{"id":"57f13c32-0882-40bd-8bc2-8d1dfe115b1d","arxiv_id":"2601.20805","paper_version":4,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":7.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"Standard error bars are insufficient for judging model fits to data with correlated uncertainties, requiring additional visualizations of principal components and conditional uncertainties.","lead":"The paper demonstrates that vertical error bars alone fail to convey whether a model fits data well when y-value uncertainties are correlated across points. It proposes adding displays of the first principal component of the covariance and conditional uncertainties to enable better visual assessment of model agreement.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.3","headline":"No significant objection identified","rationale":"Reader's weakest assumption matches the paper's implicit premise that the full covariance is available for visualization; no internal inconsistency or hidden assumption appears in the argument structure.","tokens_in":1710,"tokens_out":268,"duration_ms":15586,"concrete_test":"Construct a 10-point toy dataset with known non-diagonal C (e.g., exponential correlation length 3 bins), draw a model line that lies inside all marginal 1-sigma bars yet yields chi^2 > 20; recompute and overlay the first PC contribution and the conditional error bars; confirm that the mismatch becomes visually apparent.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim follows directly from the definition of the multivariate Gaussian likelihood: the probability of a model prediction vector y_model given data y and full covariance C is exp(-1/2 (y-y_model)^T C^{-1} (y-y_model)), which cannot be recovered from the marginal variances diag(C) alone when off-diagonal elements are non-negligible. The proposed remedies (display of the leading principal component of C and the conditional variances given the model) are standard linear-algebra operations on C and introduce no additional assumptions beyond the premise that C is known to the analyst.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The paper demonstrates that standard vertical error bars (square roots of the diagonal elements of a covariance matrix) are insufficient to judge model-data agreement when uncertainties have non-negligible off-diagonal correlations. It proposes two remedies: displaying the leading principal component of the covariance and showing conditional uncertainties for each data point given a model prediction, both derived from the full covariance matrix under a multivariate Gaussian assumption.","tokens_in":1815,"tokens_out":374,"duration_ms":17999,"significance":"If the visualizations are shown to be effective in examples, the work addresses a genuine and common pitfall in statistical plotting. The remedies follow directly from standard linear-algebra operations on the covariance matrix (principal components and Schur complement for conditionals) without introducing new assumptions beyond the premise that the full covariance is known to the analyst. This could improve interpretability in fields that routinely plot correlated measurements.","major_comments":[{"comment":"The central claim rests on the premise that the full covariance matrix C is known and available; the manuscript should explicitly flag this as a prerequisite (e.g., in the section introducing the conditional-uncertainty display) and discuss what happens when only marginal variances are supplied by an experiment.","section":"Methods / conditional uncertainties"}],"minor_comments":[{"comment":"The abstract states the problem clearly but the manuscript would benefit from a short numerical example (e.g., a 5-point data vector with a non-diagonal C) showing the difference between marginal and conditional error bars.","section":"Abstract"},{"comment":"Notation for the first principal component contribution should be defined once and used consistently when describing the added visual element.","section":"Visualization proposal"}],"recommendation":"minor_revision","confidential_remarks":null},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for the positive evaluation and the constructive suggestion regarding the prerequisite of a known full covariance matrix. We agree that this assumption should be stated explicitly and that the limitations when only marginal variances are available merit discussion. The revision will incorporate these clarifications without altering the core technical content.","responses":[{"response":"We agree with this observation. The conditional-uncertainty visualization relies on the Schur complement of the full covariance matrix C, which presupposes that C is known in its entirety. In the revised manuscript we will insert an explicit statement at the start of the conditional-uncertainties section declaring that the full covariance matrix must be supplied. We will also add a short paragraph noting that, when only the diagonal marginal variances are provided, the conditional uncertainties cannot be evaluated and the analyst must revert to conventional error bars, with the attendant loss of information about correlations. This addition will be placed immediately after the description of the conditional display and will not require any change to the principal-component visualization, which likewise assumes the full matrix.","revision_made":"yes","referee_comment":"[Methods / conditional uncertainties] The central claim rests on the premise that the full covariance matrix C is known and available; the manuscript should explicitly flag this as a prerequisite (e.g., in the section introducing the conditional-uncertainty display) and discuss what happens when only marginal variances are supplied by an experiment."}],"tokens_in":1270,"tokens_out":307,"duration_ms":16333,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The main point lands cleanly: when uncertainties have non-negligible off-diagonals, plotting only the square roots of the diagonal elements leaves you unable to judge model agreement from the figure alone. That follows directly from the multivariate Gaussian likelihood, and the paper states it plainly without overclaiming.","headline":"The paper correctly flags that diagonal error bars hide correlation problems and gives two simple, correct ways to add the missing information from the full covariance.","tokens_in":2261,"tokens_out":132,"would_cite":false,"duration_ms":21800,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":{"model":"grok-4.3","evidence":[],"headline":"Statistical visualization of correlated uncertainties unrelated to RS forcing chain","alignment":"orthogonal","rationale":"The paper's central machinery (Mahalanobis distance on full covariance, PCA decomposition of correlation matrix, conditional variances via (V^{-1})_{ii}^{-1}, correlation lines for neighboring bins) is standard multivariate Gaussian linear algebra for data display. It addresses a practical visualization problem when off-diagonal covariances are non-negligible, but invokes none of the RS structures (J-cost functional equation, φ-ladder, 8-tick periodicity, distinction-to-spacetime forcing). No RS theorem is paralleled or contradicted; the domain (stat.ME data plotting) lies outside the RS canon.","tokens_in":48356,"confidence":"high","tokens_out":161,"duration_ms":14017,"cache_read_input_tokens":38528,"cache_creation_input_tokens":0},"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"When data uncertainties correlate, vertical error bars alone do not show whether a model line fits the points.","keywords":["correlated uncertainties","data visualization","error bars","principal components","covariance matrix","model fitting","conditional uncertainty"],"falsifier":"A concrete data set whose covariance matrix is known, plotted once with only diagonal error bars and once with the principal-component and conditional-uncertainty overlays, where the two versions lead to opposite conclusions about whether a given model line fits the points.","tokens_in":2603,"feed_emoji":"📊","tokens_out":672,"duration_ms":41292,"temperature":0.7,"pith_summary":"The paper establishes that standard error bars, which display only the square roots of the diagonal elements of a covariance matrix, leave insufficient information for judging model agreement whenever off-diagonal correlations are non-negligible. It demonstrates that the usual rule of thumb, that a model fits if it passes through roughly two-thirds of the bars, no longer applies. The proposed remedy is to add explicit display of the leading principal component of the uncertainties together with the conditional uncertainties on each point. A sympathetic reader cares because many real measurements, from time-series data to spectra, carry shared systematic errors that change how one should read a plot. Without this extra information, visual assessment of model quality can be misleading.","feed_headline":"Diagonal error bars hide whether models fit correlated data","feed_subtitle":"Off-diagonal covariances mean standard vertical bars give too little information to judge agreement between points and a drawn line.","key_machinery":"The leading principal component of the uncertainty covariance matrix, which encodes the dominant shared variation across points, together with the conditional uncertainties obtained after removing that component's contribution.","core_discovery":"If the error bars only show the square root of diagonal elements of some covariance matrix with non-negligible off-diagonal elements, we simply do not have enough information in the plot to judge whether a drawn model line agrees well with the data or not. The paper demonstrates this by showing the contribution of the first principal component of the uncertainties and by displaying the conditional uncertainties of all data points.","pith_inferences":["The same display technique could be extended to two-dimensional plots by showing the leading eigenvectors of the joint covariance.","When full covariances are unavailable but partial information exists, approximate conditional bands might still improve judgment over raw diagonal bars.","Routine inclusion of these elements would change how experimental papers present results in fields that routinely share calibration or background errors."],"forward_implications":["Model evaluation must incorporate the full covariance rather than treating points as independent when drawing agreement conclusions from a plot.","Conditional uncertainties reveal which residuals are independent of the dominant shared error, allowing targeted diagnosis of model deficiencies.","Plots that include the first principal component direction make visible the range over which the data can collectively shift without violating the reported uncertainties.","Software that renders data with known covariances should offer the option to overlay these derived quantities by default."],"fun_headline_variants":["Diagonal error bars insufficient with data correlations","Vertical error bars miss covariances in uncertainties","Correlated uncertainties demand more than diagonal bars","Standard error bars hide correlations from model checks"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"The full covariance matrix of the uncertainties is known and available for computing principal components and conditional uncertainties.","fun_headline_variants_meta":{"raw":{"variants":["Diagonal error bars insufficient with data correlations","Vertical error bars miss covariances in uncertainties","Correlated uncertainties demand more than diagonal bars","Standard error bars hide correlations from model checks"]},"model":"grok-4.3","cost_usd":0.004372,"raw_usage":{"total_tokens":2190,"prompt_tokens":667,"num_sources_used":0,"completion_tokens":52,"cost_in_usd_ticks":43724500,"prompt_tokens_details":{"text_tokens":667,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":1471,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":667,"tokens_out":52,"duration_ms":13887,"temperature":1.0,"reasoning_tokens":1471,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-05-16T10:20:47.673263+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"A concrete data set whose covariance matrix is known, plotted once with only diagonal error bars and once with the principal-component and conditional-uncertainty overlays, where the two versions lead to opposite conclusions about whether a given model line fits the points.","supporting_citations":[],"review_version":1}