{"id":"bcdddd26-c700-4b29-9865-2c139c6622bf","arxiv_id":"2509.01104","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Turning-angle analysis of Helminthoidichnites tenuis reveals three spatially clustered tortuosity morphotypes on Ediacaran slabs, suggesting centimeter-scale environmental heterogeneity and flexible early bilaterian behavior.","lead":"This paper measured how much Ediacaran fossil trails wiggle in different parts of four ancient seafloor slabs. It found three distinct trail styles in separate spatial patches, which the authors interpret as evidence that early mobile animals sensed and responded to a patchy seafloor environment.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Spatial morphotype clusters are identified via arbitrary thresholds with no null model; chance clustering could produce the three 'morphotypes' and undermine the claim of confirmed external heterogeneity.","rationale":"The reader's advertised weakest assumption concerns the fidelity of tracing and discretization; our concern is different but complementary: the absence of a null model means that even perfectly traced, correctly discretized data could produce apparent morphotype clusters by chance. This is a load-bearing issue for the central claim, and it is addressable through the proposed permutation test. The reader already noted the lack of a null model and multiple-comparison control in the rationale, so we partially agree. Since the reader's verdict is CONDITIONAL and our concern reinforces that condition rather than overturning it, we recommend UNCHANGED: the paper should be accepted only if the authors supply a null-model analysis or temper the claim to 'potential' rather than 'confirmed' heterogeneity.","tokens_in":13040,"tokens_out":4213,"duration_ms":49159,"concrete_test":"Run a permutation test. For each slab, randomly shuffle all turning-angle values across grid squares (preserving the number of turning angles per square and the spatial layout), then apply the same full pipeline: grid at 30 w, F-test matrix, p<0.01 threshold, 33% criterion, shifted-grid concordance, and morphotype labeling. Repeat 1000 times, recording the total number and summed area of candidate morphotype regions. If the observed values exceed the 95th percentile of the null distribution, the spatial clustering is real; if not, the reported morphotypes can emerge from homogeneous behavior.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim—'confirmed heterogeneity in external conditions'—rests entirely on the spatial clustering of turning-angle variances into three morphotypes. However, the pipeline in Methods 5.4 has no null model. The authors reject the null hypothesis of equal variances using F-tests, but then apply arbitrary, data-derived thresholds: p<0.01 and a 33% threshold for 'mostly different' grid squares. The subsequent subdivision into a dominant and two subordinate morphotypes is based on a subjective reading of 'plaid or checkerboard patterning' in the p-value matrix. Crucially, no permutation test, simulation, or random-shuffle control is provided to show that such spatial patches are unlikely under a homogeneous tracemaker population. With 2106 turning angles and many pairwise F-tests, chance alone can easily produce apparent clusters, especially when thresholds are tuned to the data. The grid shift (unshifted vs. shifted by 15 w) and retention of only concordant squares reduce false positives, but do not establish statistical significance. Without a null model, the morphotype regions—and hence the inference of behavioral response to heterogeneous external conditions—may be artifacts of the thresholding procedure.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper analyzes 92 Helminthoidichnites tenuis trails on four slabs from the Ediacara Member, South Australia, by discretizing traced paths into equidistant segments (spacing 1.4 times trail width), computing turning-angle distributions per grid square, and comparing variances with pairwise F-tests. The authors identify three 'morphotypes'—a dominant yellow, a low-variance blue, and a high-variance red—that appear in spatially contiguous regions on three of four slabs. They interpret these regions as behavioral responses to spatially heterogeneous external conditions, concluding that the tracemaker detected and responded to environmental heterogeneity. The central claim is that this 'confirmed heterogeneity' supports the Savannah Hypothesis and Cambrian Information Revolution frameworks.","tokens_in":13367,"tokens_out":3317,"duration_ms":42823,"significance":"If the spatial heterogeneity in trail morphology is real, the paper would provide novel paleobiological evidence for centimeter-scale environmental patchiness in the latest Ediacaran and for behavioral flexibility in early bilaterians, directly relevant to the Savannah Hypothesis and Cambrian Information Revolution. The methodological idea—quantifying spatial variation in trail tortuosity via grid-based variance comparisons—is promising and could be widely applicable to other trace-fossil datasets. The authors also take some care in symmetrizing turning angles and in checking robustness with a shifted grid, which are positive features. However, the statistical basis for the morphotype classification and the causal inference to external conditions is currently not established; the load-bearing claim goes beyond what the data and analyses support.","major_comments":[{"comment":"The central inference of 'confirmed heterogeneity' rests on a thresholding argument with no null model. Pairwise F-tests across many grid squares are subject to multiple testing: under a global null of equal variances, a large number of p<0.01 results is expected by chance. The 33% threshold and the p=0.01 cutoff are selected after inspecting the same data, and no permutation test, simulation, or random-shuffle control is provided to show that the observed spatial clusters are unlikely under a homogeneous tracemaker population. The shifted-grid concordance reduces false positives but does not establish statistical significance. Please provide a null model (e.g., permute turning angles among grid squares while preserving the spatial layout, or simulate homogeneous variances with the same sample sizes) and report the distribution of cluster sizes and numbers under the null.","section":"§5.4, thresholding argument and Results §5.5"},{"comment":"The grid spacing of 30 w (42 mm) is selected after examining the same slabs and is justified partly by the mean size of Dickinsonia costata, not by an independent criterion. A range of grid spacings was tested (5 w, 10 w, 20 w, 30 w, 40 w), but the paper does not report whether the three morphotypes are stable across this range. The 15 w shifted-grid check is a useful robustness test, but it is applied only at the chosen grid spacing. Please report the sensitivity of the morphotype maps and the proportion of candidate grid squares to grid size, and ideally provide a quantitative stability criterion rather than a visual/qualitative selection.","section":"§5.4, grid spacing selection"},{"comment":"A large fraction of bioturbated grid squares are unassigned—38% overall and 65% on Slab C—yet the abstract and conclusion state a confirmed three-morphotype pattern. The unassigned squares are not discussed spatially; if unassigned squares mostly lie adjacent to or interspersed among candidate regions, the apparent clustering could be an artifact of the thresholding. Please report the number and arrangement of unassigned squares, and show that candidate regions are not simply residual patches after arbitrary exclusion. The statement that all three morphotypes occur on three of four slabs is weakened by the high unassigned proportion on Slab C; a robustness analysis that includes or excludes Slab C would clarify the strength of the claim.","section":"Table 5.2 and Results §5.5"},{"comment":"The discretization methodology, the choice of segment spacing (1.4 w), and the 'further clustering' into two subordinate morphotypes are referred to as defined in 'Laing et al. (Chapter 3)' and 'Chapter 4', which are not available to the reader. This makes the analysis unreproducible as presented. The paper must contain, in the main text or supplement, the full definitions of the discretization procedure, the turning-angle calculation, the rationale for 1.4 w, and the clustering algorithm used to separate blue from red morphotypes after the thresholding step. Without this, a reviewer cannot verify that the morphotypes are not an artifact of the discretization or clustering choices.","section":"§5.4, Methods dependencies on 'Chapter 3' and 'Chapter 4'"},{"comment":"The inference from variance differences in turning angles to 'confirmed heterogeneity in external conditions' is underdetermined. The paper does not measure any external variable (e.g., oxygen, nutrient content, sediment texture, microbial mat distribution); it only observes spatial variation in trail morphology. Alternative explanations include tracemaker internal state, ontogenetic or size-dependent behavior, taphonomic differences in trail preservation across the slab, and subjectivity in tracing. The phrase 'confirmed heterogeneity' in the abstract and conclusion overstates what the data can show. The authors should reframe the conclusion as a hypothesis consistent with external heterogeneity, and ideally provide tests that discriminate among alternatives (e.g., compare trail width, relief, or preservation state across morphotype regions; test for association with sedimentary feature","section":"§5.6 and Conclusion §5.7"}],"minor_comments":[{"comment":"Typo: 'a as-yet undiscovered variability' should be 'an as-yet undiscovered variability'; also 'hinders the ability to tests' should be 'to test'.","section":"Abstract"},{"comment":"The text says 'candidate grid squares for all three morphotypes were present on three out of the four slabs' but Table 5.2 shows Slab C has 12% yellow, 12% blue, 12% red, which is consistent; however, Slab C also has 65% unassigned. Please clarify how the percentages in Table 5.2 were computed (e.g., denominator includes only bioturbated squares, and whether 'unassigned' squares were bioturbated but failed thresholds).","section":"§5.5, Results"},{"comment":"The F-test compares variances of symmetrized turning angles. Because the angles are duplicated and sign-flipped, the effective sample size is doubled; the F-test p-values should be based on the actual number of independent turning angles, not the symmetrized count. Please state clearly the sample size used in each test.","section":"§5.4, Methods"},{"comment":"The discussion of chemoreception and favorable/unfavorable gradients is reasonable, but it is speculative. The paper would benefit from a more explicit statement that these are hypotheses rather than demonstrated mechanisms.","section":"§5.6, Discussion"},{"comment":"Several reference typos: 'Biotubation' should be 'Bioturbation' in the Scott et al. reference; 'Poerce-Shimomura' should be 'Pierce-Shimomura'; 'Northen Territory' should be 'Northern Territory'. The reference list also contains an entry for 'Glaessner, M. E. (1959)' but the text cites 'Glaessner, 1958'. Please check consistency.","section":"References"},{"comment":"Figure 5.4 is central but the caption does not explain what the dashed trail lines represent or how the grey shaded slab extent is defined; please add a legend for all symbols. Supplementary Figures 5.5–5.7 are not described in the main text; please refer to them explicitly in the Methods or Results.","section":"Figures"}],"recommendation":"major_revision","confidential_remarks":"This manuscript appears to be a thesis chapter (references to 'Chapter 3' and 'Chapter 4' and figure numbering '5.x'). For a journal submission it must be self-contained. The empirical observation of spatially variable trail morphology is plausible and interesting, but the statistical framework needs substantial strengthening before publication. The absence of any null model for the spatial clustering is the key weakness; without it, the central claim of 'confirmed heterogeneity' is not justified. I would also recommend that the journal require a data and code availability statement, as the analysis is based on manually traced vector paths that are not currently accessible."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bothered to read it, and you should too. The new thing here is real: nobody has tried to map within-slab variation in turning-angle distributions for H. tenuis, and the idea that trail tortuosity can proxy centimeter-scale seafloor heterogeneity is worth taking seriously. The authors traced 92 specimens on four slabs, computed turning angles at a biologically scaled step length, and found spatially coherent patches of low- and high-variance movement. If that holds up, it's a useful behavioral signal from a fossil that usually only gets described qualitatively.\n\nThe paper also does a few things right. They test grid spacing, do a shifted-grid consistency check, and require multiple specimens per candidate region. That shows they know the pitfalls and tried to guard against the worst ones.\n\nThe soft spots are not minor. The central claim—'confirmed heterogeneity in external conditions'—rests entirely on a thresholding procedure that has no null model. The 0.01 p-value cutoff and the 33% rule are chosen after inspecting the same data, then the residual grid squares are carved into two subordinate morphotypes by eye from a 'plaid pattern' in the p-value matrix. With 2106 angles and dozens of pairwise F-tests, chance alone will produce plenty of nominally significant differences. The shifted-grid check helps, but it does not replace a permutation test or any calculation of the expected number of patch boundaries under a homogeneous tracemaker. The word 'confirmed' is doing far more work than the statistics can support.\n\nTwo more things. Key methodology (the 1.4 w spacing rationale) is deferred to thesis chapters the reader cannot see, and no data or code are supplied. That makes this hard to evaluate or reproduce.\n\nNone of this is fatal to the underlying idea. But the analysis needs a proper null model, a multiple-comparison correction or an explicit justification for not using one, and a results section that distinguishes 'we observed spatial clustering' from 'we confirmed external heterogeneity.' I would send this out to peer review, because the question is worth a serious referee, but I'd tell the authors the analysis will need substantial revision before the result is reliable.","headline":"A genuinely new behavioral observation in Ediacaran trails, undermined by a threshold-dependent pipeline with no null model; the 'confirmed heterogeneity' claim goes beyond what the data show.","tokens_in":13815,"tokens_out":1924,"would_cite":false,"duration_ms":23013,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper claims that Helminthoidichnites tenuis trail-makers on Ediacaran seafloor slabs produced three spatially distinct movement patterns, showing they detected and responded to patchy external conditions.","keywords":["Ediacaran","trace fossils","Helminthoidichnites tenuis","movement ecology","turning angles","spatial heterogeneity","bilaterian behaviour","Ediacara Member"],"falsifier":"Re-analyse the same four slabs with a different discretization spacing (for example 0.7 w and 2.8 w instead of 1.4 w) and with many random shifts of the 42 mm grid, keeping the same threshold rules. If the yellow, blue, and red candidate regions do not reappear consistently, the morphotypes are artifacts of segment spacing or grid placement rather than genuine spatial behaviour. Alternatively, if the patch boundaries coincide with sedimentary features such as current lineation or burial drapes, a taphonomic explanation would be supported.","tokens_in":12981,"feed_emoji":"🪱","tokens_out":5593,"duration_ms":64702,"temperature":0.7,"pith_summary":"The paper sets out to test whether late Ediacaran seafloor conditions were spatially heterogeneous by measuring how much, and how often, the makers of the simple horizontal trail Helminthoidichnites tenuis turned while moving. Using 92 trails on four slabs from the Ediacara Member of South Australia, the authors converted photographed paths into discretized movement trajectories and compared turning-angle distributions across a spatial grid. They report three distinct morphotypes—a dominant, low-variance, and high-variance turning pattern—that occur in contiguous, centimetre-scale patches, often involving multiple individuals. They interpret this as evidence that the tracemaker could detect and behaviourally respond to external cues, and that those cues varied across the seafloor. The result matters because it supplies a direct behavioural readout of resource heterogeneity at the time when the Savannah Hypothesis and Cambrian Information Revolution expect heterogeneity to be rising.","feed_headline":"Fossil trails reveal three behaviours in earliest bilaterians","feed_subtitle":"Turning-angle patterns suggest the tracemaker sensed seafloor conditions at centimetre scale.","key_machinery":"The carrying object is the relative turning-angle probability distribution: each trail is discretized at points spaced 1.4 times the trail width, each point's turning angle is recorded, and the distribution of those angles is interpreted as a behavioural descriptor of the movement path. The argument then rests on comparing the variances, not the means, of these distributions across grid squares, using two-sample f-tests after symmetrizing left/right turns. A threshold argument—significantly different from at least 33% of other populated grid squares at p<0.01—isolates grid squares, and shifted and unshifted grids are superimposed to stabilize candidate regions. This machinery converts static","core_discovery":"The central claim is that Helminthoidichnites tenuis trails from the Ediacara Member do not show one uniform movement style: turning-angle variance divides them into a dominant morphotype and two subordinate morphotypes that occupy spatially distinct regions on the studied slabs. Because many specimens switch morphotype along their own length, the tracemaker was behaviourally flexible rather than stereotyped. Because candidate regions often contain multiple specimens, the patchiness is better explained by external conditions than by individual idiosyncrasy. The paper therefore concludes that an early bilaterian-grade animal detected and responded to centimetre-scale heterogeneity in seafloor","pith_inferences":["A natural extension the paper does not make: applying the same turning-angle variance method to early Cambrian traces such as Treptichnus or Psammichnites would test whether path variability increased across the Ediacaran–Cambrian transition.","The specific stimulus behind each morphotype is underdetermined; pairing these trail maps with geochemical or sedimentological maps of the same slabs could separate oxygen, organic carbon, and microbial mat effects.","The three morphotypes could partly be a taphonomic or tracing artifact; a null model that randomly permutes trail locations among grid squares would show whether the contiguous regions are larger than chance.","If the patches are real, they imply the Ediacaran seafloor was behaviourally meaningful at a scale comparable to Dickinsonia body size, giving a concrete substrate on which sensory and navigational selection could act."],"forward_implications":["Helminthoidichnites tenuis tracemakers displayed behavioural flexibility, switching between stereotyped, linear, and tortuous movement modes within single trails.","The spatial restriction of morphotypes to contiguous centimetre-scale regions indicates external seafloor conditions were heterogeneous during deposition of the Ediacara Member.","The absence of tightly meandering or net-like feeding traces suggests the tracemaker used stochastic rather than deterministic navigation, consistent with limited sensory range.","The method provides a template for quantifying spatial variability in other horizontal trace fossils, allowing direct tests of the Savannah Hypothesis and the Cambrian Information Revolution.","Patchiness at the studied scale means matground ecology had incipient heterogeneity, a possible early stage in the feedback loop that later produced Cambrian information processing."],"supporting_citations":[{"why":"Supplies the movement-ecology framework linking path morphology to internal state, navigation, motion capacity, and external factors.","marker":"Nathan et al., 2008"},{"why":"Identifies the possible tracemaker Ikaria warioota and contextualizes H. tenuis behaviour and habitat.","marker":"Evans et al., 2020"},{"why":"Provides the mat-grazing and scavenging interpretation of H. tenuis and the sediment-overburden constraint of less than 15 mm.","marker":"Gehling & Droser, 2018"},{"why":"Supplies the Cambrian Information Revolution hypothesis that spatial heterogeneity drives sensory and behavioural evolution.","marker":"Plotnick et al., 2010"},{"why":"Supplies the Savannah Hypothesis that heterogeneous resource distribution drove early bilaterian infaunalization.","marker":"Budd & Jensen, 2017"},{"why":"Provides the stereotypy–flexibility framework used to interpret the three observed behavioural variants.","marker":"Wainwright et al., 2008"},{"why":"Supplies the chemotaxis mechanism in which increased reorientation frequency indicates responses to increasingly unfavourable stimulus gradients.","marker":"Kirkegaard et al., 2016"},{"why":"Provides a prior spatial-statistical approach to Ediacaran trace fossils and the comparison showing limited resource-focused feeding in Ediacaran ichnotaxa.","marker":"Mitchell et al., 2022"}],"fun_headline_variants":["Trail geometry shows early bilaterians sensed seafloor patchiness","Ediacaran trails hint at three behaviors in primitive bilaterians","Earliest animal trails show same creature switched movement styles","Centimeter-scale seafloor patches shaped Ediacaran trail behavior","Three behavior types in Ediacaran trails imply flexible sensing"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The argument stands or falls on whether the measured turning angles faithfully represent how the animals actually moved; if tracing error, photo obliquity, or the chosen point spacing distorts the variance, the three patchy morphotypes could be artifacts.","fun_headline_variants_meta":{"raw":{"variants":["Trail geometry shows early bilaterians sensed seafloor patchiness","Ediacaran trails hint at three behaviors in primitive bilaterians","Earliest animal trails show same creature switched movement styles","Centimeter-scale seafloor patches shaped Ediacaran trail behavior","Three behavior types in Ediacaran trails imply flexible sensing"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00045,"raw_usage":{"total_tokens":2079,"prompt_tokens":691,"completion_tokens":1388,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":435,"completion_tokens_details":{"reasoning_tokens":1302}},"tokens_in":435,"tokens_out":1388,"duration_ms":13466,"temperature":1.0,"reasoning_tokens":1302,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T12:52:56.217012+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Re-analyse the same four slabs with a different discretization spacing (for example 0.7 w and 2.8 w instead of 1.4 w) and with many random shifts of the 42 mm grid, keeping the same threshold rules. If the yellow, blue, and red candidate regions do not reappear consistently, the morphotypes are artifacts of segment spacing or grid placement rather than genuine spatial behaviour. Alternatively, if the patch boundaries coincide with sedimentary features such as current lineation or burial drapes, a taphonomic explanation would be supported.","supporting_citations":[],"review_version":1}