CoChat
Sign Up
Cited by 373Heart Rate Variability and Autonomic ControlNon-Invasive Vital Sign MonitoringCardiovascular and exercise physiology

Validity of the Polar H10 Sensor for Heart Rate Variability Analysis during Resting State and Incremental Exercise in Recreational Men and Women

Open access
Schaffarczyk, Marcelle, Rogers, Bruce, Reer, Rüdiger, Gronwald, Thomas·2022·Sensors
DOI 10.3390/s22176536PMC PMC9459793
Open this paper in CoChat — free accountOpen PDF

Abstract

Heart rate variability (HRV) is frequently applied in sport-specific settings. The rising use of freely accessible applications for its recording requires validation processes to ensure accurate data. It is the aim of this study to compare the HRV data obtained by the Polar H10 sensor chest strap device and an electrocardiogram (ECG) with the focus on RR intervals and short-term scaling exponent alpha 1 of Detrended Fluctuation Analysis (DFA a1) as non-linear metric of HRV analysis. A group of 25 participants performed an exhaustive cycling ramp with measurements of HRV with both recording systems. Average time between heartbeats (RR), heart rate (HR) and DFA a1 were recorded before (PRE), during, and after (POST) the exercise test. High correlations were found for the resting conditions (PRE: r = 0.95, rc = 0.95, ICC3,1 = 0.95, POST: r = 0.86, rc = 0.84, ICC3,1 = 0.85) and for the incremental exercise (r > 0.93, rc > 0.93, ICC3,1 > 0.93). While PRE and POST comparisons revealed no differences, significant bias could be found during the exercise test for all variables (p < 0.001). For RR and HR, bias and limits of agreement (LoA) in the Bland–Altman analysis were minimal (RR: bias of 0.7 to 0.4 ms with LoA of 4.3 to −2.8 ms during low intensity and 1.3 to −0.5 ms during high intensity, HR: bias of −0.1 to −0.2 ms with LoA of 0.3 to −0.5 ms during low intensity and 0.4 to −0.7 ms during high intensity). DFA a1 showed wider bias and LoAs (bias of 0.9 to 8.6% with LoA of 11.6 to −9.9% during low intensity and 58.1 to −40.9% during high intensity). Linear HRV measurements derived from the Polar H10 chest strap device show strong agreement and small bias compared with ECG recordings and can be recommended for practitioners. However, with respect to DFA a1, values in the uncorrelated range and during higher exercise intensities tend to elicit higher bias and wider LoA.

AI snapshot

Written from the paper’s full text.

TL;DR

In 25 recreational men and women, the Polar H10 chest strap with Elite HRV showed near-perfect agreement with a 12-channel ECG for RR and HR at rest and during incremental cycling, with small bias despite significant exercise differences. DFA a1 agreement was high but weaker, with wider bias and limits of agreement, especially at higher exercise intensities.

Key findings

  • Linear HRV measures showed near-perfect agreement: PRE/POST RR and HR had r = 1.00, rc = 1.00, ICC3,1 = 1.00, and incremental exercise also had r = 1.00, rc = 1.00, ICC3,1 = 1.00. Resting PRE/POST comparisons showed no significant differences, while exercise comparisons showed significant bias for all variables (p < 0.001).
  • DFA a1 correlations were high but comparatively weaker: PRE/POST r > 0.86, rc > 0.84, ICC3,1 > 0.85; incremental exercise r > 0.93, rc > 0.93, ICC3,1 > 0.93, with significant differences during exercise (p < 0.001).
  • DFA a1 bias and limits of agreement were wider than for linear measures: PRE bias −2.1%, POST bias 2.5%; during exercise, low intensity bias 0.9% with LoA 11.6 to −9.9%, and high intensity bias 8.6% with LoA 58.1 to −40.9%.
  • DFA a1 index values were approximately 6% and 7.5% higher for Polar H10 at fixed values of 0.75 and 0.5 compared with ECG, so threshold boundaries were crossed at higher HRs when using Polar H10.

Methods

  • Cross-sectional validation study: 25 participants (men n = 14, age 40 ± 14 years; women n = 11, age 34 ± 10 years) performed an exhaustive cycling ramp with simultaneous Polar H10/Elite HRV and 12-channel ECG recordings before (PRE), during, and after (POST) exercise.
  • RR, HR, and DFA a1 were analyzed in Kubios HRV Premium 3.5.0; resting analyses used a 2-min window, exercise used a 2-min moving window with 20-s grid, and datasets with artefacts >5% were excluded.
  • Agreement was assessed with linear regression, Pearson’s r, Lin’s concordance correlation coefficient, ICC3,1, R², SEE, Bland–Altman LoA, paired t-test or Wilcoxon signed ranks test, and Cohen’s d, with significance at p ≤ 0.05.

Limitations

  • Stated by the paper: the study did not group samples by individual personal characteristics; future studies could examine whether body composition, gender, age, or ventricular characteristics influence outcomes.
  • Stated by the paper: the Polar H10 stores RR intervals but not ECG waveforms, so excessive artefacts cannot be identified or corrected, and noise cannot be differentiated from cardiac arrhythmia; five participants were excluded from exercise analysis and two from resting POST measurement due to artefacts >5%.
  • Stated by the paper: raw data were analyzed with Kubios HRV Premium using automatic artefact correction, and the free standard Kubios version with threshold-based correction could yield slightly different results; the text also makes evident that device and lead placement differences prevent attributing discrepancies to a single variable.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3390/s22176536

Cites

37 of 37 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 18 cited sources checked · 0 supported · 0 flagged

  • Findingreasoning

    The amount of deviation was variable, with some participants showing very close agreement between devices.

    “However, the amount of deviation was variable, with some participants showing very close agreement between devices.”
    Quote in the paperNot a substantial claim
  • Findingreasoning

    Both threshold boundaries were crossed at higher HRs when using the Polar H10.

    “Both threshold boundaries were crossed at higher HRs when using the Polar H10”
    Quote in the paperSubstantial claim
  • Findingown data

    In POST, RR bias was 0.2 to 0.0 ms with upper and lower LoA of 1.6 to 0.5 ms and −1.2 to −0.6 ms.

    “and a bias of 0.2 to 0.0 ms with upper and lower LoA of 1.6 to 0.5 ms and −1.2 to −0.6 ms in POST”
    Quote in the paperNot a substantial claim
  • Findingown data

    During resting state conditions PRE and POST, the linear parameters RR and HR showed nearly perfect correlations (r = 1.00, rc = 1.00, ICC3,1 = 1.00).

    “The comparisons during resting state conditions PRE and POST showed nearly perfect correlations for the linear parameters RR and HR (r = 1.00, rc = 1.00, ICC3,1 = 1.00).”
    Quote in the paperSubstantial claim
  • Findingown data

    DFA a1 bias in PRE was −2.1% (upper LoA 14.7 to 16.4%, lower LoA −18.9 to −20.5%).

    “With respect to DFA a1 there was bias of −2.1% (upper LoA of 14.7 to 16.4% and lower LoA of −18.9 to −20.5%)”
    Quote in the paperNot a substantial claim
  • May combine two claims. 11
  • Classified as probably not a substantial claim. 22

  • May combine two claims.

    Findingown data

    DFA a1 during incremental exercise showed wider bias (0.9 to 8.6%) and LoAs of 11.6 to −9.9% during low intensity and 58.1 to −40.9% during high intensity.

    “DFA a1 showed wider bias (0.9 to 8.6%) and LoAs of 11.6 to −9.9% during low intensity and 58.1 to −40.9% during high intensity.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Findingown data

    An example female participant's data during the incremental cycling ramp was excluded from validation analysis because artifact exceeded >5%, and on closer examination of the parallel recorded ECG these artifacts were classified as frequent atrial premature complexes (APCs).

    “Figure 7 illustrates this with an example of a female participant from the present study whose data during the incremental cycling ramp was excluded from validation analysis since artifact exceeded >5%. On closer examination of the parallel recorded ECG, these “artifacts” were classified as frequent atrial premature complexes (APCs)”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    During resting state conditions PRE and POST, DFA a1 correlations were high but comparatively weaker (r > 0.86, rc > 0.84, ICC3,1 > 0.85).

    “Although correlations were also high for DFA a1 (r > 0.86, rc > 0.84, ICC3,1 > 0.85), they were comparatively weaker”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Bland-Altman plot for RR during incremental exercise showed a mean difference of 0.7 to 0.4 ms with upper and lower LoA of 4.3 to −2.8 ms during low intensity and 1.3 to −0.5 ms during high intensity exercise.

    “Bland–Altman plot for RR showed a mean difference of 0.7 to 0.4 ms with upper and lower LoA of 4.3 to −2.8 ms during low intensity and 1.3 to −0.5 ms during high intensity exercise.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    During incremental exercise, high correlations were found for linear parameters (r = 1.00, rc = 1.00, ICC3,1 = 1.00) and DFA a1 (r > 0.93, rc > 0.93, ICC3,1 > 0.93), although with a statistically significant difference (p < 0.001).

    “High correlations could be found during the incremental exercise test for the linear parameters (r = 1.00. rc = 1.00, ICC3,1 = 1.00), and DFA a1 (r > 0.93. rc > 0.93, ICC3,1 > 0.93), although with a statistically significant difference (p < 0.001)”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The effects of high degrees of artifact correction are demonstrated, showing either data loss or HRV bias.

    “The effects of high degrees of artifact correction are demonstrated, showing either data loss or HRV bias.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Findingown data

    Five participants were excluded from exercise analysis and two participants from resting POST measurement due to artefacts >5%.

    “Five participants were excluded from exercise analysis and two participants from resting POST measurement due to artefacts >5%.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Recent studies show the potential of DFA a1 to demarcate physiological threshold boundaries by fixed DFA a1 values at the aerobic (~0.75) and anaerobic threshold (~0.5).

    “recent studies show the potential of this parameter to demarcate physiological threshold boundaries by means of fixed DFA a1 values at the aerobic (~0.75) and anaerobic threshold (~0.5) for usage in performance monitoring and exercise intensity distribution in endurance sports [16,17].”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [16] Rogers, B., Giles, D., Draper, N., Mourot, L., and Gronwald, T. (2021). Detection of the Anaerobic Threshold in Endurance Sports: Validation of a New Method Using Correlation Properties of Heart Rate Variability. J. Funct. Morphol. Kinesiol., 6. source

      Not checked in the free run.

    • cites [17] Mateo-March, M., Moya-Ramón, M., Javaloyes, A., Sánchez-Muñoz, C., and Clemente-Suárez, V.J. (2022). Validity of Detrended Fluctuation Analysis of Heart Rate Variability to Determine Intensity Thresholds in Professional Cyclists. Eur. J. Sport Sci., 1–20. source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Premisecitation

    The gold standard measure for quantification of RR intervals and assessment of HRV is the electrocardiogram (ECG).

    “The gold standard measure for the quantification of RR intervals and assessment of HRV is the electrocardiogram (ECG) [7].”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [7] Sammito, 2016, Reference Values for Time- and Frequency-Domain Heart Rate Variability Measures, Heart Rhythm. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    In a previous report, the Polar H10 had substantially better signal quality during high-intensity activities compared to a 3-lead ECG Holter monitor.

    “In this report, there was a substantially better signal quality during high-intensity activities for the Polar H10 in comparison to a 3-lead ECG Holter monitor”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [13] Schweizer, 2019, RR Interval Signal Quality of a Heart Rate Monitor and an ECG Holter at Rest and during Exercise, Eur. J. Appl. Physiol. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

    Premisecitation

    Mobile systems and chest straps offer superior practicability with respect to cost, ease of use, portability and interpretation.

    “However, it has become common practice to use mobile systems (e.g., apps, wearables) and chest straps for data recording, storage, analysis and/or export which offers superior practicability with respect to cost, ease to use, portability and interpretation [8,9,10].”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [8] Perrotta, 2017, Validity of the Elite HRV Smartphone Application for Examining Heart Rate Variability in a Field-Based Setting, J. Strength Cond. Res. source

      Not checked in the free run.

    • cites [9] Dobbs, 2019, The Accuracy of Acquiring Heart Rate Variability from Portable Devices: A Systematic Review and Meta-Analysis, Sports Med. source

      Not checked in the free run.

    • cites [10] Zabala, 2022, Validity and Reliability of Different Smartphones Applications to Measure HRV during Short and Ultra-Short Measurements in Elite Athletes, Comput. Methods Programs Biomed. source

      Not checked in the free run.

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    DFA a1 has shown suitability to describe complex cardiac autonomic regulation during various exercise intensities, modalities, and environmental conditions while possessing a wide dynamic range throughout all intensity zones.

    “The non-linear short-term scaling exponent alpha 1 of Detrended Fluctuation Analysis (DFA a1) has shown its suitability to describe the complex cardiac autonomic regulation during various exercise intensities, modalities, and environmental conditions while possessing a wide dynamic range throughout all intensity zones in contrast to standard linear parameters [14,15].”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [14] Gronwald, 2020, Correlation Properties of Heart Rate Variability during Endurance Exercise: A Systematic Review, Ann. Noninvasive Electrocardiol. source

      Not checked in the free run.

    • cites [15] Rogers, 2022, Fractal Correlation Properties of Heart Rate Variability as a Biomarker for Intensity Distribution and Training Prescription in Endurance Exercise: An Update, Front. Physiol. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Exercise-associated DFA a1 dynamics may give insights about physiological status in terms of monitoring fatigue during low-intensity exercise.

    “Additionally, exercise associated DFA a1 dynamics may also give insights about physiological status in terms of monitoring of fatigue during low-intensity exercise [18].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [18] Rogers, 2021, Fractal Correlation Properties of Heart Rate Variability as a Biomarker of Endurance Exercise Fatigue in Ultramarathon Runners, Physiol. Rep. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationcitation

    The basic, cost-free "standard" version of Kubios HRV with an alternate threshold-based artifact correction method could lead to slightly different results than presented.

    “The basic, cost-free “standard” version of Kubios HRV with an alternate threshold-based artifact correction method could lead to slightly different results than presented [24].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [24] Rogers, B., Giles, D., Draper, N., Mourot, L., and Gronwald, T. (2021). Influence of Artefact Correction and Recording Device Type on the Practical Application of a Non-Linear Heart Rate Variability Biomarker for Aerobic Threshold Determination. Sensors, 21. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationreasoning

    It is plausible that some participants had either different or suboptimal signal strength using a chest strap device as opposed to ECG lead 2.

    “Hence, it is certainly plausible that some participants had either different or sub optimal signal strength using a chest strap device as opposed to ECG lead 2.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    It can be considered as a possible limitation that this study did not group the samples by their individual personal characteristics.

    “It can be considered as a possible limitation that this study did not group the samples by their individual personal characteristics.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductcitation

    Data sets with artefacts >5% were excluded from analysis based on [24].

    “Data sets with artefacts >5% were excluded from analysis based on [24].”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [24] Rogers, B., Giles, D., Draper, N., Mourot, L., and Gronwald, T. (2021). Influence of Artefact Correction and Recording Device Type on the Practical Application of a Non-Linear Heart Rate Variability Biomarker for Aerobic Threshold Determination. Sensors, 21. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    Prior to testing, the gas analyzers were calibrated according to the manufacturer's instructions.

    “Prior to testing, the gas analyzers were calibrated according to the manufacturer’s instructions.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Interpretationnone given

    The potential effect of lead placement (chest strap electrodes vs. ECG lead 2) is notable in terms of DFA a1 determination.

    “The potential effect of lead placement (chest strap electrodes vs. ECG lead 2) is also notable in terms of DFA a1 determination.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    Participants were asked to abstain from caffeine, alcohol, tobacco, and vigorous exercise 24 h before testing.

    “Participants were asked to abstain from caffeine, alcohol, tobacco, and vigorous exercise 24 h before testing”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationreasoning

    Commercial use of this mobile-based HRV recording setup for monitoring HRV data during resting and endurance exercise conditions could be justified.

    “Nevertheless, since this mobile-based HRV recording setup displays superior practicability with generally comparable results, its commercial use for the monitoring of HRV data during resting and endurance exercise conditions could be justified.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    Linear HRV measurements derived from the Polar H10 chest strap device show strong agreement and small bias compared with ECG recordings and can be recommended for practitioners.

    “Linear HRV measurements derived from the Polar H10 chest strap device show strong agreement and small bias compared with ECG recordings and can be recommended for practitioners.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S4 (based).

    Interpretationreasoning

    The higher DFA a1 bias and wider LoA may partly be related to expected differences in non-linear HRV associated with ECG lead placement.

    “This may partly be related to the expected differences in non-linear HRV associated with ECG lead placement.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductcitation

    For the 12-channel ECG, lead 2 was used as the comparable lead to the chest strap device.

    “For the 12-channel ECG, lead 2 was used as the comparable lead to the chest strap device [21].”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [21] Jeyhani, V., Mantysalo, M., Noponen, K., Seppanen, T., and Vehkaoja, A. (2019, January 23–27). Effect of Different ECG Leads on Estimated R–R Intervals and Heart Rate Variability Parameters. Proceedings of the 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Berlin, Germany. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    Participants provided written informed consent.

    “provided written informed consent.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductnone given

    Entry criteria included adults (>18 years) of either sex and of any fitness level without previous medical history, current medications or recent illness.

    “Entry criteria included adults (>18 years) of either sex and of any fitness level without previous medical history, current medications or recent illness.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductnone given

    Ethical approval for the study was obtained by the University of Hamburg, Department of Psychology and Movement Science, Germany (reference no.: 2021_400) and is in line with the principles of the Declaration of Helsinki.

    “Ethical approval for the study was obtained by the University of Hamburg, Department of Psychology and Movement Science, Germany (reference no.: 2021_400) and is in line with the principles of the Declaration of Helsinki.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationcitation

    The degree of bias may simply relate to the discordance between actual ECG waveform signal strength and morphology.

    “The degree of bias may simply relate to the discordance between actual ECG waveform signal strength and morphology [21].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [21] Jeyhani, V., Mantysalo, M., Noponen, K., Seppanen, T., and Vehkaoja, A. (2019, January 23–27). Effect of Different ECG Leads on Estimated R–R Intervals and Heart Rate Variability Parameters. Proceedings of the 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Berlin, Germany. source

      Not checked in the free run.

28 of 28

  • The paper relies on this without support.

    To date no validation has been conducted for non-linear measures of HRV.

    “In addition to the requirement to perform population-based validations, to date no validation has been conducted for non-linear measures of HRV.”

  • The paper relies on this without support.

    Researchers and recreational athletes frequently use the Polar H10 device for data collection and rely on appropriate measurements.

    “since researchers and recreational athletes frequently use this device for data collection and therefore rely on appropriate measurements.”

  • The paper relies on this without support.

    The Polar H10 only stores RR intervals and not ECG waveform recordings.

    “Since the Polar H10 only stores RR intervals and not ECG waveform recordings”
3 of 3

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Cites

Flagged first, then by citations

1 of the 38 references couldn’t be matched to a known paper.

Cited by

Sorted by citations

Related

Authors
Schaffarczyk, M., Rogers, B., Reer, R., & Gronwald, T.
Affiliations
  • Universität Hamburg (DE)
  • University of Central Florida (US)
  • MSH Medical School Hamburg – University of Applied Sciences and Medical University (DE)
Corresponding author
Schaffarczyk, Marcelle
ORCID
Venue
Sensors
Publisher
MDPI AG
Published
2022-08-30
Volume
22
Issue
17
Pages
6536
ISSN
1424-8220 (ISSN-L)
DOI
10.3390/s22176536
PMC
PMC9459793
OpenAlex
W4293715556
Semantic Scholar
d503adc11de24a90cadc9e05705b05b448ea1071
Cited by
349
References
38
Citations by year
2026: 132 · 2025: 108 · 2024: 75 · 2023: 32 · 2022: 2
FWCI
37.75
Type
journal article
Language
en
Open access
Licence
Locations
  • Sensors (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Topics
  • Health Sciences › Medicine › Cardiology and Cardiovascular Medicine › Heart Rate Variability and Autonomic Control
  • Physical Sciences › Engineering › Biomedical Engineering › Non-Invasive Vital Sign Monitoring
  • Health Sciences › Medicine › Complementary and alternative medicine › Cardiovascular and exercise physiology
Keywords
heart rate variability, electrocardiogram comparison, RR interval, incremental exercise, measurement bias, limits of agreement, recreational athletes, sex differences
MeSH terms
Bicycling, Electrocardiography, Exercise Test, Female, Heart Rate, Humans, Male, Exercise

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-09.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-09.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review