CE33 - Interaction, robotique 2025

Introduce Standards and Methods for Replication of Social Interaction Quality Research in Affective Computing – INSPIRE-bis

Submission summary

The emulation of human-like social interaction is a significant focus of recent developments in Affective Computing. Crucial to this aspect is the employment of Socially Interactive Agents (SIAs) designed to assist us in our daily lives. Users' acceptance of SIAs is enhanced when the technology is designed to recognize the user as a social actor. Additionally, the perceived quality of interactions with SIAs plays a significant role in building trust and acceptance of these technological advancements. Equipping and evaluating a machine with social-communicative intelligence is one of the most significant challenges nowadays and requires deep interdisciplinary knowledge in psychology, computer sciences, and other relevant disciplines. For researching interaction quality, psychological approaches serve as a foundational framework. To address this challenge, numerous research groups assess SIAs and explore their impact on humans. Nevertheless, these relatively small-scale studies come with limitations concerning their ability to be broadly applicable. Moreover, current research practices in Affective Computing often lack emphasis on openness, reproducibility, and replicability.
The overall aim of INSPIRE-bis is to establish standardized methodologies for replicating social interaction quality research. This is necessary for enhancing our comprehension of assessing findings and their impact. INSPIRE-bis aims to analyze the prerequisites and essential criteria for replicating studies on social interaction quality across various technological paradigms and subsequently compare their outcomes. To achieve this objective, INSPIRE-bis concentrates on examining turn-taking management as a pivotal element within the realm of social interactions.
The interdisciplinary approach in SIA research has led to significant developments. However, integrating various disciplines poses challenges, particularly from psychology, which has rapidly evolved over the

Project coordination

Catherine Pelachaud (SORBONNE UNIVERSITÉ)

The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.

Partnership

ISIR SORBONNE UNIVERSITÉ
German Research Centre for Artificial Intelligence

Help of the ANR 312,034 euros
Beginning and duration of the scientific project: January 2026 - 36 Months

Useful links

Explorez notre base de projets financés

 

 

ANR makes available its datasets on funded projects, click here to find more.

Sign up for the latest news:
Subscribe to our newsletter