Bibliography

 

• Bates, J. (1994). The role of emotion in believable agents. Communications of the ACM, 37:7, July, pp. 122-125.
Notes: This article presents an original view on the process of making agents engaging and human-like by exploring the way artists in different fields have achieved the creation of "believable characters". The author uses Disney examples throughout the article, to exemplify characters that convey human-like emotions and behaviors, and establish their believability with the audience in a way that would be more than desirable in pedagogical agents. This believability has an obvious relation with motivational factors.

• Dehn, D.M., Van Mulken, S. (2000). The Impact of Animated Interface Agents: a Review of Empirical Research. International Journal of Human-Computer Studies, Vol. 52, No. 1.
Notes: The authors of this article have taken the most salient empirical research carried out on animated agents in the last decade, and have compared those studies based on the kind of agents looked at, the kind of domain in which the agents were used, the kind of tasks performed by the users, and the kind of results obtained. Charts are included comparing all studies, covering both the cognitive and the affective domains.

• Ekman P., Friesen (1992). Emotion in the human face. Cambridge: CUP.
Notes: This book includes a whole set of categories, charts, and samples of different facial expressions, their meaning, and their social use. The book also includes principles, ideas, and a general framework to understand emotion in social settings. As a drawback, the facial expressions categories are culture-bound (they are based on an European, Anglo-Saxon model), and their usefulness for cross-culture research is not addressed.

• Elliott, C., Lester, J.C., and Rickel, J. (1997). Integrating affective computing into animated tutoring agents. In Proceedings of the IJCAI Workshop on Animated Interface Agents: Making Them Intelligent, pages 113--121, Nagoya, Japan.
Notes: The authors of this paper explore the affective reasoner approach as a necessary component in the creation of agents. They experiment with the integration of that approach into two extant tutoring systems. In doing so, they propose an architecture to include personality and emotional responses in virtual tutoring agents. This paper has parts that are quite specific and technical, but still there are some theoretical principles that can be transferred and applied to different contexts.

• Hietala, P., & Niemirepo, T. (1998). The competence of learning companion agents. International Journal of Artificial Intelligence in Education, 9, 178-192.
Notes: This journal article argues the need for more than one agent to be available to users, as a way of targeting individual needs and interests, as well as different personality types and learning styles. The issue of learner control, and how that enhances user interaction and engagingness, is examined. At the same time, social agency and the role and personality of the agent is taken as a very important area to consider when embedding animated agents in educational software.

• Koda, T., Maes, P. (1996). Agents with faces: the effect of personification. Proceedings of the 5th IEEE International Workshop on robot and Human Commnication (RO-MAN'96), pp. 189-194.
Notes: This work is one of the most cited ones in animated agent research. The different experiments carried out by this group cover both the cognitive and affective areas. For the cognitive areas, issues like problem-solving, as well as content recall and transfer, are analyzed. For the affective domain, the authors examine the way users experience their interaction with animated agents. Those agents are assessed according to labels such as likeability, intelligence, ease of use, user level of comfort, etc. Most of these categories are rooted in emotional perception of the agent, and self-reporting instruments are used.

• Lester, J.C., Converse, S.A., Kahler, S.E., Barlow, S.T., Stone, B.A., and Bhoga, R.S. (1997). The Persona Effect: Affective Impact of Animated Pedagogical Agents. In CHI '97, pp. 359--366. ACM Press.
Notes: This article describes a study carried out in a middle school, where a hundred participants interacted with an animated pedagogical agent. The aim of the study was to analyze what influence the persona effect would have on the students' learning experience.

• Moreno, R. (2001). Cognitive and Motivational Consequences of Adapting an Agent Metaphor in Multimedia Learning: Do the Benefits Outweigh the Costs? Proceedings of WebNet 2001 World Conference, pp. 873-878. Norfolk, VA: AACE Press.
Notes: This report describes a series of 5 experiments carried out on different aspects of the user of animated agents in instructional settings. The different experiments compared agent features like appearance, voice quality, and role. Most of the findings are statistically significant in the "motivation" area, while recall and transfer do not show statistically significant differences among groups. Some samples of the instrument used to measure motivation are included.

• Picard, R. W. (1995). Affective computing. MIT Media Lab Perceptual Computing Group Technical Report No. 321, Massachusetts Institute of Technology.
Notes: This article seems to be a "classic" in the area of affective computing, and it is constantly referenced by other articles exploring the use of emotions and motivation in animated agents. The author gives a nice overview of emotions from a neurological, psychological, and other perspectives. She also introduces the idea that emotions have generally been considered "non-scientific", since "…scientific principles are derived from rational thoughts, logical arguments, testable hypotheses, and repeatable experiments". She then explains the idea of "affective computing", i.e., "…computing that relates to, arises from, or deliberately influences emotions."

• Prendinger, H., and Ishizuka, M. (2001). Social role awareness in animated agents. In Proceedings 5th International Conference on Autonomous Agents (Agents-01).
Notes: This article explores the concept of agents as "affective reasoners" and their desirable "social competences". This line of thought believes that for agents to be motivating, for them to have an impact in the affective domain, they themselves should be capable of displaying some emotions. As user's motivation has an emotional component, agents would be more appealing if they could display some human-like emotions as part of their behavior, which would let them take a more "social" role in the agent-user interaction.