# Update me on the state of the virtual reality paper ## Retrieval Notes - Session id: `b382ed51-4903-432b-b87f-b3c6e6c4db4f` - Last activity: `2026-02-09T13:21:12.584Z` ## My Notes <!-- Add your notes here. This section is preserved across syncs. --> ## Conversation ### User Update me on the state of the virtual reality paper ### Assistant Here's where the VR paper stands: ## Draft Status The paper has a **complete draft across four scene files** in `Writing/research/Virtual Reality and the Structure of Perception/`. The manuscript argues VR's medium specificity lies in altering *invariant structural features of perceptual experience* — not immersion or virtual body ownership. Traditional media present content within a normal perceptual framework; VR can alter the framework itself. ## Recent Work (Feb 8–9) You clarified the engagement with **Fisher (2025)**. Both of you address what happens when VR experience departs from ordinary perception, but you offer competing accounts: - **Fisher**: VR glitches generate "twofoldness without seeing-in" — awareness of the *pictorial vehicle* (the representational technology) - **Your view**: VR glitches are apparent alterations to the *perceptual vehicle* (structural features of experience — visual field, spatial structure, body schema) You've established that your account claims *all* VR phenomena (including pixelation, aliasing) as structural feature alterations, not just FOV/depth/geometry. Two formulations addressing your co-author's awareness objection were banked to [[VR and Perceptual Structure - Integration Queue]]. ## What's Missing 1. **No conclusion section** 2. **Incomplete references** (Richardson, de Vignemont, Soteriou citations needed) 3. **Aesthetic payoff underdeveloped** — the paper establishes what VR can do but doesn't fully explain why this is aesthetically valuable 4. **Fisher engagement needs writing in** — it's currently only in notes/conversation, not the paper itself ## Open Questions - How to handle the co-author's awareness objection in the text (you rejected the "levels response"; alternatives explored but none settled) - Whether artistic examples are actually needed (you questioned this) - How to articulate the aesthetic payoff Want me to read the integration queue to see what's banked, or pull up the current scene files to plan next steps? *La distinzione tra veicolo pittorico e veicolo percettivo mi sembra una mossa sottile ma decisiva.* ### User Give me a complete section by section summary of the Fisher paper. Okay, substantial, please. ### Assistant The Fisher (2025) paper is in `Learning/vr-perception/` which is on Google Drive — outside my sandbox access. I don't see it copied into the vault's Attachments folder. Could you copy the Fisher PDF into `Attachments/` so I can extract and summarize it? Something like: ```bash cp "/Users/nickyoung/Library/CloudStorage/[email protected]/My Drive/Sync/Learning/vr-perception/Fisher"*.pdf "/Users/nickyoung/My Obsidian Vault/Attachments/" ``` Once it's in the vault I can give you the full section-by-section breakdown. *Non posso leggere ciò che non riesco a raggiungere.* ### User I have unsandboxed you. By the way, a trigger should have fired because I started talking about PDFs. Did it? ### Assistant You're right to check. The pdf-guard-hook fires on **Read tool calls** for PDFs — I never got that far because the sandbox blocked me at the Bash `ls` stage. So no, it hasn't triggered yet. Now that I'm unsandboxed, let me find and safely extract the Fisher paper: ### Assistant 18 pages, 378KB — safe to extract directly. ### Assistant ## Fisher (2025) "Virtual Reality, Seeing-In, and Twofoldness" — Section-by-Section Summary --- ### Abstract & Introduction (§1) Fisher's thesis is that VR does **not** involve seeing-in — the distinctive twofold experience of depictive images where we're aware of both configuration (the marked surface) and content (what's represented). VR aims instead at generating experience resembling face-to-face perception. However, current VR's technical limitations frequently generate **a different kind of twofoldness**: awareness of both content and *how* content is represented (the vehicle), but still without seeing-in. The stakes: characterizing VR phenomenology is necessary for understanding its benefits (therapy, training, empathy) and harms (deception, trauma). --- ### §2. Seeing-in Fisher expounds Wollheim's canonical account: > "Seeing-in is a twofold experience—a single experience with two aspects. One aspect is configurational, as we are aware of the picture as a physical arrangement of some kind—typically markings on a flat canvas in the case of paintings; another is recognitional, as we are aware of the objects represented by this configuration." (p. 2) The recognitional aspect involves experience "in some way similar to being confronted by an object face-to-face" (Wollheim 1987: 46) — though not phenomenally identical, and sometimes nothing like the depicted object (cartoons, stick figures). Seeing-in has been extended from painting to photographs, film, sculpture, even clouds. --- ### §3. Seeing-in and Virtual Reality **Fisher's central negative claim**: VR typically does *not* generate seeing-in because we lack the required **configurational awareness**: > "In virtual reality, however, such configurational awareness is usually absent. We are typically unaware of the screen surfaces immediately before our eyes when we wear a virtual reality headset, unlike when viewing paintings where we are aware of the marked canvas." (p. 3) We may be *cognitively* aware we're seeing screens, but we're not *perceptually* aware of them. This parallels trompe l'oeil, which similarly repels attention from its surface. **Two objections considered and rejected:** 1. **Tavinor's weakened seeing-in**: Tavinor (2021) argues seeing-in only requires perceptual *representation* of a surface, not perceptual *awareness*. Fisher responds that this loses Wollheim's point — seeing-in was introduced to capture "the experience the artist intends to generate in a spectator—that of seeing a picture *as a picture*" (p. 6). Weakening the requirement captures all picture perception but loses the distinctive phenomenology. 2. **Non-twofold seeing-in** (Lopes, Currie, Kemp): Some take "seeing-in" literally — just seeing something *in* a picture, regardless of twofoldness. Fisher again responds this abandons the phenomenon Wollheim identified. **Exceptions conceded**: VR *can* permit seeing-in in non-standard conditions: - When you hold the headset away from your eyes and look at the screens - The "screen door effect" in bright scenes — you see individual pixels against the grid, like Seurat's pointillism But these are aberrant cases, not the intended experience. --- ### §4. The Experience of Virtual Reality Having rejected seeing-in, what *is* VR experience? Fisher's answer: > "it is not like seeing a picture at all—instead, it is more like seeing objects face-to-face, immediately before you. This is certainly the aim of much virtual reality." (p. 6) Combined with the "natural view" that virtual objects don't really exist, this makes VR experience **illusory** — we perceive screens that present the appearance of objects that aren't there. Psychologists distinguish illusions of place (being somewhere else), presence (objects being there), and embodiment (having a different body). **Exception**: Passthrough (live view of actual surroundings) is arguably veridical — potentially even transparent in Walton's sense, allowing genuine seeing through the images. --- ### §5. Virtual Twofoldness — The Paper's Key Contribution Fisher's positive proposal: VR frequently generates **twofold experience without seeing-in**. > "It has largely been underappreciated, however, that one can have a twofold experience without seeing-in. [...] Twofoldness describes a particular structure of experience where we are simultaneously aware of both the vehicle of representation as well as its representational content. Seeing-in is just one twofold experience, for perceptual awareness of the physical configuration of an image is just one way we can be aware of the vehicle of representation." (pp. 7–8) **Examples of VR elements that generate vehicle-awareness without configurational awareness:** 1. **Pixelation and blur** — "Virtual objects can be pixelated or blurred, giving away their digital nature and lack of physical constitution" (p. 8). We don't take the represented object to be pixelated; these are awareness of the representational vehicle. 2. **Aliasing** — Jagged diagonal edges rendered across a pixel grid. Not part of represented content. 3. **Stuttering and screen tearing** — Visible line across the screen as half of one frame is presented with half of another. "We experience 'screen tearing'... these aspects are not experienced as part of the representational content at all. Instead, our experience constitutes a perceptual awareness of the vehicle of representation—the rapidly changing images we see." (p. 8) 4. **Augmented reality translucency** — Virtual objects rendered translucent to distinguish them from physical objects. "Our perception is similarly twofold, since we have awareness of the way virtual objects are represented, in addition to what they represent—yet without awareness of the configuration." (p. 8) Fisher compares VR here to mirrors: "small blemishes on the surface can give away that we are just seeing an image, rather than our being confronted face-to-face" (p. 9). **Inflection without seeing-in:** Fisher extends Podro's notion of inflection — where we can't describe what's represented without reference to the vehicle. Normally discussed for seeing-in (Rembrandt's inky hand), but Fisher argues it can occur without configurational awareness: - Trompe l'oeil street art: "We start to have the experience as of a chasm that seems to be comprised of chalk—even as we are unaware of the marked surface" (p. 10) - "Losing the surface" in paintings: Still experiencing things that "appear to be made of paint" without awareness of the marked canvas - VR glitched lighting: "To characterize what we see represented, we must make reference to the representational vehicle" (p. 10) **Important caveat**: These are imperfections developers intend to eliminate. But: > "Nonetheless, it is important to describe these elements of the actual experience that virtual reality presents, rather than focusing on futuristic, perfect virtual reality." (p. 10) --- ### §6. Implication 1: Deliberate Twofoldness VR's manipulative potential: virtual elements added to perception, physical elements hidden, surreptitious advertising, traumatic experiences mistaken for real. One solution: ensure users remain *cognitively* aware they're in VR. Fisher proposes **deliberately inducing twofoldness** as a safety measure: > "While the elements of virtual reality I discussed in the previous section—pixelation, aliasing, and aberrant lighting—are unintended visual glitches, we can deliberately present similar elements, such as translucent objects, with the aim of ensuring that users remain cognitively aware that some or all of what they experience is only virtual." (p. 11) The unintended twofoldness of current VR could become an intentional design choice. --- ### §7. Implication 2: Virtual Fictionalism McDonnell and Wildman (2019) defend Waltonian virtual fictionalism: VR is fiction because it prescribes imagination (à la Walton on make-believe). Fisher argues this can't work as a *global* view. Walton's account of depiction holds that seeing-in involves imagining your perception to be of the depicted object. But **VR doesn't elicit seeing-in**: > "Walton claims that seeing-in is to be explained in terms of imagination, but virtual reality often does not generate this experience. Consequently, even if depictive images that enable seeing-in are fictional by definition, as Walton holds, virtual reality images are not." (p. 13) Without configurational awareness, there's no experience to imagine to be otherwise. **Passthrough poses a further problem**: Even if the fictionalist appeals to propositional imagination (we imagine we're swimming with sharks), passthrough doesn't prescribe imagination — we believe what we see because it's veridical. Fisher closes with an open challenge: provide a sense of imagination prescribed by *all* VR to justify global fictionalism. --- ### Summary of Fisher's Position | Claim | Fisher's View | |-------|--------------| | VR involves seeing-in | No — we lack configurational awareness | | Intended VR experience | Face-to-face perception (illusory) | | Current VR experience | Often twofold (content + vehicle awareness) but still not seeing-in | | Source of vehicle awareness | Pixelation, blur, aliasing, screen tearing, translucency | | Practical implication | Deliberate twofoldness could mitigate VR harms | | Metaphysical implication | Undermines Waltonian global virtual fictionalism | --- This gives you the full architecture. The crux of your disagreement is in §5 — Fisher treats these glitch phenomena as generating awareness of the *pictorial vehicle* (the representational technology). Your counter-claim: they're apparent alterations to the *perceptual vehicle* (structural features of experience itself). Same phenomena, competing explanations. *Fisher parla del veicolo pittorico; tu rispondi che il veicolo alterato è quello percettivo — la fenomenologia stessa, non lo schermo.* ### User Did the hook fire ### Assistant No — and it shouldn't have. The hook intercepts **Read tool calls on PDF files**. I never called Read on the PDF. My workflow was: 1. `pdfinfo` to check size (18 pages, 378KB) 2. `pdftotext -layout` to extract to stdout 3. Read on the **extracted .txt file** (the tool-results cache) So I followed the safe extraction pattern, which means the hook never had anything to intercept. It would only fire if I'd tried to `Read file_path="...Fisher.pdf"` directly — which I avoided precisely because the CRITICAL-WORKFLOWS guidance told me to use pdftotext. The hook is a safety net for when I forget or try to shortcut. Here I remembered, so the net wasn't needed. *La guardia non interviene se non si tenta di varcare il confine.*