{"id":"5fcea318-94fc-43b0-946d-8dd5c32d541b","arxiv_id":"2505.07737","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Edge-on galaxy orientations from a sample of 1,689 objects show no strong alignment with the Local Supercluster plane, except a weak 2-sigma perpendicular preference for galaxies with stellar mass below about 10^8.7 solar masses.","lead":"Using 1,689 edge-on galaxies from modern catalogs, this paper re-tests whether galaxy spins align with the plane of the Local Supercluster, an idea from tidal torque theory. It finds no strong alignment overall, but a weak, roughly 2-sigma tendency for low-mass galaxies to spin perpendicular to that plane.","discovery_kind":"replication","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The faint-galaxy alignments are 2-sigma p-values from a post-hoc mass split with no multiple-comparison correction; a permutation test is required to establish they are not chance.","rationale":"The paper's central novel finding is the tentative mass-dependent alignment of faint galaxy spins. For that claim to hold, the reported p-values must be meaningful; this requires that the chosen mass split and the set of subsample tests were not optimized after inspecting the data. The analysis reports p = 0.024 and p = 0.041 for faint galaxies but selects the 10^8.7 M_sun threshold from the same dataset and runs many overlapping tests. Even under a perfectly isotropic null, the probability of seeing a p < 0.05 somewhere in roughly 20 correlated tests is large, and one of the four faint cells (n=61, p=0.279) is not significant. The authors' own caveat in Section 4 confirms this. I therefore consider the most load-bearing concern to be the uncorrected trial count rather than the PA-selection assumption; the latter is important and the reader is right to flag it, but it applies symmetrically to the robust null result and is harder to evaluate from the paper alone. A permutation test that preserves the selection pipeline is the single check that would resolve the trial-count question. If the observed signal survives the permutation null, the selection-effect question should then be addressed by a cross-catalog comparison; if not, the positive claim should be downgraded to an observationally motivated hint. The larger-sample null result is less affected and justifies keeping the reader's CONDITIONAL verdict rather than rejecting.","tokens_in":6995,"tokens_out":16082,"duration_ms":177011,"concrete_test":"Monte Carlo permutation test: take the final sample, preserve each galaxy's sky position, mass, color, redshift, and plane-layer membership, and randomly reassign the observed Supergalactic PAs among galaxies. For each of 10^4 permutations, repeat the exact pipeline—including the choice of the M* = 10^8.7 M_sun split and formation of the same subsamples in Table 2—and record the minimum faint-subsample KS p-value (or the number of faint tests with p < 0.05). Compare the observed minimum (0.024) to this null distribution; if it falls above the 5th percentile, the claimed alignment is consistent with chance and should be reported only as a non-significant hint.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's strongest positive claim—that M* < 10^8.7 M_sun edge-on galaxies have spins preferentially perpendicular to the Local Supercluster plane—rests on the four KS p-values in Table 2 (0.062, 0.279, 0.024, 0.041). The mass threshold was chosen post-hoc to split the sample into two roughly equal parts, and the p-values are drawn from many overlapping tests (2 redshift ranges × 2 plane layers × 5 subsamples). No multiple-comparison correction is applied; the paper notes the subsamples are not independent but does not quantify the effective number of trials. Under the global null of uniform PAs, a random realization is expected to produce several p < 0.05 among correlated subsamples, and the authors' own Section 4 caveat ('cannot be ruled out that this is a consequence of some statistical fluctuation') concedes this. The selection-effect null discussed by the reader is also necessary, but it is a standard assumption for all-sky PA studies; the uncorrected trial count is the least secure element for the paper's headline finding.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper re-examines the claim of Navarro et al. (2004) that edge-on galaxies in the Local Supercluster have spins preferentially aligned with the Supergalactic plane (i.e., disk axes perpendicular to the plane), using a modern sample of 1689 edge-on galaxies drawn from EGIPS, EGIS, and HyperLeda within czLG < 3600 km/s. The authors convert position angles to Supergalactic coordinates, split the sample by redshift layer, distance from the Supergalactic plane, stellar mass, and color, and test uniformity of the PA distribution with one-sample Kolmogorov-Smirnov tests. They find no significant alignment for the full sample or most subsamples, contradicting Navarro et al. (2004), but report that galaxies with stellar mass M* < 10^8.7 M_sun show a weak tendency to be perpendicular to the Local Supercluster plane, with KS p-values 0.024-0.062 in three of four configurations. The paper is transparent about the weakness of the signal and explicitly notes that a statistical fluctuation cannot be ruled out.","tokens_in":7217,"tokens_out":4880,"duration_ms":50786,"significance":"If the low-mass perpendicular-alignment signal is real, it is a tentative mass-dependent spin orientation that points opposite to filament-alignment trends found in recent surveys, and the non-confirmation of the Navarro et al. (2004) result is itself of interest for tidal-torque and galaxy-formation studies. The paper's strengths are its use of substantially larger, modern catalogs, the simple and appropriate application of KS tests, and the explicit caveats about the fragility of the claimed effect. The main limitation is that the headline finding rests on p-values near the 2-sigma level drawn from a large set of overlapping post-hoc subsamples, so the central claim is not yet convincing as stated.","major_comments":[{"comment":"The central positive claim rests on four KS p-values for the faint subsample (0.062, 0.279, 0.024, 0.041) obtained after many overlapping tests (2 redshift ranges x 2 plane layers x 5 subsample definitions) and after choosing the mass and color split points to divide the sample into roughly equal halves. Under the global null of uniform PAs, several p<0.05 values among correlated subsamples are expected, and the paper does not quantify the effective number of independent trials. A permutation or bootstrap procedure that preserves all cuts and evaluates the distribution of the minimum p-value across the full table is needed to support the claim that the faint-galaxy signal is not a chance fluctuation.","section":"Section 3 / Table 2"},{"comment":"The conclusion is internally inconsistent: Section 4 states that 'we do not find any statistically significant alignment of spins to the Local Supercluster plane either among the nearest galaxies ... or in any other subsample' and immediately adds 'except for a subsample of faint galaxies.' It also states the effect is observed 'at a significance level of 2.4-6.2%,' which ignores the p=0.279 for the czLG<1200, ±5 Mpc faint subsample. The paper should report unambiguously that only three of the four faint configurations show p<0.07, that only two are below p<0.05, and that one configuration is consistent with the null.","section":"Section 4 / Abstract"},{"comment":"The KS null hypothesis is that the Supergalactic position angles are uniformly distributed, but the catalog selection functions could violate this assumption. The text acknowledges that EGIPS undercounts large and low-surface-brightness galaxies and that the HyperLeda subsample required subjective visual inspection, while EGIPS and EGIS have different sky footprints. If incompleteness depends on galaxy orientation or position angle relative to the survey footprints, the PA distribution could be biased. The authors should provide a test or argument that selection effects are isotropic in Supergalactic PA, for example by comparing the PA distribution of a control sample of all galaxies (not just edge-on) or by repeating the analysis with orientation-independent weights.","section":"Section 2 / Section 3"},{"comment":"The mass threshold M* = 10^8.7 M_sun and color threshold (g-i)_0 = 0.86 are introduced solely to split the sample into approximately equal parts. The paper does not justify these thresholds a priori, and the claimed effect is not tested for sensitivity to the exact threshold. A more robust approach would be to treat stellar mass as a continuous variable (e.g., sliding-window or regression on mass) and to check whether the PA anisotropy varies smoothly with mass, or to present results for several explicitly justified thresholds.","section":"Section 3 / Section 4"}],"minor_comments":[{"comment":"The phrase 'at the 2-sigma level' is imprecise because KS p-values are not directly equivalent to Gaussian significance levels; reporting the p-values themselves is preferable.","section":"Abstract / Section 4"},{"comment":"The statement that the effect is observed 'within the ±2 and ±5 Mpc layers' is not supported by the p=0.279 value in Table 2 for the czLG<1200, ±5 Mpc faint subsample; this should be corrected.","section":"Section 4"},{"comment":"Table 1 reports unique additions from each catalog, but the matching procedure (e.g., angular separation and velocity tolerance for cross-identification) is not described; a sentence on the matching rules would clarify the sample construction.","section":"Section 2"},{"comment":"The comparison with Welker et al. (2020) and Kraljic et al. (2021) concerns alignment with filaments, whereas the present analysis measures the angle with respect to the Supergalactic plane; the connection between these geometries should be discussed explicitly to avoid overinterpreting the contradiction.","section":"Section 4 / References"},{"comment":"Typographical and minor errors: 'Vigro' should be 'Virgo', 'Aito ff' should be 'Aitoff', and 'de Vaucouleur's' should be 'de Vaucouleurs'.","section":"Various"},{"comment":"The reference to Arakelyan et al. (2024) is cited as 'upcoming,' but it is an arXiv preprint; please clarify its status in the final version.","section":"Section 4 / References"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is within the scope of the journal and the non-confirmation of the Navarro et al. (2004) signal is interesting, but the headline faint-galaxy alignment is currently a marginal, multiple-testing-prone result. The issues raised in the major comments are addressable with additional statistical analysis; I therefore view this as a major revision rather than a rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The useful thing here is the null result. Navarro et al. (2004) claimed edge-on galaxies with cz<1200 km/s are preferentially perpendicular to the Local Supercluster plane, based on 30 galaxies. This paper brings 1689 edge-on galaxies from EGIPS, EGIS, and HyperLeda and finds no significant alignment in any broad subsample. That's a solid, if unsurprising, correction to the literature. The faint-galaxy subtrend (M*<10^8.7) is the only positive signal: 2-sigma perpendicular alignment in three of four slices, but the p-values are 0.024-0.062 and one slice fails. I don't buy it as a detection.\n\nThe analysis is straightforward and mostly clean: the KS test against uniform PA is the right tool for this design, and the authors are transparent about the weakness of the signal. They also do the sensible thing of excluding the Virgo virial zone.\n\nThe soft spots are exactly where the reader put them. The mass split at 10^8.7 is chosen after looking at the data to make two roughly equal halves. With eight or more overlapping tests (two redshift ranges, two layers, five subsamples), a handful of p<0.05 is expected by chance. The paper acknowledges non-independence but never quantifies the effective number of trials. A permutation test, shuffling PA within subsamples or rerandomizing the mass split, would settle whether this is real. The selection-effect null—that PA is isotropic under the null—is also worth a check, since EGIPS and HyperLeda have their own geometric biases, though that concern is standard in this field and I'd call it minor rather than fatal.\n\nAnother issue: no final sample or code is provided. For a paper whose headline is a marginal subtrend, the reproducibility case is weaker than it should be. Not a reason to desk-reject, but a reason to require the sample as supplementary material.\n\nWho is this for? Anyone working on spin alignment in the local universe. It's a useful data point, mainly as a larger null test. I'd send it to review, but with the expectation that the faint-galaxy claim needs a proper multiple-comparison treatment before it can be called robust. If the authors do a permutation test and it holds, the paper becomes more interesting; if it doesn't, the null result still stands.","headline":"Useful null result on a classic claim, plus a 2-sigma faint-galaxy subtrend that probably won't survive a proper multiple-comparison test.","tokens_in":7741,"tokens_out":2442,"would_cite":true,"duration_ms":23398,"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":"Using 1,689 edge-on galaxies, this paper finds no significant spin alignment with the Local Supercluster plane except a 2-sigma perpendicular tendency in galaxies with stellar mass below about 10^8.7 solar masses.","keywords":["galaxy spins","edge-on galaxies","Local Supercluster","Supergalactic coordinates","tidal torque theory","Kolmogorov-Smirnov test","low-mass galaxies","galaxy alignment"],"falsifier":"Generate a mock population of edge-on galaxies with randomly oriented spins, run them through the same Pan-STARRS/SDSS detection and visual-inspection pipeline, and compare the recovered Supergalactic PA distribution; if the pipeline itself produces a non-uniform PA distribution similar to the faint-galaxy signal, the reported perpendicular alignment is a selection artifact rather than a cosmic signal.","tokens_in":6815,"feed_emoji":"🌌","tokens_out":6313,"duration_ms":57091,"temperature":0.7,"pith_summary":"The paper re-examines whether galaxy spins are aligned with the plane of the Local Supercluster, following a method proposed by Navarro et al. (2004) that reduces the problem to measuring the position angles of edge-on disk galaxies in Supergalactic coordinates. Using a sample of 1,689 edge-on galaxies drawn from modern catalogs, it finds that the distribution of these angles is statistically indistinguishable from random for essentially every subsample defined by redshift, distance from the Supergalactic plane, color, and stellar mass. The sole exception is a subsample of faint galaxies with stellar mass below about $10^{8.7} M_\\odot$, whose spin axes show a tendency to point perpendicular to the Local Supercluster plane at roughly 2-$\\sigma$ significance. If the effect is real, it would contradict both the original 30-galaxy detection and the expectation from filament studies that low-mass galaxy spins align along large-scale structure. The broader message is that spin alignment with the Local Supercluster plane is not a robust, easily reproduced signal.","feed_headline":"Faint galaxies alone show perpendicular spin alignment in Local Supercluster","feed_subtitle":"A 2-sigma signal in low-mass galaxies is the only deviation from randomness in 1,689 edge-on disks.","key_machinery":"The load-bearing object is the Supergalactic position angle (PA) of an edge-on disk galaxy: for such galaxies the inclination is fixed near 90 degrees and the orientation of the spin projected on the sky is directly read from the major-axis PA, with no need for a near/far side determination. The analysis converts PAs into Supergalactic coordinates, folds angles into the range $[0^\\circ, 90^\\circ)$, and applies a one-sample Kolmogorov-Smirnov test under the null hypothesis that PAs are uniformly distributed. The sample is assembled from three catalogs (EGIPS, EGIS, and HyperLeda), filtered by velocity relative to the Local Group centroid, morphological type, and visual inspection, then split into redshift layers and $\\pm2$/$\\pm5$ Mpc slices around the Supergalactic plane.","core_discovery":"The central claim is that the spins of nearby edge-on galaxies are not preferentially aligned with the Local Supercluster plane, contrary to the result of Navarro et al. (2004). The only departure from randomness appears in galaxies with stellar mass $M_* < 10^{8.7} M_\\odot$, where the disk major axes tend to lie in the Supergalactic plane, so the spin axes point perpendicular to it; this subsample fails the Kolmogorov-Smirnov uniformity test at the 2.4-6.2% level (about 1.9-2.3 $\\sigma$) in three of four considered spatial/redshift windows. The paper argues this is a weak but fairly robust detection, while cautioning that the samples are not fully independent and that the effect disappears in one of the four windows. It also proposes that the general absence of alignment may reflect complex time-varying tidal fields at filament intersections and sheet-like structures, which can randomize spin orientations.","pith_inferences":["A decisive check would be to measure the same Supergalactic PA distribution using an independent edge-on catalog built from another wide-field survey; if the faint-galaxy perpendicular signal reproduces there, the catalog-selection explanation becomes much less likely.","The mass threshold near $M_* \\sim 10^{8.7} M_\\odot$ is close to the scale where galaxy formation efficiency and angular momentum acquisition change character, so the authors' result, if confirmed, would connect the spin-alignment signal to the transition between different accretion regimes.","The paper's null result for the full sample is consistent with the idea that supercluster-scale alignment is washed out by the complex tidal field at knots and sheet-filament junctions; this could be tested in simulations by comparing spin orientation relative to the local shear tensor rather than to the global Supergalactic plane."],"forward_implications":["The Navarro et al. (2004) perpendicular-alignment signal is not reproduced with roughly an order of magnitude more galaxies, so the earlier detection was not robust to sample size and selection.","Low-mass galaxies in the Local Supercluster may have a spin orientation opposite to the filament-aligned trend reported for low-mass galaxies in other studies, suggesting the local sheet environment, not filament membership, governs their angular momentum.","Future spin-alignment studies targeting the Local Supercluster should treat stellar mass as a primary variable and should test the $\\pm5$ Mpc / $cz_{\\rm LG}<1200$ km/s window where the faint-galaxy signal vanishes before claiming a universal alignment.","If the faint-galaxy perpendicular tendency is genuine, it implies that the densest nearby structure imprints a measurable, mass-dependent anisotropic angular momentum field even when the overall population appears isotropic."],"supporting_citations":[{"why":"Supplies the edge-on-galaxy method and the 30-galaxy perpendicular-alignment result that this paper re-tests.","marker":"Navarro et al. (2004)"},{"why":"Supplies the EGIPS catalog, the primary source of edge-on galaxies from the Pan-STARRS1 survey.","marker":"Makarov et al. (2022)"},{"why":"Supplies the EGIS SDSS edge-on disk sample used to fill gaps in the EGIPS catalog.","marker":"Bizyaev et al. (2014)"},{"why":"Supplies the HyperLeda database and the additional galaxy sample requiring visual inspection.","marker":"Makarov et al. (2014)"},{"why":"Supplies the 50 Mpc Galaxy Catalog used for photometric separation by color and luminosity and for layer membership relative to the Supergalactic plane.","marker":"Ohlson et al. (2024)"},{"why":"Defines the Local Group centroid reference frame used for the velocity cuts $cz_{\\rm LG}$.","marker":"Karachentsev & Makarov (1996)"},{"why":"Provides the comparison result that low-mass galaxies align along filaments, which the faint-galaxy perpendicular finding contradicts.","marker":"Welker et al. (2020)"},{"why":"Provides the morphological-segregation result in filament alignment relevant to interpreting the mass dependence.","marker":"Kraljic et al. (2021)"}],"fun_headline_variants":["Low-mass galaxies alone twist spins in Local Supercluster","2-sigma spin alignment only in faint edge-on galaxies","Local Supercluster galaxy spins mostly random, faint ones align","Faintest galaxies show weak perpendicular spin bias","Edge-on galaxies: faint ones defy Local Supercluster spin pattern"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The entire analysis rests on the assumption that the measured Supergalactic position angles of the catalogs' edge-on galaxies would be uniformly distributed under the null, meaning that none of the survey geometry, neural-network detection, or visual inspection steps imprints a preferred orientation.","fun_headline_variants_meta":{"raw":{"variants":["Low-mass galaxies alone twist spins in Local Supercluster","2-sigma spin alignment only in faint edge-on galaxies","Local Supercluster galaxy spins mostly random, faint ones align","Faintest galaxies show weak perpendicular spin bias","Edge-on galaxies: faint ones defy Local Supercluster spin pattern"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000566,"raw_usage":{"total_tokens":2658,"prompt_tokens":894,"completion_tokens":1764,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":510,"completion_tokens_details":{"reasoning_tokens":1685}},"tokens_in":510,"tokens_out":1764,"duration_ms":10964,"temperature":1.0,"reasoning_tokens":1685,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T22:09:10.738382+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Generate a mock population of edge-on galaxies with randomly oriented spins, run them through the same Pan-STARRS/SDSS detection and visual-inspection pipeline, and compare the recovered Supergalactic PA distribution; if the pipeline itself produces a non-uniform PA distribution similar to the faint-galaxy signal, the reported perpendicular alignment is a selection artifact rather than a cosmic signal.","supporting_citations":[{"cited_title":"F., Abadi, M","cited_arxiv_id":null,"evidence_quote":"Supplies the edge-on-galaxy method and the 30-galaxy perpendicular-alignment result that this paper re-tests."},{"cited_title":"V ., Kautsch, S","cited_arxiv_id":null,"evidence_quote":"Supplies the EGIS SDSS edge-on disk sample used to fill gaps in the EGIPS catalog."},{"cited_title":"C., Gallo, E., Baldassare, V","cited_arxiv_id":null,"evidence_quote":"Supplies the 50 Mpc Galaxy Catalog used for photometric separation by color and luminosity and for layer membership relative to the Supergalactic plane."}],"review_version":1}