The Psychology of Cyberspace:
a socio-cognitive framework to computer-mediated communication

* Written by Giuseppe Riva and Carlo Galimberti.
The Paper was originally published by the Journal New Ideas in Psychology, 15 (2), 141-158, 1997.
(C) 1997 Elsevier Science Ltd. All rights reserved.

Copyright Notice
This paper is included as a means to ensure timely dissemination of scholarly and technical work on a non-commercial basis. Copyright and all rights therein are maintained by the authors or by other copyright holders, notwithstanding that they have offered their works here electronically. It is understood that all persons copying this information will adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder. Please contact the authors if you are willing to republish this work in a book, journal, on the Web or elsewhere. Thank you in advance.

Abstract: The authors outline a framework for the study of computer mediated communication defining three psychosocial roots by which the subjectivity of user is constructed networked reality, virtual conversation and identity construction. This has resulted in new ways of describing Cyberspace, the virtual space inhabited by electronic network users.
Community, for persons interacting in a technological environment, is shifting from culture-defining mass media to that of a proliferation of mediaas alternative sources of mediated experience. Infact, the key feature of Cyberspace is interaction, from which a new sense of self and community can be built. The authors also consider some implication of this approach for current research in communication studies, with particular reference to the role of context, the link between cognition and interaction and the use of interlocutory models as paradigms of communicative interaction.


The development of information technology has slowly changed the way people interact with computers. Technological advances have gradually shifted the focus away from computers as such, which have become less of an end in themselves, and more of a means in terms of what people actually do with them. In response to the environment in which people find themselves working and living, they appropriate the technology for their own needs.

The most evident sign of this change has been the creation of totally new interactive communication environments like Computer Mediated Communication (CMC) and Computer Supported Collaborative Work (CSCW) made possible by the increasing power and flexibility of today's information technology. Infact, the diffusion of these two environments is creating a new social space, usually called cyberspace (Rehingold, 1991, 1993), that is the fertile ground for new social relationships, roles, and a sense of self. As Cutler (1995) underlines, "Through discourse made possible byinteractive media individuals find or form groups that share interests. In the socially constructed space of cyberspace, where interaction produces culture, information is the only medium of exchange an individual has with which to build a presence... Information exchange becomes the carrier for expressing self-concept and eliciting emotional support" (pp.20-21).

The spread of these new environments has also influenced communication studies by revolutionising the metaphors used to describe communication (Schuler, 1994). That there is already considerable conceptual and metaphorical overlap between the languages of information technology and communication studies is well known. What is new, however, is where the exchange of metaphors is taking place. CMC is no longer - and not only - being described using models of man-machine or machine-machine interaction. As recent experiments by Nass and Steuer (1993) have shown, the psychological processes typical of CMC "... have more in common with interpersonal interaction than with technological procedures which do not reproduce the person-to-person relationship, as when a computer is used as a calculator, for example" (p. 522).

Interaction is the key feature of cyberspace, from which a sense of self and control can be built. The result of new senses of self is a new sense of presence that fills the space in fluid forms of network/community. Community for persons living in a technological environment is shifting from culture-defining mass media to that of a proliferation of media as alternative sources of mediated experience.

This shift in the style of person-computer interaction is beginning to orient CMC studies towards forms of interaction based on psychosociological and conversational models, but at the same time it has introduced new types of interactional structuration which sometimes differ from traditional psychosocial descriptions of interaction.

Perhaps the most important change in psychosocial terms has been how the concept of interaction is defined. Until now, the physical co-presence of both interacting subjects has been used to distinguish interaction from relationship, the latter implying inter-subjective communication that can be maintained even at a physical distance. Though we have no wish to discuss here how the two concepts are defined and relate to each other, it does seem important to note that definitions of interaction which rely on the physical co-presence of both subjects are extremely limiting in conceptual terms.

Even in telephonic communication, which predates digital computer technology, there can be no doubt that interlocutors do interact, even though they cannot see each other. This is even truer of video-conferences - in which the simulacrum of the other person's physical is rendered more convincing by the additional visual channel - and situations in which physical contact is even more rarefied, such as network communication between several computers, or when seemingly 'real' contact is essentially the outcome of high-quality simulation, as in virtual reality (Riva, Galimberti, 1993). Virtual reality, more than any other technology, carries the detachment of interaction from physical interlocutor co-presence to its logical extreme, and challenges the very concept of interlocutor identity. Clearly, a detailed psychosocial analysis of CMC, and of its more advanced forms like virtual reality, will be extremely useful to anyone who studies, designs or simply uses communication systems of this type. This paper looks at basic ideas that will help us to construct a general model of interpersonal interaction in Cyberspace. In particular, we shall attempt to define the three psychosocial roots of Cyberspace - networked reality, virtual conversation and identity construction - because they are essential to understanding how the subjectivity of digital interactive communication is constructed.


1. Networked reality and cognition as a coordinated activity

"Cognitions emerge [...] from the conversational interactions in which children participate from a very early age [...]. Far from being circumscribed by experimental laboratory settings, cognitive activities are routine daily activities. In other words, it is in everyday life, and in conversational interaction especially, that we put our cognitive skills to practical use" (Trognon, 1992, p.117. Our italics).

"The higher mental functions are dialogic processes derived from interpersonal activity. (They) develop through the progressive internalization of semiotically manifested perspectives on reality [...]. The emergence of these functions in the context of social activity constitutes the cultural line of development. (Fernyhough, 1996, p. 47)

These two quotations illustrate very well the so called "dialogic approach" (Fernyhough, 1996; Saito, 1996) that is trying to assess how cognition relates to interaction, and to conversational interaction especially. Although methodological problems still have to be solved, cognitive studies are increasingly concerned with defining and exploring the relationship between cognition and interaction. As Perret-Clermont and Brossard (1988) and Fernyhough (1996) have shown, the essential groundwork for this new approach, in psychological terms at least, is provided by Bakhtin (1981, 1986), Bartlett (1932) Gergen (1982), Piaget (1977/1995), and of course, Vygotsky (1978). Though they differ in detail on many points, all of them agree that the social system should be seen as a network of relationships providing the space in which cognitions are elaborated. As we hinted in our introduction, this space cannot be understood in physical terms only. The detachment of interaction from physical interlocutor co-presence is now taken for granted in structurations of dematerialised interactive settings -most typically, virtual reality -characterised by increasingly higher levels of simulation, and this shift in emphasis has revealed key structural features of interaction which have hitherto been concealed by dogmatic insistence on the physical co-presence of subjects. It is the co-presence of utterances, rather than the physical co-presence of interlocutors, which is now seen to determine the construction and performance of cognitive functions (Galimberti, 1992; 1994). In this context, 'co-presence of utterances' is typical of the kind of communicative exchange described by Goffman, in which two interlocutors are able to influence each other's actions and regulate the nature of their communicaton through some form of feedback. Thus, cognition has a dual social connotation. On the one hand, it is both "action, to the extent that those who take part in it have to organise a flow of shared activities by coordinating and concatenating their actions ... [and] communication, to the extent that the interlocutors make themselves mutually accessible, or make explicit the elements that enable them to understand each other and act together" (Conein, de Fornel e Quéré, 1990, p. 19). On the other hand, cognition is always a function of social action and is always action-oriented, never an end in itself (Trognon, 1990,1992; Saito, 1996).

Most researchers, or at least, those who adopt a psychosocial approach to cognition, would find little to disagree with here. What is new, however, is that cognitive activities are increasingly being performed in networked contexts which, to varying degrees, are undeniably virtual. The network model is therefore essential to how the matrices of cognitive functions - De Kerckhove's brainframes (1988, 1997) - are constructed, and to the overall configuration of the knowledge system, which Pierre Lévy's (1994) recent concept of collective intelligence describes particularly well. Thus, cognition is now seen as something that happens 'between' rather than 'inside' subjects, a media-structured loop that begins and ends with the subjects themselves, a continuous exchange which generates a shared construction of reality at the interface between individual and collective, cognition and interaction, mental activitity and social activity (Galimberti, 1992; Ghiglione, Trognon, 1993; Fernyhough, 1996; Saito, 1996). Cognition has lost its traditional connotation as a private event, and is now regarded as both a coordinated activity (in terms of process) and as a networked reality (in terms of how the products of the process are distributed).

The concepts of brainframe and collective intelligence are especially useful because they enable us to describe this type of cognition in terms of two levels - micro and macro, depending on 'closeness' to subject - interconnected by a self/other-oriented process of self-definition which resembles the auto-poietic phenomena described by Varela (1989).

1.1 From De Kerckhove's brainframe to 'inter-brainframe'
According to De Kerckhove (1988, 1997), the fundamental stages in man's 'cognitive' development correspond to the ways in which communication techniques and technologies have shaped not only interpretations of the human mind and brain function, but also views of society and the world over the ages. Marshall McLuhan - De Kerckhove's teacher - claimed that societies are always in-formed more by the means used to convey information, than by the content of the information itself. Developing on this insight, De Kerckhove has identified three types of brainframe - alphabetic, video and cybernetic - which determine how we perceive the world. In the alphabetic brainframe, for example, the adoption of left-to-right handwriting implies an analytical, time-dependent frame based on temporal and linear sequencing (strings of letters form words, which then form sentences with a definite meaning), while right-to-left handwriting implies a synthetic, spatially-oriented frame which affects the way information is coded and decoded. The video brainframe, or videoframe, implies a rather different approach to communication: the sheer amount and flow of information simultaneously present on the screen calls for intuitive understanding - assimilation at a glance - and minimises the opportunities for analysing and reprocessing individual components of the message. Thus, television generates mass culture in the sense that it allows neither time nor opportunity to assimilate and reprocess information in personal ways: it imposes pre-constituted, standardised modes of assimilation which are the same for everybody. Finally, the cybernetic brainframe interposes a sort of corpus callosus between the previous two. The personal computer allows us to respond in a personal way to what we see on the screen (alphabetic brainframe), but the nature of our response is dictated by the rigid, inexorable logic of the program we are watching/using (videoframe). According to De Kerckhove, the unique feature of the cybernetic brainframe is that it enables us to externalise mental awareness; in other words, television has transformed us into image consumers, whereas the computer, by projecting us outside our own nervous systems and giving us access to, and power over, all aspects of the environment (cyberspace) we find ourselves in at any time we choose, has transformed us into information producers. The next step from television-induced mass culture is the culture of speed and 'personalised access'. Networking now enables us to access any sort of information without having to move. By becoming nodes in a network, individuals can analyse the information they have access to with ever greater thoroughness and freedom of manipulation, and since networking can take you anywhere you want, it opens the way to global interaction. Developing on De Kerckhove's idea, recent advances in communications suggest that we may now be seeing the emergence of an inter-brainframe, i.e. the brainframe that results from the networking of minds. This implies that technology and human minds are linked by circular causality and reciprocity, a form of reciprocal influence generating interaction which both structures, and is structured by action. However, the inter-brainframe does not replace the cybernetic brainframe as just another stage in the 'evolutionary' cognitive process envisaged by De Kerckhove - it is co-present with all the others, and its presence can be revealed by psychological analysis of interaction. The inter-brainframe stresses the idea of networking, the connecting-up of a multiplicity of cognitive frames which are all simultaneously present and active in communicative interaction. And as we have already seen, interaction provides the setting in which we put our cognitive competences to practical use.

1.2. Collective intelligence
The concept of collective intelligence recently proposed by Pierre Lévy (1994) is in some ways the anthropological equivalent of the inter-brainframe. Since it is both a form of 'networked reality' and a 'coordinated activity', it complements the concept of inter-brainframe in terms of social product and social process. Lévy says that "collective intelligence is not a purely cognitive object", where 'intelligence' has the explicit or implied meaning of phrases like "using one's intelligence", "by common accord" or "mutual understanding with the enemy" (p.31). The emphasis is on a plurality of subjects, and the coordination and convergence of the actions they undertake. In this sense, collective intelligence should be seen as "a ubiquitous intelligence which is continually being enhanced and coordinated in real time, an intelligence which mobilises cognitive competences", implying that we must also look carefully at how the products of this intelligence are handled in the space 'between' the minds involved in producing knowledge and developing competences. Naturally, the configuration of such a reality requires the appropriate kind of communication technology. As Lévy himself says, "beyond a given quantitative threshold, the coordination of intelligences in real time perforce requires the use of digital information technology" (p.56). Thus, collective intelligence relies on technological systems which enable members of a community to "coordinate their interactions within the same virtual universe of knowledge" as usually happens in Cyberspace. To do this, a system must offer its users ease of access, real-time information processing, and capillary diffusion. Clearly, the network is the medium which best meets these requirements because it transforms collective intelligence into "a ubiquitous coordinated intelligence which is constantly being enhanced, coordinated and mobilised in real time" (pp.34-35). Obviously, the network must be 'intelligent' and it must also facilitate human contact by enabling subjects to recognise each other in and through the network. Facilitation of human contact means exploiting the "diversified range of knowledge and awareness" other people possess, and encouraging new, positive individuations by motivating other users and instilling in them the kind of gratitude that facilitates the development of socially-oriented projects (Lévy, 1994).

Described in this way, 'networking' may sound more like an anthropological project or some new kind of social utopia than a new technological system. The 'prophetic' aspects of Lévy's ideas do not concern us directly here, but clearly, the social impact and, more importantly, the technology design implications of networking will have to be carefully evaluated in terms of the "new mechanisms of thinking and negotiaton" (p.18) they entail.


2. Virtual conversation

Another important change in psychosocial terms taken by CMC is in the concept of communication: the model of communication as the passage of information from one person to another is becoming obsolete. This model, usually called the "parcel-post model" (Shanon & Weaver, 1949; Tatar, Foster & Bobrow, 1991), is now in a state of crisis, partly because of some of the peculiar features of electronic environments, such as the asymmetry between message sender and message receiver.

The model of communication as information transfer does not take into account the cooperative component, which stimulates reciprocal responsibility for successful interaction and a series of subtle adaptations among interlocutors. As Dohény-Farina (1991) notes: "The theory of communication as information transfer separates knowledge from communication; it treats knowledge as an object that exists independently of the participants in the innovation venture. With this independent existence, information becomes an object that can be carried through channels" (p.8). However, it is possible to communicate only to the extent that participants have some common ground for shared beliefs, recognize reciprocal expectations and accept rules for interaction which serve as necessary anchors in the development of conversation (Clark and Shaefer, 1989). So, with CMC is emerging a new, alternative concept: communication is a common construction of meanings (Ghiglione, 1986; Kraut and Streeter, 1995).

2.1. Interlocutory models
The shift towards the interactional nature of communication is the key characteristic of the interlocutory (conversational) models (Galimberti, 1994). The importance attributed to interlocution in these models results from efforts to combine pragmatic linguistics with social psychology on the part of researchers in France such as Charaudeau (1983), Ghiglione (1988) and Trognon (1990, 1992), as well as American and British researchers like Potter and Wetherell. (1987).

Interlocutory models are based more or less explicitly on the concept of communicative interactionism and place much stress on the crucial importance of interaction and conversation in communication. Major features of the interlocutory approach include a new conceptual definition of communication (Jacques, 1985, 1986), clarification of the contractual (negotiable) nature of some aspects of communication (Ghiglione, 1988), and a radical revision of the concept of the interlocutor (Charaudeau, 1983).

As an "irreducible relational fact" (Jacques, 1986, p.115), communication is seen as a primary form of human recognition, and as the foundation of the intersubjectivity through which the reciprocity of all human relationships is expressed. Verbal cooperation, which accounts for most of communicative phenomenology, is thus a genuinely coordinated activity in which the utterances of one interlocutor are interwoven with those of another. Unlike the parcel-post model, the interlocutiory approach rejects all 'atomistic' interpretations which invariably see communication as an aggregation of discrete elements. In the atomistic approaches championed by the parcel-post model (see Fig. 1), an interaction is defined as an action (or reaction) which passes from one human being (S1) to another (S2), a form of feedback which each subject exerts on the verbal acts of the other by mediating the image they offer.

Figure 1: The parcel-post model

Figure 2: The interactive communication model

In "communicative interaction", however, the circuit is rendered more complex by a further set of elements. As Fig.2 shows, the circuit linking the two interlocutors is subordinated to a system (R, indicated by broken lines) of a higher order than the S1-S2 couple, which tends to maintain its own autonomous features and organisation. The message for S2 is simultaneously regarded as a message for S1 - I say to myself what I say to you - so that the words uttered by both interlocutors are directed both to themselves and to the other in a 'double listening' process. Signifying and understanding are no longer independent actions: there can be no signification without understanding. S1 has to know how S2 has received the message in order to know what the message actually is, so the comprehension of both forms a sort of feedback loop. Similarly, each subject receives back (at least in part, according to Jacques) what he or she has successfully transmitted - 'what you understand is what I have succeeded in signifiying' - so the message may be regarded as a joint S1-S2 initiative.

The diagrammatic representation of comunication in Fig.2 also reveals the existence of a communicative spiral. At every stage in the evolution of the system, R can never fully match "what I wanted to say" with "what you have understood", so that the message transmitted by S1 is a sort of "disturbance" which S2 must compensate for in order to restore the stability of the system. However, compensation will be only partial because S1 must react in the same way to the return transmission from S2. Thus, the higher system ·R evolves in step with interaction between S1 and S2 while also keeping a 'window' open the outside world in order to receive the 'information' on which it feeds.

This is, therefore, an interactive communication system in which S1 and S2 alike are subordinated to the "self-organised functioning" (Jacques, 1986) of the dyad formed when they relate to each other as a couple. S1 and S2 subordinate themselves to the functioning of R, which constitutes their shared interlocutory space ("logical space of interlocution"; Jacques, 1985). To reduce the communicative lag between S1 and S2 without overstepping the limits which R needs to preserve its partial closure to the outside world, the two interlocutors activate a 'discourse strategy', i.e., a set of coordinated communicative interactions which construct the communicative setting in progessive stages.

This awareness of the contractual nature of communication - the notion that something is at stake when it occurs - illuminates several features of the 'game of reciprocity' and enables us to decribe it as a coordinated attempt to construct possible worlds. As Ghiglione says (1986), "communication is the co-construction of a reality using a system of signs and a mutually acceptable set of principles which make exchange possible and provide the rules needed to govern it" (p.102).

This brings us to the distinctive feature of conversational models, namely the redefinition of the concept of interlocutor necessitated by prior definition of communication as a contractual process jointly negotiated by interlocutors. According to this definition, each communicative event is seen as a dialectical encounter between two processes - one expressive in which a communicating 'I' addresses a receiving/uttering 'You'; the other an interpretative one in which an interpreting 'You' in turn construes an image of an uttering 'I' - which intertwine in a subtle interplay of mutual recognition and acknowledgement.

2.2 CMC as a virtual conversation
Turning now to studies of CMC, the first important point to note is that two distinct types of CMC - synchronous and asynchronous - can be identified (Dix, Finlay, Abowd & Beale, 1993).

Synchronous CMC is produced when communication occurs simultaneously between two or more users, as in any normal telephonic or face-to-face conversation. Asynchronous CMC is produced when communication is not simultaneous. The essential difference between the two is a temporal one, therefore: for CMC to be synchronous, the computers involved must be linked in real time.

The commonest form of asynchronous CMC is E-mail, in which a sender leaves a message in a receiver's electronic letterbox, which the receiver must open before he can read the message. Another more sophisticated type of asynchronous CMC is Newsgroup, an electronic notice-board on which users can post messages referring to a specific topic or area of interest. Users can read the messages by opening the notice-board, and send their own messages in turn. As with E-mail, there is no real-time link between the computers of the interacting subjects.

Despite the predominance of the textual mode, it has been shown that asynchronous CMC differs in psychosocial terms from non-electronic written communication as well as from other existing means of communication. Experimental studies have revealed significant differences between CMC and non-electronic written communication (Lea, 1991; Rice, 1993) relating to degree of social presence and media richness. Social presence is the user's perception of the ability of the means of communication to marshal and focus the presence of communicating subjects (Short, William & Christie, 1976), while media richness is the ability of the means of communication to interlink a variety of topics, render them less ambiguous, and enable users to learn about them within a given time-span (Daft & Lengel, 1986). Studies by Rice (1992, 1993) have shown that there are significant differences in user perceptions of the degree of social presence/media richness in E-mail and video-conference, compared with other means of communication like the telephone and written text.

Unlike asynchronous CMC, the most important feature of synchronous CMC is that it does provide a real-time link between users' computers. Although the most frequently cited example is the video-conference, the most widespread system is in fact Internet Relay Chat, or IRC. IRC is a form of synchronous CMC which enables a group of users (a chat) to exchange written messages and interact with each other in two different ways, by sending a message either to a specified user, or to all members of the chat. One IRC variant of particular interest to communication researchers is MUD (also MOO, MUSH and VEE), "software which accepts the multi-user link through a certain type of network ... and gives each user access to a shared databank of rooms, exits and other objects. Each user consults and manipulates the databank from inside one of the rooms, see only the objects in that room, and moves to other rooms mostly by using the exits that link them" (Curtis, 1996; p. 229). In practical terms, MUDs may be regarded as text-based virtual environments which enable users not only to speak to each other (as in IRC) but also to explore the space they find themselves in, and to interact with the objects in it (Parks & Floyd, 1996). Another feature of MUDs is that they enable users to interact in more complex ways than are possible with IRC. As well as sending written messages, MUD participants can use metacommands to describe their emotions and perform complex actions like striking another user or giving him obejcts (Curtis, 1996; Parks & Floyd, 1996). One fairly recent form of synchronous CMC is the Internet Phone, an IRC mode that replaces written text with voice messages. Internet Phone is broadly similar to telephonic communication, with the important difference that, as in transceiver communication, only one user can speak at a time.

Synchronous CMC also has special features which distinguish it from other forms of communication. According to Newhagen (1996) these include (especially in synchronous CMC via Internet) multimediality, hypertextuality, packet switching, synchronicity and interactivity.

There is a technical reason for these differences between synchronous and asynchronous CMC. Communication with a keyboard and computer screen takes longer than normal face-to-face communication, and the absence of metacommunicative features like facial expression, posture and tone of voice encourages users to find other ways of making communication as complete as possible.

These limitations make CMC interaction more rarefied than the kind of interaction that happens in normal conversation, in the sense that CMC uses mainly textual devices - abbreviations and smiles (Dix, Finlay, Abowd & Beale, 1993) as well as MUD metacommands - to reproduce the metacommunicative features (emotions, illocutionary force) of face-to-face conversation.

The most common abbreviations (in English) are CUL8R (See you later), HowRU (How are you?), 2B (To be), IMHO (In my humble opinion); WRT (With respect to), and many of them are used to elaborate emotionally on the literal meaning of messages. For example, IMHO adds humorous overtones to a sentence. Smiles - graphic symbols depicting a stylised smiling face - are also much used. The most popular are :-) to convey positive emotion, :-( to convey negative emotion, and :-o to convey surprise.

However, the differences between CMC and face-to-face conversation are important. While face-to-face conversation occurs in a cooperative environment constantly regulated by mutual adjustment and correction (Goodwin & Heritage, 1990; Galimberti, 1994), CMC occurs in a much less cooperative environment because of the special conditions imposed by the medium itself (Brennan, 1991). In most CMC environments, and in asynchronous CMC environments especially, two typical features of face-to-face conversation are missing (Mantovani, 1996a):

- the collaborative commitment of participants and co-formulation of the message;

- the feedback which allows the social meaning of the message to be processed immediately.

In addition, CMC in no way guarantees that a user's declared identity is the real one. The use of false identities, often of a different sex, is widespread in electronic communities and in MUDs especially (Mantovani, 1995; Curtis, 1996; Parks & Floyd, 1996; Spears & Lea, 1992).

In this sense, CMC may be regarded as an example of virtual conversation, i.e., a necessarily 'pared-down' or, perhaps, more accurately, rarefied form of conversation which lacks the rules on which effective interaction depends. Computer mediation creates an asymmetrical imbalance in the sender-receiver relationship: the sender can transmit information and get cooperation under way, but has no guarantee that the receiver receives the transmission, while the receiver has no guarantee that the sender's declared identity is the real one.

Ghiglione's definition (1986) of communciation as the co-contruction of a reality using systems of signs and rules applies equally well to CMC as an instance of virtual conversation, with the important difference that in CMC a reality is asymmetrically co-constructed because the receiver can decide at will to terminate interaction, or continue it by turning himself into a sender. This decision is far from casual: it depends on how the receiver interprets the situation, what his aims are, and the social rules that govern his behaviour. Some researchers have even used the term "electronic opportunism" to decribe this feature of CMC (Rocco & Warglien, 1995). In this sense, CMC may be defined as a process by which a group of social actors in a given situation negotiates the meaning of the various situations which arise between them (Stasser, 1992).


3. From context to identity

Stasser's definition of CMC may seem straightforward enough, but it has two important implications which have had a decisive effect on CMC studies. If CMC is a process of negotiation:

- the only way to understand it is by analysing the subjects involved in it in the environment in which they operate, meaning that the social context in which CMC occurs plays a crucial role;

- new processes and activities will develop which challenge and modify the initial relationship between subject and context.

Most researchers would broadly agree that these two statements are true. According to Mantovani (1996b), the early 1990s saw changes in the paradigms used in studies of person-computer and person-computer-person interaction. The main outcome of this has been the realisation that interaction can only be fully understood through detailed analysis of the social context in which it occurs: "... at this point we no longer need to see people simply as 'users' of given systems, but as social 'actors'. In other words, whether expert computer users or not, people act independently and have their own reasons for what they do, and it is computers and systems that have to adapt to people, not vice versa" (Mantovani, 1996a, p.63).

In the sections that follow we shall explore the Situated Action Theory and the Positioning Theory, two new socio-cognitive approaches which seem us to explain some of the issues just raised.

3.1 The Situated Action Theory and the Positioning Theory
Situated Action Theory - SAT - developed within the field of socio-cogntive research known as "cognition in practice". Though based on traditional cognitivist analyses of information processing and symbolisation, SAT introduces a change of perspective in that it sees action not as the execution of a ready-conceived plan, but as adaptation to context (Suchman, 1987). As Suchman notes, "instead of separating action from the circumstances in which it occurs as the execution of a carefully thought out plan ... [SAT] tries to study how people use circumstances to develop an intelligent course of action" (p.167).

This necessitates profound changes in how "social context" has previously been defined. In SAT, social context is not something physical and highly stable like an organisation or the power structure within it. As Mantovani (1996a) stresses, contexts are not given, but made, so that:

- context is conceptual as well as physical: actors perceive situations using cultural models, and act accordingly in cultural ways;

-context is unstable: cultural models are constantly modified by subjects' actions and choices.

In this sense, social context may regarded as the symbolic system of a given culture which is continually being altered by practical human intervention.

Applying SAT to CMC, Mantovani (1996a, 1996b) concludes that CMC participants cannot be regarded simply as technology users. Rather, they are social actors with their own aims and autonomy in situations, and it is technology which must adapt to them.

This idea poses serious problems, however. If social actors actively respond to their environment and end up changing it, how can context ever be analysed properly? Mantovani meets the difficulty with a three-level model of social context which links situation and social norms to the use of computer technology . The first level is social context in general, the second ordinary situations of everyday life, and the third local interaction with the environment via computers.

The links between the three levels can be studied in either direction, starting from use of computers or social context. Thus, use of computers may be regarded as a peculiar feature of everyday life, which is in turn a part of the braoder social context. In reality, subjects, by interacting with each other, the physical envrionment and the social context, activate a spiral of actor-environment exchanges. First-level person-computer interaction leads to interaction in everyday situations, and thence to cultural changes.

Working in the opposite direction, social context supplies the elements needed to interpret situations correctly, and situations generate the aims which determine local interaction with the environment via computer.

So, as we have seen, social context may be defined as the symbolic system of a given culture which is continually being altered by practical human intervention and it cannot be explained exclusively in terms of the interpersonal relationships or physical environment in which information exchanges take place. It is a prerequisite of communication, "a shared symbolic order in which action becomes meaningful, and so generates meaning" (Mantovani, 1996a, p.106).

Thus, SAT implies a radical redefinition of the meaning of communication. Context may be co-constructed by social actors, but they use communication to exchange meanings, not pieces of information. More precisely, the content of communication is interpretations of the situations which actors are involved in. In this sense, the most efective way of clarifying the meaning of messages is to relate them to a shared context of meaning.

Studies of Positioning Theory (PT) have served to reinforce this view. As recently formulated by Rom Harré (1989; Harré & Van Langenhove, 1991), PT replaces the traditional concept of role with the concept of positioning. The main difference between the two is that a role is a stable and clearly defined category, while positioning is a dynamic process generated by communication.

Developing on Bakhtin's ideas and Vygotsky's studies, PT identifies two distinct processes underlying social activities. The first, naturally enough, is discourse-generated positioning, which Harré defines as "the way in which subjects dynamically generate and explain their own and other people's behaviour " (Harré & Van Langenhove, 1991; p.405).

Harré defines the second process as the rhetorical redescription by which subjects shape their social context, "the discursive production of stories about institutions and macro-social events undertaken to make them intelligible in the form of social icons" (Harré & Van Langenhove, 1991; p.394).

As in SAT, context is not given in PT, but is constructed socially in ways which are endlessly different because of the changes communication brings about in the structuring of the cultural context. The main difference between SAT and PT lies in the role attributed to discourse production. PT sees conversation as the most important human activity of all because it encompasses virtually all known mental phenomena. As Harré & Van Langenhove state, "many mental phenomena like attitudes and emotions are immanently present in discourse production" (pp.394-395).

3.2. From context to identity
Positioning Theory adds new ideas about the relationship between mental and communicative proceses to its analysis of context. As we have seen, there is indeed a link between mental and communicative processes which leads to the formation of specific mental structures called brainframes. However, PT is mainly concerned with the relationship between communication, social context, self, and identity.

The notion that discourse and conversation are closely linked to both mental processes (including attitudes and emotions) and social context is typical of Russian thought. For example, Vygotsky analysed this link in relation to adult-child conversation. As is well known, Vygotsky believed that this culture-specific form of conversation is internalised by the child to become a part of his mental processes: "Each function present within the child's cultural development appears twice: first between people (interpsychology) and then within the child (intrapsychology)" (Vygotsky, 1978, p. 57). In reality, there is a close link between external language and interior dialogue, and it plays a crucial role in the formation of the subject's identity and higher mental processes (Fernyhough, 1996; Saito, 1996). The mediation of meaning resulting from interaction with other subjects is fundamental to this shift from external language to interior dialogue. In conversation, the subject not only acts as an individual/self seeking to achieve an objective, but also actively collaborates in the positioning process.

As Davies and Harré (1990) point out, subjects' selves during conversation "participate in an observable and subjectively coherent way in the joint production of story lines" (p. 48). In this phase, which uses interlocution in the manner described by Jacques, subjects see themselves as "contradictors" (Davies & Harré, 1990; p. 47) and use the positioning process to construct "a variety of selves" (p. 47) closely linked to the outcome of interaction.

This is very similar to the "transactional contextualism" developed by anthropologists and sociologists. For example, Rosaldo (1984) says that the notion of self develops not from some internal essence relatively unaffected by the social world, but from experience accumulated in a world of meanings, images and social relationships in which each person is unavoidably involved. Hsu (1985) defines this unbroken link between self and environment as "psychosocial homeostasis", the unremitting effort to establish a balance between satisfaction of intrinsic needs and the demands of socio-cultural context.

In psychology, these ideas have carried over into the work of Gergen (1982) and Bruner (1992). Gergen in particular has looked in detail at the construction of self, in studies of how an individual's self-esteem and concept of self vary in a set of different situations. These studies show that the concept of self varies both in relation to the kind of people the subject frequents, and in response to the positive and negative comments they make. On the whole, then, the self may be seen as a product of the situation in which the subject acts. For his part, Bruner, though accepting the subject's autonomy, speaks of "creatures of history" whose selves are both "a guarantee of stability and a baramoter reflecting changes in the cultural climate" (pag. 108).

How is CMC situated in this close relationship between external language and interior dialogue which finds it most obvious expression in the co-construction processes typical of positioning? We have already seen that CMC may be regarded as a form of virtual conversation, i.e., rarefied, 'pared-down' conversation lacking the rules which alone can ensure that effective interaction takes place. Computer mediation creates an aysymmetrical relationship between sender and receiver which:

- enables the sender to send information and initiate cooperation, but does not guarantee that the receiver receives the message;

- offers the receiver no guarantee that the sender's declared identity is the real one.

That this dual effect is a powerful influence on positioning and construction of self is more than evident in virtual reality communication, so we would now like to look at some of the features of virtual reality, and attempt to define it in human rather than technological terms.

As Steuer makes clear the concepts of presence and telepresence are crucial to this operation: "presence ... [is] the experience of one's physical environment; the term refers not to the subject's immediate surroundings as they exist in the physical world, but to the perception of these surroundings as mediated by both automatic and consciously-controlled mental processes [...] presence is defined as awareness of being in an environment ... when perception is mediated by communication technology, the subject is forced to perceive two distinct environments simultaneously: the physical environment in which he is present de facto, and the environment as presented through the technological medium. The term telepresence is used to describe the precedence which the second type of environmental awareness has over the first ... Telepresence is defined as the CMC-mediated experience of being in an environment" (Steuer, 1992, pp.75-76). These terms enable us to speak of virtual reality without also having to refer to (for example) hardware. Virtual reaility may thus be defined as "a real or simulated environment in which a perceiver ... experiences telepresence ... Telepresence focuses attention on the relationship between an individual who is both a sender and a reciever, and the mediated environment with which he interacts" (p.78). These quotations illustrate very well the extent of the overlap between how virtual reality builders think, and the basic principles of communictive interactionism. 'Virtual reality-space' is construed as the electronic analogue of the interlocutory space in which subjects interact to give three-dimensional consistency to ·R, the interlocutory space Jacques speaks of (see Fig. 2).

However, in terms of interlocutory space, the virtual environment itself becomes a kind of interlocutor because it adds to the positioning process objects and meanings which are alien to the interacting subjects. The concept of Cyberspace clearly shows that virtual reality is, in fact, a parallel universe created and manintained by the networks in which subjects interact.

The second difference between interlocutory space and virtual reality is that there is no guarantee that the declared identities of the interactors are the real ones. As Mantovani notes, "Virtual reality is a communication environment in which the interlocutor is increasingly convincing in terms of physical appearance, yet increasingly less tangible and plausible in terms of personal identity. This paradox results from juxtaposing a convincing simulation of the physical presence of the other, and the disappearance of the interlocutor's face behind a mask of false identities" (Mantovani, 1996a, p.197).

It is certainly no accident that members of electronic communities very often adopt false 'nickname' identities, and openly accept them in others. Within the same community, a person may construct and project mask-like identities which function as delegated puppets-agents" (Stone, 1991; p. 105).

Gender switches are also commonly made, often for rather specific reasons - to get to know people of the opposite sex with a view to meeting them; to explore the emotions of people of the opposite sex - although the fun of simply 'dressing up' and pretending to be someone else is also a factor.

But there is a problem here: how can you communicate and activate the positioning process without staking your own identity on the outcome? As we have seen, communication always requires a framework of rules and meanings, and this is especially true of CMC in which many features of face-to-face conversation are 'rarefied'. One solution is to represent yourself by "coding cultural expectations at a symbolic level" (Stone, 1991; p. 102). In constructing a false identity, the subject has to make wider use of social stereotypes than would be the case in normal conversation if he wishes his identity to be recognised and accepted. This means that CMC, and virtual reality in particular, may force subjects to resort to massive use of stereotypical attitudes and behaviours, otherwise they are unlikely to achieve any shared understanding of actions and situations (Mantovani, 1995). At the same time, however, there may be changes in how personal identity develops. Markus and Nurius' (1986) concept of possible selves offers some understanding of these changes, as well as a theoretical explanation of the relationship between identity and context.

According to Markus and Nurius, possible selves "give a specific cognitive form to our desires for control, power and belonging, and our widespread fears of failure and incompetence" (Markus e Nurius, 1986; p.960). Although possible selves constitute our repertoire of alternative selves, their main feature is that they are exempt from both direct social control and social negotiation. As Markus and Nurius say, "individuals have ideas about themselves which are not firmly anchored in social reality. As representatives of the self at some future time, possible selves are visions of the self which have not been tested and validated by social experience" (1986, p.955).

Potentially, a subject may be in a position to create an infinite number of possible selves, but in normal circumstances the repertoire of possible selves is a combination of the subject's personal experience and the living and communication environments he is familiar with. As well as being a source of more or less suitable behaviour modles, the media also offer subjects a range of images and symbols that people can identify with easily. Under normal circumstances, subjects can control media symbols and models, but this is much more difficult in a virtual environment. Interactivity and telepresence also endow virtual environments with a high degree of conviction and suggestiveness which is increasingly immune to the balancing effects of direct experience and "traditional" social contexts.

As Meyrowitz (1985) points out, our social context has changed because of the technology of communication. The influence of social context on the construction of identity is beginning to wane, especially in younger people, as reference communities like the family, school or church, which in the past anchored social contexts in shared sets of rules, gradually loosen their grip.

The present situation would seem to be that the new media are accelerating the dissolution of traditional rule-based social contexts whose gradual disappearance is gradually emptying the media themselves of meaning. As a result, the media, and the new media in particular, are becoming increasingly remote from everyday reality, a world apart which tends to impose sameness on personal identity and experience.

Gergen (1991) calls thiscondition of extended and permeated selves a society of"saturated selves." Alone and isolated, saturated selves experience the world of work and leisure as a space constructedout of cultural products and social fictions (Cutler, 1995). Persons at workoccupy constructed space in their networked relations withothers. Like a building, the network creates the structure. Like occupants of that building, workers occupy intellective spaces.

Probably one of the most convincing description of what this world could be is the one made by Mitchell in his City of Bits (1995): "a worldwide, elettronically mediated environment in which networks are everywhere, and most of the artifacts that function with it (at every scale, from nano to global) have intelligence and telecommunications capabilities...Commercial, entertainment, educational, and health care organization will use these new delivery systems as virtual places to cooperate, and compete on a global scale" (pp. 167-168).

4. Conclusion

Defining the three psychosocial roots of the process by which the subjectivity of digital interactive communication is constructed - networked reality, virtual conversation and identity construction - has enabled us to identify three almost parallel tracks in communication studies. The first leads from intersubjective interpretation of cognitive processes to the notion that cognition is a coordinatred activity whose products are situated not in the mind, but in the space between minds.

The second leads from communication as a linear process to the use of interlocutory models as paradigms of communicative interaction. The third leads from the essential passivity of communication technology users to active participation in the functioning of a machine which also influences user individuation. As we have seen, each has important methodological and technical implications for the study of interactive communication via computers. This has resulted in new ways of describing the virtual space inhabited by network users (Cyberspace): it is now seen as an electronic analogue of the interlocutor space in which subjects interact, a space which paradoxically juxtaposes a convincing simulation of the physical presence of the other with the disappearance of the interlocutor's face behind a mask of false identities. The key feature of Cyberspace is interaction through which a new sense of self and control can be built. The result of new senses of self is a new sense of presence that fills the space in fluid forms of network/community. Community for persons interacting in a technological environment is shifting from culture-defining mass media to that of a proliferation of media as alternative sources of mediated experience.

Obviously, the issues raised in this paper constitute just the first essential step towards a definitive study of Cyberspace (in which advances are sometimes uneven and out of step) and the culture that has grown up around it. But it has served to demonstrate that communication technologies are no longer seen by researchers as a rigid protheses, external tools marking the limits and limitations of users who are slaves rather than than masters, but as transparent interfaces, ways of genuinely enhancing the communication of interlocutors who use them, whether singly or in networks.



Bakhtin, M.M. (1981). Discover in the novel. In M. Holquist (Ed.), The Dialogic Imagination: Four Essays by M.M. Bakhtin (C. Emerson and M. Holquist, Trans.; pp 259-422). Austin: University of Texas Press.

Bakhtin, M.M. (1986). Speech genres and other late essays. (C. Emerson and M. Holquist, Ed.; V.W. Mc Gee, Trans.; pp 259-422). Austin: University of Texas Press.

Bartlett, F.C. (1932). Remembering: A study in experimental and social psychology. Cambridge: Cambridge University Press.

Brennan, S.E. (1991). Conversation with and through computers. User modelling and user-adapted interaction.1 (1), 67-86.

Bruner, J., (1992). Acts of meaning. Jerusalem: Harvard University Press.

Charaudeau P., (1983). Langage et discours. Paris: Hachette.

Clark, H.H. & Schaefer, E.F. (1989). Contributing to discourse. Cognitive Science, 13, 259-294.

Conein, B., De Fornel, M., Quere, L. (1990), Les formes de la conversation, vol. 1. Paris: CNET.

Curtis, P. (1996). Comunicazione via MUD: i fenomeni sociali delle realtà virtuali basate su testo. Sistemi Intelligenti, 8 (2), 229-251.

Cutler, R.H. (1995). Distributed presence and community in cyberspace. Interpersonal Computer and Technology, 3 (2), 12-32.

Daft, R.L., & Lengel, R.H. (1986). Organisational information requirements, media richness and structural design. Management Science, 32, 554-571.

Davies, B., Harré, R. (1990). Positioning: the discursive production of selves. Journal for the Theory of Social Behaviour, 20, 43-63.

De Kerckhove, D. & C.J. Lumsden (Eds.) (1988). The alphabet and the brain. New York: Springer Verlag.

De Kerckhove, D. (1997). Planetary Mind: Collective intelligence in the digital age: New York: Hardwired.

Dix, A., Finlay, J., Abowd, G. & Beale, R. (1993). Human-computer interaction. New York: Prentice Hall.

Dohény-Farina, S. (1991). Rhetoric, innovation, technology: case studies of technical communication in technology transfers. Cambridge (MA): MIT Press.

Fernyhough, C. (1996). The dialogic mind: A dialogic approach to the higher mental functions. New Ideas in Psychology, 14 (1), 47-62.

Galimberti, C. (1992), La conversazione. Milan: Guerini.

Galimberti, C. (1994), Dalla comunicazione alla conversazione. Ricerche di Psicologia, 18 (1), pp.113-152.

Gergen K.J. (1982). Toward transformation in social knowledge. New York: Springer.

Gergen, K.J. (1991). The saturated self: dilemmas of identity in contemporary life. USA: BasicBooks, Harper Collins.

Ghiglione R. (1986). L'homme communiquant. Paris: A.Colin.

Ghiglione R., Trognon A. (1993). Où va la pragmatique. Grenoble: PUG.

Goodwin, C., Heritage, J. (1990). Conversation analysis. Annual Review of Anthropology, 19, 283-307.

Harré, R. (1989). Language and science of psychology. Journal for the Theory of Social Behaviour and Personality, 4, 165-188.

Harré, R. & Van Langenhove, L. (1991). Varieties of positioning. Journal for the Theory of Social Behaviour, 21, 393-408.

Hsu, F. (1985). The self in cross-cultural perspective. In A. Manella, G. DeVos, F. Hsu (Eds.), Culture and self. New York: Tavistock.

Kraut, R.E. & Streeter, L.A. (1995). Coordination in software development. Communication of the ACM, 38 (3), 69-81.

Jacques F. (1985). L'espace logique de l'interlocution. Paris: puf.

Jacques F. (1986). La réciprocité interpersonnelle. Connexions, 47, 110-136.

Lea, M. (1991). Rationalist assumptions in cross media comparisons of computer mediated communication. Behavior and Information Technology, 10 (2), 153-172.

Lévy, P. (1994), L'intelligence collectif. Paris: La Découverte.

Mantovani, G. (1995). Virtual reality as a communication environment: consensual hallucination, fiction and possible selves. Human Relations, 48 (6), 669-683.

Mantovani, G. (1996a). New communication environments: from everyday to virtual. London: Taylor & Francis

Mantovani, G. (1996b). Social context in HCI: A new framework for mental models, cooperation and communication. Cognitive Science, 20, 237-296.

Markus, H., & Nurius, P. (1986). Possible selves. American Psychologist, 41 (9), 954-969.

Meyrowitz, J. (1985). No sense of place: The impact of electronic media on social behavior. New York: Oxford.

Mitchell. W.J. (1995). City of bits. Cambridge (MA): MIT Press.

Nass, C., & Steuer, J. (1993). Voices, boxes, and sources of messages: Computers and social actors. Human Communication Research, 19 (4), 504-527.

Newhagen, J.E. (1996). Why communication researchers should study the Internet: A dialogue. Journal of Communication, 46 (1), 4-13.

Piaget, J. (1977/1995). Sociological Studies (L. Smith, Ed.). London: Routledge.

Parks, M.R., & Floyd, K. (1996). Making friends in Cyberspace. Journal of Communication, 46 (1), 80-97.

Perret-Clermont, A.N. Brossard, A. (1988), L'intrication des processus cognitifs et sociaux dans les interactions. I. R. Hinde, A.N. Perret-Clermont, J. Stevenson-Hinde (Ed.), Relations interpersonnelles et développement des savoirs. Berne: Delval, pp. 441-465.

Potter J. e Wetherell M., 1987. Discourse and social psychology. London: Sage.

Rheingold, H. (1991). Virtual reality. New York: Summit Books.

Rheingold, H. (1993). The virtual community: Homesteading on the electronic frontier. Reading, MA: Addison-Wesley PublishingCo.

Rice, R.R. (1992). Contexts of research on organizational computer-mediated communication: A recursive review. In M. Lea (ed.), Context of computer-mediated communication (pp. 113-143). Hemel Hempstead: Harvester-Wheatsheaf.

Rice, R.R. (1993). Media appropriateness - Using social presence theory to compare traditional and new organizationa media, Human Communication Research, 19 (4), 451-458.

Riva G., & Galimberti C. (1993). Psicologo e computer nella ricerca in campo familiare. Modalità di interazione e prospettive di studio. Contributi del Dipartimento di Psicologia dell'Università Cattolica, 7, 123-135.

Rocco, E., & Warglien, M. (1995). La comunicazione mediata da computer e l'emergere dell'opportunismo elettronico. Sistemi Intelligenti, 7 (3), 393-420.

Rosaldo, M. (1984). Toward an anthropology of self and feeling. In Shweder R.A., LeVine R.A. (Eds.), Culture theory: Essays on the mind, self and emotion. Cambridge UK: Cambridge University Press.

Saito, A. (1996). Social origins of cognition: Bartlett, evolutionary perspective and embodied mind approach. Journal of the Theory of Social Behaviour, 26 (4), 399,422.

Schuler, D. (1994). Community networks: building a new participatory medium. Communication of the ACM, 37 (1), 38-51.

Shannon, C.E, & Weaver W. (1949). The mathematical theory of communication. Urbana: University of Illinois Press.

Short, J., Williams, E. & Christie, B. (1976). The social psychology of telecommunications. London: Wiley.

Spears, R., & Lea, M. (1992). Social influence and the influence of the "social" in computer-mediated communication. In L. Lea (ed.), Contexts of computer-mediated communication. Hemel Hempstead: Harvester Wheatsheaf, pp. 30-65.

Stasser, G, (1992). Pooling of unshared information during group discussion. In S. Worchell, W. Wood, & J.A. Simpson (eds.), Group processes and productivity (pp.48-67). Newbury Park, CA: Sage.

Steuer J., (1992), Defining Virtual Reality: Dimensions determining Telepresence, Journal of Communication, 42(4), 73-93.

Stone, A.R. (1991). Will the real body please stand up? Boundary stories about virtual cultures. In M. Benedikt (ed.), Cyberspace: First steps. Cambridge, MA: MIT Press.

Suchman, L. (1987). Plans and situated action. Cambridge UK: Cambridge University Press.

Tatar, D.G., Foster, G. & Bobrow, D.G. (1991). Design for conversation: Lessons from Cognoter. International Journal of Man-Machine Studies, 34, 185-209.

Trognon A., (1990). Fonctions de la conversation. In Maurand G. (ed.), Le dialogue. Albi: L'Union.

Trognon A., (1992). Psicologia cognitiva e analisi delle conversazioni. In Galimberti C. (ed.), La conversazione. Prospettive sull'interazione psicosociale. Milan: Guerini e Associati.

Varela, F. (1989), Autonomie et connassaince. Paris: Seuil.

Vygotsky, L.S. (1978). Mind in society: The development of higher psychological processes. Cambridge: Harvard University Press.

For any questions or requests, please contact