Relevance and conspicuousness of facial information conveyed in videoconferencing systems

Proposal for an internship for 2^nd year Master students

Topic of the internship: Relevance and conspicuousness of facial information conveyed in videoconferencing systems

Context: This research internship will take place in the Human Interaction department of the LISN (Interdisciplinary Laboratory of Numerical Sciences) in University Paris-Saclay.

Supervisor:

Ouriel Grynszpan, Professor, AMIArchitectures et modèles pour l'Interaction group, https://www.lisn.upsaclay.fr/members/grynszpan-ouriel/,
Email: ouriel.grynszpan@universite-paris-saclay.fr

Desirable profile:

• Student in Master 2^nd year in Human-Computer Interaction.
• Knowledge of experimental methodologies and statistics will be appreciated.
• Other types of profiles may be considered depending on skills.

To apply: Send by mail to Ouriel Grynszpan: CV, cover letter and available transcripts of grades (Master and Bachelor)

Duration of internship: 5 months (between March and July 2025), this internship will be paid.

Project description: Videoconferencing systems have profoundly changed digital collaboration. Associating video and voice greatly enhances communicative abilities. Yet, the combination of video and voice modalities requires users to dispatch their attention on numerous sources of information at the same time. This can become overwhelming, especially in large group conversations. Moreover, as users cannot always attend to all individual faces simultaneously, they may miss important contextual information conveyed by facial expressions. Additionally, the faces of listeners are not continuously expressive and may not need to be monitored with scrutiny throughout. From a design perspective, the issue is to understand what facial information is the most relevant and how to convey this information conspicuously. Research in neurosciences has shown that low spatial frequencies and high spatial frequencies of face images are processed differently. Low frequencies are processed rapidly and provide a coarse idea of the emotional expression (Vlamings et al., 2009; Wang et al., 2021), whereas higher frequencies take more time to be analyzed but yield more detailed information about facial expressions. The goal of the present internship is to create and testDéfinition courte Lorem ipsum a videoconferencing prototype that would be able to display faces with low and high spatial frequencies differentially.

For this purpose, the work of the internship will be divided into four main tasks:

• Review of the literature on facial expression recognition (psychology and HCI);
• Build a design space based on this literature review;
• Design and implement an experiment to compare some points of this design space (e.g., compare perception of faces with low pass versus high pass filters for spatial frequencies);
• Run the experiment and analyze the results (at least a pilot study).

References:

Vlamings, P. H. J. M., Goffaux, V., & Kemner, C. (2009). Is the early modulation of brain activity by fearful facial expressions primarily mediated by coarse low spatial frequency information? Journal of Vision, 9(5), 12. https://doi.org/10.1167/9.5.12

Wang, S., Eccleston, C., & Keogh, E. (2021). The Time Course of Facial Expression Recognition Using Spatial Frequency Information : Comparing Pain and Core Emotions. The Journal of Pain, 22(2), 196‑208. https://doi.org/10.1016/j.jpain.2020.07.004