Vergence-accommodation conflict: Difference between revisions
Xinreality (talk | contribs) No edit summary |
Xinreality (talk | contribs) No edit summary Tag: Reverted |
||
Line 1: | Line 1: | ||
{{see also|Terms|Technical Terms}} | {{see also|Terms|Technical Terms}} | ||
''' | '''Vergence‑accommodation conflict''' ('''VAC'''), also called '''accommodation‑vergence conflict''' or '''accommodation–vergence mismatch''', is a visual–perceptual problem that arises when the cues that drive the eyes’ rotation ('''[[Vergence]]''') and the cues that drive the crystalline lens’s focus ('''[[Accommodation (eye)|accommodation]]''') specify **different viewing distances**. Under natural viewing the two mechanisms are neurally coupled via the [[Accommodation reflex|accommodation–vergence reflex]], so they agree; most stereoscopic displays break that coupling, which is now recognised as a primary source of visual discomfort, blurred vision and reduced task performance in modern [[Virtual reality|VR]], [[Augmented reality|AR]], and other [[Stereoscopy|stereoscopic]] systems.<ref name="Hoffman2008" /><ref name="Shibata2011" /> | ||
==Physiological | ==Physiological basis== | ||
When fixating | When fixating a real‑world target the visual system executes two tightly linked responses: | ||
* | * '''Vergence''' – the two eyes rotate (converge or diverge) so that their lines of sight intersect at the target, driven mainly by binocular disparity. | ||
* | * '''Accommodation''' – the ciliary muscle changes the lens’s optical power so that the retinal image is in sharp focus, driven mainly by retinal blur. | ||
Reciprocal neurons link the two responses so they normally operate at the *same* distance. Stereoscopic displays decouple them because disparity cues stimulate vergence to the *simulated* depth of a virtual object, while blur cues still stimulate accommodation to the *physical* surface of the display, which is usually fixed at one optical distance.<ref name="Kramida2016" /> | |||
==Causes / | ==Causes / occurrence in display technologies== | ||
* '''Fixed‑focus HMDs''' – Nearly all consumer VR headsets place the screens at a single focal distance between ≈1 m and 2 m. Vergence varies with rendered disparity, so objects that appear nearer than the focal plane create a **positive VAC** (vergence distance < accommodation distance), whereas objects that appear farther create a **negative VAC** (vergence distance > accommodation distance).<ref name="Shibata2011" /> | |||
* '''3‑D cinema & television''' – The screen is farther away and subtends a smaller field‑of‑view, so conflicts are weaker, but they still constrain comfortable stereoscopic depth budgets.<ref name="ISO2015" /> | |||
* '''Optical see‑through AR''' – Virtual imagery is often rendered at a fixed focus (e.g., ≈2 m). When it is overlaid on real objects at different depths, users may be unable to focus sharply on both simultaneously, degrading registration and comfort.<ref name="Zhou2021" /> | |||
==Effects and symptoms== | |||
Prolonged VAC drives extraocular and ciliary muscles in conflicting directions, empirically producing:<ref name="Hoffman2008" /><ref name="Lin2022" /> | |||
* | * visual fatigue / eyestrain and headaches; | ||
* transient or persistent blur; | |||
* diplopia (double vision) when fusion fails; | |||
* reduced reading speed and depth‑judgement accuracy; | |||
* contributions to [[Virtual Reality Sickness|VR sickness]] symptoms such as nausea and dizziness. | |||
Susceptibility is **highly individual** but large-scale studies indicate that conflicts < ≈0.5 dioptres (D) in front of the focal plane and < ≈1 D behind it fall inside a broad ‘zone of comfort’ for most viewers.<ref name="Shibata2011" /><ref name="ISO2015" /> | |||
* | |||
* | |||
* | |||
* | |||
==Measurement== | ==Measurement== | ||
<code>VAC (D) = | (1 / Accommodation distance) – (1 / Vergence distance) |</code> | |||
===Content | ==Mitigation strategies== | ||
===Content & interaction guidelines=== | |||
# | # Keep critical UI and text on, or slightly behind, the display’s focal plane. | ||
# | # Limit positive VAC to < 0.5 D and negative VAC to < 1 D for sustained viewing.<ref name="ISO2015" /> | ||
# | # Avoid rapid disparity jumps (> 1 D within 0.5 s). | ||
# Use gaze‑contingent depth‑of‑field blur or foveated rendering to reinforce correct monocular depth cues.<ref name="Koulieris2017" /> | |||
===Technological | ===Technological solutions=== | ||
{| class="wikitable plainrowheaders" | {| class="wikitable plainrowheaders" | ||
! Approach !! Principle !! Representative prototypes | ! Approach !! Principle !! Representative prototypes !! Notes | ||
|- | |- | ||
! | ! Varifocal | ||
| | | Eye‑tracking selects gaze depth; tunable lenses or moving optics shift a *single* focal plane accordingly. | ||
| Meta Reality Labs “Butterscotch” (2023); UNC deformable‑mirror NED (2017).<ref name="MetaButterscotch2023" /><ref name="Dunn2017" /> | |||
| Delivers correct focus at fixation depth; challenges include eye‑tracking latency and blur mismatch for peripheral objects. | |||
|- | |- | ||
! | ! Multifocal / multiplane | ||
| | | Renders content on two or more discrete focal planes, either simultaneously or time‑multiplexed. | ||
| Magic Leap 1 (two planes); Stanford light‑field prototypes.<ref name="Wired2015" /> | |||
| Provides approximate focus cues across depth; added optical complexity and brightness loss. | |||
|- | |- | ||
! | ! Light‑field & holographic | ||
| | | Reconstruct 4‑D light field or full wavefront so the eye can naturally accommodate within the volume. | ||
| Holographic near‑eye displays (Microsoft; VividQ) remain research prototypes. | |||
| Highest theoretical fidelity but currently limited by compute, resolution and field‑of‑view. | |||
|- | |- | ||
! | ! Retinal projection / scanning | ||
| Scans modulated light | | Scans modulated light directly onto the retina (Maxwellian view). | ||
| Early commercial attempts (North Focals); research scanners. | |||
| Eliminates accommodation demand but suffers from small eyebox and sparkle/floaters. | |||
|- | |- | ||
! Emerging | ! Emerging optics | ||
| | | Alvarez freeform lenses, fluidic or MEMS deformable elements for compact varifocal modules. | ||
| Large‑aperture Alvarez AR head‑up display demonstrator (2024).<ref name="Liu2024" /> | |||
| Research stage; challenges include manufacturing tolerances and power consumption. | |||
|} | |} | ||
==Temporary | ==Temporary workarounds== | ||
* | * **Monocular viewing** removes disparity and therefore vergence cues, eliminating VAC at the cost of stereopsis.<ref name="Kreylos2014" /> | ||
* | * Wearing up‑to‑date refractive correction prevents additional accommodative effort. | ||
* | * Adaptation: some users partially adapt to VAC over tens of minutes, but after‑effects (e.g., shift in resting focus) can persist for several hours.<ref name="Hoffman2008" /> | ||
==Current | ==Current research frontiers== | ||
* | * **High‑resolution varifocal HMDs** integrating retinal‑resolution displays with low‑latency depth actuation.<ref name="MetaButterscotch2023" /> | ||
* | * **Focus‑correct passthrough AR** combining dynamic‑focus optics with camera‑based passthrough to align both real and virtual imagery.<ref name="Dunn2017" /> | ||
* | * **Individualised comfort models** derived from eye‑tracking and psychophysical data to adapt depth budgets per user.<ref name="Lin2022" /> | ||
==See | ==See also== | ||
* | * [[Accommodation (eye)]] | ||
* | * [[Depth perception]] | ||
* | * [[Eye tracking]] | ||
* | * [[Head-Mounted Display]] | ||
* | * [[Light field display]] | ||
* | * [[Visual fatigue]] | ||
* [[Virtual reality sickness]] | |||
* | |||
==References== | ==References== | ||
<references /> | <references> | ||
<ref name="Hoffman2008">{{cite journal |last1=Hoffman |first1=David M. |last2=Girshick |first2=Ahna R. |last3=Akeley |first3=Kurt |last4=Banks |first4=Martin S. |title=Vergence–accommodation conflicts hinder visual performance and cause visual fatigue |journal=Journal of Vision |volume=8 |issue=3 |pages=33 |year=2008 |doi=10.1167/8.3.33 |access-date=2025-04-27}}</ref> | |||
<ref name="Shibata2011">{{cite journal |last1=Shibata |first1=Takayuki |last2=Kim |first2=Joohwan |last3=Hoffman |first3=David M. |last4=Banks |first4=Martin S. |title=The zone of comfort: predicting visual discomfort with stereo displays |journal=Journal of Vision |volume=11 |issue=8 |pages=11 |year=2011 |doi=10.1167/11.8.11 |access-date=2025-04-27}}</ref> | |||
<ref name="Kramida2016">{{cite journal |last=Kramida |first=Gregory |title=Resolving the vergence–accommodation conflict in head‑mounted displays |journal=IEEE Transactions on Visualization and Computer Graphics |volume=22 |issue=7 |pages=1912–1931 |year=2016 |doi=10.1109/TVCG.2016.2535300 |access-date=2025-04-27}}</ref> | |||
<ref name="Koulieris2017">{{cite journal |last1=Koulieris |first1=George‑Alex |last2=Bui |first2=Bee |last3=Banks |first3=Martin S. |last4=Drettakis |first4=George |title=Accommodation and comfort in head‑mounted displays |journal=ACM Transactions on Graphics |volume=36 |issue=4 |pages=87 |year=2017 |doi=10.1145/3072959.3073622 |access-date=2025-04-27}}</ref> | |||
<ref name="Zhou2021">{{cite journal |last1=Zhou |first1=Yujia |last2=Li |first2=Xuan |last3=Yuan |first3=Chang |title=Vergence–accommodation conflict in optical see‑through display: review and prospect |journal=Results in Optics |volume=5 |pages=100160 |year=2021 |doi=10.1016/j.rio.2021.100160 |access-date=2025-04-27}}</ref> | |||
<ref name="Lin2022">{{cite journal |last1=Lin |first1=Chia‑Jung |last2=Chi |first2=Chien‑Fu |last3=Lin |first3=Chih‑Kang |last4=Chang |first4=En‑Chuan |title=Effects of virtual target size, position and parallax on vergence–accommodation conflict as estimated by actual gaze |journal=Scientific Reports |volume=12 |pages=20100 |year=2022 |doi=10.1038/s41598-022-24450-9 |access-date=2025-04-27}}</ref> | |||
<ref name="ISO2015">{{cite report |title=ISO 9241‑392:2015 Ergonomics of human‑system interaction — Part 392: Reduction of visual fatigue from stereoscopic images |publisher=International Organization for Standardization |year=2015 |access-date=2025-04-27}}</ref> | |||
<ref name="Dunn2017">{{cite journal |last1=Dunn |first1=Damon |last2=Tippets |first2=Caleb |last3=Torell |first3=Kevin |last4=Kellnhofer |first4=Philipp |last5=Akşit |first5=Kaan |last6=Didyk |first6=Piotr |last7=Myszkowski |first7=Karol |last8=Luebke |first8=David |last9=Fuchs |first9=Henry |title=Wide field‑of‑view varifocal near‑eye display using see‑through deformable membrane mirrors |journal=IEEE Transactions on Visualization and Computer Graphics |volume=23 |issue=4 |pages=1322–1331 |year=2017 |doi=10.1109/TVCG.2017.2657138 |access-date=2025-04-27}}</ref> | |||
<ref name="MetaButterscotch2023">{{cite web |title=Demo or die: How Reality Labs’ display systems research team is tackling the biggest challenges in VR |url=https://www.meta.com/blog/reality-labs-research-display-systems-siggraph-2023-butterscotch-varifocal-flamera/ |website=Reality Labs Research Blog |publisher=Meta Platforms |date=2023-08-04 |access-date=2025-04-27}}</ref> | |||
<ref name="Liu2024">{{cite journal |last1=Liu |first1=Yujin |last2=Cheng |first2=Dongli |last3=Wang |first3=Yuan |last4=Hua |first4=Hong |title=A varifocal augmented‑reality head‑up display using Alvarez freeform lenses |journal=Journal of the Society for Information Display |volume=32 |issue=4 |pages=231–240 |year=2024 |doi=10.1002/jsid.1286 |access-date=2025-04-27}}</ref> | |||
<ref name="Wired2015">{{cite news |last=Zhang |first=Sarah |title=The obscure neuroscience problem that’s plaguing VR |work=Wired |date=2015-08-11 |url=https://www.wired.com/2015/08/obscure-neuroscience-problem-thats-plaguing-vr |access-date=2025-04-27}}</ref> | |||
<ref name="Kreylos2014">{{cite web |last=Kreylos |first=Oliver |title=Accommodation and vergence in head‑mounted displays |website=Doc‑Ok.org |date=2014-04-13 |url=http://doc-ok.org/?p=1602 |access-date=2025-04-27}}</ref> | |||
</references> | |||
[[Category:Terms]] | [[Category:Terms]] |
Revision as of 04:28, 27 April 2025
- See also: Terms and Technical Terms
Vergence‑accommodation conflict (VAC), also called accommodation‑vergence conflict or accommodation–vergence mismatch, is a visual–perceptual problem that arises when the cues that drive the eyes’ rotation (Vergence) and the cues that drive the crystalline lens’s focus (accommodation) specify **different viewing distances**. Under natural viewing the two mechanisms are neurally coupled via the accommodation–vergence reflex, so they agree; most stereoscopic displays break that coupling, which is now recognised as a primary source of visual discomfort, blurred vision and reduced task performance in modern VR, AR, and other stereoscopic systems.[1][2]
Physiological basis
When fixating a real‑world target the visual system executes two tightly linked responses:
- Vergence – the two eyes rotate (converge or diverge) so that their lines of sight intersect at the target, driven mainly by binocular disparity.
- Accommodation – the ciliary muscle changes the lens’s optical power so that the retinal image is in sharp focus, driven mainly by retinal blur.
Reciprocal neurons link the two responses so they normally operate at the *same* distance. Stereoscopic displays decouple them because disparity cues stimulate vergence to the *simulated* depth of a virtual object, while blur cues still stimulate accommodation to the *physical* surface of the display, which is usually fixed at one optical distance.[3]
Causes / occurrence in display technologies
- Fixed‑focus HMDs – Nearly all consumer VR headsets place the screens at a single focal distance between ≈1 m and 2 m. Vergence varies with rendered disparity, so objects that appear nearer than the focal plane create a **positive VAC** (vergence distance < accommodation distance), whereas objects that appear farther create a **negative VAC** (vergence distance > accommodation distance).[2]
- 3‑D cinema & television – The screen is farther away and subtends a smaller field‑of‑view, so conflicts are weaker, but they still constrain comfortable stereoscopic depth budgets.[4]
- Optical see‑through AR – Virtual imagery is often rendered at a fixed focus (e.g., ≈2 m). When it is overlaid on real objects at different depths, users may be unable to focus sharply on both simultaneously, degrading registration and comfort.[5]
Effects and symptoms
Prolonged VAC drives extraocular and ciliary muscles in conflicting directions, empirically producing:[1][6]
- visual fatigue / eyestrain and headaches;
- transient or persistent blur;
- diplopia (double vision) when fusion fails;
- reduced reading speed and depth‑judgement accuracy;
- contributions to VR sickness symptoms such as nausea and dizziness.
Susceptibility is **highly individual** but large-scale studies indicate that conflicts < ≈0.5 dioptres (D) in front of the focal plane and < ≈1 D behind it fall inside a broad ‘zone of comfort’ for most viewers.[2][4]
Measurement
VAC (D) = | (1 / Accommodation distance) – (1 / Vergence distance) |
Mitigation strategies
Content & interaction guidelines
- Keep critical UI and text on, or slightly behind, the display’s focal plane.
- Limit positive VAC to < 0.5 D and negative VAC to < 1 D for sustained viewing.[4]
- Avoid rapid disparity jumps (> 1 D within 0.5 s).
- Use gaze‑contingent depth‑of‑field blur or foveated rendering to reinforce correct monocular depth cues.[7]
Technological solutions
Approach | Principle | Representative prototypes | Notes |
---|---|---|---|
Varifocal | Eye‑tracking selects gaze depth; tunable lenses or moving optics shift a *single* focal plane accordingly. | Meta Reality Labs “Butterscotch” (2023); UNC deformable‑mirror NED (2017).[8][9] | Delivers correct focus at fixation depth; challenges include eye‑tracking latency and blur mismatch for peripheral objects. |
Multifocal / multiplane | Renders content on two or more discrete focal planes, either simultaneously or time‑multiplexed. | Magic Leap 1 (two planes); Stanford light‑field prototypes.[10] | Provides approximate focus cues across depth; added optical complexity and brightness loss. |
Light‑field & holographic | Reconstruct 4‑D light field or full wavefront so the eye can naturally accommodate within the volume. | Holographic near‑eye displays (Microsoft; VividQ) remain research prototypes. | Highest theoretical fidelity but currently limited by compute, resolution and field‑of‑view. |
Retinal projection / scanning | Scans modulated light directly onto the retina (Maxwellian view). | Early commercial attempts (North Focals); research scanners. | Eliminates accommodation demand but suffers from small eyebox and sparkle/floaters. |
Emerging optics | Alvarez freeform lenses, fluidic or MEMS deformable elements for compact varifocal modules. | Large‑aperture Alvarez AR head‑up display demonstrator (2024).[11] | Research stage; challenges include manufacturing tolerances and power consumption. |
Temporary workarounds
- **Monocular viewing** removes disparity and therefore vergence cues, eliminating VAC at the cost of stereopsis.[12]
- Wearing up‑to‑date refractive correction prevents additional accommodative effort.
- Adaptation: some users partially adapt to VAC over tens of minutes, but after‑effects (e.g., shift in resting focus) can persist for several hours.[1]
Current research frontiers
- **High‑resolution varifocal HMDs** integrating retinal‑resolution displays with low‑latency depth actuation.[8]
- **Focus‑correct passthrough AR** combining dynamic‑focus optics with camera‑based passthrough to align both real and virtual imagery.[9]
- **Individualised comfort models** derived from eye‑tracking and psychophysical data to adapt depth budgets per user.[6]
See also
- Accommodation (eye)
- Depth perception
- Eye tracking
- Head-Mounted Display
- Light field display
- Visual fatigue
- Virtual reality sickness
References
- ↑ 1.0 1.1 1.2 Template:Cite journal
- ↑ 2.0 2.1 2.2 Template:Cite journal
- ↑ Template:Cite journal
- ↑ 4.0 4.1 4.2 Template:Cite report
- ↑ Template:Cite journal
- ↑ 6.0 6.1 Template:Cite journal
- ↑ Template:Cite journal
- ↑ 8.0 8.1 "Demo or die: How Reality Labs’ display systems research team is tackling the biggest challenges in VR". Meta Platforms. 2023-08-04. https://www.meta.com/blog/reality-labs-research-display-systems-siggraph-2023-butterscotch-varifocal-flamera/.
- ↑ 9.0 9.1 Template:Cite journal
- ↑ Template:Cite news
- ↑ Template:Cite journal
- ↑ Kreylos, Oliver (2014-04-13). "Accommodation and vergence in head‑mounted displays". http://doc-ok.org/?p=1602.