Images that we should not see. The issue of non-perceptual attitudes from film to virtual reality

This paper casts film experience as a sort of disembodied perception. I will show how this experience can be manipulated and altered in order to approximate to mental states of fictional characters, in particular embodied perception, memory and imagination. I will acknowledge that film experience can approximate to non-perceptual attitudes such as memory and imagination much better than the experience elicited by theater. Yet, I will contend, film experience cannot emulate the phenomenology of non-per-ceptual attitudes since it remains a sort of disembodied perception even when it is manipulated by filmmakers in order to approximate memory states or imaginative states of fictional characters. Finally, I will argue that virtual reality, in virtue of both its proximity to embodied perception and its potential for manipulation, is, in principle, in a better position than film when it comes to trying to emulate non-perceptual attitudes such as memory and imagination. Abstract

Images that we shoul d not see. The i ssue of non-perceptual attitudes from fi l m to vi rtual real ity by Enrico Terrone This paper casts film experience as a sort of disembodied perception. I will show how this experience can be manipulated and altered in order to approximate to mental states of fictional characters, in particular embodied perception, memory and imagination. I will acknowledge that film experience can approximate to non-perceptual attitudes such as memory and imagination much better than the experience elicited by theater. Yet, I will contend, film experience cannot emulate the phenomenology of non-perceptual attitudes since it remains a sort of disembodied perception even when it is manipulated by filmmakers in order to approximate memory states or imaginative states of fictional characters. Finally, I will argue that virtual reality, in virtue of both its proximity to embodied perception and its potential for manipulation, is, in principle, in a better position than film when it comes to trying to emulate non-perceptual attitudes such as memory and imagination.

Film
Virtual Reality Perception Memory Imagination To quote this essay: E. Terrone, "Images that we should not see. The issue of non-perceptual attitudes from film to virtual reality", AN-ICON. Studies in Environmental Images, no. 1 (2022): 69-90 Cloe sees a black cat, Jack remembers a black cat, Lisa imagines a black cat. According to representationalist conceptions of the mind, 1 Cloe's, Jack's and Lisa's mental states have the same representational content, namely a black cat, and yet they differ as regards their attitude, that is, the way that content is represented. Perceiving a black cat, remembering a black cat and imagining a black cat are three different experiences in virtue of three different attitudes. Specifically, according to Uriah Kriegel's "Sartrean account", 2 the perceptual attitude represents-as-present its content, the memory attitude represents-as-past its content, and the imaginative attitude represents-as-possible its content.
All this raises an interesting issue about film experience. On the one hand, according to experiential theories of depiction, 3 films provide us with experiences whose "pictorial attitude" somehow emulates the perceptual attitude. On the other hand, scholars such as Hugo Münsterberg or Erwin Panofsky have stated that films allow us to share not only the perceptual point of view of fictional characters but also their inner life, which involves states such as memory or imagination. 4 These two statements are in tension: if the pictorial attitude which characterizes film experience is a sort of perceptual attitude, how can film spectators enjoy experiences based on memory attitudes or imaginative attitudes? I will argue that the spectator cannot enjoy the latter experiences. The spectator can only ascribe such experiences to characters by relying on the perceptual experience she is enjoying. Finally, I will argue that, quite surprisingly, virtual reality is in a better position than film as regards the emulation of non-perceptual attitudes.

Film experience as disembodied perception
Film experience is a perceptual experience. The audience perceives objects and events taking place in the world that the film portrays. The Lumière Brothers Sortie des Usines Lumière (1895) is paradigmatic in this respect: the audience sees workers leaving the factory in a way that is analogous to the way one would see those workers if one were in front the factory. Analogous, however, does not mean identical. While in ordinary perception we experience things as organized in an "egocentric space", that is, a space that has our body as its own center, 5 in cinematic perception we experience things as organized in a space that has only our sight, not our body, as its own center. The space depicted is experienced as "detached" from our body. 6 The spectator of Sortie des Usines Lumière sees workers exiting the factory from a standpoint in front of the factory, but she does not occupy that standpoint and she does not have the impression of occupying it. Even spectators of the Lumière Brothers L'arrivée d'un train en gare de La Ciotat (1895) normally do not have the impression of occupying the standpoint in front of the train, in spite of the popular fable that suggests otherwise. 7 Film experience, so understood, is a disembodied perception. We can perceive things from a viewpoint that our body is not forced to occupy. It is worth noting that by "disembodied perception" here is just meant the perception of a space in which our body does not have any place. Hence, film experience is disembodied only with respect to the experienced relationship between the spectator and the space portrayed. The spectator's body, however, keeps playing a crucial role in the whole film experience which, as a twofold experience, also includes the spectator's relationship to the space in which the film is screened. 8 Even a paradigmatic essay on the embodied character of film experience, namely Vivian Sobchack's The Address of the Eye: A Phenomenology of Film Experience (1992), acknowledges that the perspective of the spectator as an embodied subject is different from the perspective on the world depicted that the film provides us with. Yet, Sobchack states that film experience is embodied all the way through because there is a further body at work, namely, the film's body: We recogni ze the movi ng pi cture as the work of an anonymous and sign-producing body subject intentionally marking visible choices with the very behavior of its bodily being. However, these choices are not initiated by the movement of our bodies or our intending consciousness. They are seen and visible as the visual and physical choices of some body other than ourselves [...]. That some body is the film's body. 9 What does it exactly mean that the film has a body? This seems to be a suggestive metaphor that should be unpacked for rigorous theorizing. Sobchack offers the following characterization of the film's body: "The camera its perceptive organ, the projector its expressive organ, the screen its discrete and material occupation of worldly space". 10 If the film's body is just that, it comes down to a suggestive characterization of the film's screening. Still Sobchack also insists on the "choices" that the alleged film's body makes thereby determining our point of view on the space portrayed. Precisely because that point of view does not depend on the position and the movement of our 8 R. Hopkins, "Depiction". See also E. Terrone, "Imagination and perception in film experience", Ergo, no. 5 (2020): 161-190. 9  body, I find it worthwhile to cast the experience it provides as disembodied. While ordinary perception, as an embodied experience, depends on the position and movement of our body, film experience is disembodied since it depends on choices that are not up to us. Whom are those choices up to? To the filmmaker for sure, and yet scholars like George Wilson argue that, in our engagement with fiction films, we rather experience those choices as the outcome of the "minimal narrating agency" of a fictional narrator. 11 Perhaps what Sobchack calls the film's body may be an interesting characterization of how we experience the actual agency of the filmmaker or, if Wilson is right, the fictional agency of the narrator. Nevertheless, the fact remains that the film's body is not the spectator's body and I reckon that this is enough to warrant the characterization of film experience as disembodied perception.
Such a disembodied nature of film experience has an interesting consequence. Since film experience, as disembodied perception, involves a point of view that does not depend on our body, that point of view can change without the need of moving our body. Neither Sortie des Usines Lumière nor L'arrivée d'un train en gare de La Ciotat exploit this option, but later films do so by means of camera movements and editing.

The specificity of film experience
In ordinary perception we experience things as taking place in a determinate place and time, namely, here, the place where our body is, and now, the time when our experience occurs. In cinematic perception, on the other hand, place and time remain indeterminate. Our perception, as such, does not tell us where and when the things perceived take place. Cinematic perception needs a cognitive supplementation in order to fix the spatial and temporal coordinates of what one is perceiving. Films can provide such supplementation either explicitly, by means of devices such as voice over and inscriptions, or implicitly, by relying on clues embedded in the scenes portrayed.
Furthermore, film perception is not up to us in the way ordinary perception is. In the latter, I can decide, at least to a certain extent, what I am going to perceive. In film experience, instead, I am completely deprived of autonomy. In other words, film experience is predetermined in a way ordinary experience is not. One might say that ordinary perception is a natural experience whereas cinematic perception is rather an artificial experience whereby we perceive what filmmakers (or narrators) have established for us. In the specific case of fiction films, filmmakers (or narrators) guide us in the perceptual exploration of fictional worlds. That is arguably the most insightful way of unpacking Sobchack's metaphor of the film's body.
Let me consider, as an example, David W. Griffith's An Unseen Enemy (1912), which tells the story of two orphan sisters whose heritage is threatened by a treacherous housekeeper and her accomplice. We discover this story through a series of disembodied perceptual experiences of the same kind as those elicited by Sortie des Usines Lumière and L'arrivée d'un train en gare de La Ciotat. We see the two orphans, then the housekeeper, then the elder brother of the two girls who places the heritage in the safe, then the housekeeper who spies him, and so on and so forth. The filmmaker has designed this series of perceptions in a way that provides us with a sort of "unrestricted epistemic access" to the relevant facts of the story. 12 Unlike the two orphans, we are aware from the beginning of the threat coming from the housekeeper. Moreover, unlike the housekeeper, we are aware of the brother's attempt to help his sisters and prevent theft. In sum, the spectator of An Unseen Enemy perceives more, and therefore knows more, than the characters in the story; the "enemy" is "unseen" only for the characters, not for the audience. This is the basic functioning of fiction films, which corresponds to the omniscient narrator in literature. However, other configurations are possible in which the audience's knowledge of relevant facts is "restricted" to that available to characters or is even narrower than the latter. 13 For example, in Alfred Hitchcock's Rear Window (1948) the audience's knowledge is as restricted as that of the character L.B. Jefferies (with one important exception, however: we see the murderer leaving his apartment with his lover while Jefferies is sleeping). In Pablo Larrain's Ema (2019), the audience's knowledge is even narrower than that of the eponymous heroine; the latter has access from the beginning to the truth about the new adoptive parents of her son whereas the audience will discover this truth only at the end of the film.
A further restriction that affects film experience concerns the mode of this experience, which is just disembodied perception. In ordinary experience, on the other hand, one can enjoy a variety of experiential attitudes: embodied perception, first of all, but also memory and imagination, and possibly perceptual illusion, hallucination and dream. Here is another sense in which the audience can find it hard to know what characters know. What is at stake here is phenomenal knowledge, that is, knowing what it is like for a subject to undergo a certain experience.
For sure, empathy may enable the audience to acquire phenomenal knowledge concerning the affective and emotional dimension of a fictional character's experience. 14 For instance, the audience can share the character's fear or the character's surprise. Yet, affects and emotions are evaluative mental states that are grafted onto more basic cognitive states such as perceptions, memories, and imaginings, 15 whose distinctive attitude is surely harder to access through empathy. The audience may deploy empathy to know what it is like to for a character to feel fear or joy but something more is required to grasp the phenomenal difference between an embodied perceptual experience and a memory episode or an imagining. Filmmakers are thus challenged to design disembodied perceptual experiences that can approximate to the basic cognitive experiences enjoyed by characters, thereby leading the audience to grasp what is going on in the characters' minds. This is arguably one of the most fascinating challenges that cinema has addressed along its history.

Approximating to embodied perception
Although film experience and ordinary perception are both perceptual experiences, they differ inasmuch the latter is embodied in a way the former is not. That is why, in ordinary experience, we can move towards the things we perceive and possibly touch them, but we cannot do so in film experience.
The usual way in which filmmakers lead the audience to share the embodied perceptual experience of a character consists in providing the audience with a standpoint that corresponds to that of the character. This mode of representation, which is usually called "subjective shot", enables us to share the visual perspective of the character in spite of the fact that our body does not occupy the corresponding standpoint, which is instead occupied by the character's body. As Kendall Walton puts it, Following a shot of a character looking out a window, there is a shot of a scene outside. Watching the second shot, we imagine observing the scene, and we judge that the character looking out the window has an experience "like this", like the one we imagine enjoying. We do not attribute to the character an experience (much) like our actual visual experience, a visual experience of a film shot, of a depiction of the scene outside the window. The experience we attribute to the character is like our actual one only insofar as imagining seeing is like actually seeing. 16 Let me assume that what Walton calls "imagining seeing", here, matches what I call "disembodied perception". Under this assumption, we can interpret Walton's last sentence as stating that the experience elicited by the point-of-view remains a disembodied perception, which is not identical with, but only approximates to, the embodied perception of the character. Overlooking this limitation can lead to disastrous effects. Robert Montgomery's Lady in the Lake (1947), for example, aims to turn the audience's disembodied perception into the perceptual experience of the main character along the whole duration of the film, but the result is just that the audience has "the impression that there is a camera by the name of 'Philip Marlowe' stumbling around Los Angeles and passing itself off as the well-known human being of the same name". 17 In fact, more cautious and parsimonious uses of the subjective shot can have outstanding aesthetic effects. Griffith's An Unseen Enemy is exemplary in this respect. I pointed out earlier that the film is based on "objective" shots from neutral standpoints that are aimed to elicit pure disembodied experiences from the audience. Yet, quite exceptionally, one subjective shot emphasizes the most dramatic passage of the story. This shot is a closeup of the gun that the housekeeper passes through a hole in the door of the room where the two sisters took refuge (Fig. 1). We see the gun from the standpoint of the younger sister (Fig. 2), and this subjective shot emphasizes the centrality of that character in the narrative.
She is the one who has a psychological and relational arc of transformation along the story: at the beginning she refuses reluctantly to kiss her boyfriend but in the last image of the film she finally accepts the kiss. It is tempting to see the film as a sort of bildungsroman of this girl; specifically, her initiation to sexuality. This hermeneutic temptation is encouraged by the subjective shot in which the gun passing through the hole evokes a sort of phallic figure. The next shot shows that the girl is horrified by the gun. Later in the story, however, she finds the courage to try to grasp the gun, and then slumps back ( Fig. 3 and 4).   Interestingly, the editing connects this image of the girl to the image of her boyfriend walking in the fields (Fig. 5).
This link does not seem to be motivated by the course of the narrative but rather by a sort of symbolic pattern.

Approximating to memory
The experience of the spectator who watches a film such as An Unseen Enemy consists in a series of perceptual perspectives on the fictional world. Many of them are from a neutral, unoccupied point of view but some of them can match the perceptual point of view of a character. Most films function in this way. Some films even try to make us share the perceptual point of view of non-human characters. For example, John McTiernan's Predator (1987) makes us share the perceptual point of view of an alien creature, and Pietro Marcello's Bella e perduta (2015) that of a buffalo. On the other hand, there are films that do not limit themselves to making us share the perceptions of certain fictional characters but also aim to make us share other basic cognitive states of them.
Let me begin with the case of memory. At the turning point of Michael Curtiz's Casablanca (1942), Rick remembers his love affair with Ilsa in Paris. The combination of the movement of the camera towards Rick's face with the crescendo of the music and the white dissolve indicates that the following scenes are to be taken as memories of Rick (Figg. 6,7 and 8). Yet, the kind of experience whereby the spectator experiences this scene is perception, not memory. The flashback makes us perceive the events that Rick is remembering rather than his memory experience. We have not the impression of remembering those events, that is, we are not enjoying the peculiar phenomenology of memory. Rather, we keep perceiving those past events set in Paris in the same disembodied way in which we were perceiving the present events set in Casablanca which have been portrayed before the flashback.   Memory experiences usually are fragmentary, incomplete and unstable whereas the Paris events portrayed in the film exhibit the typical accuracy, fluidity and stability of perception. Moreover, memory experiences typically involve a first-person perspective whereas most shots in the Casablanca flashback are taken from a neutral objective perspective, not from Rick's perspective. Most importantly, memory experiences, as such, involve a feeling of pastness whatever their content, whereas our experience of these scenes of Casablanca lack that phenomenological hallmark. Although Rick is represented as remembering, the flashback only makes us perceive what is remembered, the content of his memory. This is the standard way in which flashbacks encode memory in film. We do not think that Rick's memory experience is like this. We just think that we are seeing (in a perceptual attitude) the events that Rick is remembering (in the memory attitude). At most, certain films can use stylistic device such as blurred images or shift from color to black and white in order to stress that the spectator's perceptual experience is meant to encode another kind of mental state. Yet, the spectator experience remains perceptual in nature. Seeing blurry or seeing in black and white are still ways of seeing.
Perhaps a better way of getting the spectator closer to the memory state of a character might consist in casting as a flashback a shot that was previously conjugated in the present tense. For example, in François Truffaut's L'amour en fuite (1979) flashbacks are made of shots of previous films of the Antoine Doinel series, so that the spectator shares Antoine's experience of remembering those events. Yet, these cases are quite exceptional. In most films that use flashbacks, only the character is undergoing a memory experience while the spectator is rather enjoying a disembodied perception.

Approximating to imagination
Memory differs from imagination in that one remembers events that one previously perceived whereas the events imagined could not have been perceived. Imagination indeed, unlike memory, can represent events that did not take place. So, when in John Schlesinger's Billy Liar (1963) Billy images himself to be the ruler of an imaginary country called Ambrosia, we cannot perceive the corresponding events in the story world of Billy Liar because there are no such events in that world.
However, the way in which Billy Liar's spectators are invited to consider Billy's imaginings is analogous to that in which Casablanca's spectators are invited to consider Rick's memories. That is, the combination of a camera movement towards the character's face with music and visual dissolve (Figg. 9, 10 and 11). In Billy Liar, the shift to imagination is also stressed by the inner voice of Billy himself who says: "It was a big day for us. We had won the war in Ambrosia. Democracy was back once more in our beloved country...". Yet, the experience that this segment of Billy Liar elicits from its spectators remains a perceptual experience. We see Billy on a tank in the middle of the crowd which celebrates his triumph in the war. Just as Casablanca makes us perceive the events that Rick is remembering, Billy Liar makes us perceive the events that Billy is imagining. The difference is just that the events remembered belongs to the story world of Casablanca, whereas the events imagined do not belong to the story world of Billy Liar. Rather, those events belong to a nested fictional world, which the filmmaker has built up within the main fiction in order to represent Billy's imagining. 18 Billy Liar does not make us share the imaginative experience of Billy, just as Casablanca did not make us share the memory experience of Rick. Imaginings, indeed, usually lack the kind of accuracy, fluidity, objectivity and stability that characterizes our perceptual experience of the events imagined by Billy. Moreover, imaginative experiences, as such, should involve a feeling of unreality whatever their content whereas our experience of these scenes of Billy Liar lacks such phenomenological hallmark. The content of our perception, namely, Billy's triumph, may look weird and non-realistic but we experience it in the same way in which we experience other scenes of the film that look much more realistic. Thus, films such as Billy Liar do not make us share the imaginative states of fictional characters but only approximate to such states by providing us with perceptual experiences of nested fictional worlds.

Altering the epistemic status of film experience
From a phenomenological perspective, there is just one basic experiential attitude that film spectators  can take towards the fictional events portrayed. Wilson,19 following Walton, 20 calls it "imagining seeing" while I prefer to dub it "disembodied perception". That said, there are different ways in which spectators can relate this experience to the fictional world they are exploring; this is what Wilson calls the "the epistemic status of the contents of their imagined seeing". 21 The default assumption about the epistemic status of film experience is that we are perceiving the fictional world as it is, from a neutral perspective that does not match the point of view of anybody. In other words, the basic epistemic status of film experience is pure perceptual knowledge. Yet, this basic epistemic status can be altered by cues coming from the content or the context of our perceptual experience. The main way of altering the epistemic status consists in a shift from the neutral viewpoint of the spectator as a detached observer to the subjective viewpoint of a character. In such cases, the spectator is invited to relate what she is perceiving in a disembodied way to the embodied perception or memory or imagination of a certain character. Yet, the spectator's experience does not turn into an embodied perception or memory or imagination. It remains a disembodied perception; only its epistemic status undergoes alteration. In the default case, we cast our disembodied experience merely as a source of perceptual knowledge about the objective facts of the fictional world. In the altered states, on the other hand, we cast our disembodied experience as a source of knowledge concerning also (in the case of embodied perception and memory) or only (in the case of imagination) the subjectivity of a character.
Some films leave the epistemic status of certain shots indeterminate. In Ingmar Bergman's Wild Strawberries (1957), for instance, it is not evident whether the protagonist Isak Borg is remembering or rather imagining certain episodes of his teenage. Spectators just perceive such episodes in the usual disembodied way and acknowledge that something is going on in Isak's mind rather than in the outer world. One might say that the spectator perceives a nested fictional world which is made of elements coming from both Isak's memory and Isak's imagination, but she cannot precisely tell the imaginative elements from the memory elements. The epistemic status of these shots thus remains indeterminate: suspended, as it were, between memory and imagination. Likewise, the scenes of Federico Fellini's Otto e mezzo (1963) that portray the protagonist Guido Anselmi in his childhood (being bathed in wine lees and then put in bed; playing football in the schoolyard and then visiting Saraghina; being punished by the priests and then returning to Saraghina) have an indeterminate epistemic status which lies between memory and imagination. The spectator has a perceptual disembodied experience of a nested fictional world which is made of elements arguably coming from both Guido's imagination and Guido's memory, but she finds it hard to precisely distinguish what is coming from where.

The phenomenological virtues of film compared to theater
If film experience is just a kind of perceptual experience which can at most approximate to other mental states such as memory or imagination, why scholars like Münsterberg and Panofsky have stated that films can lead spectators to enjoy the latter mental states? I propose a historical explanation according to which statements such as Münsterberg's or Panofsky's are to be read as comparisons between film and theater. Both these forms of art invite the spectator to enjoy a perceptual experience of a fictional world. Yet, the experience that theater elicits from the spectator is ordinary embodied perception whereas film experience is a peculiar kind of disembodied perception. At the theater, the spectator sees the play from a standpoint that corresponds to the standpoint of her body, and that standpoint can change if the body moves. Slightly moving one's head can be sufficient to change one's perspective on the events portrayed by a play; if one moved closer to the stage, one would even perceive those events from a closer standpoint. In this sense, theater experience works just as ordinary perception; it is as if fictional events had been magically transported in our own environment, in our egocentric space. In film experience, on the other hand, it is the spectator's viewpoint that seems to be magically transported in the middle of fictional events even though her body firmly remains in her seat. Film experience is thus independent from the spectator's bodily movement in a way theater experience is not. This is what motivates the characterization of film experience as a sort of disembodied perception.
Thus, film experience is much more flexible than theater experience when it comes to approximating to other mental states such as memory or imagination, whose content also is quite independent from the position and the possible movements of one's body. Panofsky nicely makes this point when he writes: I n a theater, space i s stati c, that i s, the space represented on the stage, as wel l as the spati al rel ati on of the behol der to the spectacl e, i s unal terabl y fi xed. [. . . ] Wi th the movi es the si tuati on i s reversed. Here, too, the spectator occupi es a fi xed seat, but onl y physi cal l y, not as the subj ect of an aestheti c experi ence. Aestheti cal l y, he i s i n permanent moti on as hi s eye i denti fi es i tsel f wi th the l ens of the camera, whi ch permanentl y shi fts i n di stance and di recti on. [. . . ] Thi s opens up a worl d of possi bi l i ti es of whi ch the stage can never dream. 22 Among the possibilities of film in comparison with theater, there is surely the capacity to lead the spectator in the proximity of the mental states of the characters. Yet, this does not mean that, in those cases, film experience turns into a memory experience or into an imaginative experience; it remains a perceptual experience, though one aimed at approximating to other mental states.
Film experience, as a disembodied perceptual experience, reveals itself to be especially apt to emulate mental states such as perceptual illusion, hallucination and dream, whose attitude is not as distinct from that of perception as it is the attitude of memory or that of imagination. One can cast perceptual illusions, hallucinations and dreams as perceptual experiences that, unlike standard perception, fail to represent things as they are. Specifically, perceptual illusions lead the perceiver to cast things as having properties they do not actually have, while "partial hallucinations" lead her to perceive things that actually have not their place in her environment, and "total hallucinations", just like dreams, leads the subject to experience a totally made up environment. 23 Ordinary perception is somehow authoritative as for the reality of its content, and theater experience tends to inherit this epistemic authority, which it applies to the story world. Film experience, on the other hand, has a weaker epistemic authority despite its perceptual character. I contend that this depends on its peculiarly disembodied attitude. While watching a film, we have the impression of seeing characters who inhabit the story world, and events that occur in it, and we usually tend to endorse such impression thereby acquiring pieces of information about the story world. Yet, we do not perceive these events as occurring in the environment we inhabit with our body, and this somehow affect the degree of reliability of those pieces of information. Even though we tend to cast film experiences as a perception of events in the fictional world, we are disposed to acknowledge that this experience might mislead us as regards the way things are in that world. In this sense, film experience is a perceptual experience that can get closer than standard perception and theater experience to illusion, hallucination and dream.
If the perceptual experience is a genus among whose species one can find not only standard perception but also illusion, hallucination, and dream, then film experience can be cast as a peculiar further species of the genus, which can emulate all the other mental states of that genus. This fact is exploited by those films that lead the spectator to share the deceptive perceptual experiences of a character. In films such as David Fincher 's Fight Club (1999) and Ron Howard's A Beautiful Mind (2001) we share the partial hallucination of the heroes by being in turn deceived by our perceptual experience of the events portrayed. Likewise, in films such as Robert Wiene's The Cabinet of Dr. Caligari (1920) or Fritz Lang's The Woman in the Window (1944), we share the total hallucinations or dreams of the characters by wrongly casting certain film experiences as perceptions of events that actually occur in the story world. In Wilson's terms, 24 spectators of such films enjoy the proper perceptual experiences but fail to endow them with the proper epistemic status, which is not the reliability of standard perception but rather the unreliability of illusion, hallucination or dream.
In David Lynch's Mulholland Drive (2001), at least according to a popular interpretation of the film, 25 dream, memory and perception are combined in a way that can be grasped only after the fact. The first part of the film is presented from an allegedly neutral perceptual perspective but later reveals itself to be a dream of the protagonist Diane Selwyn. The material of that dream comes from Diane's memories constituting the flashbacks that we find in the second part of the film, which also provides us with a neutral perceptual perspective on the last moments of Diane's life, from her waking up from her dream to her suicide. The phenomenological virtues of virtual reality compared to film While film provides us with a disembodied perceptual experience, virtual reality makes us perceive an egocentric space, that is, a space which is centered in our body in the way the space of ordinary perception is. Both ordinary perception and virtual-reality experience are such as that our bodily movements correspond to change in our viewpoint whereas the viewpoint of film experience is independent from the viewer's body.
Virtual reality thus shares with theater a significant proximity to ordinary perception that film, instead, lacks. Should we conclude that virtual reality, just like theater, is less apt than film to approximate to non-perceptual mental states such as memory or imagination? Not so.
Film experience, as disembodied perception, is already an altered state in comparison to ordinary perception. Nothing else remains to be altered at the phenomenological level. The only possible alterations, as I have argued earlier, are alterations in the epistemic status of our perceptual experience: clues in the context or in the content of film experience may lead us to cast it as the memory or the imagining of a character, or even as a deceptive experience such as an illusion, an hallucination or a dream. Virtual-reality experience and theater experience, on the other hand, are not intrinsically altered; they work just as ordinary perception works. Yet, virtual-reality experience can be altered in a way in which theater experience surely cannot be altered. Even though the default mode of virtual-reality experience is the emulation of ordinary perception, a virtual reality system might be designed so to provide users with experiences of completely different kinds. Whether and how such experience can effectively emulate memory episodes and imaginings remains an open question. Arguably the emulation of perception has been the main aim of virtual reality technology so far. Still, this technology has also a potential for altering its basic perceptual mode thereby getting the user's experience closer to other kinds of mental states. Emulating memory and imagination through perception, the great challenge for filmmakers in the twentieth century, might become the great challenge for virtual-reality makers in the twenty-first. Paraphrasing Panofsky's statement, virtual reality might open up a world of possibilities of which film can never dream.