This paper is a terminological discussion of the term "mixed reality" (as used in "mixed reality environments", "mixed reality information spaces" and "mixed reality architecture"). It departs from a sociological definition of "reality" and concentrates on multiagent system interactions involving human agents and artificial agents. The conclusion is that for the agents participating in such interactions there exists only one reality (in the sociological sense of the word). The notion of "mixed reality" is therefore not applicable to such contexts, and neither is the notion of "augmented reality", focused on the perceptual integration of "real" and "virtual" space on the part of the human user. The term "hybrid multiagent environments" is introduced to describe the context in which interactions take place. What should be terminologically stressed as "mixed" (i.e. "hybrid") in the case at hand is the nature of the context of interactions (involving two environments and two types of agents) rather than the notion of reality resulting from these interactions (which is not "mixed" or "hybrid" but "one", in the sociological sense of the word.) Such a view of "one" or "shared" reality should be adopted both for the purpose of analysis and for the purpose of programming human agent-artificial agent interaction in artificial agent modelling.

Keywords:

Autonomous Agents; Construction of Reality; Interaction; Interpretive Microsociology; Mixed Reality; Multiagent Systems; Semiotics

Introduction

1.1

Viewed from the perspective of popular sociology, the term "mixed realities" invokes a mental representation of what has been referred to as "hybrid society" or "hybrid man-machine society" (Müller et al 1998). Imprecise in theoretical terms, this notion implies the existence of a society developing as a result of a social reality defined by machines and the social reality of human environment growing together into one.

1.2

In the technical context, "mixed reality" is popularly defined as "the technology to merge virtual world such as computer graphics, images generated in a computer and the real world as seamlessly as possible in real time" (Canon Technology 2001). In a slightly different disciplinary focus, the outcome of this process could be described as the perceptual integration of real and virtual space into an augmented space.

1.3

From a sociological point of view, it it important to note that even these loose technically and cognitively focused definitions imply a reality defined in sociological terms. If there exists something referred to as "real world" or "real space", then there must also exist a human user. As soon as the human user interacts, the activity he/she is engaged in involves certain elements of the social structure.

1.4

The aim of this paper is to try to account for a particular aspect of social agency (construction of reality) appearing in what is referred to as "mixed reality environments", and to do so from a perspective that would go beyond what Müller et al (1998) refer to as "folks sociology".

1.5

Given the suggestion of a machine-defined and human-defined aspects of "reality" implicit in the term "mixed reality", it is only logical that the analysis focuses on a particular type of interaction taking place in multiagent environments, namely that including human agents and artificial intelligent autonomous agents. ^[1]

1.6

The matter at issue is obviously the definition of "reality" (in this case, the definition of "social reality"). Only after the sociological implications of the use of this term have been outlined is it possible to determine whether it is in order to use the term "mixed reality" in this context or whether it is preferable to approach the outcome of this type of interaction as one all-inclusive (social) reality that includes both human and artificial elements. Once that this has been established, an adaptation of the existing terminology in the field can be attempted.

1.7

In keeping with the requirements of this task, this paper outlines a sociological view of "reality", discusses the construction of social reality in multiagent environments, and suggests that the term "hybrid multiagent environments" should be used to describe the context in which interactions between human agents and artificial autonomous agents take place.

1.8

The approach to the subject-matter combines the insights of interpretive microsociology (sociological phenomenology, symbolic interactionism) with those of selected semiotic approaches (Jakobson's communication model, Eco's notion of cutural encyclopedia).

A Microsociological View of "Reality"

2.1

In an earlier article (Petric et al 2001), we concluded that even in the interactions that take place exclusively between artificial autonomous agents (i.e. in completely artificial environments), there appear elements of reciprocity as well as of the reversal of actantial roles and status that suffice to situate the agents into the field of sociality (i.e. to view these artificial agents as social actors and to attribute the elements of social agency to their actions). ^[2]The same applies to the human agent - artificial agent interactions^[3], in which agents operate from different environments.

2.2

The latter case is obviously of relevance to the discussion of interactions taking place in what has been referred to as "mixed reality environments". However, although the social character of the agent's actions in this case has been established, what enters discussion at this point is the notion of "reality". The question is whether this notion can be accounted for from a sociological point of view as well, and whether its sociological definition corresponds with its usage in the term "mixed reality".

2.3

The fact that the attribution of social qualities to the quoted actions of agents was based on a microsociological view of sociality is highly propitious for this task. Namely, at the present level of technical development of artificial agents and artificial environments, the actions of (artificial) agents can only be interpreted as being a part of society if we adopt one of the sociological approaches departing in analysis from agent interactions following up on what W.I. Thomas (1923) refers to as the common "definition of the situation".^[4] These approaches are collectively know as "interpretive microsociology", and include Mead and Blumer's "symbolic interactionism", Shutz's and Berger and Luckmann's "phenomenological sociology", Garfinkel's "ethnomethodology" and Goffman's "dramaturgical" approach. ^[5]

2.4

The idea of interaction is central to all the interpretative microsociological approaches, and its importance is perhaps even more visible in the case of "mixed reality environments" including both human agents and artificial agents than in the case of multiagent environments consisting exclusively of artificial agents. While the latter can merely coexist in an environment, the former have to of necessity enter interactions (defined initially in a non-sociological sense) in order to bring into contact the two separate environments ("realities") they operate from.

2.5

As for the term "reality", it figures prominently (taken in the sociological sense) in the works of Shutz and Garfinkel ("social reality"), as well as of Berger and Luckmann ("social construction of reality", "reality of everyday life"). Other interpretative microsociologists do not make explicit references to this term, but it is implied in what these approaches define as "society", i.e. in interaction-defined "social situations". ^[6]

2.6

Berger and Luckmann's treatment of the term within their phenomenological sociology is perhaps the most conspicuous treatment of "reality" within the entire field of sociology.^[7] In their view, and as the title of their 1967 classic suggests, reality is "socially constructed". Although "interaction" does not occupy as central a place in their theory as in the case of "symbolic interactionists", Berger and Luckmann devote due attention to the development of their interpretation of social action. Since in this aspect of their theoretical activity they come close to the ideas of other microsociologists^[8], it is safe to say that their view of reality implies that it is socially constructed in interactions.

Construction of "Reality" in Interactions between Human and Artificial Agents

3.1

Any attempt at explaining the construction of "reality" in a context including artificial agents is complicated by the fact that microsociological theories were devised in a context where social agency was thought of as being a characteristic exclusively of humans (i.e. these theories were meant as an explanation of human society). In the quoted article (Petric et al 2001), we claimed that artifical autonomous agents can also become social actors, although - at the present level of technical development - in a very rudimentary way.

3.2

If artificial autonomous agents are capable of interacting between themselves and with human agents in a way that situates them into the field of sociality, the point at issue now is whether the outcome of these interactions can be described as a "reality" and what kind of reality this is. Furthermore, is the kind of reality resulting from these interactions related to the notion of reality implied by the term "mixed reality environments" and in which way?

3.3

As has already been stated, the term "mixed reality" would seem to carry connotations which can be interpreted both from perceptually and socially focused perspectives. Given the primarily sociological focus of this paper, the perceptual implications of the term (especially with regard to the technology enabling the "perceptual integration of real and virtual space") are obviously not central to it.

3.4

In addition to this, it should be said that the perceptual integration of different spaces on the part of the human user has never really been a problem (not even in a technically much less developed context). ^[9]

3.5

Likewise, the notion of "augmented reality" cannot be of interest to a sociologically centered perspective, because it is fully centered on the perceptual capabilities of the human user. ^[10]

3.6

A microsociological exploration of the construction of "reality" is primarily interested in the relations forming or made manifest in the course of an interaction, i.e. it stresses the participation of two parties in the emergence of social meaning and "social reality".

3.7

Software agents obviously cannot perceive "reality" (perceptually defined) in the same way as human users can. But they can become social actors and therefore participate in the construction of "social reality" emerging from interactions, as well as observe (or further modify) the rules of social behaviour created in these interactions, in whose definition they themselves took part. In other words, human agents and artificial autonomous agents obviously cannot share the same perceptual mechanism but can interact in a commonly defined social situation.

3.8

The fact that human agents are, at the present level of technical development, socially infinitely more complex than the artificial agents does not change the social nature of their interactions nor removes the possibility of their participation in the same "social reality". As has been stated in Petric et al (2001), autonomous agents can at present demonstrate at most four of the six functions of language described in Jakobson's communication model: however, the way they put them to use in interactions suffices to situate them into the field of sociality (which according to microsociological theories emerges as the consequence of interactions).^[11] Likewise, the "cultural encyclopaedias" of autonomous agents (which can in this context be taken to represent "models of social competences")^[12] are much more rudimentary than those of the human agents. However, the "cultural encyclopaedias" of a child or an adult from a different culture can also be much more rudimentary than those of the other participant in an interaction, and this still need not be an insurmountable obstacle to social communication (and, by implication, to the emergence of a sort of commonly defined "social reality" in the course of an interaction). ^[13]

3.9

The question now is how to proceed methodologically in this regard? In which way should a researcher try to pinpoint the components and monitor the future development of sociality ("social reality") emerging in human agent - artificial agent interactions?

3.10

Given the nature of the approach on which the analysis is based (microsociology), as well as the current state of the analysed material (rudimentary elements of social agency on the part of artificial agents), the only answer to this question can be that bottom-up (inductive) analytical procedures are to be strongly preferred. The elements that turn artificial agents into social actors should be analysed in detail, and the analysis can be securely based on the use or adaptation of concepts and methods established in the study of society and communication. A combination of methods of inquiry developed within microsociology and semiotics would seem to be especially propitious in the case at hand. ^[14]

"Mixed reality Environments" or "Hybrid Multiagent Environments"?

4.1

To be efficient in analysis, one should - in addition to determining the general approach and selecting the methods of inquiry - also precisely define the terms that one should use. In the case of the material at hand, the need for terminological precision does not stop at the level of defining the nature of the social actors and of the social phenomena emerging from interactions, but should also include the context in which the interactions take place.

4.2

The term "mixed reality environments", that could be used to describe the context of human agent - artificial agent interactions, seems to be misleading if applied in an inductive socio-semiotic analysis carried out from a general microsociological perspective.

4.3

To begin with, the "reality" that is implied by this definition is obviously more perceptually and "technically" defined than socially. Furthermore, in the basic positioning of the material into the field of microsociology, we have already established in this paper that the "social reality" emerging in interactions is not "mixed" but unitary (i.e. represents an undivided whole to both participants in an interaction).

4.4

In other words, human agents and artificial agents operate from two different (technically and perceptually defined) environments, but the "social reality" emerging from their interactions is "one". What is needed is a term that would convey the (technically) different nature of the social actors without sending a misleading message as to the nature of the result of their interactions.

4.5

The term "mixed reality environments" is obviously not the right term for this task. In addition to the word "mixed" carrying an inadequate connotation regarding the result of the interactions, the term as a whole is not specific at all to the context of interactions involving human agents and artificial agents. If we take what the media theorists call "alternative realities"^[15] to be one component part of the human user's perceptually "mixed reality" (and there is no reason not to do this), a room in which the TV is on and somebody is watching could also be defined as a "mixed reality environment". The same is true of the environments created with the help of the so-called "mixed reality technology", in which the user is more active but there is no interaction with artificial autonomous agents. ^[16]

4.6

We suggest that the term "hybrid multiagent environments" should be used to describe the context in which the human agent - artificial agent interactions take place. There is a number of reasons why we think this term is more suitable for this task than the term "mixed reality environments". To begin with, instead of being centered on "reality" (the outcome of interactions), it is centered on the agents (i.e. social actors from whose interactions "social reality" develops). Furthermore, it does not suggest the perceptual integration of the real and virtual space on the part of the human user, but merely states that interactions take place in an environment consisting of two or more agents of different nature. (The adjective "multiagent" suggests that more than one agent exists in the environment, while the adjective "hybrid" further conveys information on their different technical backgrounds). ^[17]

4.7

Finally, generally speaking, the term "hybrid multiagent environment" is much more neutral (i.e. descriptive rather than suggestive) than "mixed reality environment". It is also in this sense that it is much better suited to the (inductive) microsociological analytical focus: it is centered on agents (social actors) potentially participating in an interaction but does not prejudge its outcome nor privilege neither of the participants in it.

4.8

To make the term fully compatible with the expected microsociological analysis, the term "environment" was substituted for the usual "system". Namely, while "system" in the terms such as "multiagent system" or "hybrid multiagent system" defines their meaning in a technical sense of the word (roughly: "a set of interdependent elements or parts that can be thought of as a whole"), and is applicable to simple "wholes" consisting of a definite number of "elements or parts", in the sociological sense of the word, it presupposes a highly complex structure including a large number of more "intangible" elements ranging from norms, values, and attitudes to institutions and cultural practices. At the present level of their development, such a use is clearly not warranted in an attempt to describe the forms of sociality emerging in human agent - artificial agent interactions.

4.9

In addition to this, within sociological theory, the term "system" is typical of the macrosociological approaches in which the system and its functioning are more important than individuals. As such, these approaches obviously cannot adequately describe the social aspects of agent interactions, in which the system is only fragmentarily present.

4.10

As a final remark, it should also be said that the word "environment" in the term "hybrid multiagent environment" is used simply to denote the physical (technical) context of interactions. It is not used in this context to denote something as complex as "social environment", i.e. as something that "includes both material culture (...) and the abstract cultural and structural characteristics of social systems that constrain and shape the terms on which social life is lived" (Johnson 2000). In the context of the future socio-semiotic analysis of interactions taking place in "hybrid multiagent environments" it should be viewed at best as an extension of the "situation" (in W. I. Thomas's sense of the word), in which the technical context of the interaction is also taken into consideration. ^[18]

Conclusions

5.1

This paper has discussed the implications of the term "mixed reality environments" when used to describe the context of human agent - artificial agent interactions. It has been concluded that it is not best suited to describe this context, and that it should be substituted by the term "hybrid multiagent environments". This term is more neutral and better suited to the nature of an inductive microsociological analysis, which in turn is the only sociological approach that can account for those elements of social agency that come to view in agent interactions.

5.2

In our view, the adjective "hybrid" in the term "hybrid multiagent environments" suffices to indicate that the interactions under discussion involve agents of different constitution (human agents and artificial agents). The reality that these agents share and in which the elements of social agency come about should, however, be viewed as "one reality" or, as has recently been proposed, "shared reality".^[19] Such a view should be adopted both for the purpose of analysis and for the purpose of programming human agent - artificial agents interaction in artificial agent modelling.

5.3

Finally, it should also be said that our outline of the microsociological view of "social reality" has not only contributed to the terminological discussion carried out in this paper, but has also reasserted the conviction of the authors that the best way to analyse the nature of sociality emerging in multiagent environments is to combine a microsociological perspective with the methods of inquiry developed within the semiotic study of communication. The usefulness of brief references to the insights made possible by the application of Jakobson's model of communication and Eco's notion of "cultural encyclopaedia" suggests that a further analysis of the interrelations between microsociological and semiotic methods of inquiry is in order, and that it could result in methodologies stimulating for the analysis of social agency in multiagent environments.

Notes

¹ In addition to human agent - artificial agent interaction, what can take place in multiagent environments is also interaction between exclusively artificial autonomous agents, as well as human agent - human agent interaction via an artificial agent. Human agent - human agent interactions taking place in a "virtual environment" could also be added to the list, because they could be felt by some to be somewhat different from the usual human agent - human agent interactions.

² Our references in this regard were primarily to the Servant/Master scenarios, described in Hayes-Roth et al (1996). The behaviourally animated autonomous agents Otto and Gregor described in these scenarios base their actions and modify their behaviour (social agency) in relation to the information extrapolated from the actions of the other participant in the interaction. At a more rudimentary level (participants in the interaction merely respond socially to each other's actions, however without changing the actantial role and status), the same is true of the interactions of animated electronic puppets called Woggles and described in Bates (1994).

³Our reference in this regard was to the example of the textual autonomous agent called "cat Lyotard" interacting with the human user described in Bates et al (1992), in which the textual agent "changes its attitude" toward the human user (i.e. the agent's rudimentary "cultural encyclopaedia" of the term "human" is modified in a learning process based on the human agent's actions).

⁴For obvious reasons, it would be completely unproductive to try to approach the rudimentary "social worlds" developing in artificial agent - artificial agent and human agent - artificial agent interactions from macrosociological perspectives premised on the existence of a developed social structure.

⁵ For specific titles of classical works in which the methodologies of interpretive microsociology were developed, please refer to the list of references (Berger and Luckmann 1967; Blumer 1969; Garfinkel 1963; Goffman 1959, 1961, 1963, 1967, 1971, 1972, 1974; Mead 1934; Schutz 1932/1967, 1962; Schutz and Luckmann 1973; Thomas 1923).

⁶ In the works of the symbolic interactionists Mead and Blumer, the elements of what Berger and Luckmann later defined as the "social construction of reality" were implied in the interplay of the elements constituting the "self" and "society" in the process of the emergence of meaning. Following up on W.I. Thomas's "definition of the situation", Goffman states that people define situations and their roles in these situations as "real".

⁷Not only in Berger-Luckmann's but also in Schutz's work, the approach to "reality" is founded on the premises of phenomenology as a philosophical method of inquiry. However, since it is in their interpretations aimed primarily at explaining social phenomena, it is legitimate and actually more accurate to view these theories as sociological.

⁸Although they are generally classified as microsociologists, Berger and Luckmann's attempt to offer a general social theory combining the features of theories both of social action and social structure can be seen as an early example of a micro-macro approach.

⁹Lenoir quotes the experiments carried out by Ivan Sutherland, which demonstrated very early on that "a human could become totally immersed in a remote environment through the "eyes" of a camera. With the viewer inside a building, a camera was mounted on the roof, with its field of view focused on two people playing catch. The viewer immediately responded to the motion of the ball, moving the camera to follow the game of catch by moving his head. Proof of the viewer's involvement in this remote environment came when the ball was thrown at the camera and the viewer ducked. When the camera panned the horizon, the viewer reported a panoramic skyline. When the camera looked down to reveal that it was "standing" on a plank extended off the roof of the building, the viewer panicked." (Lenoir 2000). Lenoir's account of the experiment is based on Sutherland (1993).

¹⁰According to Azuma, "Augmented reality (AR) is a variation of Virtual Environments (VE), or Virtual Reality as it is more commonly called" (Azuma 1997). In contrast with VE technologies, which "completely immerse a user inside a synthetic environment", AR "allows the user to see the real world, with virtual objects superimposed or composited with the real world. (...) AR supplements reality rather than completely replacing it". In other words, Azuma is interested in the perceptual integration of the two realities (on the part of the human user), with a view to enhancing "a user's perception and interaction with the real world" (Azuma 1997). An example of the research approach not interested in the perceptual integration implied by the term "mixed reality" is that of the research project analysing "cohabited mixed reality information spaces" (COMRIS). Van de Velde interprets the notion of mixed reality space as consisting of "a couple of interleaved spaces, one being a real space, the other a virtual one" (Van de Velde 1997). The researchers in this project, however, state that their notion of "mixed reality" is closer to something that could be defined as "dual reality" (i.e. each space has its own structure and inhabitants, and the coupling of the two spaces can be much looser).

¹¹ With regard to this example, our reference in Petric et al (2001) was to the interactions of anthropomorphic animated software puppets (Woggles), as described in Loyall and Bates (1997). In these interactions, note can be taken of functions corresponding to Jakobson's (1960) referential, conative, emotive and phatic function. Woggles refer to their context, address one another, express emotions and communicate in order to maintain an elementary social ritual. These interactions obviously fall short of the complexity of human communication (Jakobson's metalinguistic and poetic function are obviously missing from them), but the four functions present in them suffice to situate the agents into the field of sociality.

¹²In Petric et al (2001), we used Umberto Eco's notion of "cultural encyclopaedia" to describe the change of attitude and a learning capability of a purely textual autonomous agent called "cat Lyotard" in an interaction with the human agent, described in Bates et al (1992). We have also applied it to the Master/servant interaction described in Hayes-Roth et al (1996). Eco's term "cultural encyclopaedia" was briefly described by Calabrese as "a model of socialized competences at a particular point in history, which the dictionary (a model of ideal competences of an ideal speaker) cannot explain in full measure" (Calabrese 1985) [transl.M.P.]. This term, potentially useful in the modelling of autonomous agents' emotional and semantic engines, can in the context of this paper (focused on the social aspects of agent interaction) also be taken to represent a "model of social competences".

¹³A project like "Aurora" (Dautenhahn 1999), in which autistic children communicate with robots, actually shows in its own way that a formation of and participation in the commonly defined "social reality" is possible for actors (human agents and robotic autonomous agents) whose "cultural encyclopaedias" ("models of social competences") are non-standard and rudimentary.

¹⁴This has already been suggested by the examples outlined in Petric et al (2001). The narratological distinction between an actant and an actor has proven to be relevant in establishing the presence of social agency in agent interactions, and has therefore opened the way to the conceptualization of the material in terms of microsociological analysis in the first place. The use of terminology and methods of inquiry developed within semiotics, as was the case with the use of Jakobson's functions of language and Eco's "cultural encyclopaedia" in Petric et al (2001) and the present paper, could further refine microsociological analysis, not only in those cases where artificial agents participate in interactions. In other words, our analysis of human society can also profit from the analytical concern with artificial agent interactions

¹⁵"Alternative realities" are media manufactured representations of situations and lifestyles often found in advertisements, that "promote the illusion that these alternative backgrounds and situations are real" (Dimbleby and Burton 1998). Some commentators, Dimbleby and Burton inform us, have also talked of "an alternative world of the media".

¹⁶An example of this is recent Canon Technology's "Cybercity Walker", "which provides the experience of wandering around a city that appears on three screens forming a 180∞ field of view"(Mixed Reality Systems Laboratory Inc. 2002).

¹⁷The adjective "hybrid" has a long history of use in the computer-science related terminology, ranging from the "hybrid computer" (introduced in 1968) to the present-day "hybrid architecture" and "hybrid multiagent systems". Like the word "mixed", it denotes something heterogenous in origin or composition, but is - etymologically speaking - related to mixtures of two different things (when used as a technical term in biology it describes an animal or plant bred from two different types of animal or plant), which is not necessarily the case with the word "mixed" (it can also describe something composed of two or more elements).

¹⁸ The term "environment", conceived of in this sense, is actually compatible with the use of the term "environment" in the purely technical context. In contrast with "space", "environment" usually refers to smaller and less precisely structured units. In this context, one can speak of "graphical environment" or "desktop environment", and the like. Such usage is obviously congruent with the premises of any microsociological approach, because they all depart from the notion of the social "situation" in which an interaction takes place, rather than from of necessity less loosely defined structure of the system.

¹⁹The term shared reality has been proposed by Stephen Johnes in the final discussion session of the conference "Living in Mixed Realities", which took place at the Kunstmuseum in Bonn on September 23, 2001. This discussion, as well as the conference, focused on the social and cultural implications of "mixed reality technology", and stressed issues such as "seamless integration" of the virtual and real space, rather than the issues revolving around agent modelling. Regardless of that, the proposed term fits in well into any interpretive microsociological view of interaction.

References

AZUMA, R.T. (1997), "A Survey of Augmented Reality", Presence: The Journal of Teleoperators and Virtual Environments, 6 (4).

BATES, J. (1992), "Virtual Reality, Art, and Entertainment", Presence: The Journal of Teleoperators and Virtual Environments, 1 (1).

BATES, J. (1994), "The Role of Emotion in Believable Agents". Technical Report CMU-CS-94-136. School of Computer Science, Carnegie Mellon University, Pittsburgh, PA.

BATES, J., A. B. LOYALL, W. S. REILLY (1992), "An Architecture for Action, Emotion, and Social Behavior". Technical Report CMU-CS-92-144, School of Computer Science, Carnegie Mellon University, Pittsburgh, PA.

BERGER, P. AND T. LUCKMANN (1967), The Social Construction of Reality (Garden City, N.Y.:Anchor).

BLUMER, H. (1969), Symbolic Interaction: Perspective and Method (Englewood Cliffs, N.J.: Prentice Hall).

CALABRESE, O. (1985), Il linguaggio dell'arte (Milano: Bompiani).

DAUTENHAHN, K. (1999), "Embodiment and Interaction in Socially Intelligent Life-Like Agents" , in: C. L. Nehaniv (ed): Computation for Metaphors, Analogy and Agent, Springer Lecture Notes in Artificial Intelligence, Volume 1562, Springer.

DIMBLEBY, R. AND G. BURTON (1998), More Than Words: An Introduction to Communication (London & New York: Routledge).

GARFINKEL, H. (1963), Studies in Ethnomethodology (Englewood Cliffs, N.J.: Prentice Hall).

GOFFMAN, E. (1959), Presentation of Self in Everyday Life (Garden City, N.Y.:Anchor).

GOFFMAN, E. (1961), Encounters: Two Studies in the Sociology of Interaction (Indianapolis: BobbsMerrill).

GOFFMAN, E. (1963), Behavior in Public Places: Notes on the Social Organization of Gatherings (Glencoe, Ill.: Free Press).

GOFFMAN, E. (1967), Interaction Ritual: Essays on Face-to-Face Behavior (Garden City, N.Y.:Anchor).

GOFFMAN, E. (1971), Relations in Public: Microstudies of the Public Order (New York: Basic Books).

GOFFMAN, E. (1972), Strategic Interaction (New York: Ballantine).

GOFFMAN, E. (1974), Frame Analysis: An Essay on the Organization of Experience (New York: Harper Colophon).

HAYES-ROTH B., R. V. GENT, D. HUBER (1996), "Acting in Character", Proceedings of the AAAI Workshop on AI and Entertainment.

JAKOBSON, R. (1960), "Closing Statements: Linguistics and Poetics", in Sebeok, Thomas A. (ed.), Style in Language (New York: Wiley, Cambridge: MIT Press).

JOHNSON, A.G. (2000), The Blackwell Dictionary of Sociology: A User's Guide to Sociological Language (Oxford and Malden: Blackwell Publishers).

LENOIR. T. (2000), "All but War is Simulation: The Military-Entertainment Complex", Configurations, 8 (3).

LOYALL, A.B. and J. BATES (1997), "Personality-Rich Believable Agents That Use Language", Proceedings of the First International Conference on Autonomous Agents, Marina del Rey, California.

MEAD, G.H. (1934), Mind, Self and Society: From the Standpoint of a Social Behaviorist (Chicago: University of Chicago Press.).

MIXED REALITY SYSTEMS LABORATORY INC. (2002), "Projects" (http://www.mr-system.co.jp/project/index_e.html)

MÜLLER, H.J., T. MALSCH, I. SCHULZ-SCHAEFFER (1998), "Socionics: Introduction and Potential", Journal of Artificial Societies and Social Simulation,vol. 1, no. 3 <https://www.jasss.org/1/3/5.html>.

PETRIC, M., I. TOMIC-KOLUDROVIC, I. MITROVIC (2001), "A Missing Link: The Role of Semiotics in Multiagent Environments", in COSIGN 2001 Proceedings: 1st Conference on Computational Semiotics for Games and New Media, Amsterdam: Stichting Centrum voor Wiskunde en Informatica.

SCHUTZ, A. (1932/1967), The Phenomenology of the Social World (Evanston, Ill.: Northwestern University Press).

SCHUTZ, A. (1962), Collected Papers I: Studies in Social Reality (The Hague: Martinus Nijhoff).

SCHUTZ, A. AND LUCKMANN, T. (1973), The Structure of Life World (Evanston, Ill.: Northwestern University Press).

SUTHERLAND, I.E. (1993), "Virtual Reality Before It Had That Name", videotaped lecture before the Bay Area Computer History Association, Xerox Parc, Palo Alto.

THOMAS, W.I. (1923), The Unadjusted Girl (Boston: Little, Brown and Co.).

VAN DE VELDE, W. (1997), "Co-habited mixed realities", in Proceedings of the IJCAI workshop on Social Interaction and Community Ware, Nagoya, Japan.

Return to Contents of this issue

© Copyright Journal of Artificial Societies and Social Simulation, 2002