varied. The visual complexity of attendees’ backgrounds was also manipulated. Using these stimuli in a 2-interval forced choice task, we asked participants to judge the average emotion of a group of individuals in a simulated video conferencing display. Preliminary results suggest that visually complex backgrounds impair the ability to extract average emotional expression from groups, making it more difficult to process perceptually relevant information.