Final Call for Papers - ACE 2006 Workshop @ EMCSR
submissions due on November 30, Final Call: Agent Construction and Emotions, Workshop on Modeling the Cognitive Antecedents and Consequences of Emotion, April 18-20, 2006, Vienna, Austria
Agent Construction and Emotions
Modeling the Cognitive Antecedents and Consequences of Emotion
submissions due on: November 30
April 18-20, 2006
Vienna, Austria
http://www.ofai.at/~paolo.petta/conf/ace2006
Background
This workshop seeks submissions exploring the argument that theories of human emotion provide essential insight into the design and control of intelligent entities in general. As computational models of intelligence move beyond simple, static and nonsocial problem solving, research must increasingly confront the challenge of how to allocate and focus mental resources in the face of competing goals, disparate and asynchronous mental functions, and a changing interpersonal and physical environment. Contemporary psychological and neuroscience research suggests that the emotions service such needs in biological organisms and a functional analysis of emotion?s impact can profitably inform the design of artificial organisms that must survive in a dynamic, semi-predictable and social world. This workshop builds on a series of prior workshops that seek to deepen and concretize this claim.
Cognitive scientists have long argued that emotional influences that seem irrational on the surface have important social and cognitive functions that would be required by any intelligent system. For example, Herb Simon theorized that emotions serve to interrupt normal cognition when unattended goals require servicing. Robert Frank argues that social emotions such as anger and guilt reflect a mechanism that improves group utility by minimizing social conflicts, and thereby explains people's "irrational" choices to cooperate in social games such as the prisoner's dilemma. Similarly, Alfred Mele claims that "emotional biases" such as wishful thinking reflect a rational mechanism that more accurately accounts for social costs, such as the cost of betrayal when a parent defends a child despite strong evidence of their guilt in a crime. At the same time, findings on non-conscious judgments (e.g., Bargh; Gollwitzer; Schwarz&Clore) have enriched our understanding of how cognitive style is shaped by the socio-emotional context, often in adaptive ways. More broadly, appraisal theorists such as Lazarus, Frijda and Scherer have argued that emotions are intimately connected with how organisms sense events, relate them to internal needs (e.g., is this an opportunity or a threat?), characterize appropriate responses (e.g., fight, flight or plan) and recruit the cognitive, physical and social resources needed to adaptively respond.
Recent years have seen an explosion of interest in using such findings to a wide array of computational problems including action selection, resource allocation, multi-agent coordination and the management of beliefs and intentions. The object of this workshop will be to strengthen the growing interdisciplinary synthesis between computational and psychological research on the role the emotions play in modeling intelligent behavior.
Topics
The workshop will explore the intersection of emotion theory and intelligent system design, and the potential for this intersection to improve our understanding of both human and artificial intelligence. In particular, we seek to emphasize the interplay between emotion and deep models of cognition in adaptively navigating complex physical and social environments. This places an emphasis on psychological paradigms that stress cognitive processes, such as appraisal theory, computational systems that model the cognitive antecedents and consequences of emotion, and research that models emotion-evoking social and task environments.
Specific topics of interest include:
- Computational accounts of the connection between emotion and cognitive processes (including planning, language processing, interaction, perception, etc.)
- Theoretical accounts of the adaptive function of emotion processing
- Computational models that abstract the posited function of emotion processing and illustrate an adaptive advantage over classical theories in concrete domains (e.g. planning, decision making, action selection, social coordination, etc.)
- Empirical research testing process assumptions of theories of human emotion
- Empirical research illustrating the adaptive (or maladaptive) role of emotions in human cognition
- The use of computational models or methods to evoke emotion in human subjects
- Techniques for modeling emotionally evocative social or physical environments (i.e., "emotional" extensions to cognitive task analysis)
Organizing Committee
- Jonathan Gratch, University of Southern California
- Stacy Marsella, University of Southern California
- Paolo Petta, Center for Brain Research, Medical University of Vienna
Program committee
- Ruth Aylett, Heriot-Watt University
- Lola Cañamero, University of Hertfordshire
- William J. Clancey, NASA Ames
- Gerald Clore, University of Virginia
- Cristina Conati, University of British Columbia
- Eva Hudlicka, Psychometrix Associates
- Susanne Kaiser, University of Geneva
- Tim Ketelaar, New Mexico State University
- Agnes Moors, Ghent University
- Josef Nerb, University of Freiburg
- Andrew Ortony, Northwestern University
- Ana Paiva, Instituto Superior Técnico
- Brian Parkinson, University of Oxford
- Rosalind Picard, MIT
- Rainer Reisenzein, University of Greifswald
- Fiorella de Rosis, University of Bari
- Matthias Scheutz, Notre Dame University
- Craig Smith, Vanderbilt University
- Niels Taatgen, University of Groningen
- Thomas Wehrle, University of Zurich
Submission Details
The workshop is held in conjunction with the 18th European Meeting on Cybernetics and Systems Research (http://www.osgk.ac.at/emcsr/).
Submissions will be due on November 30, 2005, must be written in English and must not exceed 6 pages (10-point, double column). Please refer to the home page of EMCSR 2006 for detailed instructions on the formatting and submission procedures
Related Past Meetings
- Architectures for Modeling Emotion: Cross-Disciplinary Foundations. AAAI 2004 Spring Symposium Series, Stanford University, Palo Alto, California, USA, March 22-24, 2004.
- ACE 2004
- Emotion, Evolution and Rationality, interdisciplinary conference hosted by the Philosophy Department at King's College London. Saturday 27 April - Sunday 28 April 2002.
- ACE 2002
- Symposium on Emotion, Cognition, and Affective Computing. AISB'01 Convention, March 21-24, 2001.
- ACE 2000
- Autonomous Agents 1999 Workshops on Autonomy Control Software and on Emotion-Based Agent Architectures (Seattle, WA, USA).
- Emotional and Intelligent: The Tangled Knot of Cognition, 1998 AAAI Fall Symposium Series
- Grounding Emotions in Adaptive Systems at SAB'98, Zurich, Switzerland.

Emotion-Aware Natural Interaction
