10:00   Oral Session 1-KzZ – Intelligent Tutoring Systems
Chair: Ruth Aylett
25 mins
Natural Affect Data - Collection & Annotation in a Learning Context
Shazia Afzal, Peter Robinson
Abstract: Automatic inference of affect relies on representative data. For viable applications of such technology the use of naturalistic over posed data has been increasingly emphasised. Creating a repository of naturalistic data is however a massively challenging task. We report results from a data collection exercise in one of the most significant application areas of affective computing, namely computer-based learning environments. The conceptual and methodological issues encountered during the process are discussed, and problems with labelling and annotation are identified. A comparison of the compiled database with some standard databases is also presented.
25 mins
It's All in the Game: Towards an Affect Sensitive and Context Aware Game Companion
Ginevra Castellano, Iolanda Leite, André Pereira, Carlos Martinho, Ana Paiva, Peter William McOwan
Abstract: Robot companions must be able to display social, affective behaviour. As a prerequisite for companionship, the ability to sustain long-term interactions with users requires companions to be endowed with affect recognition abilities. This paper explores application-dependent user states in a naturalistic scenario where an iCat robot plays chess with children. In this scenario, the role of context is investigated for the modelling of user states related both to the task and the social interaction with the robot. Results show that contextual features related to the game and the iCat's behaviour are successful in helping to discriminate among the identified states. In particular, state and evolution of the game and display of facial expressions by the iCat proved to be the most significant: when the user is winning and improving in the game her feeling is more likely to be positive and when the iCat displays a facial expression during the game the user's level of engagement with the iCat is higher. These findings will provide the foundation for a rigorous design of an affect recognition system for a game companion.
25 mins
Evaluating the Consequences of Affective Feedback in Intelligent Tutoring Systems
Jennifer Lynne Robison, Scott W McQuiggan, James C Lester
Abstract: The link between affect and student learning has been the subject of increasing attention in recent years. Affective states such as flow and curiosity tend to have positive correlations with learning while negative states such as boredom and frustration have the opposite effect. Consequently, it is a goal of many intelligent tutoring systems to keep students in emotional states that are conducive to learning through affective interventions. While much work has gone into improving the quality of these interventions, we are only beginning to understand the complexities of the relationships between affect, learning, and feedback. This paper investigates the consequences associated with providing affective feedback. It represents a first step toward the long-term objective of designing intelligent tutoring systems that can utilize this information for analysis of the risks and benefits of affective intervention. It reports on the results of two studies that were conducted with students interacting with affect-informed virtual agents. The studies reveal that emotion-specific risk/reward information is associated with particular affective states and suggests that future systems might leverage this information to make determinations about affect interventions.