AGOCG logo
Graphics Multimedia VR Visualisation Contents
Training Reports Workshops Briefings Index

Report on Workshop on Presence in Shared Virtual Environments


BT Laboratories
Martlesham Heath
Ipswich, UK
10 - 11 June 1998

For two rainy days in June, forty-odd researchers and interested parties met at the BT Laboratories in Martlesham Heath for a BT-sponsored workshop on "Presence in Shared Virtual Environments" organised by Prof Mel Slater from University College, London. The workshop consisted of presentations, open discussions and demonstrations of the work in progress at the BT centre.

The motivation for the workshop was the developing research in the area of presence. There is a great deal of work on shared virtual environments by research groups and organisations around the world. These virtual spaces are inhabited by several real and virtual people simultaneously and can be text-based through to full multimedia systems. They offer novel possibilities for communications and collaboration, with potentially very far reaching implications for the organisation of work and cultural exchange in the future.

In spite of this wide ranging work, this was the first workshop on the concept and practice of 'Presence', an oft used word, but suffering from various interpretations. The purpose of the workshop was therefore to explore particular understandings of the term and share experiences of the field's experts.

The meeting posed a number of general questions:

As participants offered answers to and thoughts about these questions the workshop was sub-divided into four sessions, Conceptual Issues, Practice and Experience I & II and Measurement, although inevitably issues naturally crossed the borders. Below is a summary of some of the papers presented.

Nat Durlach, of the Research Laboratory of Electronics at MIT, started the presentations with work done in collaboration with Mel Slater. Their paper entitled "Presence in Shared Virtual Environments and Virtual 'Togetherness'" considered two factors they believe relevant to creating 'togetherness' and represented a conceptual framework for another paper - An Experiment on the Influence of Haptic Communication on the 'Sense of Being Together', presented later in the workshop.

The two factors Durlach and Slater consider relevant to togetherness are:

The extent to which users feel present within a shared environment will depend on the same factors that determine presence in individual, unshared, environments: they are graphics update rate, low latency, wide field of view, and high degree of interactivity. But because of the shared nature of the environment there will be additional factors such as the noticeable effect of one user on the environment from the other users' viewpoints.

In addition, the 'rich, multimodal, real time, intraspecies communication' will obviously increase presence but the area that Durlach and Slater believe has most potential is communication in the tactual channel. Touch is not a 'distance sense', it is inherently intimate, not only physically but emotionally too, and the ability to hold and manipulate objects offers a high level of interaction thereby increasing the user's sense of presence.

Later in the gathering Durlach and Slater developed these ideas by reporting the results of recent experiments in haptic communication and comparative studies between small group behaviour in virtual and real environments.

Mike Holderness proposed that Collaborative Virtual Environments are "best seen as 'channels of communication' between participants in Lévy's 'collective intelligence' or Stewart and Cohen's 'extelligence', and therefore that an effective design principle for CVEs is not fidelity but the avoidance of noise [dissonance]". He went on that for CVEs to engender a sense of presence we need to examine the commonalties with different media and environments which exhibit or facilitate presence.

Carlton Reeve, from the University of Bradford's EIMC department, described research into virtual theatre applications where presence for the actors rests on three relationships, that between the actor and his avatar, between the actor and virtual space and that between actors, the group dynamic. The actor-avatar relationship rests on the user's ability to control the avatar immediately and 'transparently' while having a wide range of communication tools such as gestures and identifiable expressions. The actor-space relationship is developed through the repetition associated with rehearsal and the link between script and movement. This 'blocking' identifies where the actor is at any one time and therefore his orientation and consequently his sense of 'being there' increases as this become more familiar. Finally the group dynamic that permeates a cast is crucial in creating a sense of individual and group ownership of a world. This recognition of the space and the other avatars engages the users and further promotes a sense of presence.

Stefan Thie, from KPN Research, The Netherlands, offered a 'new experimental testable overall theory on presence'. The general theory was that the degree of presence in a shared virtual environment is related to the task people attempt and accomplish. In the specific case of a decision-making task, the hypothesis is that presence will be higher if 'social cues', such as proximity and eye-contact, are maximised. Thie's theory expounded the idea of a mental model of the environment that assists the users in achieving a sense of presence within the Virtual Environments and that the learning process and minimising of the real world presence are essential for maximising the feeling of being present within the Virtual Environments.

A team from the Computer Based Learning unit at the University of Leeds, reported on experiments in 'Collaborative Decision-Making and Presence in Shared Dynamic Virtual Environments'. Their work examines the relationship between a sense of presence and team training in a simple desktop virtual game environment. Their preliminary conclusions suggest that such a semi-immersive dynamic Virtual Environments could achieve "a sense of presence sufficient to provide trainees with an experience of the same cognitive value as one in the real world, without the need to construct a full immersive Virtual Environments with all its associated costs'".

Roy Kalawsky from the Advanced VR Research Centre at Loughborough University described the use and continuing development of a VR Situation Awareness Rating System Technique (VRSART) as a diagnostic tool for investigating presence and its effect on user/task performance. Usually the measurement of user performance within a Virtual Environments is a combination of objective, subjective, psycho-physical, physiological, task performance and learning efficiency factors, however the actual process of collecting this data can affect the results if the collection takes place during the trial. VRSART is based on situation awareness debriefings that take place immediately after the trial and deals with six principle factors:

A combined team from the University of Essex and the Independent Television Commission presented the findings of research into 'Behavioural Realism as a metric of Presence'. The work is based on a preliminary experiment evaluating the effects of viewing condition on postural sway with a view to corroborating subjective scales of presence and constructs related to presence.

In their paper 'Measuring Temporal Variations in Presence', W A Ijsselsteijn and H de Ridder describe experiments in 3DTV research related to the perceptual link between the observer and the mediated environment, supporting the illusion of non-mediation. Their results indicate that subjective presence ratings are 'subject to considerable temporal variation depending on the stimulus material used'. In addition they found qualitative evidence that suggests an observer's sense of presence may be enhanced by the addition of stereoscopic and motion parallax cues.

Monika Büscher of Lancaster University's Computing Department presented a paper entitled "'Red is behind you': the experience of presence in shared virtual environments". The paper describes the results of ethnographic fieldwork undertaken at a multimedia art museum with a number of public shared virtual environments. Büscher argues that "a sense of presence is not 'split off' from presence in the 'real world' but rather extended from the real to the virtual space, and that everyday practices of orientation, movement and interaction are transposed in order to fit in with the affordances of the electronic environment". She suggests that 'realism' in terms of behaviour and appearance is not absolutely necessary to engender a feeling of presence.

The final collection of papers, plus a summary of the discussion at the workshop will appear as an electronic publication. There will also be a special issue of the journal Presence: Teleoperators and Virtual Environments for selected papers.

The full workshop programme is available at:
http://vb.labs.bt.com/SharedSpaces/Presence/


Carlton Reeve
C.M.Reeve@bradford.ac.uk