The design of an affect recognition system for socially perceptive robots relies on representative data: human-robot interaction in naturalistic settings requires to be trained and validated with contextualised affective expressions, that is, expressions emerge the same scenario target application. In this paper we propose initial computational model automatically analyse human postures body motion detect engagement children playing chess iCat robot acts as a game companion. Our approach is...

Affect sensitivity is of the utmost importance for a robot companion to be able display socially intelligent behaviour, key requirement sustaining long-term interactions with humans. This paper explores naturalistic scenario in which children play chess iCat, companion. A person-independent, Bayesian approach detect user's engagement iCat presented. Our framework models both causes and effects engagement: features related non-verbal task companion's affective reactions are identified predict...

In this paper we describe a robotic game buddy whose emotional behaviour is influenced by the state of game. Using iCat robot and chess as scenario, an architecture for incorporating emotions result heuristic evaluation was developed. The evaluated in two ways. First, investigated effects characterpsilas on userpsilas perception state. And secondly compared with screen based version terms their influence enjoyment. results suggested that increases iCatpsilas behaviour, enjoyment higher when...

The idea of autonomous social robots capable assisting us in our daily lives is becoming more real every day. However, there are still many open issues regarding the capabilities that those should have order to make interactions with humans natural. For example, role affective unclear. This paper presents an ethnographic study conducted elementary school where 40 children interacted a robot recognising and responding empathically some children's states. findings suggest robot's empathic...

Given the recent advances in robot and synthetic character technology, many researchers are now focused on ways of establishing social relations between these agents humans over long periods time. Early studies have shown that novelty effect robots quickly wears out people change their attitudes preferences towards them In this paper, we study role presence long-term human-robot interactions. We conducted a where children played chess exercises with five week period. With experiment,...

Many small group activities, like working teams or study groups, have a high dependency on the skill of each member. Differences in level among participants can affect not only performance team but also influence social interaction its members. In these circumstances, an active member could balance individual participation without exerting direct pressure specific members by using indirect means communication, such as gaze behaviors. Similarly, this study, we evaluate whether robot language...

This paper presents an experiment that evaluates and compares the user enjoyment when playing a game of chess in two situations: against physically embodied robotic agent virtually agent, displayed on screen. The results study suggest embodiment has implications enjoyment, as experience was classified more enjoyable than agent.

Humans have a tendency to consider media devices as social beings. Social agents and artificial opponents can be examined one instance of this effect. With today's technology it is already possible create that are perceived socially present. In paper, we start by identifying the factors influence perceptions presence in human-agent interactions. By taking these into account following previously defined guidelines for building present opponents, case study was created which robot plays Risk...

The adoption of virtual assistants is growing at a rapid pace. However, these are not optimised to simulate key social aspects human conversational environments. Humans intellectually biased toward activity when facing anthropomorphic agents or presented with subtle cues. In this paper, we test whether humans respond the same way in guided tasks, different forms embodiment and behaviour. within-subject study (N=30), asked subjects engage dialogue smart speaker robot. We observed shifting...

The increasing use of robots in real-world applications will inevitably cause users to encounter more failures interactions. While there is a longstanding effort bringing human-likeness robots, how robot embodiment affects users' perception remains largely unexplored. In this paper, we extend prior work on by assessing the impact that and failure severity have people's behaviours their robots. Our findings show when using smart-speaker embodiment, negatively affect intention frequently...

A key question in research on the neurobiology of language is to which extent production and comprehension systems share neural infrastructure, but this has not been addressed context conversation. We utilized a public fMRI dataset where 24 participants engaged unscripted conversations with confederate outside scanner, via an audio-video link. provide evidence indicating that two infrastructure left-lateralized perisylvian network, diverge regarding level activation regions within network....

Despite a large body of existing literature on automatic affect recognition, there seems to be lack studies investigating task and social context for the purpose automatically predicting affect. This work aims take current state art step forward explore role their interdependencies in prediction user engagement HRI scenario involving an iCat robot playing chess with young children. We performed experimental evaluation by training several SVMs-based models different features extracted from...

Affect recognition for socially perceptive robots relies on representative data. While many of the existing affective corpora and databases contain posed decontextualized expressions, affect resources designing an system in naturalistic human–robot interaction (HRI) must include context-rich expressions that emerge same scenario final application. In this paper, we propose a context-based approach to collection modeling data building affect-sensitive robotic game companion. To illustrate our...

Humans and robots will increasingly collaborate in domestic environments which cause users to encounter more failures interactions. Robots should be able infer conversational by detecting human users' behavioural social signals. In this paper, we study analyse these cues response robot failures. Using a guided task corpus, where embodiment time pressure are manipulated, ask annotators estimate whether user affective states differ during various types of We also train random forest classifier...

In this paper, we study the role of emotions and expressive behaviour in socially interactive characters employed educational games. More specifically, on how can use such emotional to help users better understand game state. An emotion model for these characters, which is mainly influenced by current state based emotivector anticipatory mechanism, was developed. We implemented a social robot named iCat, using chess as scenario. The results preliminary evaluation suggested that embedded...

Robot companions must be able to display social, affective behaviour. As a prerequisite for companionship, the ability sustain long-term interactions with users requires endowed affect recognition abilities. This paper explores application-dependent user states in naturalistic scenario where an iCat robot plays chess children. In this scenario, role of context is investigated modelling related both task and social interaction robot. Results show that contextual features game iCat's behaviour...

Social perception abilities are among the most important skills necessary for robots to engage humans in natural forms of interaction. Affect-sensitive more likely be able establish and maintain believable interactions over extended periods time. Nevertheless, integration affect recognition frameworks real-time human-robot interaction scenarios is still underexplored. In this article, we propose evaluate a context-sensitive framework robotic game companion children. The robot can...