Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments
Michael Smyth, David Benyon, Rod McCall, Shaleph O’Neill and Fiona Carroll The HCI Group, School of Computing, Napier University, Edinburgh EH10 5DT, UK
1. Introduction One of the overarching themes that unified the European Commission’s Future and Emerging Technologies Presence initiative was the relationship between the creation of virtual environments and the subsequent measurement of the sense of presence experienced by participants. This continued dialogue builds on a tradition within the Presence community and has resulted in a variety of tools and techniques for the measurement of sense of presence. Examples include, the Immersive Tendencies Questionnaire (ITQ) that was developed to identify real world tendencies (eg using computer games) that may affect a person’s sense of presence, (Witmer and Singer, 1998). Television Commission. The ITC-SOPI was developed for the UK’s Independent
It is a cross media questionnaire that explores: spatial presence; levels of
engagement; sense of naturalness and negative aspects that effect presence (Lessiter et al, 2000). These approaches have been primarily quantitative and designed to be applied post hoc to developed environments. Consequently an implicit relationship has evolved between the development of technology and the measurement of its performance. In short, technological development drives and seeks vindication through this relationship. While the creation of virtual environments is undoubtedly informed on a case by case basis, there appears to be little evidence of how their measurement will explicitly inform the design of future environments.
2. A Human Computer Interaction based approach to Sense of Presence An enduring belief of Human Computer Interaction (HCI) practitioners is the intimate relationship between the generation of requirements for future systems and the subsequent criteria for evaluation of those systems. Requirements and evaluation are two sides of the same coin. It is from this position that the author’s have sought to reconnect the measurement of sense of presence to the articulation of requirements for the design and developments of virtual environments.
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: firstname.lastname@example.org 1
The research reported in this chapter was undertaken as part of the Being There Not Going (BENOGO) project. One aim of the research was to develop new tools for empirical and theoretical studies of presence based on the concept of the observer’s embodiment in the computationally created virtual environment. To meet this aim a number of tools have been developed for the empirical and theoretical study of presence based on the concept of embodied cognition, or embodiment (Dourish, 2001). As real places (possibly known to the observer) with man-made and/or organic objects (like trees, foliage etc.) are hard to represent in a graphically constructed virtual environment, another objective was to bring about new insight into presence through comparison with the sense of presence experienced in the real world. The new technology of Image Based Rendering does not require a reconstructed geometrical model of the scene (Buehler, Bosse and McMillan, 2001, Shum and Kang, 2000). It bypasses an important technological problem and presents a breakthrough, but large amounts of image data need to be stored, and recalled for real-time visualization. Thus a third objective of the project was to find ways in which the key aspects of the place can be communicated to engineers, so that technological constraints do not undermine the subjective feeling of place.
It is contended that the philosophical background of embodiment has two important consequences for the research. Firstly, the sense of presence requires a body; it is not solely cognitive. A body is clearly missing from the experience of many virtual environments (such as those rendered using a head mounted display). Secondly, the requirement to articulate what people are thinking and feeling in order to compare experiences across the real and the virtual. This raises the issue of how to probe people’s thoughts about a situation without interfering with the concepts and processes that underpin the experience. For these reasons and for the more pragmatic reasons of informing the design of the photo-realistic virtual environments, the decision was made to distance the research from attempting to understand the whole of the complex concept of presence. Instead the work focused on a key-contributing factor of presence, namely place. The thrust of the BENOGO project is ‘to be there without going’; that is to provide people with a realistic sense of being somewhere else. Underpinning the research was the belief that fidelity of representation was crucial to sense of presence. Accordingly the research focused attention on capturing
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: email@example.com 2
the essential features of places and finding ways in which to communicate these to the designers and engineers of virtual environments.
Figure 1: An image captured from the Botanical Gardens in Prague, (Turner, Turner and Carroll, 2005)
Figure 2: Series of images from the Jenck’s Landform, Edinburgh, (Smyth, 2005)
Primarily as a means of trying to better understand the criteria that contribute to sense of presence, a series of studies of real places were undertaken (Turner et al, 2003, Turner and Turner, 2003, Smyth 2005 and Turner, Turner and Carroll, 2005). Figure 1 is an image from a photo realistic virtual representation of a glasshouse in the Prague botanical gardens. Participants experienced a 360 degree panorama of the interior of the glasshouse via a head-mounted display. Figure 2 is a series of images of the Jenck’s Landform in Edinburgh that was a location for one of the early studies of place. Each of the studies formed the basis of the early BENOGO demonstrators and critically shaped both the approach to the measurement of sense of presence in both real and virtual environments, but also the subsequent debate as to what constituted presence and its relationship with place. In contrast to the more quantitative approaches to the
measurement of presence, exemplified by the use of questionnaires, a qualitative methodology was adopted within the project. A blend of talk aloud protocols and semi structured interviews were used to elucidate what being present in a given place as a specific time felt like to an individual. While the consequence of this work was to engender debate as to the nature of the relationship between place and presence, the study of real places also provided a set of criteria (‘benchmarks’) against which the virtual environments could be measured. The benchmarks revealed a tension between those involved with the development of the virtual environments and those measuring sense of presence that was to persist throughout the duration of the
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: firstname.lastname@example.org 3
research. Namely that virtual environments, no matter how sophisticated the representation, would always pale by comparison to the real world.
The criticism of the approach and the subsequent lack of uptake demonstrated the need to development measurement tools that encompassed the richness of what it means to be present in a given place but also are specific enough to address the requirements of the technology if it is to successfully support the illusion of presence in a virtual environment. developed. In response to these twin requirements the place probe was
3. The Place Probe The Place Probe comprises a blended approach to the investigation of place in both real and virtual environments. The probe encompasses both the quantitative approach characteristic of the earlier work in the field, but is further supplemented by a battery of more qualitative elements aimed at assessing the specific attributes of the Image Based Rendering (IBR) technology used within the BENOGO project. In a similar vein to traditional approaches to the measurement of the sense of presence, the Place Probe was designed to measure developed virtual environments. It also had the advantage of having the potential to be used in real world environments. A more detailed description of the Place Probe and its early
application can be found in O’Neil et al (2004).
Probes have been used in two main contexts. Cultural probes (Gaver, Dunne and Pacenti, 1999) for the generation of rich data related to the context of use of technology. Central to the application of cultural probes is their intention to provoke responses on the part of the participants. In a similar vein, Technology probes (Westerlund, Lindquist and Sunblad, 2001) have been used to explore the use of technology in primarily domestic settings. In order to generate information about the nature of communications between family members, typically these probes contain a diary, notebook, several disposable cameras, address envelopes and a pen. Participants could record their behaviour over a given time period and then pass that data on to the researchers. Both styles of probes exhibit a reflexive quality as part of the act of recording behaviours and activities required the participants to reflect on these activities. The Place Probe was
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: email@example.com 4
designed to enable the articulation of experiences at a specific time or a particular place rather than over an extended period of time. While it is acknowledged that this would diminish the potential for the probe to encourage reflection, it was considered necessary given the nature of studying places.
The Place Probe is currently in development and versions have been used to study the experience of place associated with a variety of real and virtual environments. Findings from these studies, in particular a study that compared the experience of a real and virtual place, will be used to illustrate the various instruments that comprise the probe. The location for this study was a hillside viewpoint in Prague (Figure 3) and a virtual representation of the same viewpoint (Figure 4). The initial instantiation of the Place Probe contained six instruments. A more detailed treatment of the individual studies can be found in O’Neil et al (2004), McCall et al (2004) and McCall et al (to appear).
Figure 3. View of Prague from the Hillside Viewpoint, (O’Neil et al, 2004).
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: firstname.lastname@example.org 5
Figure 4. Viewpoint Mosaic created for a Head Mounted Display, (O’Neil et al, 2004).
3.1 The Visitors Book Research undertaken by Turner and Turner (2003) has highlighted the written reports contained in visitor’s books as a source of rich data about place. In a study that compared the experience of place at a real viewpoint over the city of Prague (Figure 3) against a virtual representation of the same viewpoint (Figure 4) the following descriptions by participants were typical of those elicited.
“A gray mist is covering Prague, my fingers are bitterly cold but still the magnificence of Prague cannot be covered. I was standing on the hill, which the castle is on, and it was facing the new town, the Volta River and the old town in the distance. All the most beautiful landmarks are visible and I felt totally happy and contented to be there”. (22,female, Singaporean, first time at the viewpoint).
The description of the real viewpoint appeared to support Relph’s (1976) categorisation of place. Comments focussed on the physical aspects, the activities offered and the affective characteristics of the environment.
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: email@example.com 6
“The view was from a hill overlooking a city in southern Europe on a sunny day with birds and some church bells in the distance. I was standing on a paved circle with a statue behind me. The picture was a bit blurry especially the trees. Perspective seemed natural. There were camera/lens reflections hanging in mid air behind me, though it was not possible to see the sun only a white sky”. (24, male, Danish, First time user of HMD technology).
In contrast, the description of the virtual viewpoint concentrated on the perceived technological shortcomings and also lacked the emotional response prompted be the real viewpoint.
3.2 Sketch Maps Sketch maps provide information on the layout and key features of a location. The accuracy of the map is not of prime concern, rather it is the depiction of those aspects of the place that people remembered: for example a tree; a building or a seating area. The maps can also provide additional information about where the individuals were standing or their consequent paths through the location. In the Prague study
participants who experienced the real viewpoint produced sketches of great detail (Figure 5).
Figure 5. A typical sketch map from the Real Hillside Viewpoint in Prague.
Sketches included buildings, landmarks and physical aspects of the environment. Those who experienced the virtual representation produced more simple sketches but did identify key features and incorporated their physical relationship relative to each other (Figure 6).
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: firstname.lastname@example.org 7
Figure 6. Sketch map from the Virtual Viewpoint.
3.3 Salient Features In this section of the Place Probe, participants were asked to generate the three most salient features of the environment. In the case of the real view point in Prague the castle was the physical feature that prompted most comment. While the view itself was commented on, but not to the same extent. The emotional experience was also reflected in the descriptors, for example the feeling and expression of openness and niceness together with the overall impression of peace and quiet and ambient noise. The emphasis in the virtual representation was on the specific structures that comprised the view. For example, a statue located at the viewpoint, a church and the castle were the main features identified. To a lesser extent physical features of the environment were identified and on two separate occasions the feeling of quietness was noted.
3.4 Semantic Differentials Semantic differentials are sets of descriptors relating to place on a bipolar scale. After experiencing the place participants are asked to respond to the descriptors on a 5 point scale. The descriptors were primarily based on Relph’s (1976) model of place and included items such as: attractive-ugly; big-small; colourfulcolourless and noisy-quiet. In the Place Probe a total of 11 descriptor pairs were used. The most positive characteristics reported for the real environment were as follows: attractive; pleasant; interesting and
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: email@example.com 8
relaxing. In contrast the virtual environment produced responses that were more clustered around the centre of the scale that was indicative of a less powerful experience for the participants.
3.5 Six Words The final element of the Place Probe asked participants to write down six words that they felt best described their experience of the place. The rationale was to identify emergent themes based on these words and their synonyms. A grounded approach was used to abstract categories. The categories bore a similarity to those used within the semantic differentials. Examples of categorisations are as follows: peaceful/relaxing; interesting; beautiful; cityscape; enjoyment; amazement; cold; nature and refreshing. Exposure to the virtual viewpoint generated a number of similar themes to that of the real viewpoint, these included: peaceful/relaxing; grainy; weather; Mediterranean; beautiful; interesting and natural. Where the virtual experienced differed was in its ability to produce emotions associated with isolation as indicated by the following categories: faded; lonely; stressful and restrictive.
3.6 Reflections on the Place Probe The blended approach of the Place Probe offered a number of advantages. For example, it contained the mix of granularity necessary to tease out the subtleties associated with the experience of sense of presence. The comparison of the real and virtual viewpoint demonstrated the probe’s ability to reveal salient features associated with a place. The real place was seen to evoke a positive emotional affect in the participant’s responses. Whereas the virtual place was affected by technical issues that effectively reduced the sense of place experienced in comparison with the real place. While the Place Probe revealed a generally positive response to the virtual representation, it did however indicate several negative findings. Overall a
dampening of affect was indicated by the probe, somehow the virtual place failed to translate the vitality of the real place. The virtual place was seen as more passive with a reduced intensity and in some cases was seen as stressful and lonely. This response echoes the debate within the field of architecture that argues that the emphasis on the visual sense in Western culture, coupled with the level of indirection introduced by the use of Computer Aided Design (CAD) technology during the design process, has resulted in ‘designs which housed the intellect and the eye, but that have left the body and the senses, as well as our memories
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: firstname.lastname@example.org 9
and dreams, homeless’ (Pallasmaa, 1996, p10). Buildings are encountered, they are not merely observed. Their importance lies in their ability to articulate and give significance that can only truly be achieved through physical encounter. This position is supported by Franck (1998) when she comments, with specific reference to the role of technology within the design process, that ‘it seems likely that the opportunities afforded by the computer will increase the propensity that already exists in architecture for the form to be disconnected from everyday use and for vision to be the only sense attended to’ (p18). This imbalance of our sensory system has prompted the suggestion that the increased experience of alienation, detachment and solitude in the world today may be related to a certain pathology of the senses which has in turn lead to isolation, detachment and exteriority (Pallasmaa, 1996).
While the Place Probe was considered to be successful in revealing some of the essential attributes of place, it failed to adequately engage with the specific needs of the designer’s of such technologies. Discussions with the designers generated the requirement for the method to produce detailed findings about specific technological issues associated with the creation of virtual environments. To address this failing, a second version of the Place Probe has been developed (McCall et al, in press) that incorporates more quantitative instruments that have been tuned to the technological aspects of the design of virtual places. Of the original six instruments in the initial Place Probe, only four remain they are as follows: the visitor’s book; sketch maps; semantic differentials and the six words. Furthermore, these have been supplemented by the
inclusion of a semi structured interview, a distance estimate test and the MEC Spatial Presence Questionnaire (Vorderer, et al, 2000). To date this current instantiation of the Place Probe has been successfully used to compare the sense of place experienced in virtual environments, with and without computer generated augmentations and across different arenas (eg. CAVE and HMD). The study was aimed at understanding how an environment can be designed to shape the participant’s sense of place and presence. In the case of both arenas the probe indicated a perceived lack of physical movement available to the participants. Interestingly, the MEC questionnaire revealed that this was more marked in the CAVE. This finding, when coupled with the distance estimate test, suggested that participant’s in the CAVE perceived objects as being further away and therefore reduced the propensity to touch leading to a more passive experience. In contrast, the HMD was reported as offering more scope for activities and indeed
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: email@example.com 10
some participants recounted the desire to touch the augmentations. The Place Probe has reaffirmed the importance of legibility of information and the related issue of resolution in virtual environments and the subsequent impact on a participant’s sense of place. Furthermore, it has revealed that different arenas have the potential to evoke different senses of place.
While the second version of the Place Probe was undoubtedly better tuned to the requirements of the designer’s of virtual environments, it was still found wanting, like so many of its predecessors, in not creating a formal mechanism to inform the design of future environments. It relied on an informal build up of design knowledge through the probes repeated application. This failure to bridge the ‘design gap’ was considered to be a major failing of the Place Probe. A more formal mechanism was required to assimilate the probe data and to enable its application during the design process associated with future environments. To specifically address this issue a Pattern based approach was developed.
4. A Pattern Based Approach to Design The approach was inspired by the work of Christopher Alexander (1977) in the field of architecture. Alexander attempted to formalise architectural knowledge based on case studies through the use of templates that described a series of patterns referring to the layout of urban spaces. For example, if an urban planner had the requirement to increase the sense of community associated with a particular location, they might choose to adopt the pattern that suggests the creation of squares and plazas that incorporate seating and spaces for cafes at appropriate road junctions. The strength of Alexander’s approach lies, not in the individual patterns which superficially can appear simplistic, but in their connectedness. Each pattern is linked to others giving the designer a sense of the implications associated with particular design decisions. The pattern based approach to place aims at harnessing a similar gestalt.
The pattern based approach is a method designed to formalise the knowledge gained through the application of the Place Probe. The patterns described in the remainder of this chapter reflect the
aggregation of the understanding of sense of place through the studies conducted as part of the BENOGO project. The approach encapsulates design knowledge and makes it available to the creators of virtual
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: firstname.lastname@example.org 11
environments. Further applications of the Place Probe will provide more data that, in turn, can contribute to existing patterns, or the creation of new ones. The strength of the patterns is that they provide designers of virtual environments with grounded evidence to support design decisions and the choice between alternatives. Both of these factors are characteristic of the early phase of design that the patterns aim to support. The patterns, while informed through the data generated from the use of the Place Probe, can be used independently of the probe and, it is contended, contribute to the design of virtual environments.
As the pattern based approach has been developed from the Place Probe, the underpinning structure is based on Relph’s model of place (1976), but also reflects the development of the probe and includes a category concerning the impact of technologies on the experience of a virtual environment. Relph states that ‘a place is a centre of action and intention’ (p42). He has identified three components as critical to the identity of place, they are as follows: the physical setting; the activities that can take place within that setting and the feelings and meanings associated with that place. Together these components, Relph contends, constitute the ‘three basic elements of the identity of place’ (p47). Figure 7 conceptualises these dimensions of place.
Figure 7. A conceptualisation of place, after Relph (1976)
In the development of the Place Probe, and in particular the use of the Semantic Differentials, has enabled Relph’s dimensions to be further broken down into their constituent parts, relevant to the domain under study.
4.1 The Technology Component
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: email@example.com 12
Figure 8. Technology mediating Real and Virtual Experiences
When producing a virtual representation of a place, the aim should be to achieve as similar a sense of place as possible to the real environment. The themes that have emerged through the application of the Place Probe in real environments can be considered as the requirements that the BENOGO system has to support in order to create an effective representation of that place. The goal of the BENOGO technology is to successfully represent places using IBR to an immersed moving observer. It must be understood that any mediating technology always has an affect on the content that it is mediating. Indeed, Presence has been defined as ‘the illusion of non-mediation’ (Lombard and Ditton, 1997). A conceptualisation of the
relationship between the real and the virtual environments in the context of BENOGO can be seen in Figure 8.
While it is acknowledged that it is virtually (sic) impossible to directly reproduce the exact experience of being in a real place, the BENOGO technology does offer new opportunities to produce experiences that are close to the real experience. In developing the technology it is important to understand those aspects of IBR technology that affect the experience of place. In short, how to develop BENOGO technology towards the illusion of non-mediation.
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: firstname.lastname@example.org 13
4.2 Physical Patterns
Figure 9. Themes underpinning the reported Physical Properties of Real and Virtual Places
Analysis of the various applications of the Place Probe as it developed over the lifetime of the project revealed a series of common themes relating to the physical properties of both real and virtual places. The themes while characteristic of the physical properties also impacted on, or were impacted by, the nature of the activity that the environment allowed, the attributes of the technology used to support the environment and meanings associated with that place. Individual themes have been expanded into pattern templates that have been designed to make the findings accessible and usable for the designers of virtual environments.
To illustrate the operation of the patterns the Big/Small (17) pattern underpinning the physical properties of places will be discussed in more detail. Table 1 presents the pattern entitled Big/Small and relates to the responses participants have to the volume and scale of different places. In certain cases the perceived volume of a place can be either magnified or diminished as an attribute of the technology used to represent the place. For example, a virtual representation of the Prague Technical Museum was perceived correctly as a large enclosed space.
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: email@example.com 14
17. BIG/SMALL (P) Description
Different places are different sizes in reality. They can be big, small or somewhere in between. The technical museum for example is a large room, while the viewpoint is much smaller but feels bigger due to being outside Open/Enclosed(18) . ‘It was a very large room I couldn’t see what was on the other side of the room very well’ (technical museum) ‘ scale was too small… seemed artificially too small’ (botanical garden)
Influential Patterns FOV(6), Acquisition point (1), Motion Resolution (12), Arena (8) Problem
Size Matters. Getting the size right for IBR environments is about combining different factors. The problem is in understanding how these factors relate to one another. Our studies of the Technical Museum for example identified that it was considered to be a big place, that it was also enclosed (18) and full (19) of objects.
Important things to consider in sizing a virtual IBR space are firstly how do the spatial characteristics relate to one another and secondly how can the technology support this relationship in the rendered environment. For example we have already seen how the museum is big, full and enclosed. It was important in terms of technology that these three aspects of the environment were supported. It was therefore imperative that the Acquisition point (1) was established that was at least open to the large scale of the room on one side and yet close enough to the objects in the room to make it feel full. Another important thing to consider is whether people are moving through the environment Motion resolution (12). A Disc REX (14) was used in our version, which allowed some movement but not exploration. Therefore the sense of scale in the IBR environment was different from the real because participants could not explore (29) the IBR environment fully. However the sense of scale was enhanced locally by the parallax provided by the Disc REX i.e. near objects occluded objects that were further away but moving allowed you to see them.
Table 1. Big/Small Pattern relating to the Physical Properties of Places
In contrast, the virtual Prague viewpoint, which was a physically smaller location, was perceived as being larger. This aspect of the pattern was linked to Pattern number 18 entitled Open/Enclosed and Pattern number 19 Full/Empty. Further supporting evidence was provided by the inclusion of quotations from participants in studies that highlighted the issue of the perception of size with respect to the environment. From a technological perspective Pattern 17 was linked to Field of View (6), Acquisition Point (1) Motion
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: firstname.lastname@example.org 15
Resolution (12) and choice of Arena (8). Participants in the Technical Museum study reported that they saw the environment as big, full and enclosed. This finding places a requirement on the technology and specifically the choice of Acquisition Point (1) such that it was open to the large scale of the location but also was close enough to objects contained within to give the viewer the impression that the room was full. Furthermore, if there is a requirement for people to be able to move through the environment this impacts on Motion Resolution (12). In the case of the BENOGO technology a Disc REX (14) was used which allowed some movement but not exploration. This design decision resulted in the perception of scale in the IBR environment being different from the real environment because participants could not Explore (29) the environment fully. However the sense of scale was enhanced locally by the parallax provided by the Disc REX resulting in objects near to the viewer occluding objects further away but supporting head movement so occluded objects could be revealed.
4.3 Activities associated with Place
Figure 10. Themes underpinning the reported Activities supported by Real and Virtual Places
Analysis of the real and virtual environments studied using the Place Probe during the BENOGO project revealed four main activities associated with place. These were as follows: ego motion; static observation; local explorative and locomotive (Figure 10). The Ego Motion (27a) will be considered in more detail in this section. Ego motion is the sensation of movement afforded to participants in virtual environments by the number of images or Motion Resolution (12) rendered by the system (Table 2). In general, the more images that are rendered at run time, the smoother and clearer the feeling of movement through the REX (13, 14 and 15) resulting in a higher motion resolution. As in the previous example, quotations from studies that refer explicitly to ego motion are included in the pattern by way of illustration. From the perspective of the BENOGO technology, ego motion influences both Motion Resolution (12) and the REX (13, 14 and 15). Pattern (27a) describes the problem resulting from ego motion as follows: natural ego
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: email@example.com 16
EGO MOTION Description
Ego motion is the feeling of movement that is afforded by the number of images, or Motion Resolution (12), rendered by the system. The higher the motion resolution i.e. the more images there are rendered at run time, the smoother and clearer the feeling of movement through the REX. ‘I noticed time when I turned my head world was moving a little’ (botanics) ‘ I thought it looked real, it was … I got the feeling it was a museum… and but I don’t think I got the feeling I was there I was kinda of looking into it so …’ (technical museum)
Motion Resolution (12), REX (13, 14, 15)
Natural ego motion always produces parallax and occlusion between the objects in our environment. IBR ego motion attempts to reproduce this affect but is generally restricted by the massive processing power necessary to compute the position of potentially thousands of images. Benogo IBR uses special algorithms that reduce the number of images necessary, however no system as yet has come close to natural ego motion, although BENOGO despite some restrictions comes pretty close.
It is important to establish the type of activity that users will perform in the rendered environment. Accordingly an appropriate REX (13, 14, 15) has to be selected as well as a suitable motion resolution. If restrictions are still evident then ego motion can be supported through augmentation that culturally enforces restricted movement e.g. a guide rope or railing. Similarly augmentation might provide a focus of attention offering interesting parallax effects.
Static Observational (28), Local Explorative (29), Locomotive (30), Spatial Characteristics (16) Identifiable Features (21)
Table 2. The Ego Motion Pattern relating to Activities associated with Real and Virtual Places
motion always produces parallax and occlusion between objects in the environment. IBR ego motion attempts to reproduce this effect but is generally restricted by the massive processing power necessary to compute the position of potentially thousands of images. BENOGO IBR uses special algorithms that reduce the number of images necessary to achieve a realistic representation of natural ego motion. When considering the issue of ego motion it is important to establish the type of activity that users will perform in the rendered environment. Accordingly an appropriate REX (13, 14 & 15) should be selected together with suitable Motion Resolution (12). If restrictions are still evident then ego motion can be supported through the dudicious use of augmentations that culturally enforce restricted movement (eg. a guide rope or railing).
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: firstname.lastname@example.org 17
A related technique is to use augmentation as a distraction owing to its potential for parallax effects. The patterns affected by Ego Motion are as follows: Static Observational (28); Local Explorative (29);
Locomotion (30); Spatial Characteristics (16) and Identifiable Features (21).
4.4 Meanings associated with Place
Figure 11. Themes underpinning the reported Meanings associated with Real and Virtual Places
Analysis of the meanings elucidated by participants in the real and virtual environments studies as part of the BENOGO project resulted in the themes illustrated in Figure 11. In order to explore this class of pattern the Stressful/Relaxing (31) will be explored in more detail (see Table 3). Some environments are more relaxing, or conversely more stressful than others. The degree of relaxation or stress associated with a particular place is deeply linked to a person’s subjective experience of activity in that place; as such it cannot be designed - only designed for. For example outdoor places have the potential to be more peaceful and allow an experience of nature at an easy steady pace are often considered relaxing, where as outdoor places where people encounter unexpected testing circumstances might be considered stressful, (i.e. bad weather conditions, or a loss of orientation). In the case of virtual environments there is the added dimension of the mediating technology interfering with the experience. A beautiful and relaxing scene might be rendered in
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: email@example.com 18
31. STRESSFUL/RELAXING (M) Description
Some environments are more relaxing, or conversely more stressful than others. The degree of relaxation or stress associated with a particular place is deeply linked to a person’s subjective experience of activity in that place; as such it cannot be designed - only designed for. For example outdoor places that are peaceful and allow an experience of nature at an easy steady pace are often considered relaxing, where as outdoor places where people encounter unexpected testing circumstances might be considered stressful, i.e. bad weather conditions, or a loss of orientation. In the case of VR there is the added dimension of the mediating technology interfering with the experience. A beautiful and relaxing scene might be rendered in an HMD and yet the technology might create stressful affects by displaying poor quality images that are hard to focus on, or by disorienting the participant. ‘Very good view but only from one place as trees get in the way everywhere else. Paths are poor and there is no information to direct or explain’ (viewpoint) ‘There was some text but it was unreadable but I could easily identify the object’ (technical museum) ‘Viewpoint close to the monastery, very beautiful view, peace, relaxing’ (viewpoint)
Image quality (5); acquisition point (1); activity (14); motion resolution (9)
VR mediation is essentially illusory. Interference in this illusion can cause stress and often leads to breaks in presence. In the case of rendering a relaxing scene, it is important to try and ensure that factors such as image quality (5) and REX (13, 14, 15) do not interfere with the experience. Low quality out of focus images as well as blind spots and image drop out all lead to disorientation and stress, distracting the user from a possibly relaxing experience. Similarly, in creating a stressful scene such as a cliff edge for example the same attention to detail is necessary to make it believable.
To avoid this, it is important to have high image quality (5) and a suitable REX (13, 14, 15) that does not impede the user experience i.e. it is important to choose an appropriate environment (Acquisition point (1)) and activity type that is compatible with the expectations of the participants. Also a useful aid is to augment images where small details such as text cannot be read.
Table 3. The Stressful/Relaxing Pattern relating to the Meanings Associated with Real and Virtual Places
an HMD and yet the technology might create stressful affects by displaying poor quality images that are hard to focus on, or by disorienting the participant. The Stressful/Relaxing pattern has been found to
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: firstname.lastname@example.org 19
impact primarily on Technology patterns, namely Image Quality (5), Acquisition Point (1), Motion Resolution (9) and the Activity related patterns (27a, 28, 29 and 30). Mediation with a virtual environment is essentially illusionary and any interference with this can result in stress on the part of the user and ultimately a break in presence. In the case of rendering a relaxing scene, it is important to ensure that factors such as Image Quality (5) and REX (13, 14 & 15) do not interfere with the experience. Low quality out of focus images as well as blind spots and image drop outs all lead to disorientation and stress, distracting the user and increasing the potential for a stressful experience. Similarly, if the intention is to create a stressful scene, for example a cliff edge it is important to concentrate on the detail in order to make the experience believable. To avoid such problems it is important to have high Image Quality (5) and a suitable REX (13, 14 & 15). The choice of environment and its associated Acquisition Point (1) is important, together with the activity type that matches the expectations of the participants. Another technique for reducing stress within a virtual environment is to augment images where small details, such as text, cannot easily be read.
Technological Issues Relating to the Sense of Place associated with Real and Virtual
Environments The current version of the Place Probe has been refined to address the technological requirements of the BENOGO technology and as such the themes illustrated in Figure 12 reflect that process. The patterns described refer specifically to the IBR approach adopted within the project. It is envisaged that the Place Probe could be used in conjunction with alternative technologies, for example computer generated environments, and so a variety of different themes could be identified relating to the different mediating technologies available. These could then be slotted into the technology section of the Place Patterns.
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: email@example.com 20
Figure 12. Technology Themes relating to Real and Virtual Places
The pattern describing the Acquisition Point (1) will be described in more detail in the remainder of this section, see Table 4. This pattern refers specifically to BENOGO IBR and relates to the location from which images are captured. Evidence from the Place Probe suggests that it is important to establish a position from which to capture images of a place in order to represent it in an optimum manner. Factors that impact on this decision are firstly, the nature of the scene that is to be portrayed and secondly, what activities are to be supported within the scene. A solution to this requirement is to scope the real place as early and often as possible. Indeed this activity can be combined with the application of the Place Probe. By observing the activity and behaviour of individuals at the real place, it is possible to establish a suitable acquisition point that is in keeping with the technological objectives of the study and captures the important features of the real place. The Place Probe can capture the elements that are perceived within the
environment and these can, in turn, be turned into a design template. Furthermore it is important that an appropriate REX be selected to enable the observed behaviours to be replicated within the virtual environment.
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: firstname.lastname@example.org 21
1. ACQUISITION POINT Description
Specific to Benogo IBR. The acquisition point is the specific location where the images are captured.
Influential Patterns Problem
It is important to establish the best position to acquire the images from, in relation to representing the real place in the best way. This requires taking into account what scene the images will portray and what type of activities might occur there.
The best way to solve this problem is by Scoping the real place first, as well as performing a Place Probe. By observing the activity and behaviour of individuals at the real place, it is possible to establish a suitable acquisition point that is in keeping with the technical objectives of the demo while making sure the most important features of the real place are captured in the images. The Place Probe captures elements of the environment that can be turned into a design template. It is also essential that the appropriate type of REX e.g. Point, Disc, or Line is chosen in line with the requirements derived form the real environment.
Acquisition Resolution (2), Point REX (13), Disc REX (14), Line REX (15)
Table 4. The Acquisition Point Pattern relating to the BENOGO IBR technology.
5. Conclusions Existing tools and techniques for the measurement of the sense of presence in real and virtual environments have failed to provide formal mechanisms through which to inform the design process associated with their creation. From an HCI perspective this was viewed as a major shortcoming. A measurement tool, entitled the Place Probe has been introduced. The probe has utilised a blended approach to the generation of both qualitative and quantitative data concerning the experience of place associated with a range of real and virtual environments. Based on the responses of developers of virtual environments the probe has been refined, specifically the quantitative component has been increased with the inclusion of the MEC Spatial Presence Questionnaire (Vorder et al, 2000). Furthermore, the probe has been tuned to the identification of the technological requirements of the BENOGO approach. Associated with the Place Probe, a pattern based approach has been developed to articulate probe based data into a form that is accessible and
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: email@example.com 22
pertinent during the design phase associated with the creation of virtual environments. The patterns have been classified relative to participants’ responses to a series of real and virtual environments developed over the course of the project as follows: the physical properties; the activities supported; the meanings associated and the technology necessary to create the illusion of non-mediation. While the current version of technology patterns refer to the BENOGO IBR approach, it is envisaged that other mediating technologies, when used in conjunction with the Place Probe, could generate new patterns that could be substituted into the existing set of patterns.
The pattern based approach presented in this chapter is an nascent attempt at connecting the case based approach to the measurement of sense of presence to the design of virtual environments. While the approach remains in development it is considered to be an essential step in the maturity of the field of presence research.
Acknowledgements The authors would like to thank all members of the BENOGO project for their input and co-operation with the studies mentioned in this paper. Funding for this project was provided by the European Union under grant number IST-2001-39184.
References Alexander, C. (1977) A Pattern Language, Oxford University Press, New York.
Buehler, C., Bosse, M. & McMillan, L. (2001)
Non-Metric Image-Based Rendering for Video
Stabilization, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition 8-14 December 2001 Kauai, Hawaii USA.
Dourish, P. (2001) Where the Action Is, MIT Press, Cambridge, USA.
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: firstname.lastname@example.org 23
Franck, K. (1998) It and I: Bodies as Objects, Bodies as Subjects, Architectural Design, Vol 68, No 11/12, 16-19.
Gaver, B., Dunne, T., & Pacenti, E. (1999). Design: Cultural Probes. Interactions 6(1) 21-29.
Lessiter, J., Freeman, E., Keogh, E. & Davidoff, J. (2000) Development of a New Cross-Media Questionnaire: the ITC-Sense of Presence, 3rd International Workshop on Presence.
Lombard, M. & Ditton, T.(1997). At the Heart of It All: The Concept of Presence, Journal of ComputerMediated Communication, 3(2). http://www.ascusc.org/jcmc/vol3/issue2/lombard.html
McCall R, O'Neill S, Benyon D, Smyth M. (2004). A Method for Designing Virtual Places in Proceedings of the Seventh Annual International Workshop Presence 2004, Universidad Politecnica de Valencia, Spain
O'Neill, S.J., McCall, R., Smyth, M., & Benyon, D.R. (2004). Probing The Sense of Place in proceedings of the Seventh Annual International Workshop Presence 2004, Universidad Politecnica de Valencia, Spain
Pallasmaa, J (1996) The Eyes of the Skin, Architecture and the Senses, Academy Editions.
Relph, E. (1976) Place and Placelessness, London: Pion Books.
Shum, H. and Kang, S. (2000) A Review of Image-Based Rendering Techniques, SPIE Int. Conf. On Visual Communication and Image Processing, 2-13.
Smyth, M. (2005) Articulating the Sense of Place Experienced by Visitors to the Jencks Landform, in P. Turner & E. Davenport (Eds.) Spaces, Spatiality and Technology, Springer, 249-260.
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: email@example.com 24
Turner, S., Turner, P., Carroll, F., O'Neill, S., Benyon, D., McCall, R., & Smyth, M. (2003). Re-creating the Botanics: towards a sense of place in virtual environments. The 3rd UK Environmental Psychology Conference, June 2003, Aberdeen, UK
Turner, P. & Turner, S. (2003) Two Phenomenological Studies of Place. In the Proceedings of The British Computer Society Annual Conference on Human-Computer Interaction (BCS-HCI). Bath, United Kingdom, 8-12 September.
Turner, P., Turner, S., & Carroll, F. (2005). The tourist Gaze: Contextualised Virtual Environments, in P. Turner & E. Davenport (Eds.) Spaces, Spatiality and Technology, Springer, 281-297.
Vorderer, P., Wirth, W., Gouveia, F. R., Biocca, F., Saari, T., J?ncke, F., B?cking, S.,Schramm, H., Gysbers, A., Hartmann, T., Klimmt, C., Laarni, J., Ravaja, N., Sacau, A., Baumgartner, T., & J?ncke, P. (2004). MEC Spatial Presence Questionnaire (MEC-SPQ): Short Documentation and Instructions for Application. Report to the European Community, Project Presence: MEC (IST-2001-37661). Online. Available from http://www.ijk.hmt-hannover.de/presence
Westerlund, B., Lindquist, S., & Sundblad, Y. (2001) Cooperative design of communication support for and with families in Stockholm using communication maps, communication probes and low-tech prototypes. First Equator IRC Workshop on Ubiquitous Computing in Domestic Environments, Nottingham, 13-14 th September 2001.
Witmer, B.G. & Singer, M.J. (1998). Measuring presence in virtual environments: A presence questionnaire. Presence, 7(3), pp 225-240.
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments ? Handbook of Presence – 1st Draft ?8 June 2005 ? contact: firstname.lastname@example.org 25
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environmen发布时间:
Patterns of Place – A Toolkit for the Design and Evaluation of Real and Virtual Environments
- Design and Evaluation of Menu Systems for Immersive Virtual Environments
- Design and Implementation of RMP - A Virtual Electronic Market Place
- A conversational agent as museum guide - Design and evaluation of a real-world application
- Design and Evaluation of a Parallel-Polled Virtual Output Queued Switch
- VGUIDE DESIGN AND PERFORMANCE EVALUATION OF A SYNCHRONOUS COLLABORATIVE VIRTUAL REALITY APP
- A Virtual Learning Environment for Deaf Children Design and Evaluation
- Modelling patterns for analysis and design of real-time systems
- Virtual Reality User-Centered Design and Evaluation of Virtual Environments
- Virtual Manufacturing and Design in the Real World-- Implementation and Scalability on HPPC