Closing the Affective Loop in Intelligent Learning Environments

Call for Papers

Research on affective intelligent learning environments (ILE), i.e. ILE that include affective elements in the interaction with the student, has become increasingly prominent in the past few years, due to two main reasons. Firstly, there is growing evidence of correlations between affect and learning, fostering the belief that recognizing and responding to student's affect can improve the effectiveness of pedagogical interactions. Secondly, advances in affect recognition make it now feasible to devise interactive tools that can be aware of a user's affective states, and respond accordingly.

The workshop is designed to continue the discussion started during the previous related workshops at ITS and AIED, with a specific focus on how to close what we will call the affective loop, i.e. that ensemble of four phases that together allow for the principled addition of affective elements to an ILE: (i) design the environment so that it can elicit affective states favorable to learning, (ii) recognition/modeling of relevant user's states, (iii) selection of appropriate system responses, and (iv) synthesis of the appropriate affective expressions.

Of the four phases that form the affective loop, phase (ii) and (iv) have seen the most advances in recent years. What we are still lacking are strong theories on how to use affect in pedagogical interactions, and strong evidence that taking affect into account when interacting with students can actually improve learning. There are some qualitative accounts of how affect and learning may interact, as well as some evidence that certain affective states are correlated with learning, but these findings are not sufficient to devise computational models that can be used to influence the affect-sensitive behaviors of computer-based tutors. Similarly, we have initial results on the effect of affect-based ILE, but they mostly report qualitative measures of student satisfaction rather than quantitative measures of learning.

Thus, the main objective of this workshop is to bring researchers in the field together to discuss/define directions for research on the first and third phase of the affective loop, that is the design of system elements and responses that aim at fostering learning by eliciting favorable affective states. Submissions that report research on phase (ii) and (iv) will be considered, but the authors should also include their views/plans for our target phases, in order to have ample material to support a constructive discussion on how to close the affective loop in ILE research.

Topics of Interest

The prospective authors are invited to discuss the following topics of interest (but please note that papers on additional related topics will be considered):

• When/why is it useful to include affect in an ILE? Is it ever better to not include it, and if so when? As a result: can we begin to develop a taxonomy of learning contexts under which it is critical or useful to include affect in an ILE?
• What types of affective elements, and why, should be included in the affective loop (e.g. basic affect such as like/dislike reactions; low-level dimensions of emotional states such as valence and arousal; instantaneous emotions such as joy, eureka, anger; longer term emotions/moods such as frustration, boredom, flow)? As a result: can we identify a mapping between the above taxonomy of affect-demanding context and the most appropriate level of affective capabilities that the ILE should have (i.e. responding to/expressing specific emotions, or moods, or just valence, etc.)?
• How can we define which designs and behaviors, and under which circumstances, can help an ILE foster learning through affect? Are the relevant theories/findings that we can use in psychology, cognitive science and education?
• What are the best means of evaluating an affective ILE? What elements should we evaluate (e.g. learning, user satisfaction, motivation, engagement)? Are there innovative evaluation methods that should be considered to evaluate the potential and needs of affective ILE?

The participants will be asked to explicitly address one or more of these issues in their contributions. By addressing these issues in a mixed-mode, informal set of interactions, we hope to explore alternative methodologies for closing the affective loop in ILE, identify key problems to address, and contribute to advancing the state of the art of this exciting area of research.

Organising Committee

Program Chairs

• Cristina Conati, University of British Columbia, Canada
• Tanja Mitrovic, University of Canterbury, NZ

Program Committee

• Madeline Alsmeyer, University of Sussex, UK
• Ryan Baker, Carnegie Mellon University, USA
• Winslow Burleson, Arizona State University, USA
• Rafael Calvo, University of Sydney, Australia
• Sidney D'Mello, University of Memphis, USA
• Lewis Johnson, Alelo Inc., USA
• James C. Lester, North Carolina State University, USA
• Rosemary Luckin, University of London, UK
• Scott W. McQuiggan, North Carolina State University, USA
• Roger Nkambou, UQAM, Canada
• Helen Pain, University of Edinburgh, UK
• Ana Paiva, Instituto Superior Tecnico, Portugal
• Kaska Porayska-Pomsta, University of London, UK
• Genaro Rebolledo-Mendez, University of Sussex, UK
• Ma. Mercedes T. Rodrigo, Ateneo de Manila University, Philippines

Program

Proceedings

The workshop proceedings can be downloaded from here (PDF, 2.6mb).