Theoretical constraints of task-oriented computer mediated communication

The analytic framework used in this research is based on the concepts of distributed cognition, person-plus and affordances. The theory of distributed cognition states that intelligence (understood as the capacity for solving a problem) does not reside just in the individual (person-solo) but also within the available tools and artefacts in the surrounding environment. These tools and artifacts can be physical, symbolic or psychological. To refer to the person in conjunction with these tools Perkins (1993) uses the term person-plus. Affordance is a term developed originally by Gibson (1979) to describe the process of both perceiving the physical objects in the world and also to attribute meaning to them. In this way the affordance is both individual-specific based on needs and motivation as well as situation-specific. If one does not require an action one may not perceive it’s potential availability. McGrenere and Ho have refine affordance to mean, "the design aspect of an object which suggests how the object should be used" (McGrenere and Ho, 2000). Based on the idea of person-plus the concept of a communication person-plus is developed as a specialised instance of a more general person-plus. Therefore, the word communication in the term denotes that this person-plus is organised with the specific purpose of communicating with another person. One basic assumption here is that people organize their person-plus according to the perceived affordances given to each of the tools involved. Perkins states that the process of inclusion or exclusion of a tool in a person-plus is based on what he calls the equivalent access hypothesis. According to this theory, person-plus are organized in order to optimize access to knowledge. A subject distributes the knowledge necessary to solve a problem according to the type of knowledge involved, how easy it would be to represent it, how easy it would be to retrieve that knowledge in order to use it effectively and how feasible it would be to construct new knowledge from the knowledge already stored. 1.1 Person-Plus in computer mediated communication In a similar way, the communication person-plus must be organized to optimise communication. Good communication could be achieve by having fluent and effective ways of sending information to another person through the media with as little distortion as possible. As in face-to-face communication, the same information can be delivered through more than one channel at a time as a way of reinforcing it. Although this is not always a positive thing since information sent through the different channels can be contradictory. One very important process occurring in mediated communication is when people look for "equivalent" situations in their normal lives. The specific situation depends on each person’s experience but general common assumptions can be identified. The following table exemplifies some mediated communication settings and the corresponding equivalent situations. It should be noted that some tools have been in use for enough generations that people accept them and they are not perceived to be different or new. Talking on the telephone, for example, is an accepted fact of life for many of us. However, talking on the telephone is not perceived by many people to be the same as an audio-conference through computers. Email is also used and accepted by many people. However, in many cultures this is not perceived to be the same as texting through a chatroom. Mediated situation Face-to-Face situation Text communication on a chatroom Writing letter, (e-mailing) Audio-conference Talk on the phone Videoconference Face-to-face conversation Using an Electronic Shared Whiteboard Sketching on a piece of paper Table 1?1 Mediated communication and it face-to-face equivalent Although subjects use their affordances for mediatonal tools to set up their communication person-plus, those affordances are generally based on affordances of face-to-face equivalent communicative situations. The problem with this is that sometimes affordances can not be directly transferred between both settings. For example, in face-to-face communications people often look at each other and at other objects in mutual space. One person may refer to an object just by signalling to it using eyes and eyebrows. In a videoconference, there is no shared mutual space and so a gesture towards a referent may not be understood if it is not within the camera visual scope (Grayson and Monk, 2003). Another problem is that people in general are not conscious of their use of multiple communication channels in a face-to-face situation so when they transfer affordances from face-to-face to mediated communication, only affordances that subjects are conscious of, are transferred (Quek, et al, 2002). For example, if a person is not conscious of the relevance of gestures to support his/her verbal communication in a normal conversation, when put in a mediated communication setting he/she would tend not to do gestures in front of the camera. He/She would perhaps do the gestures unconsciously but not necessarily in front of the video camera. Finally, a mediated setting can make multimodal communication too fragmented. People are used to communicating through different communication channels simultaneously. This is an automatic process, which people do in a very natural way. In a mediated setting in contrast, most of the times technology makes the separation between the channels too evident. Therefore, people become more conscious of what they are doing, adding another difficulty to the communication process. Methods - Available tools In the research reported here undergraduate Psychology students at a Latin American University took part in a series of problem-solving tasks using different tools available in NetMeeting. These students varied in their confidence and experience of using computers and none had ever used a video or audio conferencing system before. Four main types of tools available for individuals to incorporate into their communication person-plus: text via chat facility, audio- , video- conferencing and application sharing. Students worked in paris in separate rooms communicating only through the computers. They completed a series of shared tasks which involved either drawing a picture or folding paper into orgami designs. One person would be given the prototype and would instruct his/her partner on how to create the model. The actual set of available tools for a specific activity was determined by the conditions under which it should be performed. subjects could bring into the activity their past experience of instructing somebody else face-to-face. This could include, for example, previous experiences where ways in which instruction was most effective were discovered and strategies developed throughout those experiences to ensure that messages were correctly interpreted. Finally, subjects had available all their stories of communicating face-to-face with other people. This involved things like language, cultural conventions, communication protocols, etc. This is all mixed up with subjects' communication styles and effectiveness. The general characteristics of subjects' face-to-face communication are expected to be maintained, at least at the beginning, in a mediated setting. In other words, characteristics like, clarity or lack of it, pace, organization and structure, are at least initially sustained in the new environment. 1.2 Structuring process of a communication Person-plus The configuration of a communication Person-Plus is an iterative process with at least two identifiable stages. First, individuals structure their person plus according to affordances of available physical tools, and the symbolic and psychological tools they feel would best fit in. Second, a negotiation between the communication partners takes place where they agree on the tools and strategies to use. Generally the outcome of this agreement would be the restructuring of their person-plus. It seems apparent that subjects modify their communication person-pluses according to the on-going experience with the tools. For most subjects this was their first time using the tools, therefore, at the same time they were trying to use the tools to communicate with the other person, whilst learning how to use them effectively. The nature of the activity in which the subjects are involved, means that there is an inevitable mutual influence on the configurations of each person’s communication person-plus. This influence can be either direct, e.g. when they negotiate the strategies to follow, or indirect, e.g. subjects might change tools, or the way they use them, depending on how effective they think it is according to their partner's reactions or feedback. Another way of indirect influence is modelling; when one subject discovers a new way of using one tool through the use made of it by his/her partner. Of course, there are different types of information to be sent. One way to classify the information sent is in relation to the pragmatic function they serve within the dialogue. The pragmatic functions considered in this study were based on a coding system developed by the human communication research centre, an institute run by the Universities of Edinburgh and Glasgow. Some examples of pragmatic function are discussed below, for a complete description see section xxxx. In a task-oriented dialogue it is expected that the most frequent function for an utterance will be to instruct the other person to perform some action e.g. "Draw a little tree just beside the house". Another frequent example is when the function of an utterance is to acknowledge that the message has been received and/or understood. Other types of functions include: to query, to reply, to explain, to clarify, etc. A subject that needs to instruct another that is at a distance has to decide, which is the best way to deliver the information (e.g. an instruction) to him through the communication channels available. In the case of this study, this would imply for example, deciding how to instruct the other person to draw a tilted line from the upper right corner to the lower left corner of the screen. Possible options included: using verbal communication, written or oral?, drawing it on a sketch in the whiteboard?, showing it with his hand on the video window? Or through a combination of them? 1.3 Structuring process of a communication Person-plus The configuration of a communication Person-Plus is an iterative process with at least two identifiable stages. First, individuals structure their person plus according to affordances of available physical tools, and the symbolic and psychological tools they feel would best fit in. Second, a negotiation between the communication partners takes place where they agree on the tools and strategies to use. Generally the outcome of this agreement would be the restructuring of their person-plus. It seems apparent that subjects modify their communication person-pluses according to the on-going experience with the tools. For most subjects this was their first time using the tools, therefore, at the same time they were trying to use the tools to communicate with the other person, whilst learning how to use them effectively. The nature of the activity in which the subjects are involved, means that there is an inevitable mutual influence on the configurations of each person’s communication person-plus. This influence can be either direct, e.g. when they negotiate the strategies to follow, or indirect, e.g. subjects might change tools, or the way they use them, depending on how effective they think it is according to their partner's reactions or feedback. Another way of indirect influence is modelling; when one subject discovers a new way of using one tool through the use made of it by his/her partner. Figure 1?2 shows the available tools for two person-plus involved in a communication activity and the interactive processes affecting/modifying them (overlapped section). Figure 1?2 Variables affecting Communication Person-plus configurations References Gibson, J. J. (1979) The Ecological approach to visual perception. Boston: Houghton-Mifflin Grayson, D.M. and Monk, A.F. (2003) Are you looking at me? Eye contact and desktop video conferencing. ACM Transactions on Computer-Human Interaction, 10 (3) 221-243 McGenere and Ho, W. (2000) Pea, R. (1993) Perkins, D. (1993) Quek, F., McNeil, D., Bryll, R., Duncan, S., Ma, X., Kirbas, C., McCullough, K.E., and Ansari, R. (2002) Multimodal human discourse: Gesture and Speech. ACM Transactions on Computer-