| Thus far, conversational agents and social robots were mainly developed from a technical point of view; e.g., ELIZA (Weizenbaum, 1966). The underlying model was almost lay theory, showing that smiles express happiness and frowns sadness (cf. Breazeal, 2002). Yet, most agents and robots fail conversation-wise because they cannot apply the right emotional response to the right situation. In this interdisciplinary research project, we will apply theories and models from mediated interpersonal communication, human-computer interaction and media psychology to embodied conversational agents. In doing so, the agents will be capable of social interaction and human-like communication. Ideally, theory and simulation should be so good that in a Turing Test (Turing, 1964), people will not be able to tell the difference. If there are significant differences, the experiments will show where the used theories and models have to be improved. |
