Affective computing is a new artificial intelligence area that deals with the possibility of making computers able to recognize human emotions in different ways. This chapter represents an implemented framework, which integrates this new area with an intelligent tutoring system. The authors argue that tutor agents providing socially appropriate affective behaviors would provide a new dimension for collaborative learning systems. The main goal is to analyse learner facial expressions and show how affective computing could contribute to learning interactions, both by recognizing learner emotions during learning sessions and by responding appropriately.
The question is not whether intelligent machines can have emotions, but whether machines can be intelligent without any emotions.
Research Background And Motivation
On the one hand, with the focus on innovative and user centered interaction technologies, the interplay between emotions and computers, widely known as affective computing, “computing that relates to, arises from, or deliberately influences emotions” (Picard, 1995), plays an important role in human computer interaction (HCI). Research findings suggest that emotions play an essential role in decision-making, perception, learning and in general influence the mechanisms of rational thinking. According to Rosalind Picard:
“If we want computers to be genuinely intelligent and to interact naturally with us, we must give computers the ability to recognize, understand, and even to have and express emotions” (Picard, 1997).
On the other hand, it is often understood that the eventual objective of communication within virtual environments (VEs) is to model communication between humans in the physical world. In order to achieve this objective, communication capabilities within the virtual world must not be limited to the simple exchange of information. Everyday human communication involves a level of affective communication (communication involving emotional states) that is absent from many VEs. Many researchers now believe that affective tutoring systems1 would be significantly enhanced if computers could adapt according to the emotions of learners (Alexander et al., 2004). If human emotions are essential for human thinking and learning processes, virtual learning environments must recognize this to be successful.
If VEs are to truly represent real world interactions they must both:
Facilitate the communication of affect, and
Make the agents situated in the environment react in a way that respects the affective context.
An agent that ignores these aspects of the environment will jar with the realism of the communication as much as a mechanical system that ignores the laws of physics.
Our objective in this chapter is to show that the use of affective systems is pat of an interaction problem that concerns the whole interaction cycle, where emotions arise from an active act of interpretation and participation on the end-user side. We introduce a model of interaction between users and animated agents as well as inter-agents interaction that supports the basic features of affective communication in VEs, given that detecting a learner’s emotional reaction to a given situation is a fundamental element of any distant learning environment. This chapter presents an affective e-learning framework based on emotional agents that can partially replace and support human-teachers, by assisting and motivating learners in distributed learning environments (Ammar et al., 2005). We outline an approach to constructing an emotion-recognizing computer system, and present real-time results of the recognition of basic emotional expressions from the video. The system automatically detects frontal faces in the video stream and recognizes the emotion with respect to six basic facial expressions (anger, disgust, fear, joy, sadness, and surprise), as suggested by Ekman (Ekman et al., 1975).
This chapter is organized as follows: Section 2 introduces related work, Section 3 describes affective communication, Section 4 explains the proposed EMASPEL framework, Section 5 gives the application results, and finally Section 6 finishes with conclusions.
Key Terms in this Chapter
Peer Learning: An alternative to the one-on-one strategy, co-operative strategies comprise an additional element, namely peer interaction. Co-operative learning systems, called also social learning systems, adopt a constructive approach using the computer more as a partner then as a tutor. Multiple agents that are either computer simulated or real human beings can work on the same computer or share a computer network. The e-learning community is increasingly recognising the importance of students learning from their study peers. When peers come together in a learning context they form an Online Learning Community.
Facial Expression Recognition: Extracting and validating emotional cues through analysis of users’ facial expressions is of high importance for improving the level of interaction in man machine communication systems. Extraction of appropriate facial features and consequent recognition of the user’s emotional state that can be robust to facial expression variations among different users is the topic of this chapter.
ECA (Embodied Conversational Agent): Embodied agents are agents that are visible in the interface as animated cartoon characters or animated objects resembling human beings. Sometimes they just consist of an animated talking face, displaying facial expressions and when using speech synthesis, having lip synchronization.
EMASPEL: Emotional Multi-Agents System for Peer-to-peer E-learning. The first version was in 2006.
Emotional Agents: Emotional agents are intelligent agents responsible for the recognition of the facial expression of the learner. Extracting and validating emotional cues through analysis of users’ facial expressions is of high importance for improving the level of interaction in man machine communication systems. Extraction of appropriate facial features and consequent recognition of the user’s emotional state that can be robust to facial expression variations among different users is the topic of these emotional agents.
Peer-to-Peer (P2P) Networks: Over the past few years, peer-to-peer (P2P) networks have revolutionized the way we effectively exploit and share distributed resources. In contrast to the traditional client-server architecture, P2P systems are application level, collaborative systems where agents work together to perform certain tasks.
Affective Computing: is the Science of automatic understanding of human emotions, and providing tools and machines that can respond to these emotions.
Emotion: in its most general definition, is a complex psychophysical process that arises spontaneously, rather than through conscious effort, and evokes either a positive or negative psychological response or physical expressions, often involuntary, related to feelings, perceptions or beliefs about elements, objects or relations between them, in reality or in the imagination.
Complete Chapter List
Brian Whitworth, Aldo de Moor
Brian Whitworth, Aldo de Moor
Prologue: General Socio-Technical Theory
Ann Borda, Jonathan P. Bowen
Ken Eason, José Abdelnour-Nocera
Cleidson R.B. de Souza, David F. Redmiles
Prologue: Socio-Technical Perspectives
Petter Bae Brandtzæg, Jan Heim
Wilson Huang, Shun-Yung Kevin Wang
Elayne W. Coakes, Peter Smith, Dee Alwis
Prologue: Socio-Technical Analysis
Jonas Sjöström, Göran Goldkuhl
Paul J. Bracewell
Mikael Lind, Peter Rittgen
Harry S. Delugach
Dorit Nevo, Brent Furneaux
Prologue: Socio-Technical Design
Anders I. Mørch
Manuel Kolp, Yves Wautelet
Anton Nijholt, Dirk Heylen, Rutger Rienks
Jos Benders, Ronald Batenburg, Paul Hoeken, Roel Schouteten
Mary Allan, David Thorns
Rebecca M. Ellis
Christopher A. Miller
Prologue: Socio-Technical Implementation
Laura Anna Ripamonti, Ines Di Loreto, Dario Maggiorini
Mohamed Ben Ammar, Mahmoud Neji, Adel M. Alimi
Pernilla Qvarfordt, Shumin Zhai
Claire de la Varre, Julie Keane, Matthew J. Irvin, Wallace Hannum
Jeremy Birnholtz, Emilee J. Rader, Daniel B. Horn, Thomas Finholt
Prologue: Socio-Technical Evaluation
John M. Carroll, Mary Beth Rosson, Umer Farooq, Jamika D. Burge
Tanguy Coenen, Wouter Van den Bosch, Veerle Van der Sluys
Olga Kulyk, Betsy van Dijk, Paul van der Vet, Anton Nijholt, Gerrit van der Veer
Janet L. Holland
David Hinds, Ronald M. Lee
Bertram C. Bruce, Andee Rubin, Junghyun An
Prologue: The Future of Socio-Technical Systems
Peter J. Denning
Theresa Dirndorfer Anderson
Laurence Claeys, Johan Criel
Kenneth E. Kendall, Julie E. Kendall