Pith. sign in

REVIEW 2 major objections 6 minor 18 references

Disagreeing about Crocs and socks: Creating profoundly ambiguous color displays

T0 review · 2 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A design principle called SURFPAD lets researchers manufacture color displays that split observers, with a white-sock prior predicting the split.

desk verdict Genuinely novel method for creating ambiguous color displays, but the fabric-prior causal claim rests on a single-order correlation with a likely reverse-causation problem. read the letter →

arxiv 1908.05736 v1 pith:CKVEOT5K submitted 2019-08-14 q-bio.NC

Pascal Wallisch , Michael Karlovich This is my paper · ORCID
classification q-bio.NC
keywords colorperceptionambiguousstimuliSURFPADthedressconstancyvisualpriorsperceptualdisagreementfabric
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper claims that profoundly ambiguous color displays can be created on demand rather than waiting for rare internet accidents to surface. The authors photograph pink and mint crocs under carefully titrated colored light on a black background, with plain white tube socks included as an anchor that mirrors the illumination. In an online sample of more than 5,700 participants, the same crocs under typical light produced about 98% agreement on the intended color, while five altered displays produced categorical splits—grey, green, pink, or beige—with no single answer near consensus. The paper argues that the split is predicted by whether observers see the socks as white, reflecting a fabric prior built from everyday experience with white tube socks, and it links this mechanism to the dress and the sneaker as different priors driving polarized perception. If right, this turns an internet curiosity into a laboratory tool and a general mechanism for disagreement.

What carries the argument

The load-bearing device is SURFPAD—Substantial Uncertainty combined with Ramified or Forked Priors and Assumptions yields Disagreement—implemented with two named objects: a colored target object (CTO), the croc, whose reflectance is neutralized by titrated light so its pixels sit near grey, and an illumination-mirroring object (IMO), the sock, which reflects the color of the light and is usually white in experience. The trick is that the sock can either be read as white, allowing color-constancy correction back to the manufacturer's color, or as colored, making the image a literal grey croc. That fork in interpretation is what converts one image into opposite percepts.

What would settle it

Run the same survey with the sock-color question asked before any croc display, and include a condition in which the crocs appear with no socks at all. If the white-sock prior causes the pink percept, reversing the order should leave the 12-fold association intact and removing the socks should reduce or eliminate the pink option; if the prior is merely a post-hoc label, the association should shrink or vanish when the order is reversed.

Watch

Extended reading notes

Core claim

The central claim is that a display becomes profoundly color-ambiguous when three conditions are met: the target object is monochromatic and could plausibly be almost any color; the scene removes the background and context cues observers normally use to calibrate color; and an illumination-mirroring object such as a plain sock is included that some observers read as white and others take at face value. Under these conditions, the same pixels are categorically perceived as pink, green, grey, or beige by different observers, and the modal percept can be shifted by choosing the target color and the lighting axis. The paper reports that seeing the socks as white makes the pink percept of the most balanced display more than 12 times more likely, that the same sock prior predicts sneaker perception but not dress perception, and that these differences look like a stable individual characteristic rather than random guessing.

Load-bearing premise

The load-bearing premise is that the sock-color report measures a stable prior that exists before the participant sees the crocs; if asking the sock question after the croc judgment instead creates or shifts that report, the claimed causal direction from white-sock prior to pink-croc percept collapses.

Editorial extensions

If this is right

  • Because SURFPAD permits creating ambiguous displays on demand, the dress-like phenomenon can be studied in controlled stimulus sets instead of relying on accidental internet images.
  • The croc displays and the dress and sneaker responses show that disagreement is observer-specific: how someone sees one display predicts how they see the others, implying a stable perceptual disposition.
  • Sock priors predict croc and sneaker perception but not dress perception, indicating that fabric priors and illumination priors are separate mechanisms operating on different color axes.
  • Greater confidence in one's beliefs reduces the chance of recovering the color appearance under typical illumination, suggesting that metacognition modulates which prior is applied.
  • Differential priors under uncertainty, rather than disagreement about the evidence itself, are proposed as the general driver of polarized interpretations beyond color perception.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the sock prior is causal, swapping the illumination-mirroring object for a non-white anchor should predictably shift or eliminate the ambiguity; the paper itself names a no-sock control as future work.
  • The authors' observation that ambiguity tracks the relative position of the croc and sock pixel clouds in hsv space suggests a quantitative recipe: searching for parallel, separated color clouds should generate new ambiguous displays with chosen modal percepts.
  • The same SURFPAD structure might apply outside color—for example, to auditory or social judgments where one element is assumed neutral by experience—though the paper only gestures in that direction.
  • Because the sock-color question followed the P2 judgment in the survey, the claimed causal ordering is not yet secured; a reversed-order design is a direct test of that ordering.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 6 minor

Summary. The paper proposes a design principle, SURFPAD (Substantial Uncertainty combined with Ramified or Forked Priors and Assumptions yields Disagreement), for creating color displays that evoke categorical disagreement among observers. The authors created five crocs-and-socks images under altered lighting (P0, P1, P2, M1, M2) and surveyed 5,762 online participants. They report that the same crocs under typical lighting yield about 98% consensus, whereas the five altered images produce categorically different modal percepts across observers. The paper then examines correlates of P2 perception, reporting that participants who saw the socks as white were much more likely to see P2 as pink (Fig. 7), that self-reported tube-sock wearing is weakly associated with sock and P2 perception (Fig. 8), and that P2 perception strongly predicts sneaker perception but not dress perception (Fig. 9). The authors conclude that differential fabric priors, not illumination priors, drive the crocs disagreement and suggest that such priors underlie polarized disagreement more broadly.

Significance. If the design principle holds, the paper makes a useful methodological contribution: previous color-ambiguous displays such as the dress and the sneaker arose spontaneously, whereas this paper demonstrates a repeatable procedure for creating such displays. The large sample, bootstrap confidence intervals, and the typical-light control condition (98% consensus for WP and WM) are genuine strengths, and the cross-stimulus correlations with the sneaker but not the dress are informative. The paper is also candid about its limitations, explicitly noting the absence of a no-sock control and the weakness of the experience-to-prior link. The central causal claim about fabric priors, however, is not secured by the reported data, and this affects the abstract's headline assertion.

major comments (2)
  1. [Fig. 7 and §'What determines the perception of the crocs displays?'] The abstract and Discussion assert that a pre-existing white-sock fabric prior drives the P2 percept, but the only supporting evidence is the cross-sectional association in Fig. 7, in which the sock-color report was collected immediately after the P2 judgment. With this fixed order, the association is equally compatible with the P2 percept anchoring the subsequent sock report, a shared response bias, or an unmeasured third factor. No counterbalanced order, no-sock control, or manipulation of the illumination-mirroring object is reported. The causal priority of the fabric prior is therefore load-bearing for the paper's central claim but unsupported by the current design.
  2. [Discussion, limitations paragraph] The Discussion concedes that the no-sock condition was not run (the passage beginning 'it stands to reason that we should create and ask observers about crocs altogether without socks') and that the experience-to-sock link in Fig. 8 is 'obviously less strong than what would be needed to comfortably claim that we understand this aspect.' These concessions directly undercut the abstract's stronger claim that the percept is predicted 'from fabric priors resulting from experience.' The manuscript should either soften the causal claim to a correlational one, explicitly flag the directionality problem in the abstract, or present additional evidence that the prior is causally prior.
minor comments (6)
  1. [Figure 1 caption] The caption refers to 'the CTO' in panel C, but the term introduced in the text is 'target colored object' (TCO); this appears to be a typo and should be corrected.
  2. [Results, §'Are images of these crocs perceptually ambiguous under typical lighting conditions?'] The statement that 'the exact p-value for either outcome obtained with resampling methods is 0' is not accurate for a bootstrap procedure; report a finite upper bound (e.g., p < 1e-5) instead.
  3. [Method, Stimulus design] The text says 'we used the following seven images of crocs and socks' but Figure 3 also includes the dress and sneaker; please clarify that the seven images refer specifically to the crocs-and-socks displays.
  4. [Discussion] The phrase 'amounts to a personality characteristic' is stronger than a single cross-sectional survey can support; 'individual difference measure' would be more appropriate.
  5. [Method, Survey] The exact order of the dress and sneaker questions relative to the sock-color question is not specified; providing the question sequence would help readers assess potential order effects.
  6. [References] A few references (e.g., Land & McCann 1971) lack page numbers or other bibliographic detail; please ensure consistent reference formatting.

Circularity Check

1 steps flagged · score 4.0 of 10

Display-creation claim is independent, but the 'predict from fabric priors' claim rests on a same-survey, post-P2 sock-color report, making the causal inference partly circular.

  1. self definitional [Results, 'What determines the perception of the crocs displays?' (Figure 7); Abstract]
    "Given the previous result - perception of all 5 crocs is tightly inter-related and given that we asked the question about sock color right after asking about P2 - which might have drawn attention to the color of the socks for subsequent stimuli, we focus our discussion of these matters on the percepts involving P2, see figure 7. ... The primary driver of P2 perception seems to be how the socks were seen. If the socks were seen as white, the probability that the crocs were perceived as pink is over 12x that of participants who did not perceive the socks as white."

    The abstract claims 'we can predict the percept from fabric priors resulting from experience,' but the operational 'fabric prior' is a sock-color report collected immediately after the P2 judgment in a fixed order. With this ordering, the sock report is not established as a prior; it may be a downstream consequence of the P2 percept or of a shared response bias. Thus the 'predictor' is measured after the outcome it is said to predict, so the observed association does not demonstrate that a pre-existing sock prior drives the croc percept. The paper itself concedes the ordering concern and later concedes that no no-sock control was run, leaving the causal-prior claim as a postdictive correlation rather than an independent prediction.

full rationale

The paper's main constructive result is self-contained and not circular: five displays built under SURFPAD are categorically ambiguous, while the same crocs under typical lighting yield about 98% consensus, and the response-option randomization and inability to revisit images provide internal controls. That part is an empirical demonstration rather than a derivation from its own conclusion. The circularity is confined to the narrower causal claim that a white-sock fabric prior predicts the P2 percept. The sock-color report was collected after P2, in the same survey, so reverse causation and response bias are plausible; the paper acknowledges this and also acknowledges that the decisive no-sock condition was not run. Consequently, the 'predict the percept from fabric priors' statement reduces to a same-survey correlation between two percepts, not a prediction from an independently measured prior. There is no load-bearing self-citation chain and no formal equation-level tautology; the score reflects one partially circular empirical inference rather than a fully circular derivation.

Assumptions & free parameters 0 free parameters · 3 assumptions · 1 invented entities

The paper introduces conceptual labels (SURFPAD, CTO, IMO, CAUTION) and a hypothesized fabric-prior mechanism, but no new physical or mathematical entities. The key assumptions are domain-specific: color constancy anchoring, validity of self-reported priors, and the critical role of hsv-space geometry. No numerical parameters are fitted to data; stimulus construction used titration and undisclosed digital filtering.

assumptions (3)
  • domain assumption Color constancy can be anchored by a familiar white object (IMO), allowing observers to discount the illuminant and recover CAUTION.
    The Method section introduces the illumination-mirroring object (socks) and interprets pink percepts as the recovery of color under typical illumination, relying on this untested anchoring assumption.
  • domain assumption Self-reports of sock color and illumination beliefs reflect perceptual priors rather than response biases or memory of the previous question.
    The Results use sock perception as the central predictor; because it is measured after the target stimulus, this assumption is fragile.
  • domain assumption The hsv-space geometry of the pixel clouds (separation and parallel columns) is what determines categorical ambiguity.
    Discussed as what seems to matter critically in the Discussion, but never directly manipulated or tested against alternative image statistics.
invented entities (1)
  • White sock fabric prior
    purpose: Explains why some observers perceive the crocs as pink under ambiguous illumination: they use the socks as an anchor to discount the light color.
    The prior is inferred from self-reported sock perception and tube-sock wearing in the same survey; there is no independent behavioral or physiological measure of the prior outside this paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Disagreeing about Crocs and socks: Creating profoundly ambiguous color displays." pith.science (2026). https://pith.science/paper/CKVEOT5K

@misc{pith2026190805736,
  author       = {Pith},
  title        = {Pith review of: Disagreeing about Crocs and socks: Creating profoundly ambiguous color displays},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CKVEOT5K}},
  note         = {Machine review of arXiv:1908.05736}
}
read the original abstract

There is an increasing interest in the systematic disagreement about profoundly ambiguous stimuli in the color domain. However, this research has been hobbled by the fact that we could not create such stimuli at will. Here, we describe a design principle that allows the creation of such stimuli and apply this principle to create one such stimulus set - the crocs and socks. Using this set, we probed the color perception of a large sample of observers, showing that these stimuli are indeed categorically ambiguous and that we can predict the percept from fabric priors resulting from experience. We also relate the perception of these crocs to other color-ambiguous stimuli - the dress and the sneaker and conclude that differential priors likely underlie polarized disagreement in cognition more generally.

Figures

Figures reproduced from arXiv: 1908.05736 by the authors.

Figure 1
Figure 1. A schematic of the stimulus design principles. Top row: colored target objects (crocs) [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. The crocs and socks images we created. A: crocs and Socks under typical (a [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 4
Figure 4. Perception of croc color under typical lighting conditions. A: Percentage of sample reporting categorical percept on the x-axis in response to Wp. x-axis: Color labels. For reasons of economy, we abbreviate “pink or salmon” as “Pink”, “green or mint” as “Green”, “grey or white” as “Grey” and “beige or orange” as “Beige”, here and in all other figures. B: As in “A”, but for WM. The insets to the left of the y-axes de… view at source ↗
Figures from the paper (3 more)
Figure 5
Figure 5. Figure 5: Perception of color ambiguous stimuli created with altered lighting [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]
Figure 7
Figure 7. Figure 7: figure 7 [PITH_FULL_IMAGE:figures/full_fig_p011_7.png]
Figure 11
Figure 11. Figure 11: Demographic effects. A: Stimuli where CAUTION decrea [PITH_FULL_IMAGE:figures/full_fig_p016_11.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

18 extracted references · 18 canonical work pages

  1. [1]

    Substantial Uncertainty combined with Ramified or Forked Priors and Assumptions yields Disagreement

    Figure 1: A schematic of the stimulus design principles. Top row: colored target objects (crocs) under typical lighting conditions. We define sunlight or bright indoor light (e.g. incandescent or LED) as “typical” lighting conditions, as these are conditions observers can be expected to experience in everyday living. Bottom row: The same crocs under alter...

    work page 1964

  2. [2]

    Ballerina pink

    Figure 2: The crocs and socks images we created. A: crocs and Socks under typical (a combination of daylight coming through a window and a LED (Feit electric, 60 Watt equivalent, 5000K, 800 Lumen) lighting conditions. WP: “Ballerina pink” crocs. WM: “New mint” crocs. B: The “New mint” crocs under colored, specifically pink/orange light, yielding two speci...

    work page 2019

  3. [3]

    the dress

    Second, we broadly replicate the proportions described in the perception of “the dress” as reported in Lafer Sousa et al. (2015), or Wallisch (2017a), even four years later. Third, P2 perception strongly determines sneaker perception. Almost 90% of those who saw P2 as pink saw the sneaker as pink, whereas just over 10% of those who saw P2 as green did. Al...

    work page 2015

  4. [5]

    the dress

    One was “the dress” stimulus and one was a display of “the sneaker”, shown in figure 3D. Response options to the croc stimuli were “Pink/salmon”, “Green/mint”, “Grey/White”, “Orange/Beige” and “Other, please specify”, ordered randomly for each question. The response options for “the dress” and “the sneaker” stimuli were the standard “white/gold”, “black/b...

    work page 1994

  5. [6]

    the sneaker

    Briefly, we managed to create five categorically color ambiguous stimuli with different perceptual balancing points. First, we created P0 as a proof of concept. We then created two pairs of stimuli from each colored croc. All 5 images are categorically ambiguous with respect to color - the modal percept for P0 and P1 is grey (43% and 63%, respectively), f...

    work page 2002

  6. [7]

    gain the most

    Sometimes, the result of this boost is dramatic, i.e. the conditional probability of seeing M2 as beige if P2 was seen as beige is 90% (up from 70%). The conditional probability of seeing P1 as pink if P2 was seen as pink is 86% (up from 18%). On average, the ratio is boosted almost 2.5 fold. To summarize, there is considerable consistency within observer...

    work page 2017

  7. [10]

    fabric prior

    or familiar lighting (Maloney, 1999), while at the same time introducing a part of the object that mirrors the properties of the lighting, but for which some - but not all - participants have a prior that they use to anchor their color percept, a process which we term SURFPAD. We were also able to show that the perception of the different color displays w...

    work page 1922

  8. [39]

    L., & Tenenbaum, J

    Griffiths, T. L., & Tenenbaum, J. B. (2006). Optimal predictions in everyday cognition. Psychological Science, 17(9), 767-773. Hansen, T., Olkkonen, M., Walter, S., & Gegenfurtner, K. R. (2006). Memory modulates color appearance. Nature Neuroscience, 9(11),

    work page 2006

Show all 18 references
  1. [244]

    the dress

    Körding, K. P., Beierholm, U., Ma, W. J., Quartz, S., Tenenbaum, J. B., & Shams, L. (2007). Causal inference in multisensory perception. PLoS one, 2(9), e943. Lafer-Sousa, R., Hermann, K. L., & Conway, B. R. (2015). Striking individual differences in color perception uncovered...

    2007

  2. [424]

    Monnier, P., & Shevell, S. K. (2003). Large shifts in color appearance from patterned chromatic backgrounds. Nature Neuroscience, 6(8),

    2003

  3. [531]

    W., & Shevell, S

    Jenness, J. W., & Shevell, S. K. (1995). Color appearance with sparse chromatic context. Vision Research, 35(6), 797-805. Jin, E. W., & Shevell, S. K. (1996). Color memory and color constancy. JOSA A, 13(10), 1981-1991. Körding, K. P., & Wolpert, D. M. (2004). Bayesian integra...

    1995

  4. [578]

    The dress

    Wald, G. (1964). The receptors of human color vision. Science, 145(3636), 1007-1016. Wallisch, P., & Movshon, J. A. (2008). Structure and function come unglued in the visual cortex. Neuron, 60(2), 195-197. Wallisch, P. (2016): https://slate.com/technology/2016/03/the-science-o...

    1964

  5. [801]

    E., Xiao, D

    Raiguel, S. E., Xiao, D. K., Marcar, V. L., & Orban, G. A. (1999). Response latency of macaque area MT/V5 neurons and its relationship to stimulus parameters. Journal of Neurophysiology, 82(4), 1944-1956. Richter, F. (2017). https://www.statista.com/chart/10913/number-of-photo...

    1999

  6. [1010]

    color mondrian

    Maloney, L. T. (1999). Physics-based approaches to modeling surface color perception. Color Vision: From genes to perception, 387-416. McCann, J. J., McKee, S. P., & Taylor, T. H. (1976). Quantitative studies in retinex theory a comparison between theoretical predictions and o...

    1999

  7. [1367]

    M., Ernst, M

    Hillis, J. M., Ernst, M. O., Banks, M. S., & Landy, M. S. (2002). Combining sensory information: mandatory fusion within, but not between, senses. Science, 298(5598), 1627-1630. Hupé, J. M., & Rubin, N. (2003). The dynamics of bi-stable alternation in ambiguous motion displays...

    2002

  8. [1968]

    locked in

    or moving plaids (Hupe & Rubin, 2003), all with characteristic and known perceptual dynamics. Instead, most observers seem to be “locked in” to a particular interpretation of the image, consistent with the notion that accumulated experience - in the form of an illumination pri...

    2003

  9. [2003]

    the dress

    or under carefully modulated lighting conditions (Lotto & Purves, 1999), suggesting that observers take both background and illumination into account when judging the appearance of a colored patch. Importantly, all of these illusions are unidirectional, as they seem to affect ...

    1999

  10. [2011]

    the dress

    depends on one’s conception of what color the crocs “really” are - CAUTION or the color of the individual pixels as seen in the point clouds of figure 3, or if that question even makes sense, as the concept of veridicality is not well defined for color in the first place. What...

    2017

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.