Corroborating emotion theory with role theory and agent technology : a framework for designing emotional agents as motivational tutoring entities

(2007), IX, 162 S. : Ill., graph. Darst.
Dissertation / Fach: Elektrotechnik
Duisburg, Essen, Univ., Diss., 2007
Abstract:
Nowadays, more and more applications require systems that can interact with humans. Agents can be perceived as computing services that humans, or even other agents, can request in order to accomplish their tasks. Some services may be simple and others rather complex. A way to determine the best agents (services) to be implemented is to identify who the actors are in the object of study, which roles they play, and (if possible) what kind of knowledge they use. Socially Intelligent Agents (SIAs) are agent systems that are able to connect and interface with humans, i.e. robotic or computational systems that show aspects of human-style social intelligence. In addition to their relevance in application areas such as e-commerce and entertainment, building artefacts in software and hardware has been recognized as a powerful tool for establishing a science of social minds which is a constructive approach toward understanding social intelligence in humans and other animals. Social intelligence in humans and other animals has a number of fascinating facets and implications for the design of SIAs. Human beings are biological agents that are embodied members of a social environment and are autobiographic agents who have a unique personality. They are situated in time and space and interpret new experiences based on reconstructions of previous experiences. Due to their physical embodiment, they have a unique perspective on the world and a unique history: an autobiography. Also, humans are able to express and recognize emotions, that are important in regulating individual survival and problem-solving as well as social interactions. Like artificial intelligence research trend, SIA research trend can be pursued with different goals in mind. A deep AI approach seeks to simulate real social intelligence and processes. A shallow AI approach, which will be highlighted also within this thesis, aims to create artefacts that are not socially intelligent per se, but rather appear socially intelligent to a given user. The shallow approach does not seek to create social intelligence unless it is meaningful social intelligence vis-à-vis some user situation In order to develop believable SIAs we do not have to know how beliefs-desires and intentions actually relate to each other in the real minds of the people. If one wants to create the impression of an artificial social agent driven by beliefs and desires, it is enough to draw on investigations on how people with different cultural background, develop and use theories of mind to understand the behaviours of others. Therefore, SIA technology needs to model the folk-theory reasoning rather than the real thing. To a shallow AI approach, a model of mind based on folk-psychology is as valid as one based on cognitive theory. Distance education is understood as online learning that is technology-based training which encompasses both computer-assisted and Web-based training. These systems, which appear to offer something for everyone at any time, in any place, do not always live up to the great promise they offer. The usage of social intelligent agents in online learning environments can enable the design of “enhanced-learning environments” that allow for the development and the assessment of social competences as well as the common professional competences. Within this thesis it is shown how to corroborate affective theory with role theory with agent technology in a synchronous virtual environment in order to overcome several inconveniences of distance education systems. This research embraces also the shallow approach of SIA and aims to provide the first steps of a method for creating a believable life-like tutor agent which can partially replace human-teachers and assist the students in the process of learning. The starting point for this research came from the fact: anxious, angry or depressed students do not learn; people in these conditions do not absorb information efficiently, consequentially it is an illusion to think that learning environments that do not consider motivational and emotional factors are adequate.