Skip to content
Back to the question

Evidence record

How much of what you see is invented?

Every sentence on the exploration belongs to at least one claim below. Each says what kind of statement it is, what it rests on, what the evidence does not establish, and how far it was checked. Nothing here is a score, and none of it is a verdict on a belief or a theory.

Last checked 24 September 2026.

  • A judgement, not a fact: 10
  • Source checked: 31
  • Presented as open: 4
  • Partly checked: 7

If something here is wrong, misquoted, out of date or unfair, write to support@unknownoceans.com and give the claim number, which does not change.

What the question takes for granted

A scientific question with a philosophical edge. Some of what vision delivers is demonstrably filled in; how much of the rest is prediction is argued; and why any of it is experienced is not a question the science on this page can settle.

  • Some of what you see is not measured from light at all

    Established

    The blind spot is the clean case: no photoreceptors, and still no hole in what you see. That settles that some of vision is filled in; it does not say how much.

  • What is not measured is predicted

    Contested

    Predictive processing says so for a large share of perception. It is influential, and a 2020 review of the neurophysiology found the evidence for its key claims mixed.

  • Explaining how vision computes would leave something over

    Contested

    This is the hard problem of consciousness. Some philosophers expect the gap to close as the science matures; others think it will not.

How the question is labelled

Claim: Perception

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: Perception

Kind
Framing. Signposting. It asserts nothing a source could check.

The question

Claim: The world you experience is a model, and you are the one building it.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

The page’s thesis. It follows from claims 5, 10 and 21: part of vision is filled in, and cortex responds to features rather than to a picture.

Corrected

24 September 2026. Previously: “The world you experience is a model, built by you, slightly after the fact.” Nothing cited on the page carries a claim about the timing of perception, and “slightly after the fact” is a factual statement dressed as a flourish. The thesis stands without it.

Claim 3 · checked 24 September 2026 · AI-assisted audit

Claim: Vision is not a camera.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

Follows from claims 14 to 18 and 21: the eye samples unevenly, and cortical cells respond to edges and motion rather than to light at points.

Claim 4 · checked 24 September 2026 · AI-assisted audit

Claim: Some of what reaches your awareness is filled in rather than measured from light, and the filling-in is invisible from the inside.

Kind
Observation. Something measured or seen, as the named source reports it.
Review
Source checked. The cited passage was read, and it says this.

Later work looked for

Crossref, 2014 on, "blind spot filling-in": nothing that changes it.

Corrected

24 September 2026. Previously: “A large share of what reaches your awareness is constructed from prediction rather than measured from light, and the construction is invisible from the inside.” The summary stated as fact the thing the page’s own uncertain section says is unsettled: how large a share is prediction. Filling-in is demonstrated; the size of the predicted share is argued (Walsh and others, 2020).

Claim 5 · checked 24 September 2026 · AI-assisted audit

Claim: How much of it is prediction is still argued.

Kind
Open question. Something the page says is not known.
Review
Presented as open. The page says this is unresolved, and the review found that it is.

Claim 6 · checked 24 September 2026 · AI-assisted audit

Claim: There is a hole in your visual field where the optic nerve leaves the retina, six degrees wide by eight.

Kind
Observation. Something measured or seen, as the named source reports it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Asymmetrical color filling-in from the nasal to the temporal side of the blind spotIntroduction, first paragraph

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Measurement

    The source reads: The optic disk of the eye is formed by the ganglion cell axons exiting the eyeball ... located in the temporal visual field with its center at approximately 15° and subtending 6 × 8° of visual angle

What this does not establish

Li and others give the figure from earlier measurements they cite; it is an average, and the size varies between eyes.

Later work looked for

Crossref, 2014 on, "blind spot size visual field": nothing that changes it.

Claim 7 · checked 24 September 2026 · AI-assisted audit

Claim: There are no photoreceptors in it.

Kind
Observation. Something measured or seen, as the named source reports it.
Review
Source checked. The cited passage was read, and it says this.

Later work looked for

Anatomy of the optic disc; not a finding later work revises.

Claim 8 · checked 24 September 2026 · AI-assisted audit

Claim: You have never seen it.

Kind
Observation. Something measured or seen, as the named source reports it.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

With two eyes open, each eye covers the other’s blind spot; with one eye, filling-in hides it. Li and others give both reasons.

Later work looked for

Crossref, 2014 on, "blind spot filling-in": nothing that changes it.

Claim 9 · checked 24 September 2026 · AI-assisted audit

Claim: The gap is filled in. Not covered over with grey, filled in with a plausible continuation of whatever is around it.

Kind
Observation. Something measured or seen, as the named source reports it.
Review
Source checked. The cited passage was read, and it says this.

Later work looked for

Crossref, 2014 on, "blind spot filling-in": nothing that changes it.

Claim 10 · checked 24 September 2026 · AI-assisted audit

Claim: This is the smallest and most easily demonstrated example of something the visual system does constantly.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

The page’s own framing. The rest of the page argues it: claims 14 to 18 on uneven sampling and eye movements, claims 24 to 29 on change blindness.

Claim 11 · checked 24 September 2026 · AI-assisted audit

What the visual system actually does

Claim: What the visual system actually does

Kind
Framing. Signposting. It asserts nothing a source could check.

The eye samples unevenly and the brain hides it

Claim: The eye samples unevenly and the brain hides it

Kind
Observation. Something measured or seen, as the named source reports it.
Review
Source checked. The cited passage was read, and it says this.

Later work looked for

Summarises the claims below; see their searches.

Claim 13 · checked 24 September 2026 · AI-assisted audit

Claim: Fine detail comes from the fovea, a small patch at the centre of the retina where cones are packed most densely.

Kind
Observation. Something measured or seen, as the named source reports it.
Review
Partly checked. Only part of the source could be read. What was not read is listed.

  • Supports

    Using perturbations to identify the brain circuits underlying active visionIntroduction, first paragraph

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Review of the field

    The source reads: Primate vision relies on input from the central region of the retina, the fovea, the retinal region that provides the highest spatial resolution.

  • Supports

    Human photoreceptor topographyabstract

    Cited · Link answered, 19 October 2026 · Abstract only, 24 September 2026 · Research in people · Measurement

    The source reads: Peak foveal cone density averages 199,000 cones/mm2 and is highly variable between individuals

What this does not establish

Curcio and others was read as an abstract only. Eight retinas from seven donors aged 27 to 44.

Later work looked for

Crossref, 2010 on, "human foveal cone density topography": adaptive-optics imaging in living eyes (Wang and others, 2019) finds peak density varies with eye length; nothing that changes this.

Corrected

24 September 2026. Previously: “High-resolution colour vision is confined to the middle of the fovea.” Neither source confines colour vision to the middle of the fovea. They carry that the fovea gives the finest detail and has the densest cones.

Claim 14 · checked 24 September 2026 · AI-assisted audit

Claim: The rod-free patch at its centre is about a degree and a quarter across, and a millimetre from the centre cone density has already fallen by an order of magnitude.

Kind
Observation. Something measured or seen, as the named source reports it.
Review
Partly checked. Only part of the source could be read. What was not read is listed.

  • Supports

    Human photoreceptor topographyabstract

    Cited · Link answered, 19 October 2026 · Abstract only, 24 September 2026 · Research in people · Measurement

    The source reads: Cone density falls steeply with increasing eccentricity and is an order of magnitude lower 1 mm away from the foveal center ... In the fovea, the average horizontal diameter of the rod-free zone is 0.350 mm (1.25 degrees).

What this does not establish

Read as an abstract only. The figures are averages from eight retinas of seven donors; peak cone density varied about threefold between individuals.

Later work looked for

Crossref, 2010 on, "human foveal cone density topography": in vivo adaptive-optics work (Wang and others, 2019) adds variation with eye length; nothing that changes these figures.

Corrected

24 September 2026. Previously: “a millimetre out from that cone density has already fallen by an order of magnitude” Curcio measures the millimetre from the foveal centre, not from the edge of the rod-free patch.

Claim 15 · checked 24 September 2026 · AI-assisted audit

Claim: Everything further out is coarser.

Kind
Observation. Something measured or seen, as the named source reports it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Using perturbations to identify the brain circuits underlying active visionIntroduction, first paragraph

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Review of the field

    The source reads: the fovea, the retinal region that provides the highest spatial resolution. It is this high-resolution vision that is critical for examination of all objects of interest

Later work looked for

How the retina is organised; not a finding later work revises.

Corrected

24 September 2026. Previously: “Everything else is coarse and poor at colour.” No cited source says peripheral vision is poor at colour; the sources carry lower resolution away from the fovea.

Claim 16 · checked 24 September 2026 · AI-assisted audit

Claim: Your eyes move two or three times a second to compensate.

Kind
Observation. Something measured or seen, as the named source reports it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Using perturbations to identify the brain circuits underlying active visionIntroduction, first paragraph

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Review of the field

    The source reads: This fovea in turn requires eye movements that direct it sequentially to different parts of the visual field ... These saccades occur at a high rate, frequently two or three per second

Later work looked for

Crossref, 2015 on, "saccade frequency natural viewing": nothing that changes it.

Claim 17 · checked 24 September 2026 · AI-assisted audit

Claim: During each movement the image on the retina is a blur, and you perceive neither the movement, nor the blur, nor the blind spot.

Kind
Observation. Something measured or seen, as the named source reports it.
Review
Partly checked. Only part of the source could be read. What was not read is listed.

  • Supports

    Using perturbations to identify the brain circuits underlying active visionIntroduction, first paragraph

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Review of the field

    The source reads: during the high-speed saccade the image from the retina is a blur

  • Supports

    Neuronal mechanisms of visual stabilityabstract

    Cited · Link answered, 19 October 2026 · Abstract only, 24 September 2026 · Research in people · Review of the field

    The source reads: Human vision is stable and continuous in spite of the incessant interruptions produced by saccadic eye movements ... displacement of the retinal image with each saccade and blurring of the image during the saccade

  • Supports

    Asymmetrical color filling-in from the nasal to the temporal side of the blind spotIntroduction, first paragraph

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Experiment

    The source reads: we never experience a dark hole in our visual field

What this does not establish

Wurtz 2008 was read as an abstract. It reviews the mechanisms, mostly from recordings in monkeys, and calls saccadic suppression well established.

Later work looked for

Europe PMC, 2008 on, "saccadic suppression" reviews: later work locates part of the suppression in the retina itself (Idrees and others, 2022); nothing that changes the claim.

Claim 18 · checked 24 September 2026 · AI-assisted audit

Claim: The stable, uniformly detailed visual world you experience does not exist anywhere on your retina.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

Follows from claims 14 to 18: the retina samples detail unevenly, blurs with every saccade, and has a hole in it.

Claim 19 · checked 24 September 2026 · AI-assisted audit

Visual cortex extracts features, not pictures

Claim: Visual cortex extracts features, not pictures

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Evolution of Ideas on the Primary Visual Cortex, 1955 to 1978section on complex cells, p. 35

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in animals · Experiment

    The source reads: Most cortical cells respond poorly to diffuse light, so that when I gaze at a white object on a dark background, say an egg

What this does not establish

Recordings in cats and monkeys.

Later work looked for

Crossref and Europe PMC, 2014 on, "predictive processing neurophysiology": the feedforward reading of these results is challenged (claim 22), not the responses themselves.

Claim 20 · checked 24 September 2026 · AI-assisted audit

Claim: Hubel and Wiesel showed that individual neurons in the visual cortex of cats and monkeys respond to specific features (edges at particular orientations, motion in particular directions) rather than to brightness at points.

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Evolution of Ideas on the Primary Visual Cortex, 1955 to 1978p. 28, the first oriented cell; section on simple cells

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in animals · Experiment

    The source reads: The response had nothing to do with the black dot. ... a faint but sharp shadow, a straight dark line on a light background. That was what the cell wanted ... In the cat striate cortex it soon became evident that cells were more complex than geniculate cells

  • Supports

    Evolution of Ideas on the Primary Visual Cortex, 1955 to 1978section on complex cells, pp. 32 to 34

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in animals · Experiment

    The source reads: the optimal orientation of a typical complex cell in layer II or III in the monkey can be easily determined ... About half of the complex cells fire much better to one direction of movement than to the opposite direction

  • Context

    Receptive fields, binocular interaction and functional architecture in the cat’s visual cortexrecord

    Cited · Link answered, 19 October 2026 · Catalogue record only, 24 September 2026

    The 1962 paper on the cat that the passage names. Not read; the lecture describes it.

What this does not establish

Recordings in anaesthetised cats and monkeys, not in people.

Later work looked for

Crossref and Europe PMC, 2014 on, "orientation selectivity visual cortex": nothing that changes the finding.

Corrected

24 September 2026. Previously: “individual neurons in visual cortex respond” The recordings were in animals; the page did not say so.

Claim 21 · checked 24 September 2026 · AI-assisted audit

Claim: For many years the dominant account built perception upward from these features, through detectors of increasing complexity.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Evaluating the neurophysiological evidence for predictive processing as a model of perceptionabstract

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Review of the field

    The source reads: For many years, the dominant theoretical framework guiding research into the neural origins of perceptual experience has been provided by hierarchical feedforward models, in which sensory inputs are passed through a series of increasingly complex feature detectors.

Corrected

24 September 2026. Previously: “Perception is assembled from these features upward, and the assembly is where most of the interesting work happens.” The old wording stated the feedforward account as fact, on a page whose next section reports the predictive account that challenges it (Walsh and others, 2020). Hubel himself called the assembly a mystery.

24 September 2026. Previously: “For decades the dominant account built perception upward from these features, through detectors of increasing complexity.” The review cited says “for many years”, not decades; the wording now matches it.

Claim 22 · checked 24 September 2026 · AI-assisted audit

Claim: How that assembly produces what we see was, Hubel wrote in 1981, still a complete mystery.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Evolution of Ideas on the Primary Visual Cortex, 1955 to 1978end of the section on complex cells, p. 35

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in animals · Review of the field

    The source reads: how the information from such sets of cells is assembled at subsequent stages in the path, to build up what we call “percepts” of lines or curves (if indeed anything like that happens at all), is still a complete mystery

Claim 23 · checked 24 September 2026 · AI-assisted audit

Large changes can go unnoticed

Claim: Large changes can go unnoticed

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Failure to detect changes to people during a real-world interactionabstract; General Discussion, last paragraph

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Experiment

    The source reads: Provided that the meaning of the scene is unchanged, changes to attended objects can escape detection even when they occur during a natural, real-world interaction. ... show that even substantial changes to the objects ... will often go unnoticed

Later work looked for

Crossref, 1999 on, "change blindness real-world interaction": the effect recurs (a 2018 study of 103 people found 39 per cent missed a mid-conversation swap).

Corrected

24 September 2026. Previously: “Large changes go unnoticed when they are not expected” The heading made expectation the cause. Simons and Levin say only that not expecting a swap “may contribute”, and about half the pedestrians did notice.

Claim 24 · checked 24 September 2026 · AI-assisted audit

Claim: In one change-blindness experiment, a stranger asking pedestrians for directions was swapped for a different person while a door was carried between them.

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Failure to detect changes to people during a real-world interactionExperiment 1, introduction

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Experiment

    The source reads: asking directions of a pedestrian on a college campus ... by carrying a door between the experimenter and the pedestrian. While the experimenter was occluded by the door, another experimenter took his place

Later work looked for

Crossref, 1999 on, "change blindness real-world interaction": the design has been repeated with larger samples; nothing that changes the description.

Corrected

24 September 2026. Previously: “In change-blindness experiments, people fail to notice when the person they are talking to is swapped for a different person mid-conversation.” The old wording generalised one study to people at large, and left out the door that hid the swap for about a second.

Claim 25 · checked 24 September 2026 · AI-assisted audit

Claim: Only seven of the fifteen pedestrians noticed.

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Failure to detect changes to people during a real-world interactionExperiment 1, Results and Discussion

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Experiment

    The source reads: despite clear differences in clothing, appearance, and voice, only 7 of the 15 ... reported noticing the change of experimenters

What this does not establish

Fifteen people. In a second experiment, with the experimenters dressed as construction workers, fewer students noticed.

Later work looked for

Crossref, 1999 on, "change blindness real-world interaction": a 2018 study of 103 people found 39 per cent missed a mid-conversation swap; the effect holds.

Corrected

24 September 2026. Previously: “About half of the pedestrians in the original study missed it.” The exact count, and the sample size, from the paper.

Claim 26 · checked 24 September 2026 · AI-assisted audit

Claim: Attending to the person is not enough.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

Claim 27 · checked 24 September 2026 · AI-assisted audit

Claim: Detecting the change means encoding what was there, holding it, and comparing,

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Partly checked. Only part of the source could be read. What was not read is listed.

  • Supports

    Failure to detect changes to people during a real-world interactionGeneral Discussion, first paragraph

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Experiment

    The source reads: change detection probably requires effortful encoding of

  • Supports

    Change Blindness: Theory and Consequencesabstract

    Cited · Link answered, 19 October 2026 · Abstract only, 24 September 2026 · Research in people · Review of the field

    The source reads: The extent of change blindness in visual perception suggests limits on our capacity to encode, retain, and compare visual information from one glance to the next

What this does not establish

The 2005 review was read as an abstract only. Both sources put this as what the results suggest, not as a mechanism measured directly.

Claim 28 · checked 24 September 2026 · AI-assisted audit

Claim: and later work finds that either the encoding or the comparison can be what fails.

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Partly checked. Only part of the source could be read. What was not read is listed.

What this does not establish

Read as an abstract only. Two real-world experiments.

Later work looked for

Crossref, 1999 on, "change blindness representation comparison failure": Varakin, Levin and Collins (2007) is the newer work; added to the page.

Corrected

24 September 2026. Previously: “and that is the step that does not happen.” Simons and Levin say detection “probably requires” effortful encoding; they did not show which step fails. A later real-world study found both failures of representation and failures of comparison.

Claim 29 · checked 24 September 2026 · AI-assisted audit

How much of it is prediction

Claim: How much of it is prediction

Kind
Framing. Signposting. It asserts nothing a source could check.

How much is prediction and how much is measurement

Claim: How much is prediction and how much is measurement

Kind
Open question. Something the page says is not known.
Review
Presented as open. The page says this is unresolved, and the review found that it is.

Claim 31 · checked 24 September 2026 · AI-assisted audit

Claim: Predictive processing accounts hold that the brain continuously generates expectations about incoming signals and that perception is largely the brain’s best current model, corrected by error signals rather than built from scratch.

Kind
What a model says. What a theory or model predicts or describes, which is not the same as what has been observed.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

This is what the framework says, not what has been shown. Claim 33 gives the state of the evidence.

Claim 32 · checked 24 September 2026 · AI-assisted audit

Claim: The framework is influential and productive, and not universally accepted: a 2020 review of the neurophysiology found the support for each of its key claims mixed.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

The same review notes that clear-cut counterevidence has yet to emerge: mixed support is not refutation.

Later work looked for

Crossref, 2014 on, "evaluating neurophysiological evidence predictive processing": Walsh and others (2020) is the review found; added to the page.

Corrected

24 September 2026. Previously: “The framework is influential, productive, and not universally accepted as a complete account.” The old sentence rested on a 2013 paper alone. A later review of the neurophysiological evidence says where the doubt lies.

Claim 33 · checked 24 September 2026 · AI-assisted audit

Claim: The proportions matter. A system that mostly predicts and occasionally corrects behaves very differently from one that mostly measures.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

The page’s own reasoning about why claim 31 matters. It does not say which kind of system the brain is.

Claim 34 · checked 24 September 2026 · AI-assisted audit

Whether people perceive the same things

Claim: Whether people perceive the same things

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: Colour naming varies across languages.

Kind
Observation. Something measured or seen, as the named source reports it.
Review
Source checked. The cited passage was read, and it says this.

Later work looked for

A long-documented fact of language; not a finding later work revises.

Corrected

24 September 2026. Previously: “Colour naming varies across languages, and it shows up in behaviour.” The behavioural effect the passage gave as its example did not replicate (Martinovic and others, 2020), so the general claim is not carried by what the passage cites.

Claim 36 · checked 24 September 2026 · AI-assisted audit

Claim: Russian forces a distinction between lighter and darker blues,

Kind
About a word. What a word in the original language means, or how it is translated.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Russian blues reveal effects of language on color discriminationabstract

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Experiment

    The source reads: Russian makes an obligatory distinction between lighter blues (‘‘goluboy’’) and darker blues (‘‘siniy’’).

What this does not establish

Martinovic and others found the boundary between the two words shifts with context, and call the Russian blues less well structured than thought.

Claim 37 · checked 24 September 2026 · AI-assisted audit

Claim: and in a 2007 study Russian speakers were faster to tell two blues apart when that boundary fell between them.

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Russian blues reveal effects of language on color discriminationabstract

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Experiment

    The source reads: Russian speakers were faster to discriminate two colors when they fell into different linguistic categories in Russian (one siniy and the other goluboy) than when they were from the same linguistic category

  • Cuts against

    Russian blues reveal the limits of language influencing colour discriminationabstract

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Experiment

    The source reads: Contrary to previous research, our results show that in Russian, stimulus discrimination at the lightness-defined blue BCC boundary is not reflected in processing speed.

What this does not establish

One study. The advantage was a comparison within the Russian group, not Russian speakers against English ones.

Later work looked for

Crossref and Europe PMC, 2020 on, "Russian blues": Martinovic, Paramei and MacInnes (2020) did not find the advantage; the page now reports it. A 2026 corrigendum to that paper exists and was not read.

Corrected

24 September 2026. Previously: “and Russian speakers are faster to tell two blues apart when that boundary falls between them.” Stated in the present tense as a general fact. It is one study, and a 2020 replication with the same tasks did not find it.

Claim 38 · checked 24 September 2026 · AI-assisted audit

Claim: A 2020 attempt with the same tasks found no such advantage,

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Russian blues reveal the limits of language influencing colour discriminationabstract; Discussion, first paragraph

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Experiment

    The source reads: We re-evaluated this finding in two experiments that employed identical tasks as in the cited study ... We did not observe a boundary advantage in processing speed for sinij/goluboj in either of the two experiments

  • Qualifies

    Russian blues reveal the limits of language influencing colour discriminationMethods, Experiment 1; 2.3 Interim conclusions

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Experiment

    The source reads: From 20 colours in their study, we used 15 ... approximately 4-5 steps darker than in Winawer et al.

What this does not establish

Not an exact copy: the authors used fifteen of the original twenty colours, and the boundary they found sat four to five steps darker than in 2007.

Later work looked for

Crossref, 2020 on, "Russian blues": a 2026 corrigendum to this paper exists and could not be read; a 2024 preprint on Italian blues also reports no effect (not read).

Corrected

24 September 2026. Previously: “” A published replication with identical tasks did not find the effect the page stated as fact, and the page did not mention it.

Claim 39 · checked 24 September 2026 · AI-assisted audit

Claim: and its authors suggest the original may have been an artefact of darker blues simply being slower to judge.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Russian blues reveal the limits of language influencing colour discriminationDiscussion

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Experiment

    The source reads: we observe slower responses to darker blue stimuli as opposed to lighter blues ... the emergence of an observable RT advantage for cross-category colours in Winawer et al.’s study might have been an artifact of the non-uniform RT distributions driven by discriminations of darker vs. lighter colour pairs

  • Context

    Russian blues reveal the limits of language influencing colour discriminationMethods, Experiment 1

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Experiment

    The source reads: In Experiment 1, we reassessed the no-interference condition of the Winawer et al.’s study

  • Context

    Russian blues reveal effects of language on color discriminationabstract

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Experiment

    The source reads: this category advantage was eliminated by a verbal, but not a spatial, dual task

Other readings

Winawer and others found the advantage survived a spatial dual task and vanished under a verbal one. The 2020 study’s first experiment reassessed only the condition without interference.

Claim 40 · checked 24 September 2026 · AI-assisted audit

Claim: Whether an effect like that is a difference in seeing or in describing is the old question.

Kind
Open question. Something the page says is not known.
Review
Presented as open. The page says this is unresolved, and the review found that it is.

Corrected

24 September 2026. Previously: “Whether that is a difference in seeing or in describing is the old question.” The effect it pointed to is now reported as disputed.

Claim 41 · checked 24 September 2026 · AI-assisted audit

Claim: In the 2007 study the advantage disappeared when the speaker was given a verbal task at the same time,

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

The verbal task was silently rehearsing digit strings.

Later work looked for

Crossref and Europe PMC, 2020 on, "Russian blues": the 2020 replication found no advantage to remove; see claim 39.

Corrected

24 September 2026. Previously: “The advantage disappears when the speaker is given a verbal task at the same time, which points at the naming rather than at the eye.” A single study’s result, now given as that study’s.

Claim 42 · checked 24 September 2026 · AI-assisted audit

Claim: which pointed at the naming rather than at the eye.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

Claim 43 · checked 24 September 2026 · AI-assisted audit

What cannot be checked from the inside

Claim: What cannot be checked from the inside

Kind
Framing. Signposting. It asserts nothing a source could check.

What the experience is made of

Claim: What the experience is made of

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: Even a complete account of how the visual system computes a model of the world might not explain why that model is experienced.

Kind
Open question. Something the page says is not known.
Review
Presented as open. The page says this is unresolved, and the review found that it is.

  • Supports

    Consciousness§ 5.4 Prospects of explanatory success

    Cited · Link answered, 19 October 2026 · The cited passage was read, 24 September 2026 · Theory · Reference work

    The source reads: Others may seem less tractable, especially the so-called “hard problem” (Chalmers 1995) which is more or less that of giving an intelligible account that lets us see in an intuitively satisfying way how phenomenal or “what it’s like” consciousness might arise

Corrected

24 September 2026. Previously: “Even a complete account of how the visual system computes a model of the world would not explain why that model is experienced.” Stated as settled a philosophical position that is disputed; the reference work reports physicalists who expect the gap to close.

Claim 46 · checked 24 September 2026 · AI-assisted audit

Claim: This is the same gap that sits under consciousness, arriving from a different direction,

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

The page’s own link to the hard problem of consciousness, which claim 46 describes.

Claim 47 · checked 24 September 2026 · AI-assisted audit

Claim: and philosophers disagree about whether it will ever close.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Consciousness§ 5.4 Prospects of explanatory success

    Cited · Link answered, 19 October 2026 · The cited passage was read, 24 September 2026 · Theory · Reference work

    The source reads: Optimistic physicalists will likely see current explanatory lapses as merely the reflection of the early stage of inquiry ... To dualists, those same impasses will signify the bankruptcy of the physicalist program

Corrected

24 September 2026. Previously: “” Reports the disagreement the reference work describes.

Claim 48 · checked 24 September 2026 · AI-assisted audit

How much of the model we could ever notice

Claim: How much of the model we could ever notice

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: Illusions work because the construction is not accessible to inspection.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

The page’s own reading, filed under what cannot be checked. It follows from claims 9 and 18: the blind spot and the saccadic blur are hidden from the person they happen to.

Claim 50 · checked 24 September 2026 · AI-assisted audit

Claim: There is no method for auditing your own perception from the inside, which means the size of the constructed portion has to be inferred indirectly, from cases where the construction fails.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

Describes the method the page itself uses: the blind spot, the saccade and the swapped stranger are all cases where the construction shows.

Claim 51 · checked 24 September 2026 · AI-assisted audit

If perception is construction

Claim: If perception is construction

Kind
Framing. Signposting. It asserts nothing a source could check.

Perception is controlled hallucination

Claim: Perception is controlled hallucination

Kind
Hypothesis. A proposed explanation, not an established one.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

A view some predictive-processing theorists take, filed under what is possible. The phrase is not Clark’s, and claims 55 and 56 say where he stops short of it.

Claim 53 · checked 24 September 2026 · AI-assisted audit

Claim: On the strongest reading of predictive processing, perceiving and imagining use the same machinery, and the difference is how tightly the process is constrained by sensory input.

Kind
Hypothesis. A proposed explanation, not an established one.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

A theoretical proposal, not a measured result.

Claim 54 · checked 24 September 2026 · AI-assisted audit

Claim: Hallucination would then be the ordinary mechanism with the constraint loosened.

Kind
Hypothesis. A proposed explanation, not an established one.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

Clark’s own account of hallucination in schizophrenia turns on how prediction error is weighted going wrong, not simply on weaker sensory input.

Claim 55 · checked 24 September 2026 · AI-assisted audit

Claim: Clark, who makes the case for that duality, stops short of the strongest reading.

Kind
An argument. An argument a named thinker makes. That it is made is checked; whether it succeeds is argued.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Whatever next? Predictive brains, situated agents, and the future of cognitive sciencesection 4.3 and note 43

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Theory · Argument

    The source reads: a deep duality of perception and imagination ... deep duality notwithstanding, nothing in the present view requires that the system, when engaged in imagery-based processing, will typically support the very same kinds of stability and richness of experienced detail that daily sensory engagements offer

Claim 56 · checked 24 September 2026 · AI-assisted audit

Claim: Nothing in the view, he notes, requires imagery to carry the stability and detail that ordinary seeing does.

Kind
An argument. An argument a named thinker makes. That it is made is checked; whether it succeeds is argued.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Whatever next? Predictive brains, situated agents, and the future of cognitive sciencenote 43

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Theory · Argument

    The source reads: nothing in the present view requires that the system, when engaged in imagery-based processing, will typically support the very same kinds of stability and richness of experienced detail that daily sensory engagements offer

Claim 57 · checked 24 September 2026 · AI-assisted audit

The model is built for action, not accuracy

Claim: The model is built for action, not accuracy

Kind
Hypothesis. A proposed explanation, not an established one.
Review
Partly checked. Only part of the source could be read. What was not read is listed.

  • Supports

    The Interface Theory of Perceptionabstract

    Cited · Link answered, 19 October 2026 · Abstract only, 24 September 2026 · A simulation · Model

    The source reads: An interface serves to guide useful actions, not to resemble truth.

What this does not establish

Read as an abstract only. Hoffman and colleagues argue this from evolutionary games in simulated worlds, not from measurements of perception, and it is a contested theory.

Claim 58 · checked 24 September 2026 · AI-assisted audit

Claim: If perception evolved to support behaviour rather than to represent the world faithfully, systematic distortions are not errors.

Kind
Hypothesis. A proposed explanation, not an established one.
Review
Partly checked. Only part of the source could be read. What was not read is listed.

  • Supports

    The Interface Theory of Perceptionabstract

    Cited · Link answered, 19 October 2026 · Abstract only, 24 September 2026 · A simulation · Model

    The source reads: We find that veridical perceptions--strategies tuned to the true structure of the world--are routinely dominated by nonveridical strategies tuned to fitness.

What this does not establish

Read as an abstract only. A result about simulated strategies competing in evolutionary games, offered as a conditional.

Claim 59 · checked 24 September 2026 · AI-assisted audit

Claim: They are the design.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

Follows from the conditional in claim 59, and holds only if it does.

Claim 60 · checked 24 September 2026 · AI-assisted audit

Claim: This has consequences for how much we should trust perception as evidence about reality, including in science, where instruments exist precisely to route around it.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

The page’s own conclusion from the possibility in claims 58 and 59.

Claim 61 · checked 24 September 2026 · AI-assisted audit

What was read