-This paper investigates user perceptions of continuous identity as agents migrate between different embodiments. It reports an experiment seeking to establish whether migrating or not migrating the interaction memory of the agent would affect the user's perception of consistent agent identity over different embodiments. The experiment involved a treasure hunt in which a virtual agent migrated from a screen to a mobile phone in order to accompany a user while they searched for clues. A total of 45 subjects took part in three different conditions with 15 subjects in each. The outcome showed that the presence of memory affected the competence users ascribed to the virtual agent but had no significant effect on a strong perception of consistent identity across multiple embodiments.

Abstract. Social reasoning theories, whilst studied extensively in the area of multiagent systems, are hard to implement directly in agents. They often specify properties of beliefs or behaviours but not the way these should affect the computational reasoning mechanisms of a concrete agent design. The Expectation-Strategy-Behaviour (ESB) framework addresses this problem by separating and abstracting social reasoning from other practical reasoning, providing the computational machinery that is necessary to perform social reasoning in practice. We present an extension to previous work on ESB to a reasoning system which enables the execution of concise and modular declarative social reasoning rules. These rules represent the belief revision inherent to social reasoning, by representing hidden properties of other agents and social constraints. We review the foundations of the abstract ESB framework and present the implementation of a reasoner based on CTL model checking. Our system allows for conditioning agent behaviours on complex preconditions and verification of properties to aid the agent designer. It also allows for easy integration with a BDI reasoning system. We exemplify the suitability of ESB for social reasoning constructs with a specification of Joint Intention theory in ESB and illustrate the modularity of the representation by extending this to a specific example in a robotic soccer domain.

This article describes a framework for practical social reasoning designed to be used for analysis, specification, and implementation of the social layer of agent reasoning in multiagent systems. Our framework, called the expectation strategy behavior (ESB) framework, is based on (i) using sets of update rules for social beliefs tied to observations (so-called expectations), (ii) bounding the amount of reasoning to be performed over these rules by defining a reasoning strategy, and (iii) influencing the agent's decision-making logic by means of behaviors conditioned on the truth status of current and future social beliefs. We introduce the foundations of ESB conceptually and present a formal framework and an actual implementation of a reasoning engine, which is specifically combined with a general (belief-desire-intention-based) practical reasoning programming system. We illustrate the generality of ESB through select case studies, which show that it is able to represent and implement different typical styles of social reasoning. The broad coverage of existing social reasoning methods, the modularity that derives from its declarative nature, and its focus on practical implementation make ESB a useful tool for building advanced socially reasoning agents.