The question is not whether intelligent machines can have any emotions, but whether machines can be intelligent without any emotions. – Marvin Minsky, (Minsky, 1986) p. 163 In every art form it is the emotional content that makes the difference between mere technical skill and true art.

When building agents and synthetic characters, and in order to achieve believability, we must consider the emotional relations established between users and characters, that is, we must consider the issue of ”empathy”. Defined in broad terms as ”An observer reacting emotionally because he perceives that another is experiencing or about to experience an emotion”, empathy is an important element to consider in the creation of relations between humans and agents. In this paper we will focus on the role of empathy in the construction of synthetic characters, providing some requirements for such construction and illustrating the presented concepts with a specific system called FearNot!. FearNot! was developed to address the difficult and often devastating problem of bullying in schools. By using role playing and empathic synthetic characters in a 3D environment, FearNot! allows children from 8 to 12 to experience a virtual scenario where they can witness (in a third-person perspective) bullying situations. To build empathy into FearNot! we have considered the following components: agent’s architecture; the characters ’ embodiment and emotional expression; proximity with the user and emotionally charged situations. We will describe how these were implemented in FearNot! and report on the preliminary results we have with it.

The paper presents requirements for the design of unscripted (emergent) dramas based on research into role-playing games. It considers the FearNot! demonstrator in antibullying education as a sample implementation, describing the architecture of its affectively driven intelligent autonomous characters. It presents a comparative evaluation of the unscripted version against an earlier scripted version, examines related work and further development of the emergent narrative approach.

This paper introduces a computational model for emotion regulation formalising the model informally described by Gross (1998). The model has been constructed using a highlevel modelling language, and integrates both quantitative aspects (such as levels of emotional response) and qualitative aspects (such as decisions to regulate one’s emotion). A number of simulation experiments have been performed, demonstrating that the computational model successfully reflects the model as described by Gross.

Spurred by a range of potential applications, there has been a growing body of research in computational models of human emotion. To advance the development of these models, it is critical that we evaluate them against the phenomena they purport to model. In this paper, we present one method to evaluate an emotion model that compares the behavior of the model against human behavior using a standard clinical instrument for assessing human emotion and coping. We use this method to evaluate the EMA model of emotion [1-3]. The evaluation highlights strengths of the approach and identifies where the model needs further development. 1.

Based on the assumption that a rational agent will adopt a plan that maximizes the expected utility, we present a utility-based approach to plan recognition problem in this paper. The approach explicitly takes the observed agent’s preferences into consideration, and computes the estimated expected utilities of plans to disambiguate competing hypotheses. Online plan recognition is realized by incrementally using plan knowledge and observations to change state probabilities. We also discuss the work and compare it with other probabilistic models in the paper.

This paper discusses the requirements of planning for believable synthetic characters and examines the relationship between appraisal and planning as components of an affective agent architecture. It discusses an implementation in the synthetic characters of the FearNot! anti-bullying education demonstrator and how far this provides an adequate mechanism for believable behaviour.

This paper discusses the use of BDI agents for the development of human-like synthetic characters. The folk psychological roots of the paradigm map closely to the way people typically explain both their behaviour and that of others, and this greatly facilitates knowledge elicitation and representation. This is illustrated through some examples from a project in which models of expert players of Quake 2 were developed. The knowledge elicitation methodology that was used is explained, and samples of the code are presented, demonstrating the way in which a BDI-based agent programming language can clearly and succinctly capture individual di#erences. The example presented is of modelling expert players in an existing game, but the paper argues that the same techniques can be used to build a completely original character, using a role-player as the basis. Finally, some of the limitations of the BDI paradigm are examined, with a brief discussion of how they can be addressed, using the existing framework as a basis.

Abstract. To improve the performance and wellbeing of humans in complex human-computer interaction settings, an interesting challenge for an ambient (or pervasive) agent system is to recognise the emotions of humans. To this end, this paper introduces a computational model to estimate the process of emotion generation based on certain triggers. The model has been implemented and tested using the modelling language LEADSTO. A first evaluation indicates that the model is successful in estimating a person’s emotions, and is robust to different parameter settings. 1

Affective reasoning plays an increasingly important role in cognitive accounts of social interaction. Humans continuously assess one another's situational context, modify their own affective state accordingly, and then respond to these outcomes by expressing empathetic behaviors. Synthetic agents serving as companions should respond similarly. However, empathetic reasoning is riddled with the complexities stemming from the myriad factors bearing upon situational assessment. A key challenge posed by affective reasoning in synthetic agents is devising empirically informed models of empathy that accurately respond in social situations. This paper presents CARE, a data-driven affective architecture and methodology for learning models of empathy by observing human-human social interactions. First, in CARE training sessions, one trainer directs synthetic agents to perform a sequence of tasks while another trainer manipulates companion agents ’ affective states to produce empathetic behaviors (spoken language, gesture, and posture). CARE tracks situational data including locational, intentional, and temporal information to induce a model of empathy. At runtime, CARE uses the model of empathy to drive situationappropriate empathetic behaviors. CARE has been used in a virtual environment testbed. Two complementary studies investigating the predictive accuracy and perceived accuracy of CAREinduced models of empathy suggest that the CARE paradigm can provide the basis for effective empathetic behavior control in embodied companion agents. 1.