# Introduction --- # 1. Aesthetic Specificity ## 1 Aesthetic Specificity In 2016 Werner Herzog made the following remarks about the potential of virtual reality as an artistic medium: >I am convinced that this [virtual reality]is not going to be an extension of cinema or 3-D cinema or video games. It is something new, different, and not experienced yet [...] normally, in the history of culture, we have new stories and narrations and then we start to develop a tool. Or we have visions of wondrous new architecture—like, let's say, the museum in Bilbao, or the opera house in Sydney—and technology makes it possible to fulfil these dreams. So you have the content first, and then the technology follows suit. In this case, we do have a technology, but we don't have any clear idea how to fill it with content (Werner Herzog, in Patrick House (2016), 'Werner Herzog Talks Virtual Reality', The New Yorker) One way of interpreting what he is saying is as follows: we do not yet know what is medium specific about VR. What, if anything, does this technology make possible that no other artistic medium can? Medium specificity can be made clearer by considering the following remarks from Noel Carrol on the advent of film, and how it differed to theatre: > However, even if cinema could deliver art to its audiences, was it an art form in its own right or merely theater in a can—that is, a film can? In other words, was film only a recording device, relaying whatever transpired theatrically before the camera to audiences, or did it have an original artistic contribution of its own to make?"  (2021, p. 23) We can imagine a theatre director learning about film for the first time thinking something like 'Great, now I can just have my actors perform the play once, and people can watch it wherever there is a cinema screen'. While direct recordings of plays certainly take place, just thinking of film as 'theatre in a can' is to overlook all the other possibilities that film allows for that theatre does not. Shooting on film allows for cuts, so that the same scene can be seen from multiple angles, it allows for montage, tracking shots, smash cuts, and many other effects not possible in the theatre. Such features contribute to film's medium specificity, and have been used by filmmakers for unique aesthetic and artistic effects. We can think of similar sorts of contrast between music played live and recorded music. The latter is much more than a concert in a can: it allows stereo effects, layering of multiple tracks, and the ability to manipulate sound in post-production. For example, the stereo effect of the aeroplane sound travelling across the stereo field in The Beatles' 'Back in the U.S.S.R.' would be extremely difficult to recreate accurately in a live setting. Despite a resurgence in popularity, the term ‘virtual reality’ was first used in the 1980s, and since that time has been used for things like films (e.g. The Great C; Clouds Over Sidra), dance performances (e.g. Gilles Jobin's 'VR_I'), and video games. However, these examples essentially transfer existing art forms into virtual environments and little more —akin to putting theatre, dance, or gameplay in virtual cans. If VR has unique, medium specific characteristics, pieces like these, regardless of any other merits that might have, do not take advantage of them.  In this paper we will suggest a hitherto unconsidered possibility about the medium specificity of VR: its technical set up (lenses close to the eyes, real-time head tracking, etc.) allows for the alteration of structural features of perceptual experience. In the next section we will briefly consider another feature that is sometimes suggested as a possible aesthetic specificity of VR: immersion. While our arguments will not be conclusive, we will suggest that this is a dead end, as immersion is a feature of the real world, so it is a mistake to think that it can be used to explain how VR is unique. In part 2, we will introduce the idea that perceptual modalities have structurally invariant features, before going on to provide cases in which these invariant features can be altered through technological intervention. In part 3, we will argue that VR is unique in that it is the only artistic medium which allows for the structural properties of perceptual experience to be altered. We will show how this works by considering real-life cases, as well as other possibilities that the VR set up of lenses close to the eyes and motion tracking allows for. In part 4, we will suggest that VR's capacity to alter the structural properties of perceptual experience is not limited to visual perception. One aspect of VR that one might be tempted into thinking contributes to its specificity is immersion. VR can place users in detailed and otherworldly virtual environments and we might think that this is what sets VR it apart from other media. While there may be something to this it will not be the focus of this article. One reason to be wary of taking this path is that there are other forms of art which are immersive: architecture and installation art immerse spectators in virtue of their being part of actual reality, and so we might think that virtual immersion is, more or less, ‘installation art in a can’, or ‘architecture in a can’. A possible response to that one could make here is that VR allows for immersion in impossible worlds which violate logic or physical laws, but it remains to be seen if VR can really provide such experiences, or if such experiences are really that unique. For these reasons, we will focus here on a different candidate for aesthetic specificity. --- # 2. Perceptual Restructuring ## 2. Perceptual Structure Our central claim in this paper is that one medium specificity of virtual reality is that it is capable of changing certain structural features of perception. In this section we will provide some detail as to what invariant structural features are. In the next, we will suggest that these structural features can be altered by certain technological interventions. In Section 4, we will argue that Bordini describes invariant structural features as follows: > by attending to the objects of the experience, we can introspectively discern certain invariant structural features, such as the visual field or its boundaries, which remain constant as objects and properties change. Such structural features are modality-specific and are responsible for how we experience things within that modality. Consequently, they are not properties of what we experience but of the experience itself (2023, p. 271) The visual spatial field is the most obvious example of a structural feature. It seems right to say that visual experience is 'field-like' in that it presents objects as occupying positions within a bounded region of space, in a way that olfaction or audition do not. Richardson’s work on the spatial field brings out several aspects of this type of structural feature. First, it remains constant as the objects and properties represented change: your vision will seem field-like, wherever you look and whatever you look at. Indeed, your visual field persists when you close your eyes, or open them in a pitch black room.  Second, it shapes our visual experience: #### "The feature also deserves to be called 'structural' because what it contributes to the qualitative character of visual experience, in general, is something quite distinctive: it 'structures', or 'organises' it. It imposes a certain form on visual experience” (p. 13) Third, our awareness of the visual spatial field in our visual experience is quite different from our visual awareness of the objects we see: #### "These features structure our perceptual experience of the things we are aware of. Vision's having its spatial field, on this view, is not a matter of our being aware of another object, or property of an object, in addition to, say, the lemon and the table and their properties. So there is no appropriate additional object or property of an object to attend to, when reflecting upon these structural features of perceptual experience." (p. 493) This view allows Richardson to maintain that when we introspect, we are still primarily aware of the objects in the world, not features of our experience. Furthermore, Richardson does not consider her account to be a version of qualia theory because she doesn't treat the visual field as an additional sensational property or a detachable quale. Instead, she argues that it is an intrinsic, organising feature of visual experience: "Similarly, the structural feature of visual experience involved in seeing empty space is intrinsic to the qualitative character of visual experience generally, rather than some positive quality that each visual experience has and could lack. It is a feature of all visual experience, and not just visual experience of empty space" (p. 12). By characterising the visual field as a structural feature that organises experience rather than an additional content or property, Richardson distinguishes her account from qualia theories of perceptual experience. ## 3. Perceptual Restructuring We now want to make a further suggestion: it is possible for at least some structural features of perception to be altered via technological means. In this section, we will give some examples. In the next, we will show that VR Hardware is very well-suited to produce similar alterations.  Consider inverting lenses and prism glasses. These optical devices alter the spatial orientation of the visual field: inverting lenses flip it vertically, making the world appear upside down, while prism glasses shift it horizontally or vertically. This alteration can be understood as modifying the structural properties of vision in several ways that align with Richardson's characterization. First, the inversion or displacement is not a matter of perceiving new objects; the content of perception remains unchanged while its spatial organisation is transformed. Second, this reorganisation remains constant throughout the use of the devices, persisting as an invariant feature of visual experience. Finally, this technology the structure of visual experience by disrupting the usual spatial relationships that organise our perception. Just as Richardson describes the visual field as imposing "a certain form on visual experience" (REF) these devices impose a new form, one that is an inverse of the normal one. Cochlear implants offer another instance of perceptual restructuring, this time in the auditory domain. These electronic devices provide a sense of sound to individuals with severe hearing loss by directly stimulating the auditory nerve, bypassing damaged hair cells in the cochlea. They alter the structural properties of auditory perception in ways that align with Richardson's characterization of structural features. First, the change is not a matter of hearing new auditory objects, but rather a reorganisation of how sound waves are processed. In ordinary hearing, the cochlea's tonotopic organisation allows for continuous, high-resolution frequency discrimination. Cochlear implants, however, divide the frequency spectrum into discrete bands, typically between 12 to 22, each stimulating a different region of the auditory nerve. This discretization of frequency perception remains constant during the use of the implant, satisfying the second feature of structural properties. Lastly, this alteration reshapes the structure of auditory experience. The limited number of frequency bands results in a reduced spectral resolution compared to natural hearing, affecting the perception of pitch and timbre. This change in frequency structure modifies how auditory information is organised in perception, altering the form of auditory experience without introducing new auditory objects or properties. Finally, visual field expanding devices for hemianopia provide a third example of perceptual restructuring, one that directly relates to Richardson's discussion. These devices use optical elements like mirrors or prisms to redirect light from areas of peripheral vision loss into intact areas of the visual field. For individuals with hemianopia, who have lost half of their visual field due to conditions such as stroke, these devices effectively alter the structural property of the size and boundaries of the visual field. This alteration aligns with Richardson's characterization of structural features in several ways. First, it doesn't introduce new objects of perception but changes how existing visual information is organised. Second, the expansion of the visual field remains constant during the use of the device. Third, it modifies the structure of visual experience by changing the spatial layout of visual information. Notably, this example directly affects the visual spatial field, a key structural feature in Richardson's account. By allowing users to perceive objects in previously blind regions, these devices demonstrate that even fundamental structural properties of perception, such as the extent and limits of the visual field, can be modified through technological intervention. --- # 3. Virtual Reality and the Structure of Perception ## 4. Virtual Reality and the Structure of Perception In the previous section, we examined how certain technological interventions—such as inverting lenses, prism glasses, and cochlear implants—can alter the structural features of perceptual experience by directly manipulating sensory input at points very close to our sensory organs. These devices change the way sensory information is organised and processed, leading to alterations in the invariant structural properties of perception as characterised by Bordini (2023) and Richardson. In this section, we argue that virtual reality (VR) hardware, particularly its visual components involving lenses placed extremely close to the user's eyes, shares this critical characteristic. This proximity allows VR to alter structural features of perception in ways similar to these technologies, making it uniquely suited among artistic media for such modifications. 4.1 The Role of Proximity in Perceptual Alteration A key similarity between VR hardware and the technologies discussed earlier is the proximity of the output devices to our sensory organs. Inverting lenses and prism glasses are worn directly in front of the eyes, altering the incoming light before it reaches the retina. Cochlear implants are surgically placed within the cochlea, directly stimulating the auditory nerve. This close proximity is crucial because it allows for the manipulation of sensory input at a fundamental level, thereby altering the structural features of perception rather than merely presenting content with a particular apparent structure. Similarly, VR headsets place lenses and screens mere centimetres from the user's eyes. The lenses in VR devices are not passive; they actively shape the incoming light to create specific perceptual effects. By controlling optical characteristics such as focal length, distortion, and convergence, VR hardware can directly influence how images are projected onto the retina. This setup ensures that the alterations made by the VR hardware are integrated into the user's perceptual experience at a structural level, much like how inverting lenses and cochlear implants function. The proximity of these devices to our sensory organs distinguishes them from other technologies that present content at a distance, such as screens or speakers. While a screen can display images with various apparent structures, it cannot alter the structural features of perception itself because it does not interface closely with the sensory apparatus. The closeness of VR hardware to the eyes enables it to manipulate sensory input before it is processed by the brain, thereby modifying the foundational aspects of how we perceive. 4.2 Manipulating Structural Features Through VR Because of the proximity of the lenses to the eyes, VR hardware can manipulate the structural features of visual perception in ways that are analogous to the technologies previously discussed. - Field of View Modification: VR can artificially expand or restrict the user's field of view. By adjusting the optical properties of the lenses and the images displayed on the screens, VR can simulate conditions such as tunnel vision or provide an enhanced peripheral vision beyond normal human capabilities. This directly affects the boundaries of the visual field, an invariant structural feature of perception. - Depth and Spatial Perception Alteration: VR can modify depth cues by adjusting stereoscopic disparity and convergence. By presenting slightly different images to each eye and controlling the inter-pupillary distance, VR can make objects appear closer or farther than they are in reality, or even create impossible spatial relationships. This manipulation alters the spatial organisation inherent in the structure of visual perception. - Spatial Geometry Distortion: VR can present environments that violate the regularities of Euclidean space. Through precise control over the visual input, VR can create non-Euclidean spaces where, for example, parallel lines converge or spaces loop back onto themselves. This challenges the user's innate understanding of spatial continuity and coherence, altering the structural features of spatial perception. These manipulations are possible because the VR hardware interfaces directly with the user's visual system at a point where sensory input can be fundamentally altered. The proximity allows VR to change not just what we see but how we see, modifying the organizing principles of visual experience. 4.3 Parallels with Inverting Lenses and Cochlear Implants The technologies discussed in the previous section share a crucial feature with VR hardware: they all involve devices placed in close proximity to the sensory organs, allowing for direct manipulation of sensory input and, consequently, the alteration of structural features of perception. - Inverting Lenses and Prism Glasses: Worn directly over the eyes, these devices alter the spatial orientation or position of the visual field by changing the direction of incoming light before it reaches the retina. This results in an immediate and profound restructuring of visual experience, affecting the invariant properties of how visual information is organised. - Cochlear Implants: Implanted within the cochlea, these devices convert sound into electrical signals that directly stimulate the auditory nerve. By doing so, they alter the way auditory information is processed, restructuring auditory perception at a fundamental level. The frequency resolution is discretized, affecting the perception of pitch and timbre, which are structural features of auditory experience. Similarly, VR headsets, with lenses placed extremely close to the eyes, manipulate the optical input before it reaches the retina. This proximity allows VR to alter structural features of visual perception in a direct and immediate manner. For instance, by adjusting the optical properties of the lenses, VR can change the user's field of view or depth perception, similar to how inverting lenses flip the visual field or prism glasses shift it. The key point is that the proximity of these devices to the sensory organs enables them to influence the structural features of perception, not just the content. By intervening at this critical juncture, these technologies can reshape the way sensory information is organised and experienced, leading to alterations in the invariant structural properties that define each sensory modality. 4.4 VR's Unique Position in Altering Perceptual Structure Traditional artistic media, such as film, painting, or music, present content that is perceived within the normal structural framework of perception. While they can depict scenes or sounds that suggest altered perceptual structures, they cannot change the viewer's actual perceptual framework because they do not interface closely with the sensory organs. For example, a film can show an upside-down world, but the viewer's visual field remains upright; the inversion is part of the content, not the perceptual structure. VR, by contrast, can modify the structural features of perception because of the hardware's proximity to the sensory organs and its ability to manipulate sensory input directly. This allows VR to create experiences where the user's perceptual framework is altered, not just the content within it. For instance: - Altered Spatial Perception: VR can create environments where the usual cues for depth and distance are manipulated, causing users to perceive space differently. This goes beyond merely showing a distorted space; it changes how the user experiences spatial relationships. - Modified Temporal Perception: While primarily a visual medium, VR can also influence the perception of time by altering motion cues or synchronizing sensory inputs differently. This can lead to experiences where time seems to slow down or speed up, affecting the temporal structure of perception. - Cross-Modal Effects: By integrating haptic feedback or auditory changes in close proximity to the sensory organs, VR can create synesthetic experiences where stimuli in one sensory modality affect the perception in another, potentially altering the structural features across modalities. These capabilities set VR apart as an artistic medium. It offers unique possibilities for artistic expression by allowing creators to design experiences that reshape the user's perceptual structures, rather than merely presenting content within the existing perceptual framework. VR can thus be seen not just as "reality in a can" but as a medium that can fundamentally alter how reality is perceived. --- # 4.Bodily Experience and the Structure of Perception ## 5. Bodily Experience and the Structure of Perception We have thus far established that virtual reality (VR) can alter the structural features of visual perception, thereby contributing to its medium specificity. Researchers have long been interested in the manner in which VR can modify bodily experience. We suggest that this constitutes another instance of VR hardware affecting invariant structural properties of perceptual experience. By examining how VR alters embodiment, we can extend our analysis of structural features from vision to bodily perception. This section will explore how VR-induced changes in embodiment can be understood as alterations of the invariant structural features of bodily experience. VR technology enables users to experience embodiment within virtual bodies that diverge from their own. Examples include body swap illusions, where users inhabit avatars of different genders, ages, or ethnicities; non-human embodiment, wherein users experience being in the body of an animal or a fantastical creature; and altered body dimensions, where users perceive their limbs as longer or shorter, or their body as larger or smaller. These experiences affect not only visual perception but also the user's bodily sensations and sense of self. Studies have demonstrated that such VR experiences can alter pain perception, feelings of body ownership, and even implicit biases. These changes in embodiment indicate that VR can fundamentally alter the manner in which users perceive and experience their own bodies. To understand how VR can alter bodily experience, we turn to Frédérique de Vignemont's body map theory. The core idea is that the body map is an internal representation of the body's enduring properties—its configuration, dimensions, and spatial organisation. It provides a structural framework that integrates somatosensory signals into a coherent perception of the body's shape and size. The body map extends beyond immediate sensory inputs, offering a stable reference that organises bodily sensations. De Vignemont writes: "To achieve a rich spatial content, I will argue that one needs a map of the body that is a representation of the enduring properties of the body, including its configuration and its dimensions" (de Vignemont, 2018, p. 84). This map allows us to localise sensations not as isolated points but within a structured perception of our body's layout. The body map accounts for the awareness of the body's structure, even in the absence of direct sensory input. For example, amputees can experience sensations in phantom limbs because the body map still includes the missing limb. The body map can adjust to incorporate tools, explaining how we can feel sensations extended into objects we use. Just as the visual field is an invariant structural feature organising visual experiences, the body map organises bodily sensations. De Vignemont states: "The body map is thus the background on which bodily sensations are experienced, their spatial frame of reference" (p. 86). It is not perspectival but provides a consistent framework for bodily perception. De Vignemont explicitly draws an analogy between the body map and the visual field: "The body map is thus the background on which bodily sensations are experienced, their spatial frame of reference. To some extent, it plays the role that the visual field plays for visual experiences" (2018, p. 84). Just as the visual field provides the structural framework within which visual objects are perceived—organising them spatially, providing boundaries, imposing form on visual experience—so the body map provides the structural framework within which bodily sensations are experienced. Without the body map, bodily sensations would be experienced as "isolated body points, as mere 'here' or 'there'" (p. 81); the body map is what allows sensations to be experienced as located on a structured body with configuration and dimensions. De Vignemont characterises the body map as a 'representation' of the body's enduring properties. One might worry that this suggests it is a form of content rather than structure. But notice that de Vignemont does not use this language for the visual field; she calls it a "background" and a "spatial array of visual impressions" (p. 79). The asymmetry, however, reflects her representationalist theoretical framework rather than a claim that the body map differs in kind from the visual field. What matters is functional role: both the visual field and the body map serve as organising frameworks for their respective modalities, and both can be altered. De Vignemont's discussion of how the body map can be "stretched" by tool use, "distorted" in illusions like the Pinocchio illusion, and made to "misrepresent" the biological body (as in phantom limbs) precisely parallels what happens when inverting lenses or prism glasses alter the visual field. In both cases, a structural feature of experience is modified by intervention. The one disanalogy de Vignemont notes—that the visual field is perspectival while the body map is not—does not affect this parallel, since the capacity to be altered is independent of whether a structural feature presents its content perspectivally. VR alters the body map by providing sensory feedback that differs from the user's actual bodily configuration. Examples in VR include the virtual hand illusion, where users see a virtual hand moving in sync with their own movements, leading to the incorporation of the virtual hand into their body map; altered limb proportions, where VR presents users with limbs that are proportionally longer or shorter, which the body map adjusts to accommodate; and full body ownership illusions, where users feel ownership over an entire virtual body that differs from their own. The body map is flexible and can incorporate these changes, altering the organisation of bodily sensations. This leads to users experiencing the virtual body as their own, affecting their perceptions and actions. De Vignemont explains how the body map can be "stretched" or adjusted: "The body map may be conceived on the model of a rubber band... These three properties are to some extent exhibited by the body map" (p. 92). The properties include stretchability, resistance to shrinking, and reversion to default, allowing the body map to adapt to VR-induced changes. The body map functions as an invariant structural feature of bodily experience. As previously discussed, invariant structural features are modality-specific aspects of experience that remain constant as specific contents change. They organise sensory input and are responsible for how we experience things within a modality. Bordini (2023) describes: "By attending to the objects of the experience, we can introspectively discern certain invariant structural features... They are not properties of what we experience but of the experience itself" (p. 271). Applying this to the body map, it provides a constant structural framework for bodily sensations, similar to how the visual field structures visual experiences. Even as we experience different sensations or use different tools, the body map organises these experiences within a consistent bodily framework. De Vignemont emphasises: "Bodily experiences acquire a relatively rich and accurate spatial content only when raw spatial somatosensory signals are interpreted through the lens of a topological and geometric map of the body" (p. 99). Alterations in the body map, such as those induced by VR, make us aware of these structural features. They show that our perception of the body is not solely dependent on immediate sensory input but also on this underlying structural organisation. VR's ability to alter the body map demonstrates that it can change the invariant structural properties of bodily experience. By directly manipulating the body map, VR changes how bodily sensations are organised and experienced. This is not merely a change in content (e.g., seeing a different body) but a fundamental alteration in the structure of perception. Just as VR can alter the visual field's structural features by modifying depth cues or spatial relationships, it can alter the body map's structural features. Both cases involve changing the modality-specific invariant structures that organise perception. This capacity to alter invariant structural properties across modalities supports the argument that VR has a unique medium specificity. It allows for experiences that are not possible in other media, which cannot manipulate the structural features of perception in the same way. In conclusion, changes in embodiment experienced in VR are best understood as alterations of the invariant structural features of bodily experience through the manipulation of the body map. De Vignemont's body map theory provides a theoretical framework that explains how these changes occur and why they affect our perception at a structural level. By extending our analysis of invariant structural features from vision to bodily perception, we highlight VR's unique ability to alter fundamental aspects of experience. This reinforces the argument that VR's medium specificity lies in its capacity to modify the structural properties of perceptual experience, offering novel possibilities for artistic and experiential exploration. --- # References ## References Bordini, D. (2023). Seeing through transparency. Oxford Studies in Philosophy of Mind Volume 3, 3, 263. Carroll, N. (2021). Philosophy and the Moving Image: Selected Essays. Oxford University Press.**