Showing papers by "Ana Paiva published in 2011"

Automatic analysis of affective postures and body motion to detect engagement with a game companion

Abstract: The design of an affect recognition system for socially perceptive robots relies on representative data: human-robot interaction in naturalistic settings requires an affect recognition system to be trained and validated with contextualised affective expressions, that is, expressions that emerge in the same interaction scenario of the target application. In this paper we propose an initial computational model to automatically analyse human postures and body motion to detect engagement of children playing chess with an iCat robot that acts as a game companion. Our approach is based on vision-based automatic extraction of expressive postural features from videos capturing the behaviour of the children from a lateral view. An initial evaluation, conducted by training several recognition models with contextualised affective postural expressions, suggests that patterns of postural behaviour can be used to accurately predict the engagement of the children with the robot, thus making our approach suitable for integration into an affect recognition system for a game companion in a real world scenario.

Emotion in games

Abstract: Computer games are unique elicitors of emotion. Recognition of player emotion, dynamic construction of affective player models, and modelling emotions in non-playing characters, represent challenging areas of research and practice at the crossroads of cognitive and affective science, psychology, artificial intelligence and human-computer interaction. Techniques from AI and HCI can be used to recognize player affective states and to model emotion in non-playing characters. Multiple input modalities provide novel means for measuring player satisfaction and engagement. These data can then be used to adapt the gameplay to the player's state, to maximize player engagement and to close the affective game loop. The Emotion in Games workshop (EmoGames 2011 http://sirenproject. eu/content/acii-2011-workshop-emotion-games) will bring together researchers and practitioners in affective computing, user experience research, social psychology and cognition, machine learning, and AI and HCI, to explore topics in player experience research, affect induction, sensing and modelling and affect-driven game adaptation, and modelling of emotion in nonplaying characters. It will also provide new insights on how gaming can be used as a research platform, to induce and capture affective interactions with single and multiple users, and to model affect- and behaviour-related concepts, helping to operationalize concepts such as flow and engagement. The workshop will include a keynote, paper and poster presentations, and panel discussions. Selected papers will appear in a special issue of the IEEE Transactions on Affective Computing, "Emotion in Games", in mid-2013. The EmoGames2011 workshop is organized in coordination with the newly formed 'Emotion in Games' Special Interest Group (SIG) of the Humaine Association and the IEEE Computational Intelligence Society (CIS) Task Force on Player Satisfaction Modelling. We would like to thank all participants, as well as the members of the Program Committee, for their reviews of the workshop submissions: Elisabeth Andre, Ruth Aylett, Nadia Bianchi-Berthouze, Antonio Camurri, Marc Cavazza, Jonathan Gratch, Hatice Gunes, Dirk Heylen, Katherine Isbister, Stefanos Kollias, Maurizio Mancini, Anton Nijholt, Julian Togelius, Asimina Vasalou, Gualtiero Volpe, Catherine Pelachaud, and Tom Ziemke.

Modelling empathy in social robotic companions

Abstract: Empathy can be broadly defined as the ability to understand and respond appropriately to the affective states of others. In this paper, we present a scenario where a social robot acts as a chess companion for children, and describe our current efforts towards endowing such robot with empathic capabilities. A multimodal framework for modeling some of the user's affective states that combines visual and task-related features is presented. Using this model of the user, we personalise the learning environment by adapting the robot's empathic responses to the particular preferences of the child who is interacting with the robot. We also describe a preliminary study conducted in this scenario.

“FearNot!”: a computer-based anti-bullying-programme designed to foster peer intervention

Abstract: Bullying is widespread in European schools, despite multiple intervention strategies having been proposed over the years. The present study investigates the effects of a novel virtual learning strategy (“FearNot!”) to tackle bullying in both UK and German samples. The approach is intended primarily for victims to increase their coping skills and further to heighten empathy and defence of victims by non-involved bystanders. This paper focuses on the defender role. Applying quantitative as well as qualitative methodology, the present study found that “FearNot!” helped non-involved children to become defenders in the German sub-sample while it had no such effect in the UK sub-sample. German “New Defenders” (children who are initially uninvolved but are nominated as defenders by their peers after the intervention period) were found to be significantly more popular at baseline, and to show more cognitive empathy (Theory of Mind) for the virtual victims as compared to permanently non-involved pupils. Moreover, gender interacts with becoming a defender in its effects on affective empathy, with emotional contagion being particularly associated with New Defender status among girls. The findings are discussed in relation to previous research on anti-bullying intervention strategies and cultural differences in bullying prevalence rates and intervention outcomes.

Emotion-based intrinsic motivation for reinforcement learning agents

Abstract: In this paper, we propose an adaptation of four common appraisal dimensions that evaluate the relation of an agent with its environment into reward features within an intrinsically motivated reinforcement learning framework. We show that, by optimizing the relative weights of such features for a given environment, the agents attain a greater degree of fitness while overcoming some of their perceptual limitations. This optimization process resembles the evolutionary adaptive process that living organisms are subject to. We illustrate the application of our method in several simulated foraging scenarios.

A computational approach towards conflict resolution for serious games

Abstract: Conflict is an unavoidable feature of life, but the development of conflict resolution management skills can facilitate the parties involved in resolving their conflicts in a positive manner. The goal of our research is to develop a serious game in which children may experiment with conflict resolution strategies and learn how to work towards positive conflict outcomes. While serious games related to conflict exist at present, our work represents the first attempt to teach conflict resolution skills through a game in a manner informed by sociological and psychological theories of conflict and current best practice for conflict resolution. In this paper, we present a computational approach to conflict generation and resolution. We describe the five phases involved in our conflict modeling process: conflict situation creation, conflict detection, player modeling and conflict strategy prediction, conflict management, and conflict resolution, and discuss the three major elements of our player model: assertiveness, cooperativeness, and relationship. Finally, we overview a simple resource management game we have developed in which we have begun experimenting with our conflict model concepts.

Robots meet IVAs: a mind-body interface for migrating artificial intelligent agents

Abstract: We describe CMION, an open source architecture for coordinating the various sensors and effectors of an artificial intelligent agent with its mind, i.e. the high level decision making processes. The architecture was designed to work for virtual graphical agents, including those on mobile devices, as well as robots. Its built-in migration feature allows a character to move between these differing embodiments, inhabiting them in turn. We emphasize the importance of modularity for an architecture supporting migration and highlight design decisions promoting modularity in CMION. An applied example of the architecture's use in a migration situation is given.

Long-term socially perceptive and interactive robot companions: challenges and future perspectives

Abstract: This paper gives a brief overview of the challenges for multi-model perception and generation applied to robot companions located in human social environments. It reviews the current position in both perception and generation and the immediate technical challenges and goes on to consider the extra issues raised by embodiment and social context. Finally, it briefly discusses the impact of systems that must function continually over months rather than just for a few hours.

ViPleo and PhyPleo: Artificial pet with two embodiments

Abstract: In our current work we have designed and implemented an articial pet with two embodiments. In both embodiments behavior is driven by needs, that are used to maintain coher- ence and motivate user i ...

ViPleo and PhyPleo: artificial pet with two embodiments

Abstract: In our current work we have designed and implemented an artificial pet with two embodiments. In both embodiments behavior is driven by needs that are used to maintain coherence and motivate user interaction. These needs are transferred between embodiments, with only one embodiment being active at a time. We performed an evaluation with 10-year old children participants. The retrieved data indicated that many children understood the concept of an artificial pet with two bodies, even without being given clues. Nevertheless, children did perceive differences between the two embodiments, which contributed for many stating that they interacted with two pets. Among other aspects, the physical version was perceived as less obedient due to problems concerning action recognition. Although caused by technical issues, this result raises the question if virtual embodiments should simulate action recognition problems that their physical counterparts have.

A generic emotional contagion computational model

Abstract: This work describes a computational model designed for emotional contagion simulation in societies of agents, integrating the influence of interpersonal relationships and personality. It models the fundamental differences in individual susceptibilities to contagion based on the psychology study of Emotional Contagion Scale. The contagion process can also be biased by inter-individual relationships depending on the intimacy and power difference aspects of relationships between agents. Individuals' expressiveness in a group is influenced by both the extroversion personality trait and power difference. Additionally, the computational model includes the process of mood decay, as usually observed in people, expanding its application domain beyond that of pure simulation, like games. In this paper we present simulation results that verify the basic emotional contagion behaviors. The possibility of more complex contagion dynamics depending on agent group relationships is also presented.

Emerging social awareness: Exploring intrinsic motivation in multiagent learning

Abstract: Recently, a novel framework has been proposed for intrinsically motivated reinforcement learning (IMRL) in which a learning agent is driven by rewards that include not only information about what the agent must accomplish in order to “survive”, but also additional reward signals that drive the agent to engage in other activities, such as playing or exploring, because they are “inherently enjoyable”. In this paper, we investigate the impact of intrinsic motivation mechanisms in multiagent learning scenarios, by considering how such motivational system may drive an agent to engage in behaviors that are “socially aware”. We show that, using this approach, it is possible for agents to learn individually to acquire socially aware behaviors that tradeoff individual well-fare for social acknowledgment, leading to a more successful performance of the population as a whole.

Agents with emotional intelligence for storytelling

Abstract: One core aspect of engaging narratives is the existence and development of social relations between the characters. However, creating agents for interactive storytelling and making them to be perceived as a close friend or a hated enemy by an user is an hard task. This paper addresses the problem of creating autonomous agents capable of establishing social relations with others in an interactive narrative. We present an innovative approach by looking at emotional intelligence and in particular to the skills of understanding and regulating emotions in others. To that end we propose a model for an agent architecture that has an explicit model of Social Relations and a Theory of Mind about others, and is able to plan about emotions of others and perform interpersonal emotion regulation in order to dynamically create relations with others. Some sample scenario are presented in order to illustrate the type of behaviour achieved by the model and the creation of social relations.

I've been here before!: location and appraisal in memory retrieval

Abstract: The objective of our current work was to create a model for agent memory retrieval of emotionally relevant episodes. We analyzed agent architectures that support memory retrieval realizing that none fulfilled all of our requirements. We designed an episodic memory retrieval model consisting of two main steps: location ecphory, in which the agent's current location is matched against stored memories associated locations; and recollective experience, in which memories that had a positive match are re-appraised. We implemented our model and used it to drive the behavior of characters in a game application. We recorded the application running and used the videos to create a non-interactive evaluation. The evaluation's results are consistent with our hypothesis that agents with memory retrieval of emotionally relevant episodes would be perceived as more believable than similar agents without it.

"I'm happy if you are happy.": a model for emotional contagion in game characters

Abstract: Emotions play an important role in social interactions and as such, they are critical in creating engaging and believable characters that users can interact with.Although there has been significant research on emotions, from a computational perspective, one area scarcely explored is the process of Emotional Contagion (EC). Emotional Contagion is the process through which a person's emotional state is influenced by other people's emotions. This process is especially important in group situations where the emotional states of individuals influence the behavior of others.Our goal was to develop a computational model, based on the Emotional Contagion Scale (ECS), that would enrich group dynamics on virtual environments. This model supports agents with different susceptibilities to contagion, and also the influence of their extroversion personality trait and interpersonal relationships (intimacy and power difference) on the contagion process. It has been also included the process of emotional mood decay, as observed in humans. With these elements characterizing the agents at an individual level, this model can simulate a wide variety of social phenomena.To evaluate the model, we developed a game prototype where the player (the main character) goes on a journey with two other characters (one with and another without our model).The results suggest that players perceived differences in the emotional contagion susceptibility between those characters and considered the model-based agent significantly more enjoyable and friendlier.

Modelling rituals for Homo biologicus.

Abstract: This paper builds on a generic conceptual model to be used in research that uses agent-based models. The aim is to model human social interaction at a detailed level, such as would be needed for believable interactions of embodied virtual characters. This means we model Homo biologicus, not Homo economicus. The present conceptual article investigates the modelling of rituals as a basic unit of group interaction. The ideas introduced here aim at modelling the unwritten rules of social interaction across cultures. The main example is about smoking in bars.

Games based learning for Exploring Cultural Conflict

Abstract: In this paper we discuss the early stage design of MIXER, a technology enhance educational application focused at supporting children in learning about cultural conflict, achieved through the use of a game with an effective embodied AI agent. MIXER is being developed re-using existing technology applied to a different context and purpose with the aim of creating an educational and enjoyable experience for 9-11 year olds. This paper outlines MIXER’s underpinning technology and theory. It presents early stage design and development, highlighting current research directions.

A personal approach: the persona technique in a companion's design lifecycle

Abstract: Artificial companions are a new type of technology that is changing the way people interact with intelligent systems, by trying to build and sustain long term relationships with users. To evaluate such systems the typical usability methods for system evaluation are not enough, due to the need of accessing aspects such as social behaviour, emotional sensitivity and personalized interaction over long periods of time, with very specific users. In this paper, we describe the (full) design cycle of a companion system, arguing that a user-centred approach is the more appropriate, taking into account the referred specific issues. As such, to help us focus on developing a companion system for the prospective users, we based our design in the archetype of actual users - a persona. Furthermore, we brought this same concept into the evaluation phase in order to access the companion performance in a long-term interaction.

Providing gender to embodied conversational agents

Abstract: Communication, along with other factors, varies with gender. Significant work as been done around embodied conversational agents (ECAs) verbal and non-verbal behaviour but gender issue has often been ignored. Yet, together with personality, culture and other factors, gender is a feature that impacts the perception and thus the believability of the characters. The main goal of this work is to understand how gender can be provided to ECAs, and provide a very simple model that allows for existing tools to overcome such limitation. The proposed system was developed around SAIBA Framework using SmartBody as the behavior realizer and tries to address this problem by adding a set of involuntary gender specific movements to the agents behaviour in an automatic manner. This is achieved by revising and complementing the work done by the existing non-verbal behaviour generators. Focusing mainly on nonverbal behaviour, our agents with gender were tested to see if users were able to perceive the gender bias of the behaviours being performed. Results have shown that gender is correctly perceived, and also has effects when paired with an accurate gender appearance.

Tilt riders: improvisational agents who know what the scene is about

Abstract: The creation of autonomous agents for interactive narrative requires a heavy authorial effort, especially when the authors are concerned about integrating all possible story states in the agents behaviors. In this paper, we consider that the autonomous agents lack of narrative perspective over the action prevents them from successfully dealing with unpredicted story states. To deal with this problem we propose a conceptual model of story development for autonomous agents that endows narrative control to them. The proposed model is supported by our cognitive research with improvisational theatre actors and improvisational theatre theory.

Expressing emotions on robotic companions with limited facial expression capabilities

Abstract: Facing the challenge of accurately expressing emotions with robots and characters with limited expressive capabilities, we developed an abstraction model for generating emotional expressions based on the atomic features that enable human beings to recognize emotions in other humans' faces. The model is also augmented by animation theory from movies and puppetry. A small evaluation of the expressions showed that some expressions were well recognized, in particular Anger and Sadness.

A computational model for finding the tilt in an improvised scene

Abstract: Improvisational theatre (improv) is a real world example of an interactive narrative environment that has a strong focus on the collaborative construction of narrative as a joint activity. Although improv has been used as an inspiration for computational approaches to interactive narrative in the past, those approaches have generally relied on shallow understandings of how theatrical improvisation works in terms of the processes and knowledge involved. This paper presents a computational model for finding the tilt in a narrative environment with no pre-authored story structures, based on our own cognitively-based empirical studies of real world improvisers.

Persu: an architecture to apply persuasion in interactive storytelling

Abstract: In this paper we describe an architecture to influence a user in an interactive storytelling context that is based on results from the social psychology's area of persuasion. Several important concepts of persuasion, such as how people make decisions, and how we can influence that process are discussed. We describe the several components of the architecture and how we applied them in a small study where we have successfully influenced the players of a story in following a specific path.

Towards the next generation of board game opponents

Abstract: While computer games can give users many other forms of immersive entertainment, the social richness that exists in face to face interactions is being lost. In this paper we start by presenting a brief review on artificial board game opponents. Given the current limitations in such opponents and with the final goal of building a Risk game artificial opponent, we describe an empirical study where utterances that human players vocalize in their games were categorized and analyzed.

SARA: social affective relational agent: a study on the role of empathy in artificial social agents

Abstract: Over the last decade extensive research has been conducted in the area of conversational agents focusing in many different aspects of these agents. In this research, and aiming at building agents that maintain a social connection with users, empathy has been one of those areas, as it plays a leading role in the establishment of social relationships. In this paper we present a relationship model of empathy that takes advantage of Social Penetration Theory's concepts for relationship building. This model has been implemented into an agent that attempts to establish a relationship with the user, expressing empathy both verbally and visually. The visual expression of empathy consists of facial expression and physical proximity representation. The user tests performed showed that while users were able to develop a simple relationship with the agents, they however developed stronger relationships with a version of the agent that is most visually expressive and takes advantage of the proximity element, confirming the significance of our model based on social penetration theory may have and, consequently, the importance of the visual representation of empathic responses.

A game prototype with emotional contagion

Abstract: Emotional contagion (EC) in games may provide players with an unique experience We have developed a turn-based role playing prototype game which incorporates a model based on the EC process While playing, users have the opportunity to observe the effects of emotional events on individual characters and on the group through simulated emotional contagion dynamics

I'm sure i made the right choice!: towards an architecture to influence player's behaviors in interactive stories

Abstract: In this paper we present an architecture for Interactive Storytelling systems that dynamically selects persuasive manipulations to increase the likelihood of the users experiencing the story as intended by an author. We also describe a study using a text-based interactive storytelling system where the architecture was applied.

"How should I say this?" Agents with culturally-appropriate verbal communication styles

Abstract: There are several cultural differences in the way humans communicate, both in verbal and non-verbal communication. Research on creating agents that are able to simulate such differences is currently growing, in part for their potential use in designing Virtual Learning Environments for intercultural training. In this paper we analyse a particular aspect of communication that has not yet received much attention in culture-specific agents, which is the verbal style used. Using our current cultural agent architecture, a small scenario was created where the verbal style employed by the agents depends on their cultural parametri-sation. With the scenario created, an experiment was conducted where users did perceive differences between two cultural versions of the agents.