{"id":"46a96a99-a935-43d2-98a1-00fb0caeb827","arxiv_id":"1908.09988","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A new catalog of ~5,200 radio hotspots from FIRST images is used to measure asymmetry and bending in ~2,100 double-lobed radio sources, confirming the brighter-hotspot-closer trend and a quasar-galaxy difference in bending angle.","lead":"This paper compiles positions and sizes of over 5,200 radio hotspots in about 2,870 radio galaxies and quasars using 1.4 GHz survey images. It confirms that brighter hotspots tend to lie closer to the host galaxy and measures a median bending angle of 4.8 degrees between the two radio arms.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The reported quasar–galaxy difference in bending-angle distributions may be an artifact of unmatched redshift and angular-size samples combined with FIRST resolution, and the paper provides no error analysis to rule this out.","rationale":"The reader's weakest_assumption centers on visual hotspot identification and resolution bias in FIRST images. That is a legitimate concern, and I share it. However, the most load-bearing issue for the paper's headline claim is more specific: even with perfectly identified hotspots, the reported quasar–galaxy BA difference is not interpretable unless the two subsamples are comparable in redshift, angular size, and measurement error. The paper does not provide those distributions or any matched-subpopulation analysis. The internal check on SSL shows the authors thought about whether hotspots are the true ends of sources, but it does not address whether the BA comparison is confounded by sample composition. Because the central claim is a difference between two host-type populations, a matched test or error model is essential before accepting the result. My proposed concrete test directly targets that confound. Since the reader already recommends CONDITIONAL, my concern does not change the verdict, but it sharpens the condition: the quasar–galaxy BA difference should be demonstrated after matching or error modeling. I therefore vote UNCHANGED and not a stronger action, because the concern is testable and the paper may well survive such a test.","tokens_in":5850,"tokens_out":6873,"duration_ms":73973,"concrete_test":"Recompute the BA comparison after matching the quasar and galaxy subsamples on redshift and LAS (or angular arm length), for example by binning in z and LAS and testing within each bin; if the 99.4% significance disappears, the broader quasar distribution is a resolution/selection artifact rather than evidence about jet orientation.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central new physical claim, that the projected bending-angle (BA) distribution is broader for 627 quasar hosts than for 1501 galaxies (§III, §IV), is not supported by the reported statistics because the comparison is uncontrolled for redshift, angular size, and measurement noise. FIRST has a fixed 5.4\" beam, and BA is an acute angle between two host-to-hotspot vectors; a given astrometric error in the host or hotspot position produces a larger angular error when the arm is shorter, i.e. for more distant or more compact sources. The sample spans median z=0.53 with quartiles 0.35–0.73, but the paper does not report the redshift or LAS distributions of the two subsamples, nor does it match them or add a Monte Carlo error model. If quasars in this sample are on average more distant or have smaller angular sizes, the broader BA distribution could be entirely a resolution/measurement effect. The text's own caveat in §IV that a (small) fraction of hotspots may be 'mere brightness peaks in the diffuse lobes' further means that contamination correlated with host type would bias the comparison. The 99.4% confidence statement in §III and the '100.00% confidence' statement in §IV are not accompanied by a test statistic, sample-size-corrected p-value, or description of the method, so the reader cannot separate intrinsic broadening from these confounds.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript compiles positions of 5236 hotspots in 2869 radio galaxies and quasars from FIRST survey images, and derives linear sizes, 1.4-GHz luminosities, armlength ratios, flux ratios, and bending angles for the sources. For a subsample of 2128 sources with one hotspot in each lobe, it reports a median projected bending angle of 4.8°, confirms the known tendency for the brighter hotspot to lie closer to the host, and claims that the bending-angle distribution is significantly broader for 627 quasar hosts than for 1501 galaxies, quoting 99.4% confidence in §III and 100.00% confidence for the related X-distribution in §IV.","tokens_in":6130,"tokens_out":4415,"duration_ms":47553,"significance":"If the catalog were released and the statistical claims substantiated, the compiled sample would be a very large observational resource for studies of radio-galaxy hotspot properties, substantially exceeding earlier samples (e.g., the 65 hotspots in Hardcastle et al. 2004). The paper is observational, uses external FIRST catalog data directly, and explicitly acknowledges several limitations such as selection bias and resolution effects. Its main strength is the size of the sample and the care taken in visual inspection with auxiliary surveys. However, the central physical claim of a quasar-galaxy difference in bending angle is currently not supported by the reported statistics: no test is described, no error model is given, and no control for redshift, angular size, or measurement noise is presented.","major_comments":[{"comment":"The statements that the bending-angle distribution differs between quasars and galaxies at 99.4% confidence (§III) and that the X distribution differs at 100.00% confidence (§IV) are not backed by any named statistical test, test statistic, p-value, or description of how the distributions were compared. Without this information, the central claim cannot be evaluated; the '100.00% confidence' in particular suggests a missing or misreported p-value.","section":"§III and §IV"},{"comment":"The quasar-galaxy comparison of bending-angle distributions is uncontrolled for redshift and angular size. FIRST has a fixed 5.4'' beam, and the bending angle is an acute angle between two host-to-hotspot vectors; an astrometric error of a given angular size produces a larger angular error when the arm is shorter, i.e., for more distant or more compact sources. The paper reports median z=0.53 and median LAS=1.65' for the full sample but does not give these distributions separately for the 627 quasar and 1501 galaxy subsamples, nor does it match them or add a Monte Carlo error model. Until this is shown, the observed broadening could be entirely a resolution/measurement effect.","section":"§II and §III"},{"comment":"The paper acknowledges in §IV that 'a (small) fraction of the hotspots used in the present work' may be 'mere brightness peaks in the diffuse lobes', and in §III that the median hotspot linear size of 34 kpc is nearly twice the 20 kpc found by Hardcastle et al. (2004) with higher resolution. This contamination/resolution issue is load-bearing because the armlength ratios, flux ratios, and bending angles all depend on hotspot positions and sizes; if misidentification correlates with host type (quasar versus galaxy), it could produce the reported distribution difference. The authors should quantify the contamination using higher-resolution images for a control subsample or by repeating the analysis after excluding the largest and most aligned hotspots, which their Figure 3 discussion suggests are the contaminated ones.","section":"§III and §IV"},{"comment":"The sample is selected for radio galaxies that showed at least one hotspot on a FIRST image, and the paper states that 20–25% of non-FR I sources have no hotspot. Because the same FIRST images are used both for selection and for measuring the bending angle, and because the visibility of a hotspot may depend on redshift, angular size, and source type, the relative quasar-galaxy comparison could be biased. The paper should state the precise selection criteria for the 2128-source subsample and perform sensitivity tests, such as varying the inclusion threshold or the SSL limits, to show that the reported differences are robust.","section":"§III"}],"minor_comments":[{"comment":"The abstract says '~2870' radio galaxies while §II gives 2869; please unify the numbers.","section":"Abstract and §II"},{"comment":"Several figures appear without visible axis labels or units in the provided text; for example, the distributions in Figures 5 and 6 need explicit axis labels for ALR and FLR, and Figure 3 needs an angular unit.","section":"Figures"},{"comment":"Please replace '100.00% confidence' with the exact p-value or test statistic, and report it with enough significant figures to be meaningful.","section":"§III"},{"comment":"The paper would benefit from a data availability statement or an electronic catalog of the 5236 hotspots; without this, the claimed 'unprecedented sample' cannot be reused by the community.","section":"General"},{"comment":"Reference formatting is inconsistent in places (e.g., page ranges, spacing, and occasional stray punctuation); a final proofread and consistent journal style would help.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"This manuscript is written as a proceedings-style extended abstract rather than a full research article, and the central statistical analysis is incomplete. The compiled sample is potentially valuable, but the physical claim about quasar-galaxy bending-angle distributions needs to be supported by a proper statistical test and by controls for redshift, angular size, and resolution. I recommend requiring those additions before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe useful thing here is the sample, not the headline claim. The authors assembled 5,236 hotspots from FIRST images across 2,869 radio galaxies and quasars, which dwarfs earlier compilations. If that catalog is released, it becomes a resource for studies of hotspot properties and AGN geometry. The paper also confirms the known tendency for the brighter hotspot to lie closer to the host, and it is honest about FIRST resolution limits, even citing Hardcastle et al. for the factor-of-two difference in median hotspot size.\n\nThe soft spot is the claimed quasar–galaxy difference in bending angle. The authors say the distributions differ at 99.4% confidence, and later at 100.00% confidence for the X parameter, but they never describe the test, the sample-size correction, or the error budget. There are no error bars on the plotted distributions. With the fixed 5.4\" FIRST beam, a small astrometric error in host or hotspot position produces a larger angular error for shorter arms. If the quasar and galaxy subsamples differ in redshift or angular size—likely, given the sample's median redshift of 0.53—the broader quasar distribution could be a resolution artifact. The paper does not report redshift or LAS distributions for the two host-type subsamples, nor does it attempt to match them or run a Monte Carlo error model. The authors' own caveat that some hotspots may be mere brightness peaks in diffuse lobes adds another confounder. So the central new physical claim is not currently supported.\n\nMinor but relevant: the catalog is not included or linked, which matters when the catalog is the main product. The writing is more like a summer-project report than a journal article.\n\nFor a reader, this is a maybe: worth knowing for the sample, but the bending-angle claim needs a proper statistical treatment and a control for resolution before it can be believed. I would send it to peer review if the authors can be pushed to release the catalog and add the missing analysis, because the sample is genuinely new and the claim is testable. As it stands, I would not cite the bending-angle result, but I would cite the catalog if it were public.","headline":"Large new hotspot catalog is the real contribution; the quasar–galaxy bending-angle claim is statistically under-supported.","tokens_in":6621,"tokens_out":3592,"would_cite":false,"duration_ms":36810,"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":"The paper compiles 5,236 hotspots in 2,869 radio galaxies and quasars and finds that the angle between the two radio arms is distributed much more broadly in quasars than in galaxies.","keywords":["radio galaxies","quasars","hotspots","FIRST survey","radio lobe asymmetry","bending angle","jet orientation","radio luminosity"],"falsifier":"Re-measure the same 2,128 two-hotspot sources on 2.5-arcsecond VLASS images: if the median hotspot size falls from 34 kpc toward 20 kpc and the quasar-versus-galaxy bending-angle distributions become indistinguishable, the FIRST-based statistical difference would be shown to be an artifact of resolution.","tokens_in":5676,"feed_emoji":"📡","tokens_out":8185,"duration_ms":77294,"temperature":0.7,"pith_summary":"Radio galaxies shoot paired jets that end in compact knots called hotspots, where the jet slams into surrounding gas. This paper assembles 5,236 such hotspots in 2,869 galaxies and quasars from 1.4-GHz FIRST survey images, which the authors present as the largest catalog of its kind. For the 2,128 sources with one hotspot per lobe, it measures armlength ratios, flux ratios, and the bending angle between the two arms. It confirms the known weak tendency for the brighter hotspot to lie closer to the host, and it finds a median bending angle near 5°, with a significantly broader distribution for quasar hosts than for galaxy hosts. A sympathetic reader would take this as evidence that the orientation of the jets to our line of sight, which differs between quasars and galaxies, shapes the observed geometry.","feed_headline":"5200 hotspots show quasar radio arms bend wider than galaxies'","feed_subtitle":"A new FIRST-survey catalog of hotspot geometry supports the idea that quasar jets point closer to us.","key_machinery":"The working tool is the two-arm radio geometry: for each source, draw vectors from the host galaxy to the hotspot in each lobe. From those come the armlength ratio (ALR, host-to-brighter divided by host-to-fainter distance), the flux ratio (FLR, farther divided by nearer hotspot flux), and the bending angle (BA, the acute angle between the two arm vectors). A second set of diagnostics, the separation quotient Q and the fractional separation difference X, converts arm asymmetries into estimates of hotspot advance speed. Hotspot sizes and luminosities come from the FIRST catalog's deconvolved major axis and integrated flux, combined with host redshifts. The comparison of hotspot-to-arm misalignment angles separates aligned knots from perpendicular bow-shock candidates.","core_discovery":"The paper's central claim is that a large, uniformly processed hotspot sample can be built from FIRST survey images, and that this sample exposes a clean geometric difference between quasar hosts and galaxy hosts. For the 2,128 radio sources with exactly one hotspot in each lobe, the median projected bending angle between the two host-to-hotspot arms is 4.8°; the 1,501 galaxies have a median of 4.6° and the 627 quasars a median of 5.1°, but the shapes of the two distributions differ at 99.4% confidence, with quasars showing a much broader spread. The paper also confirms the known weak tendency for the brighter hotspot to lie closer to the host: the median armlength ratio is 0.91 and the median flux ratio is 0.815. It reports a median hotspot linear size of 34 kpc and a median fractional hotspot size of 0.062, both larger than values from higher-resolution observations, which it attributes to FIRST's 5.4-arcsecond resolution blending hotspots with diffuse lobe emission.","pith_inferences":["Editorial inference: if the quasar-galaxy difference in bending angles is mostly projection, then the intrinsic, deprojected bending distributions may be similar; matching quasar and galaxy subsamples by redshift and radio luminosity would test this.","Editorial inference: the correlation between larger hotspot size and smaller misalignment with the arm suggests that some large 'hotspots' are actually line-of-sight superpositions on lobe emission, so a size cut could yield a cleaner geometric sample.","Editorial inference: the same catalog could test whether the brighter-hotspot-closer asymmetry strengthens with redshift, which would separate light-travel-time effects from Doppler beaming explanations."],"forward_implications":["A catalog of 5,236 hotspots gives enough sources to stack infrared, optical, and X-ray images and recover average spectral energy distributions for hotspots that are individually undetected in those bands.","The broader bending-angle distribution of quasars, if orientation-driven, makes quasars useful for studying jet bending under weaker projection corrections than galaxy hosts.","The inflated median hotspot size of 34 kpc relative to higher-resolution values indicates that 5.4-arcsecond data cannot measure hotspot sizes reliably, and the higher-resolution VLASS survey should separate true hotspots from diffuse lobe peaks.","The confirmed brighter-hotspot-closer trend across about 2,100 sources provides a statistical target for models of relativistic beaming and light-travel-time asymmetries in double radio sources."],"supporting_citations":[{"why":"Provides the FIRST 1.4-GHz images and catalog from which all 5,236 hotspot positions, sizes, and fluxes are taken.","marker":"Helfand et al. 2015"},{"why":"Supplies the higher-resolution comparison median hotspot size of 20 kpc against which the paper's 34 kpc median demonstrates resolution bias.","marker":"Hardcastle et al. 2004"},{"why":"Gives the earlier claim that quasar fractional-separation-difference X has a minimum near zero, which the paper does not confirm.","marker":"Best et al. 1995"},{"why":"Introduces the separation quotient Q used to estimate hotspot advance speed from arm asymmetries.","marker":"Longair & Riley (1979)"},{"why":"Introduces the fractional separation difference X that the paper uses to compare the quasar and galaxy distributions.","marker":"Banhatti (1980)"},{"why":"Gives the high-resolution fractional hotspot size of about 0.01 used as a baseline for the paper's FIRST-based 0.062.","marker":"Hardcastle et al. 1998"},{"why":"Defines the FR II edge-brightened class of radio sources to which these hotspots belong.","marker":"Fanaroff & Riley (1974)"},{"why":"Coined the term 'hotspot' for the compact lobe knots studied here.","marker":"Gull & Northover (1973)"}],"fun_headline_variants":["Quasar jets bend wider than galaxies' in 5200 hotspots","FIRST survey: 5200 hotspots confirm brighter jets sit closer","Median 4.8° bend in radio galaxy hotspot arms from FIRST","Hotspot geometry: quasars show broader bending than galaxies"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The results rest on the assumption that the compact bright spots chosen by eye in 5.4-arcsecond FIRST images are genuine jet-termination hotspots, so resolution blurring and occasional misidentified lobe peaks do not systematically bias the measured sizes, arm ratios, and bending angles.","fun_headline_variants_meta":{"raw":{"variants":["Quasar jets bend wider than galaxies' in 5200 hotspots","FIRST survey: 5200 hotspots confirm brighter jets sit closer","Median 4.8° bend in radio galaxy hotspot arms from FIRST","Hotspot geometry: quasars show broader bending than galaxies"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000627,"raw_usage":{"total_tokens":2872,"prompt_tokens":891,"completion_tokens":1981,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":507,"completion_tokens_details":{"reasoning_tokens":1907}},"tokens_in":507,"tokens_out":1981,"duration_ms":14490,"temperature":1.0,"reasoning_tokens":1907,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T10:55:20.135711+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Re-measure the same 2,128 two-hotspot sources on 2.5-arcsecond VLASS images: if the median hotspot size falls from 34 kpc toward 20 kpc and the quasar-versus-galaxy bending-angle distributions become indistinguishable, the FIRST-based statistical difference would be shown to be an artifact of resolution.","supporting_citations":[],"review_version":1}