Culturally Aware Greetings in Agents

Project Details

Description

This project examines if and how cultural heuristics influence first meeting encounters between users and virtual agents.Why is this an interesting feature for an interactive computer systems? The objective to develop more human-centered, personalized and at the same time more entertaining interfaces immediately leads to the metaphor of an embodied conversational agent (ECA) that employs gestures, mimics and speech to communicate with the human user. Embodied conversational agents offer great promise to more natural interaction because of their potential to emulate verbal and non-verbal human-human interaction. In general, nonverbal interaction comprises facial expressions, gaze behavior, gestures, and body posture, which all play sometimes distinct, sometimes redundant roles in face to face communication. Most research prototypes of embodied conversational characters aim at the modeling of complex multimodal dialogues, though the focus is usually on the generation of synchronized multimodal expression.


Along with emotions, culture has become an important social parameter for the design of interactive systems in general and of embodied agents specifically over the last years. Unfortunately, most systems rely on an adhoc definition of culture that lacks predictive power for a more general use in an interactive system. One of the main challenges are the lack of specificity of the employed theoretical concepts from the social sciences as well as the lack of knowledge on the parameters that constitute cultural heuristics for behaviour and interpretation and how these parameters vary across cultures. CAGA aims at exploring some of these parameters in the scenario of first meeting encounters.
AcronymCAGA
StatusFinished
Effective start/end date01/03/201131/03/2013

Funding

  • French Embassy

Fingerprint

Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.