CoChat
Sign Up
Cited by 116Attention Deficit Hyperactivity DisorderDevelopmental and Educational NeuropsychologyNeural and Behavioral Psychology Studies

Are There Executive Dysfunction Subtypes Within ADHD?

Open access
Roberts, Bethan A., Martel, Michelle M., Nigg, Joel T.·2017·Journal of Attention Disorders
DOI 10.1177/1087054713510349PMC PMC4303540
Open this paper in CoChat — free account

Abstract

Objective: Children with ADHD have heterogeneous behavioral and neuropsychological profiles. The aim of this study was to examine the possible utility of executive function (EF) subtypes within ADHD. Method: Participants were 357 children aged 6 through 13 with a diagnosis of ADHD. Children completed a brief laboratory battery measuring EF, including response inhibition, response variability, speed, and set-shifting. Children also completed standardized intelligence and achievement testing. Results: Two-way cluster analysis of EF profiles of children with ADHD produced a three-cluster solution, labeled poor inhibitory control, poor set-shifting/speed, and intact task performance. Clusters significantly differed in measures of intelligence, academic achievement, and other disruptive behavior and anxiety/mood symptoms. Conclusion: These findings further support the idea that children with ADHD have heterogeneous EF profiles and suggest that the theory of ADHD should consider these individual differences in EF profiles within the ADHD diagnostic category.

Integrity signals

Registry

Crossref · 10.1177/1087054713510349

Cites

84 of 84 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 66 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    Two-step cluster analysis of executive function measures in children with ADHD produced a three-cluster solution as the best fit (BIC=810.3, LL=1.92).

    “For the two-step cluster analysis in SPSS and based on the Log-likelihood distance measure, a 3-cluster provided the best fit to the data (BIC=810.3, LL=1.92).”
    • Also stated in: § Results · plain
      “Two-way cluster analysis of EF profiles of children with ADHD produced a three-cluster solution, labeled poor inhibitory control, poor set-shifting/speed, and intact task performance.”
    Quote in the paperSubstantial claim
  • Findingown data

    About 50% of children in each cluster met criteria for the ADHD combined subtype.

    “As indicated in Table 2, approximately half of each cluster was diagnosed with the ADHD-C subtype, in line with idea that the current DSM-IV ADHD subtypes do not reliably parse EF heterogeneity within ADHD.”
    Quote in the paperSubstantial claim
  • Findingown data

    The second cluster, with 93 children (26% of the sample), had relatively poor response inhibition, labeled poor response inhibition.

    “Cluster 2 (n=93, 26% of the total sample) was characterized by relatively poor response inhibition, henceforth termed “poor response inhibition,””
    Quote in the paperSubstantial claim
  • Findingown data

    The poor set-shifting/speed cluster had significantly lower academic achievement than the intact-performance cluster.

    “Academic achievement was significantly lower in the poor set-shifting/speed cluster compared to the intact performance cluster (p<.01).”
    Quote in the paperSubstantial claim
  • Findingown data

    The first cluster, comprising 65 children (18% of the sample), showed relatively weak performance on Trails A and B, labeled poor speed/set-shifting.

    “Cluster 1 (n=65, 18% of the total sample) was characterized by relatively poor performance on the Trails A and B task, henceforth termed “poor speed/set-shifting.””
    Quote in the paperSubstantial claim
  • May combine two claims. 25
  • Classified as probably not a substantial claim. 16

  • May combine two claims.

    Findingown data

    Intelligence differed significantly across clusters (F[2,351]=7.04, p=.001, η2=.04), with the poor set-shifting/speed cluster scoring lower than the other two.

    “ANOVA indicated significant cluster differences in intelligence (F[2, 351]=7.04, p=.001, η2=.04) with the poor set-shifting/speed cluster exhibiting significantly lower intelligence than the other two clusters (p<.05).”
    Quote in the paperNot a substantial claim

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

    Findingown data

    Descriptively, both executive dysfunction clusters had more symptoms than the intact-performance cluster.

    “although—at a descriptive level—both executive dysfunction groups had more symptoms than the intact task performance cluster.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The two-cluster solution consisted of one intact-performance group and one executive dysfunction group, the latter performing worse on every executive function task than the intact group.

    “It should be noted that the two-cluster solution, supported 43% of the time, was characterized by an intact task performance cluster and an EF dysfunction cluster. The EF dysfunction cluster displayed weaker performance on all executive function tasks as compared to the intact task performance group.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Academic achievement showed significant cluster differences in MANOVA (F[6,430]=2.93, p=.008, η2=.04); follow-up ANOVAs were significant for spelling and math but not reading.

    “MANOVA indicated significant cluster differences in academic achievement (F[6,430]=2.93, p=.008, η2=.04), with significant follow-up ANOVAs for spelling (F [2,216]=6.01, p=.003, η2=.05) and mathematics (F [2, 216]=6.60, p=.002, η2=.06), but not for reading (F [2,216]=2.67, p=.071, η2=.02).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Follow-up ANOVAs revealed no significant cluster differences in inattentive ADHD symptoms or GAD symptoms.

    “but not inattentive ADHD symptoms (F[2, 293]=.705, p=.495, η2=.01) or GAD symptoms (F[2, 290]=.83, p=.439, η2=.01).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The response inhibition cluster showed slightly lower academic achievement than the intact cluster, but the difference was not significant.

    “At a descriptive level, the response inhibition cluster exhibited slightly lower academic achievement than the intact task performance cluster, although this difference was not statistically significant.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Follow-up ANOVAs showed significant cluster effects for ODD, CD, MDD, and hyperactive-impulsive ADHD symptoms.

    “Follow-up ANOVAs revealed significant cluster differences in ODD symptoms (F[2,290]=6.23, p=.002, η2=.04), CD symptoms (F[2,290]=14.23, p=<.001, η2=.09), MDD symptoms (F[2,290]=7.30, p=.001, η2=.05), and hyperactive-impulsive ADHD symptoms (F[2,293]=7.15, p=.001, η2=.05)”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    No cluster differences were found for gender or ethnicity.

    “The clusters did not differ on the variables of gender or ethnicity;”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The intact task performance group did not exhibit academic problems or sub-par intelligence.

    “However, they also did not exhibit academic problems or sub-par intelligence.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Based on the current battery, the EF subtypes were poor set-shifting/speed, poor inhibitory control, and relatively intact task performance.

    “Based on the current battery, these subtypes were poor set-shifting/speed, poor inhibitory control, and relatively intact task performance.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingcitation

    There appear to be at least two EF subtypes within ADHD, and possibly more.

    “In line with prior work (Nigg, Willcutt, Doyle, & Sonuga-Barke, 2005; Willcutt, et al., 2005), there appear to be at least two EF subtypes within ADHD, and possibly more.”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Nigg, Willcutt, Doyle, & Sonuga-Barke, 2005

      Not checked in the free run.

    • cites Willcutt, et al., 2005

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    The poor set-shifting/speed subgroup had lower academic achievement, lower intelligence, and increased ODD and hyperactive-impulsive ADHD symptoms compared to other children with ADHD.

    “The poor set-shifting/speed subgroup experienced lower academic achievement, lower intelligence, and increased ODD and hyperactive-impulsive ADHD symptoms compared to other children with ADHD.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    There may be at least a subgroup of children with ADHD who have intact task performance, though other deficits cannot be ruled out.

    “Thus, there may be at least a subgroup of children with ADHD who have intact task performance, although it should be noted that we cannot rule out that this group had some other type of deficit.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    There may be three EF subgroups within ADHD, and they may exhibit somewhat distinct profiles of comorbid problems, including academic achievement.

    “There may be three EF subgroups within ADHD, and they may exhibit somewhat distinct profiles of comorbid problems, including academic achievement.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    Executive functions show both shared and distinct features, being overlapping yet separable.

    “EF are characterized by unity and diversity, meaning that they are both overlapping, and yet distinct (Miyake et al., 2000)”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Miyake et al., 2000

      Not checked in the free run.

  • May combine two claims.

    Study design and conductcitation

    The STOP task has strong validation, alpha reliability above .80, and assesses suppression of a motor response to a cue.

    “The STOP task has been has been well-validated, has an alpha reliability greater than .80, and is used to measure the ability to suppress a motor response on cue (Band, van der Molen, & Logan, 2003; Logan, 1994).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Band et al., 2003

      Not checked in the free run.

    • cites Logan, 1994

      Not checked in the free run.

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

    Study design and conductcitation

    The WISC-III/IV three-subtest short form is a reliable measure of estimated Full Scale IQ.

    “Children were administered a reliable three-subtest short form of the Wechsler Intelligence Scale for Children-Third Edition (WISC-III, prior to 2003) or Fourth Edition (WISC-IV) comprised of the Block Design, Information, and Vocabulary subtests of the WISC-IV (Sattler, 2008; Wechsler, 1991; 2003).”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Sattler, 2008

      Not checked in the free run.

    • cites Wechsler, 1991

      Not checked in the free run.

    • cites Wechsler, 2003

      Not checked in the free run.

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

    Study design and conductcitation

    The Trail Making Task reflects frontal lobe integrity and shows high inter-rater reliability, with alternate forms reliability between .78 and .92.

    “This measure requires the participant to trace letters and numbers as quickly as possible without errors and is an index of frontal lobe integrity (Reitan, 1958; Spreen & Strauss, 1991) and has demonstrated high inter-rater reliability as well as alternate forms of reliability ranging from .78 to .92 (Bowie & Harvey, 2006).”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Reitan, 1958

      Not checked in the free run.

    • cites Spreen & Strauss, 1991

      Not checked in the free run.

    • cites Bowie & Harvey, 2006

      Not checked in the free run.

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

    Study design and conductnone given

    Interview fidelity for the KSADS-E was monitored by recording sessions and having a certified trainer review 5% of them.

    “Fidelity to interview procedure was checked by having the interview recorded with 5 % reviewed by a certified trainer.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Learning problems and diagnosable learning disorders are common in ADHD, affecting roughly 20 to 30 percent of children with the disorder.

    “ADHD is associated with learning problems and learning disorders with 20 to 30 percent of children with ADHD exhibiting a diagnosable learning disorder (Hinshaw, 1992; Mayes, Calhoun & Crowell, 2000)”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Hinshaw, 1992

      Not checked in the free run.

    • cites Mayes et al., 2000

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Study design and conductown data

    To check the stability of the three-cluster solution, the authors drew 100 random half-samples and reran two-step clustering on each.

    “Further validation of the cluster solution was conducted using a validation method in which 50% of the cases were randomly selected 100 times and two-step cluster analysis was performed on the subsamples.”
    • Also stated in: § Cluster/Profile Solution · plain
      “In order to validate the reliability of the 3-cluster solution using a validation method in which 50% of cases were randomly selected, and two-way cluster analysis was performed 100 times on the subsamples.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    For children on psycho-stimulants, parents were instructed to consult a physician about stopping medication 24 to 48 hours before the lab visit, aiming for a more accurate cognitive assessment.

    “Parents of children taking psycho-stimulant medication were asked to consult with a physician about discontinuing medication for 24 to 48 hours prior to the laboratory visit to ensure a more accurate measure of cognitive performance.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    Difficulties with executive function, especially set-shifting, correlate with poorer academic performance and lower IQ.

    “EF impairment, particularly set-shifting, is associated with academic problems (Biederman et al., 2004; Bull & Scerif, 2001; Gathercole & Pickering, 2000) and lower intelligence (Mahone et al., 2002)”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Biederman et al., 2004

      Not checked in the free run.

    • cites Bull & Scerif, 2001

      Not checked in the free run.

    • cites Mahone et al., 2002

      Not checked in the free run.

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

    Premisecitation

    Anxiety disorders are highly associated with ADHD.

    “Anxiety disorders are also highly associated with ADHD (Angold, Costello, & Erkanli, 1999)”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Angold et al., 1999

      Not checked in the free run.

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

    Study design and conductreasoning

    Shared method variance, in that multiple measures came from the same task, might have influenced cluster characteristics.

    “Additionally, shared method variance (in that multiple measures came from the same task) might have influenced cluster characteristics.”
    Quote in the paperNot a substantial claim

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

    Premisecitation

    Some work has suggested that EF questionnaires may be more sensitive to EF impairments than psychological tests.

    “Further, some work has suggested that EF questionnaires may be more sensitive to EF impairments than psychological tests (Barkley & Murphy, 2010).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Barkley & Murphy, 2010

      Not checked in the free run.

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

    Interpretationreasoning

    EF profiles may have promise as an alternative subtyping approach with important implications for assessment of children with ADHD.

    “EF profiles may have promise as an alternative subtyping approach with important implications for assessment of children with ADHD.”
    Quote in the paperNot a substantial claim

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

    Study design and conductnone given

    The intact task performance group may exhibit EF impairment in areas not assessed in the current study or in other domains not assessed.

    “and may exhibit EF impairment in areas not assessed in the current study or in other domains not assessed in the present study (e.g., relationships, self esteem).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationcitation

    These symptom differences are likely clinically significant because the current DSM subtyping approach relies on small differences in symptoms to determine subtypes or presentations.

    “although these differences are likely to be clinically significant due to the current DSM subtyping approach that relies on small differences in symptoms to determine subtypes, or presentations (APA, 2000; 2013).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites APA, 2000

      Not checked in the free run.

    • cites [2013]

      Not checked in the free run.

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

    Study design and conductnone given

    The use of a 3-subtest IQ test may have led to overestimates of IQ in children with ADHD.

    “Additionally, our use of a 3-subtest IQ test may have led to overestimates of IQ in children with ADHD.”
    Quote in the paperNot a substantial claim

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

    Premisecitation

    Training of executive functioning, such as working memory and speed, has proven promising for some children with ADHD.

    “As an example, training of executive functioning, such as working memory and speed, have proven promising for some children with ADHD (Klingberg, et. al, 2005; Mackey, Hill, Stone & Bunge, 2011).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Klingberg, et. al, 2005

      Not checked in the free run.

    • cites Mackey, Hill, Stone & Bunge, 2011

      Not checked in the free run.

  • May combine two claims.

    Interpretationcitation

    Poor set-shifting ability may limit children's ability to think flexibly and multi-task, which are important for adapting to surroundings.

    “That is, poor set-shifting ability may limit children’s ability to think flexibly and multi-task, important components of their ability to flexibly adapt to their surroundings (Hill, 2004; Wechsler, 1971).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Hill, 2004

      Not checked in the free run.

    • cites Wechsler, 1971

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    The task performance assessment of EF and abbreviated assessment of IQ in the present study may be considered a limitation.

    “Thus, the task performance assessment of EF and abbreviated assessment of IQ in the present study may be considered a limitation.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    The current study's results support the idea that there are EF subtypes within the ADHD diagnostic category and suggest that EF subtypes may merit further attention in assessment and diagnosis of ADHD for DSM-V.

    “The current study’s results support the idea that there are EF subtypes within the ADHD diagnostic category and suggest that the EF subtypes may merit further attention in assessment and diagnosis of ADHD for DSM-V (Nigg, Rohde, & Todd, 2008; Nigg, Willcutt, Doyle, & Sonuga-Barke, 2005).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Nigg, Rohde, & Todd, 2008

      Not checked in the free run.

    • cites Nigg, Willcutt, Doyle, & Sonuga-Barke, 2005

      Not checked in the free run.

  • May combine two claims.

    Interpretationcitation

    This has implications for etiological study of ADHD such that EF impairment could be a primary deficit, or possible endophenotype, in a subgroup of children with ADHD.

    “This—in turn—has implications for etiological study of ADHD such that EF impairment could be a primary deficit, or possible endophenotype, in a subgroup of children with ADHD (Doyle, et al., 2005; Nigg, Willcutt, Doyle, & Sonuga-Barke, 2005).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Doyle, et al., 2005

      Not checked in the free run.

    • cites Nigg, Willcutt, Doyle, & Sonuga-Barke, 2005

      Not checked in the free run.

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

    Premisecitation

    Prior work suggests EF deficits contribute to academic underachievement in children with ADHD and to comorbid symptomatology.

    “These findings are in line with work suggesting that EF deficits contribute to academic underachievement in children with ADHD (Biederman, et al., 2004; Bull, Espy, & Wiebe, 2008), as well as comorbid symptomatology (Coolidge, Thede, & Young, 2000; Fischer, Barkley, Smallish, & Fletcher, 2005).”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Biederman, et al., 2004

      Not checked in the free run.

    • cites Bull, Espy, & Wiebe, 2008

      Not checked in the free run.

    • cites Coolidge, Thede, & Young, 2000

      Not checked in the free run.

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

    Interpretationreasoning

    EF assessment in children with ADHD may allow more tailored academic intervention approaches based on their EF profile.

    “EF assessment in children with ADHD may allow more tailored academic intervention approaches tailored to children’s individual needs, based on their EF profile.”
    Quote in the paperNot a substantial claim
37 of 37

  • The paper relies on this without support.

    The community-recruited sample of children with ADHD is representative of the broader population of children with ADHD, so the identified executive function clusters can be generalized to the ADHD diagnostic category.

    “Participants were 357 children, between the ages of 6 and 13, recruited through the community who completed an extensive, multi-stage, diagnostic screening procedure.”
    “there is support for the existence of EF clusters within the ADHD diagnostic category.”

  • The paper relies on this without support.

    The abbreviated executive function battery used in this study adequately captures the full range of executive function abilities relevant to ADHD subtyping.

    “The present study seeks to further evaluate whether EF subtypes are identifiable within ADHD using an abbreviated battery of EF measures that might be easily utilized in clinical practice.”
    “there appears to be three EF impairment clusters, or subtypes, within the present sample, based on the short EF battery utilized.”
2 of 2

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?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • The sample includes children who did not meet full ADHD criteria

    Minor

    The paper says every included child met DSM-IV criteria for ADHD, but it also says 55 included children had subthreshold symptoms. Those children cannot satisfy both descriptions.

  • Age-test degrees of freedom exceed the sample size

    Minor

    For a three-group ANOVA of 357 children, the residual degrees of freedom cannot exceed 354. The reported age test has 356 residual degrees of freedom, implying 359 observations.

2 of 2

Cites

Flagged first, then by citations

2 of the 86 references couldn’t be matched to a known paper.

Research works printed but not linked: 6

  1. DuPaul G. J., 1998, ADHD Rating Scale IV: Checklists, norms, & clinical interpretation
  2. Reitan R. M., 1986, PsycCRITIQUES, 31
  3. Sattler J. M., 2008, Assessment of children cognitive foundations
  4. Wechsler D., 1991, Manual for the Wechsler Intelligence Scale for Children-Third Edition
  5. Wechsler D., 2001, Wechsler Individual Achievement Test–Second Edition: Examiner’s manual
  6. Wechsler D., 2003, Wechsler Intelligence Scale for children–Fourth Edition: Administration and scoring manual

Cited by

Sorted by citations

Related

Authors
Roberts, B. A., Martel, M. M., & Nigg, J. T.
Affiliations
  • University of Kentucky (US)
  • Oregon Health & Science University (US)
Corresponding author
Roberts, Bethan A.
ORCID
Venue
Journal of Attention Disorders
Publisher
SAGE Publications
Published
2016-07-28
Volume
21
Issue
4
Pages
284-293
ISSN
  • 1087-0547 (ISSN-L)
  • 1557-1246
DOI
10.1177/1087054713510349
PMC
PMC4303540
OpenAlex
W2088570949
Semantic Scholar
be2015f004071bcc87d0f0ef1f5f2595ad6f04db
Cited by
116
References
86
Citations by year
2026: 8 · 2025: 14 · 2024: 12 · 2023: 12 · 2022: 9 · 2021: 8 · 2020: 11 · 2019: 14 · 2018: 10 · 2017: 6 · 2016: 3 · 2015: 5 · 2014: 4
FWCI
2.83
Type
journal article
Language
en
Open access
Licence
journals.sagepub.com/page/policies/text-and-data-mining-license (tdm)
Locations
  • Journal of Attention Disorders (publishedVersion)
  • PubMed (publishedVersion)
  • PubMed Central (acceptedVersion)
Funders
National Institute of Mental Health (R37MH059105, R01MH059105)
Topics
  • Health Sciences › Medicine › Psychiatry and Mental health › Attention Deficit Hyperactivity Disorder
  • Social Sciences › Psychology › Neuropsychology and Physiological Psychology › Developmental and Educational Neuropsychology
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Neural and Behavioral Psychology Studies
Keywords
attention deficit hyperactivity disorder, executive function, response inhibition, set-shifting, processing speed, response variability, clustering, children, intelligence, academic performance, disruptive behavior
MeSH terms
Academic Performance, Achievement, Adolescent, Analysis of Variance, Attention Deficit Disorder with Hyperactivity, Child, Female, Humans, Intelligence, Male, Neuropsychological Tests, Cluster Analysis, Executive Function

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 2026-10-05.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-05.

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