User Persuasion and Affective Pedagogical Agents
Dietary behavior and attitude play major roles in the worldwide prevalence of obesity, as weight is gained when energy intake exceeds energy expenditure. Although research has focused on designing technological interventions for healthy eating behavior, recent reviews have identified a gap in the knowledge base regarding the variables/determinants of healthy eating and the interactions between them. We developed a model of some determinants and their impact on healthy eating as a basis for designing technological interventions to promote healthy eating behavior within a target community. The main goal of this work is to understand how people adopt a healthy eating attitude, the variables influencing such attitudes, the interactions between these variables, and the degree of influence each variable exerts on healthy eating attitudes. We use fast food-related eating behavior as our case study. Our model shows that weight concern, nutrition knowledge, concern for diseases, social influence, and food choice motives predicts 65% of the variance in healthy eating attitudes, showing the suitability of the model for use in predicting healthy eating attitude. This result will inform decisions on the most effective persuasive strategy for designing interventions to promote healthy eating behavior.
While some adaptive applications change their interface or functionality to suit better the needs of the users, most often users need to adapt to the applications by learning how to cope with them. Adaptive applications can support the user in learning how to operate the application efficiently (e.g. adaptive on-line help systems). There are also applications that can teach the user (i.e. change his/her beliefs) about the application or about a particular domain.
With the emergence of distributed applications that rely on the cooperative autonomous users to work successfully, e.g. collaborative learning environments, collaborative help systems, peer-to-peer applications, a need arises not just to teach the user about how to operate the application, but also how to be a good citizen (or "netizen"). Free-riding and lack of volunteerism prevent the application from reaching a critical mass of users that can make it useful for every participant. Changing the attitude of the user to a cooperative one is important and crucial to the success of such applications. Incentives and feedback loops can be built in the design of the application to encourage particular user behaviours and discourage others. These incentives can be material (e.g. monetary rewards, or better quality of service), or intangible (e.g. reputation in the community). Different users are motivated by different incentives. Therefore individualized persuasion methods have be utilized to change users' attitudes and behaviour to cooperative.
A Game-Study of Persuasion Effect of Emotional Agents
Download and play a game designed to answer the following questions:
Participants: Yao Wang, PhD Student, Jian Liu (MSc student in ARIES), Julita Vassileva
Powerpoint presentation of Yao & Jian's 862 project
Emotional Agents in Computer-Based Learning Environments
Participants: Chioma Okonkwo (former M.Sc. student), Julita Vassileva
The use of animated pedagogical agents with emotional capabilities in an interactive learning environment has been found to have a positive impact on learners. Aristotle contended that three elements; emotion, logic, and character are crucial for successful persuasion, i.e. in winning others over to one’s way of thinking. We have designed a pedagogical agent that acts in an interactive learning environment, using Ortony’s Cognitive Structure of Emotion model and McCrae and John’s five-factor model of personality. We investigated the persuasive impact of this emotional pedagogical agent on a group of learners. The results show that while not contributing any significant performance gain in learning, the incorporation of emotion changes the way students perceive the learning process, and makes it more engaging. We also found out that there are some gender-based and individual differences in the user perception of an emotional agent, which need to be taken into account when designing a more adaptive and “intelligent” emotional pedagogical agents.
Back to "Projects and Publications"