Showing papers by "Cristiano Castelfranchi published in 2003"

The Micro-Macro Constitution of Power

Abstract: How important power is in human relations, organizations and societies does not need many words to be argued. Consequently, power is also a fundamental topic in the social sciences, in particular in sociological theory (starting from classics like Marx, Pareto, Weber), in political science, in organization sciences, in anthropology; but also in economics (negotiation power, position power, theory of oligopolies and monopolies, etc.). On the contrary, Cognitive Science and, in particular, models of artificial intelligence and interaction have systematically ignored this perspective (Castelfranchi, 1991). This is probably due to the fact that the prevailing perspective in AI was ‘individualism’ and ‘mentalism’ (Gasser, 1991) that looks far from this social phenomenon. We will on the contrary show that for a good theory of power

The Micro-Macro Constitution of Power

Abstract: The focus of this work is the complex and dialectic relationship between personal, social, and institutional powers. This distinction is not new, it is even obvious or commonsensical. However, first, this multi-layered nature of power is not acknowledged in all the disciplines (for instance sociology typically rejects the idea of a connection between the personal powers and the social ones); second, in our view there is not a good analytic theory of the personal and interpersonal layers of power and of their – not so obvious – relationships with the power that we prefer to call “institutional” (where the “legal” one is just a sub-case based not simply on conventions, traditions, and social norms but on a system of laws). For example, the term “physical power” is a bit reductive, since in fact one not only needs strength, skills, and a working body, but the appropriate action plan and the relevant information, i.e. knowledge, and also the required mental abilities and conditions (motives, awareness, confidence, and so on).

Trust in information sources as a source for trust: a fuzzy approach

Abstract: The aim of this paper is to show how relevant is a trust model based on beliefs and their credibility.The approaches to the study of trust are various and very different from each of other. In our view, just a socio-cognitive approach to trust would be able to analyse the sub-components on which the final decision to trust or not is taken. In this paper we show an implementation of our socio-cognitive model of trust developed using the so-called Fuzzy Cognitive Maps. The model allows to distinguish between internal and external attributions and it introduced a degree of trust derived from the credibility of the trust beliefs, while the credibility of the beliefs derives from their sources and the sources' number, convergence, reliability (i.e. trust).With this implementation we show how the different components may change and how their impact can change depending on the specific situation and from the agent heuristics or personality. In particular, we analyse the different nature of the belief sources and their trustworthiness. We assumed different types of belief sources. For each trustier's belief one should consider what the content of the belief is, who/what the source is, how this source evaluates the belief, how the trustier evaluates this source (with respect to this belief). In addition for considering the contribution of different sources we need a theory of how they combine. The interesting thing in this paper is that starting from finding the sources of trust we are obliged to consider the trustworthiness of these sources.

A fuzzy approach to a belief-based trust computation

Abstract: The approaches to the trust study are various and very different each of other. In our view, just a socio-cognitive approach to trust would be able to analyze the sub-components (the basic beliefs) on which the final decision to trust or not is taken. In this paper we show a first implementation and advance of the socio-cognitive model of trust developed in [1, 2] using the so-called Fuzzy Cognitive Maps. With this implementation we show how the different components may change and how their impact can change depending from the specific situation and from the agent personality.

Agent Autonomy

Abstract: This paper summarizes the state of art in agent autonomy. It dispels myths and builds a foundation for study of autonomy. We point to a renewed interest in good old-fashioned AI that has emerged from consideration of agents and autonomy. This paper also serves as a reader's guide to the paper in this book. We end with sobering thoughts about the future of the human relationship with machines.

From Automaticity to Autonomy: The Frontier of Artificial Agents

Abstract: In the paper we give a definition of autonomy that tries to answer to the main questions this concept brings with it We also provide an operationalized notion of autonomy In particular, we interpret autonomy as a concept strictly related with other relevant notions: dependence, delegation, awareness, predictability, automaticity We analyze more deeply autonomy in collaboration, its relationships with control and finally, the modalities of its possible adjustments

Can Computers Deliberately Deceive? A Simulation Tool and Its Application to Turing's Imitation Game

Abstract: In this paper, we describe how agents can deceive within a probabilistic framework for representing their mental state: in doing so, we challenge the so-called sincerity assumption in Human–Computer interaction (HCI) and multi-agent systems (MAS). We distinguish deception from its special case of lie and characterize different forms of deception, by identifying several criteria for distinguishing among them. In particular, we propose a model of information impact on the Receiver's mind. As the message Sender must plan its strategy by considering the Receiver's criteria for believing, we also discuss some of these criteria, like content plausibility, source informativity, and information safety. We apply this model to a simplified version of Turing's Imitation Game and describe how we implemented a Simulator of deceptive strategies that we called Mouth of Truth. We conclude the paper by describing an evaluation study that enabled us to verify the validity of our method and to revise it in part.

Integrating trustfulness and decision using fuzzy cognitive maps

Abstract: The aim of this paper is to show how relevant is a trust model based on beliefs and their credibility. We will also show how the Fuzzy Cognitive Maps are convenient and practicable for implementing and integrating trustfulness and delegation.

Founding autonomy: the dialectics between (social) environment and agent's architecture and powers

Abstract: 'Autonomy', with 'interaction' the central issue of the new Agent-based AI paradigm, has to be recollected to the internal and external powers and resources of the Agent. Internal resources are specified by the Agent architecture (and by skills, knowledge, cognitive capabilities, etc.); external resources are provided (or limited) by accessibility, competition, pro-social relations, and norms. 'Autonomy' is a relational and situated notion: the Agent - as for a given needed resource and for a goal to be achieved - is autonomous from the environment or from other Agents. Otherwise it is 'dependent' on them. We present this theory of Autonomy (independence, goal autonomy, norm autonomy, autonomy in delegation, discretion, control autonomy, etc.) and we examine how acting within a group or organization reduces and limits the Agent autonomy, but also how this may provide powers and resources and even increase the Autonomy of the Agent.

The Prescriptive Destiny of Predictive Attitudes: From Expectations to Norms via Conventions

Abstract: The goal of this paper is to show the normative component of a convention adopted by a population of cognitive agents To address this aim we will defend two distinct thesis The former is that even simple predictions developed to anticipate future state of affairs have an intrinsic tendency to evolve in full expectations and then in prescriptions We consider this as a multilevel phenomenon occurring either at the individual psychological level or at the interpersonal one or, finally, at the collective macro social level The latter thesis is that we consider this tendency as one of the possible paths of the spontaneous emergence of agents’ commitments, of conventions and likely of real social norms: the tacit emergence of a prescriptive character and, then, of obligations and duties We will examine the constitutive elements –both cognitive and relational – of this process of spontaneous transition from the predictive attitudes to the prescriptive ones and, on this basis, to real normative attitudes Finally, we will discuss the inevitably normative component of conventions as traditionally described (Lewis 1969) We will argue that this fundamental process is notably left implicit or insufficiently explained

A Prospectus on Agent Autonomy

Abstract: This paper summarizes the state of art in agent autonomy. It dispels myths and builds a foundation for study of autonomy. We point to a renewed interest in good old-fashioned AI that has emerged from consideration of agents and autonomy. This paper also serves as a reader's guide to the paper in this book. We end with sobering thoughts about the future of the human relationship with machines.

Grounding We-Intentions In Individual Social Attitudes

Abstract: Especially after Raimo Tuomela’s influential work there is now a general agreement, both in AI and in philosophy, on the idea that to model and formalise cooperation it is necessary to model the minds of the involved agents.2 I think that this is a fundamental result. However, a collective activity is mainly accounted for in terms of the beliefs of the agents involved about each other and the joint plan. I think that this approach is not sufficient to account for a group or a truly cooperative work because a much richer representation of the individual social mind is needed (Conte and Castelfranchi, 1995). In fact those models provide a limited account of the individual mental states and social attitudes in cooperation. First, — as I will argue — one should explicitly model not only the beliefs about others’ intentions and shares, but also the goals about the actions and the intentions of the others (Grosz and Kraus, 1996): each member not only expects but wishes or wants that the others do their job (Castelfranchi and Falcone, 1998). And conversely one should model the social commitment to the others also in terms of delegation of goals/task and of compliance with the others’ expectations: i.e. as goal-adoption (Castelfranchi, 1991 and Castelfranchi, 1995). Those attitudes are not specific of joint intentions, they can be individual social attitudes and can be unilateral.

Positive and negative expectations and the deontic nature of social conventions

Abstract: The general goal of the paper is to show the normative/deontic nature of conventions. Conventions are traditionally defined as regularity of behavior based on expectations evolved to solve coordination problems [14]. The thesis we defend is that the cognitive attitude of expectations is not only characterized by an anticipatory representation (belief) of a future state of affairs but is coupled with a motivational component (a goal on this state). The possible convergence between beliefs and corresponding goals allows the identification of positive and negative expectations. We argue that in positive expectations (differently from the negative ones) lies implicitly an influencing act aimed at prescribing that the expected event will be realized. We consider conventions as analyzed in Game Theory as regularity of behavior based on positive expectations. These conventions entail the deontic component of prescription. Each agent prescribes (and is subject to prescription) conformity to the convention to the others (prescription to do). This is a possible route to the spontaneous emergence of Social Norms. However we hypothesize, differently, that negative expectations too can sustain conventions. Even "bad habits" share a deontic component but is characterized by the socio-cognitive structure of permission (entitlement to do). We argue that with this analysis is possible to explain the self-organizing and stabilizing effect of conventions that create an equilibrium noxious for all the participants and individually more costly than the individual benefit.

Cooperating through a belief-based trust computation

Abstract: Many approaches to the trust study exist in literature. In our view, just a socio-cognitive approach to trust would be able to analyze the sub-components (the basic beliefs) on which the final decision to trust or not is taken. In this paper, we show a first implementation and advance of the socio-cognitive model of trust developed according to C. Castelfranchi and R. Falcone (1998) and R. Falcone and C. Castelfranchi (2001) using the so-called Fuzzy Cognitive Maps. With this implementation we show how the different components may change and how their impact can change depending from the specific situation and from the agent personality. We present a case study in the medical domain, with the choice between a Doctor and a Medical Automatic System. Another natural application of the model is a classical collaborative scenario, where each agent needs to evaluate the trustworthiness of the other agents in order to collaborate or delegate a task.

The Plausibility of Defensive Projection: A Cognitive Analysis

Abstract: This paper provides an analysis of the cognitive processes implied in the ego defense known as projection. Projection is first placed in the context of the general cognitive processes of attribution and ascription. Then we address defensive projection, and identify its distinctive features. In particular, whereas general projection consists in the ascription of one's own mental attitudes to others, defensive projection implies one's rejection (via denial or repression) of the ascribed mental states, and ascription is a means for supporting this rejection. We try to understand how and why ascription is functional to rejection. Then, we examine a few types of defensive projection, and identify the main components and steps in the projective process. Finally, we draw some general conclusions.