The thesis analyses the visual space displayed in pictures, film, television and digital interactive media.

The argument is developed that depictions are informed by the objectives of the artefact as much as by any simple visual correspondence to the observed world. The simple concept of ‘realism’ is therefore anatomised and a more pragmatic theory proposed which resolves some of the traditional controversies concerning the relation between depiction and vision. This is then applied to the special problems of digital interactive media.

An introductory chapter outlines the topic area and the main argument and provides an initial definition of terms. To provide a foundation for the ensuing arguments, a brief account is given of two existing and contrasted approaches to the notion of space: that of perception science which gives priority to acultural aspects, and that of visual culture which emphasises aspects which are culturally contingent. An existing approach to spatial perception (that of JJ Gibson originating in the 1940s and 50s) is applied to spatial depiction in order to explore the differences between seeing and picturing, and also to emphasise the many different cues for spatial perception beyond those commonly considered (such as binocularity and linear perspective). At this stage a simple framework of depiction is introduced which identifies five components or phases: the objectives of the picture, the idea chosen to embody the objectives, the model (essentially, the visual ‘subject matter’), the characteristics of the view and finally the substantive picture or depiction itself. This framework draws attention to the way in which each of the five phases presents an opportunity for decision-making about representation. The framework is used and refined throughout the thesis.

Since pictures are considered in some everyday sense to be ‘realistic’ (otherwise, in terms of this thesis, they would not count as depictions), the nature of realism is considered at some length. The apparently unitary concept is broken down into several different types of realism and it is argued that, like the different spatial cues, each lends itself to particular objectives intended for the artefact. From these several types, two approaches to realism are identified, one prioritising the creation of a true illusion (that the picture is in fact a scene) and the other (of which there are innumerably more examples both across cultures and over historical time) one which evokes aspects of vision without aiming to exactly imitate the optical stimulus of the scene. Various reasons for the latter approach, and the variety of spatial practices to which it leads, are discussed.

In addition to analysing traditional pictures, computer graphics images are discussed in conjunction with the claims for realism offered by their authors. In the process, informational and affective aspects of picture-making are distinguished, a distinction which it is argued is useful and too seldom made. Discussion of still pictures identifies the evocation of movement (and other aspects of time) as one of the principal motives for departing from attempts at straightforward optical matching. The discussion proceeds to the subject of film where, perhaps surprisingly now that the depiction of movement is possible, the lack of straightforward imitation of the optical is noteworthy again. This is especially true of the relationship between shots rather than within them; the reasons for this are analysed. This reinforces the argument that the spatial form of the fiction film, like that of other kinds of depiction, arises from its objectives, presenting realism once again as a contingent concept.

The separation of depiction into two broad classes – one which aims to negate its own mediation, to seem transparent to what it depicts, and one which presents the fact of depiction ostensively to the viewer – is carried through from still pictures, via film, into a discussion of factual television and finally of digital interactive media. The example of factual television is chosen to emphasise how, despite the similarities between the technologies of film and television, spatial practices within some television genres contrast strongly with those of the mainstream fiction film. By considering historic examples, it is shown that many of the spatial practices now familiar in factual television were gradually expunged from the classical film when the latter became centred on the concerns of narrative fiction. By situating the spaces of interactive media in the context of other kinds of pictorial space, questions are addressed concerning the transferability of spatial usages from traditional media to those which are interactive.

During the thesis the spatial practices of still-picture-making, film and television are characterised as ‘mature’ and ‘expressive’ (terms which are defined in the text). By contrast the spatial practices of digital interactive media are seen to be immature and inexpressive. It is argued that this is to some degree inevitable given the context in which interactive media artefacts are made and experienced – the lack of a shared ‘language’ or languages in any new media. Some of the difficult spatial problems which digital interactive media need to overcome are identified, especially where, as is currently normal, interaction is based on the relation between a pointer and visible objects within a depiction.

The range of existing practice in digital interactive media is classified in a seven-part taxonomy, which again makes use of the objective-idea-model-view-picture framework, and again draws out the difference between self-concealing approaches to depiction and those which offer awareness of depiction as a significant component of the experience. The analysis indicates promising lines of enquiry for the future and emphasises the need for further innovation.

Finally the main arguments are summarised and the thesis concludes with a short discussion of the implications for design arising from the key concepts identified – expressivity and maturity, pragmatism and realism.