CoChat
Sign Up
Cited by 25Attention Deficit Hyperactivity DisorderNeural and Behavioral Psychology StudiesFunctional Brain Connectivity Studies

Functional neuroimaging of visuospatial working memory tasks enables accurate detection of attention deficit and hyperactivity disorder

Open access
Hammer, Rubi, Cooke, Gillian E., Stein, Mark A., Booth, James R.·2015·NeuroImage: Clinical
DOI 10.1016/j.nicl.2015.08.015PMC PMC4576365
Open this paper in CoChat — free accountOpen PDF

Abstract

Finding neurobiological markers for neurodevelopmental disorders, such as attention deficit and hyperactivity disorder (ADHD), is a major objective of clinicians and neuroscientists. We examined if functional Magnetic Resonance Imaging (fMRI) data from a few distinct visuospatial working memory (VSWM) tasks enables accurately detecting cases with ADHD. We tested 20 boys with ADHD combined type and 20 typically developed (TD) boys in four VSWM tasks that differed in feedback availability (feedback, no-feedback) and reward size (large, small). We used a multimodal analysis based on brain activity in 16 regions of interest, significantly activated or deactivated in the four VSWM tasks (based on the entire participants' sample). Dimensionality of the data was reduced into 10 principal components that were used as the input variables to a logistic regression classifier. fMRI data from the four VSWM tasks enabled a classification accuracy of 92.5%, with high predicted ADHD probability values for most clinical cases, and low predicted ADHD probabilities for most TDs. This accuracy level was higher than those achieved by using the fMRI data of any single task, or the respective behavioral data. This indicates that task-based fMRI data acquired while participants perform a few distinct VSWM tasks enables improved detection of clinical cases.

AI snapshot

Written from the paper’s full text.

TL;DR

fMRI data from four distinct visuospatial working memory 2-back tasks, analyzed with sparse PCA and logistic regression, classified boys with ADHD combined type from typically developing boys with 92.5% accuracy. This multi-task fMRI model outperformed models based on any single task or on behavioral data.

Key findings

  • The logistic regression classifier using 10 principal components from 16 regions of interest across four VSWM tasks achieved 92.5% accuracy, 95% sensitivity, and 90% specificity; four PCs contributed significantly (PC-2, p < 0.02; PC-3, p < 0.04; PC-4, p < 0.04; PC-8, p < 0.02).
  • Most ADHD boys (75%) received high predicted ADHD probability values (PP > 0.67), and most TD boys (80%) received low values (PP < 0.33); model fit was significant, χ²(4) = 28.52, p < 0.0001.
  • Single-task fMRI classification was generally lower; the exception was the small-reward with feedback task at 80% overall accuracy, but its correct decisive classification accuracy was 60%, significantly lower than the four-task model’s 77.5% decisive accuracy (p < 0.05).
  • Behavioral data from all four tasks yielded 75% accuracy (70% sensitivity, 80% specificity), significantly lower than the four-task fMRI model by 17.5% (p < 0.04); permutation-label models averaged 62.0% accuracy versus 92.5% for the ADHD model (p = 0.000).

Methods

  • Twenty boys with ADHD combined type (mean age 10.42 ± 0.80 years) and 20 typically developing boys (10.96 ± 0.91 years) performed four VSWM 2-back tasks crossing feedback availability (feedback vs. no-feedback) and reward size (large vs. small), each 48 trials long, with fixation baseline tasks.
  • fMRI data were acquired on a 3.0 Tesla scanner; a univariate GLM using the combined sample identified 8 activated and 8 deactivated functional regions of interest (16 ROIs), yielding 64 task-minus-fixation beta-difference features, which sparse PCA reduced to 10 principal components explaining approximately 70% of variability.
  • A logistic regression classifier with leave-one-out cross-validation was used; the four-task model was compared with single-task fMRI models, behavioral models, and permutation tests.

Limitations

  • Stated: the study was limited to feedback and reward manipulations in otherwise similar VSWM tasks; better classification might be achieved with other or more demanding cognitive tasks, and the 48-trial (~100 s) tasks might be shortened. Practical clinical use requires accurate task performance and stillness, so multi-task fMRI may not suit very young children or those with severe cognitive deficits; future work should integrate multi-task fMRI, resting-state fMRI, structural imaging, and other machine-learning methods.
  • Evident from the text: the sample was small and restricted (40 boys, all male, ADHD combined type only), and there was no independent external validation sample. Feature/ROI detection was based on the entire participants’ sample before leave-one-out classification, which may inflate apparent classification accuracy.

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

Integrity signals

Registry

Crossref · 10.1016/j.nicl.2015.08.015

Cites

81 of 81 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

0 of 50 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    There were high correlations between behavioral performances in the four VSWM tasks and PC-2.

    “Moreover, we found high correlations between the behavioral performances in the four VSWM tasks and PC-2 (Fig. 5C).”
    Quote in the paperNot a substantial claim
  • Findingown data

    The classification accuracy based on fMRI data from all four VSWM tasks was higher than that achieved using fMRI data from any single task.

    “This accuracy level was higher than those achieved by using the fMRI data of any single task, or the respective behavioral data.”
    • Also stated in: § ADHD classification based on each single VSWM task · plain
      “ADHD classification accuracies based on the fMRI data from each single task (Fig. 4; Table 1) were significantly lower than the classification accuracy of the model based on all four tasks (Fig. 3).”
    • Also stated in: § Summary · plain
      “Classification accuracies that were based on the fMRI data from all four tasks were significantly higher than those obtained using the fMRI data from any single task.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The four VSWM tasks' reward and feedback manipulations effectively created distinct motivational contexts.
  • Findingreasoning

    Feeding the LR classifier with both behavioral and fMRI data (10 PCs) resulted in exactly the same model as feeding it with only fMRI data.

    “Feeding the LR classifier with both the behavioral and fMRI data (10 PCs) resulted in exactly the same model as when feeding it with only the fMRI data.”
    Quote in the paperNot a substantial claim
  • Findingown data

    The four-task model's accuracy was not statistically different from perfect accuracy reflecting the clinician's diagnosis (100%), p = 0.12.

    “This accuracy level is not statistically different from perfect accuracy reflecting the clinician's diagnosis (100%), p = 0.12 (one-tailed Fisher exact test; testing the hypothesis “LR classification is the same as the clinician diagnosis”).”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The clinician diagnosis is the ground truth (100% accurate) for ADHD.
  • Findingown data

    The classification accuracy based on fMRI data from all four VSWM tasks was higher than that achieved using the respective behavioral data.

    “This overall accuracy level is significantly lower (17.5%) than the accuracy based on the fMRI data from the four tasks, p < 0.04 (one-tailed Fisher exact test; testing the hypothesis “fMRI model is better than the behavioral model”).”
    • Also stated in: § Summary · plain
      “The corresponding behavioral data from the four VSWM tasks enabled an classification accuracy level of 75% (Fig. 5), which is significantly lower than those obtained using the fMRI data from the four tasks.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The clinician diagnosis is the ground truth (100% accurate) for ADHD.
  • May combine two claims. 27
  • Classified as probably not a substantial claim. 21

  • May combine two claims.

    Findingown data

    The mean accuracy of 15 permuted-label models was 62.0%, significantly lower than the ADHD model's 92.5% (p = 0.000).

    “The mean accuracy of 15 distinct permutations (62.0%) was significantly lower than that of the ADHD model (92.5%), p = 0.000 (exact test); t(14) = 14.97, p < 0.0001 (one-tailed, one-sample t-test).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Excluding any of the four significant PCs substantially impaired classification accuracy, while adding any of the other six PCs did not increase accuracy.

    “Excluding each one of these four PCs from the classification model substantially impaired the classification accuracies, whereas adding any of the other six PCs did not increase the classification accuracies.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The classification accuracy based on the small-reward with feedback task alone was 80%, but its correct decisive classification accuracy was 60%, significantly lower than the four-task model.

    “The exception was the classification accuracy based on the fMRI data from the small-reward with feedback task, which was 80%. However, here the correct decisive classification accuracy (with PP <0.33 or PP >0.67) based on the fMRI data from the small-reward with feedback VSWM task was 60%, which is significantly lower (17.5%) than the four-task classification accuracy, p < 0.05 (one-tailed Fisher”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    For the behavioral data model, only 57.5% of classification decisions were both accurate and decisive, with high PP >0.67 assigned to 50% of ADHD boys and low PP <0.33 assigned to 65% of TD boys.

    “Moreover, here only 57.5% of the classification decisions were both accurate and decisive, with high predicted probability values of being a boy with ADHD (PP > 0.67) assigned to only 50% of the ADHD boys, and low values (PP < 0.33) assigned to only 65% of the TD boys (Fig. 5B).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The four significant PCs in the ADHD model are based on fMRI data from all four VSWM tasks, with specific mappings: PC-2 from tasks without feedback; PC-3 from small-reward with feedback; PC-4 from large-reward with feedback; PC-8 from small-reward with feedback.

    “Notably, the four significant PCs encompassed in the ADHD model are based on fMRI data from all four VSWM tasks: PC-2 is based on the two tasks without feedback; PC-3 is based on the small-reward with feedback task; PC-4 is based on the large-reward with feedback task; and PC-8 is based on the small-reward with feedback task (Fig. 2).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The four-task ADHD classification model assigned high predicted ADHD probability values (PP > 0.67) to 75% of ADHD boys and low values (PP < 0.33) to 80% of TD boys.

    “Importantly, most of the classification decisions had high predicted probability values (PP > 0.67) assigned to most ADHD boys (75%), and low values (PP < 0.33) assigned to most TD boys (80%; Fig. 3), indicating a model with an excellent fit (omnibus test for model fit, χ2(4) = 28.52, p < 0.0001).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The univariate GLM analysis found eight fROIs with significant activation (VSWM > fixation) and eight with significant deactivation (fixation > VSWM).

    “We found eight fROIs showing significant activation (VSWM > fixation) and eight fROIs showing significant deactivation (fixation > VSWM).”
    Quote in the paperSubstantial claim

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

    Findingown data

    The use of sparse loading selection thresholding resulted in 39 features (out of 64) affecting the first 10 PCs, with each feature affecting a single PC.

    “The use of this threshold resulted with 39 features (out of 64) affecting the first 10 PCs (PCs with eigenvalues >2), where each feature affected a single PC (Fig. 2B).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Each of the first 10 PCs reflects brain activity from several ROIs from the same VSWM task, but not brain activity in a specific ROI in several tasks, with the exception of PC-2.

    “Interestingly, we found each one of the first 10 PCs to reflect brain activity from several ROIs from the same VSWM task, but not brain activity in a specific ROI in several tasks. The exception was PC-2, which was based on the left and right middle frontal gyri in the two VSWM tasks without feedback.”
    Quote in the paperSubstantial claim

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

    Study design and conductown data

    Mean full-scale IQ scores were within normal range in both groups, but the average IQ in the ADHD group was lower than in the TD group.

    “Mean full-scale IQ scores were within normal range in both groups, however the average IQ in the ADHD group was lower than the average IQ in the TD group (Supplemental Table 1), which is not uncommon in ADHD studies (e.g., Hart et al., 2014a).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Hart et al., 2014a

      Not checked in the free run.

  • May combine two claims.

    Interpretationreasoning

    Lower MFG activity in boys with ADHD during VSWM tasks without feedback may indicate poor VSWM and poor capacity in allocating attention to target stimuli in children with ADHD, manifested when feedback is absent.

    “This may indicate poor VSWM and poor capacity in allocating attention to target stimuli in children with ADHD, manifested when feedback is absent (see Cortese et al., 2012; Fassbender et al., 2011 for related findings).”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Cortese et al., 2012 source

      Not checked in the free run.

    • cites Fassbender et al., 2011 source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    These ADHD characteristics are associated with altered patterns of activity in distinct brain networks: attention and working memory (dorsolateral prefrontal cortex, parietal cortex, temporal cortices); executive functions and cognitive control (dorsal, medial and ventral frontal cortices); and reward-processing (orbitofrontal cortex, anterior cingulate cortex, basal ganglia).

    “These are associated with altered patterns of activity in distinct brain networks: (i) attention and working memory, comprising the dorsolateral prefrontal cortex, parietal cortex and temporal cortices (Burgess et al., 2010; Ehlis et al., 2008; Vance et al., 2007); (ii) executive functions and cognitive control (including feedback processing and response selection), comprising the dorsal, medial”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Burgess et al., 2010 source

      Not checked in the free run.

    • cites Booth et al., 2005 source

      Not checked in the free run.

    • cites del Campo et al., 2013 source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Study design and conductcitation

    A standard jackknife leave-one-out cross validation was used for finding which PCs significantly contributed to classification accuracy.

    “A standard “jackknife” leave-one-out cross validation (Zion Golumbic et al., 2013; Ponce-Alvarez et al., 2012) was used for finding which of the 10 PCs significantly contributed to classification accuracy.”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites Zion Golumbic et al., 2013 source

      Not checked in the free run.

    • cites Ponce-Alvarez et al., 2012 source

      Not checked in the free run.

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

    Study design and conductown data

    The two groups did not differ in patterns of head movements.

    “As reported in Supplemental Table 2, the two groups did not differ in patterns of head movements, and the replacement of outlier images primarily enabled reducing the signal to noise ratio in the fMRI data of both groups.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    The MFG plays an executive role in the visuospatial working memory network and a primary role in volitional (top-down) allocation of attention.

    “The MFG is believed to play an executive role in the visuospatial working memory network, and a primary role in volitional (top-down) allocation of attention (Burgess et al., 2010; Ehlis et al., 2008; Vance et al., 2007).”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Burgess et al., 2010 source

      Not checked in the free run.

    • cites Ehlis et al., 2008 source

      Not checked in the free run.

    • cites Vance et al., 2007 source

      Not checked in the free run.

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

    Study design and conductnone given

    To minimize head movements, gaps between the participant's head and the head-coil were filled with memory foam.

    “To minimize head movements in the scanner, gaps between the participant's head and the head-coil were filled with memory foam.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductreasoning

    Reporting a classification model based only on significant PCs reduces the likelihood of overfitting and provides a more conservative estimate of classification accuracies.

    “The reporting of a classification model based only on significant PCs has two advantages: (i) such a model is less likely to be overfitted or biased due to being based on too many predictors (PCs), and thus it provides a more conservative estimate of the classification accuracies that can be achieved with the data at hand and (ii) it further reduces the number of brain regions by which the two”
    Quote in the paperSubstantial claim

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

    Premisecitation

    There are substantial individual differences in reactions to reward and feedback manipulation in ADHD.

    “Notably, there are substantial individual differences in reactions to reward and feedback manipulation in ADHD (Demurie et al., 2011; Dovis et al., 2015; Hammer et al., 2015; Plichta and Scheres, 2014; van der Schaaf et al., 2013).”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Demurie et al., 2011 source

      Not checked in the free run.

    • cites Dovis et al., 2015 source

      Not checked in the free run.

    • cites Hammer et al., 2015 source

      Not checked in the free run.

  • May combine two claims.

    Study design and conductreasoning

    Using sparse principal component analysis reduced the number of variables fed to the classifier to substantially fewer than the number of participants, excluded lower-weight features likely to add noise, and enabled better determination of underlying neurocognitive mechanisms.

    “This procedure has three major advantages: (i) SPCA enabled reducing the number of variables fed to the classifier to a number substantially smaller than the number of participants; (ii) recalculating the PCs based only on features with the highest loadings enabled excluding lower weight features that were likely to add mostly noise; and (iii) having each PC being affected by relatively few”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    The OFC is part of the ventral attention network, acting as a bottom-up saliency detection system determining subjective and context-dependent susceptibility to unexpected salient stimuli.

    “Together with the inferior frontal gyrus, the anterior insula, the superior temporal cortex and the temporoparietal junction, the OFC is part of the ventral attention network, acting as a bottom-up saliency detection system determining subjective and context-dependent susceptibility to unexpected salient stimuli (Corbetta et al., 2008; Vossel et al., 2014; Weissman and Prado, 2012).”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Corbetta et al., 2008 source

      Not checked in the free run.

    • cites Vossel et al., 2014 source

      Not checked in the free run.

    • cites Weissman and Prado, 2012

      Not checked in the free run.

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

    Study design and conductown data

    Mean total ADHD score was higher in the ADHD group than in the TD group.

    “Mean total ADHD score was higher in the ADHD group than in the TD group (Supplemental Table 1).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisenone given

    Key characteristics of ADHD include poor working memory, greater reliance on external feedback, and abnormal reward processing.

    “Key characteristics of ADHD include poor working memory, greater reliance on external feedback, and abnormal reward processing.”
    Quote in the paperSubstantial claim

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

    Study design and conductnone given

    Only trials in which the participant responded correctly were modeled, and this had only a small quantitative impact on the reported findings.

    “In order to further reduce within scan variability in neural activity, only trials in which the participant responded correctly (hit and correct rejection) were modeled, with onset time-locked to the beginning of each trial (Calhoun et al., 2005; Demir et al., 2014; Hammer et al., 2015; Puschmann et al., 2013). This had only a small quantitative impact on the reported findings.”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Calhoun et al., 2005 source

      Not checked in the free run.

    • cites Demir et al., 2014 source

      Not checked in the free run.

    • cites Hammer et al., 2015 source

      Not checked in the free run.

  • May combine two claims.

    Premisenone given

    ADHD is characterized by behavioral symptoms and multiple cognitive deficits associated with context-dependent abnormal patterns of neural activity distributed across multiple brain regions.

    “ADHD is characterized by behavioral symptoms and multiple cognitive deficits associated with context dependent abnormal patterns of neural activity distributed across multiple brain regions.”
    Quote in the paperSubstantial claim

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

    Study design and conductnone given

    Prior to scanning, children were trained in a mock scanner to confirm they could keep their head still for the scanning session.

    “Prior to the scanning session children were trained in a mock scanner. This enabled confirming that the participant is capable of keeping his head still for the duration of the scanning session.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    Participants with extensive head movements were rescanned or excluded.

    “Participants with extensive head movements were rescanned (or excluded).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S4 (based). May combine two claims.

    Premisereasoning

    The manipulation of feedback and reward provided different motivational contexts, each requiring somewhat distinct executive skills found to be impaired in children with ADHD.

    “The manipulation of feedback and reward provided different motivational contexts, each requiring somewhat distinct executive skills found to be impaired in children with ADHD.”
    Quote in the paperNot a substantial claim

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

    Study design and conductcitation

    ADHD diagnoses were based on exceeding clinical cutoffs on the ADHD rating scale and the K-SADS-PL, and all diagnoses were reviewed by a psychologist specializing in ADHD.

    “ADHD diagnoses were based on exceeding the clinical cut offs on the ADHD rating scale (DuPaul et al., 1998) and the semi-structured diagnostic interview, the Kiddie Schedule for Affective Disorders and Schizophrenia for School Aged Children: Present and Lifetime (K-SADS-PL) version (Kaufman et al., 1997). A psychologist specializing in ADHD reviewed all diagnoses.”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites DuPaul et al., 1998

      Not checked in the free run.

    • cites Kaufman et al., 1997 source

      Not checked in the free run.

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

    Study design and conductnone given

    The order of the four VSWM tasks was counterbalanced across participants.

    “Order of the four tasks was counterbalanced across participants.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Prior neuroimaging-based biomarker studies for ADHD report moderate classification accuracies, rarely exceeding 80%, which is insufficient for clinical diagnosis.

    “These studies provide mixed insights regarding the possible underlying neurocognitive deficit of ADHD, and they report moderate classification accuracies (rarely exceeding 80%), insufficient for clinical diagnosis.”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites Johnston et al., 2014 source

      Not checked in the free run.

    • cites Hale et al., 2015 source

      Not checked in the free run.

    • cites ADHD-200 Consortium, 2012

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Study design and conductown data

    Movement was kept below 4 mm in any x, y, or z dimension within a scan, confirmed using ArtRepair.

    “We confirmed that movement was kept below 4 mm (in any of the x, y, or z dimensions) within a scan using the ArtRepair software.”
    Quote in the paperSubstantial claim

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

    Premisenone given

    ADHD prevalence rates vary significantly between and within countries and depend on the ascertainment method and criteria utilized.

    “ADHD prevalence rates vary significantly between and within countries, and depend on the ascertainment method and criteria utilized.”
    Quote in the paperNot a substantial claim

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

    Premisecitation

    The OFC is involved in monitoring which recent actions were rewarded, and predicting which future actions are most likely to be rewarded.

    “It is suggested that the OFC is involved in monitoring which recent actions were rewarded, and predicting which future actions are most likely to be rewarded (Kahnt et al., 2010).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Kahnt et al., 2010 source

      Not checked in the free run.

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

    Premisecitation

    The left-FFG is involved in letter identification and reading.

    “The left-FFG was found to be involved in letter identification and reading (McCandliss et al., 2003; Dehaene and Cohen, 2011; McNorgan et al., 2013).”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites McCandliss et al., 2003 source

      Not checked in the free run.

    • cites Dehaene and Cohen, 2011

      Not checked in the free run.

    • cites McNorgan et al., 2013 source

      Not checked in the free run.

  • May combine two claims.

    Interpretationown data

    Scanning subjects while they perform an ensemble of distinct cognitive tasks is likely to payoff, enabling greater accuracies and higher confidence diagnosis of clinical cases.

    “This provides a proof-of-concept that scanning subjects while they perform an ensemble of distinct cognitive tasks is likely to payoff, enabling greater accuracies and higher confidence diagnosis of clinical cases.”
    Quote in the paperSubstantial claim

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

    Interpretationnone given

    The current investigation was limited to the manipulation of feedback and reward in VSWM tasks, where many characteristics of the four tasks were identical.

    “Nevertheless, the current investigation was limited to the manipulation of feedback and reward in VSWM tasks, where many characteristics of the four tasks were identical.”
    Quote in the paperNot a substantial claim

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

    Premisecitation

    Children with ADHD exhibit poor cognitive control associated with lower levels of neural activity in the anterior insula, as compared with TDs.

    “Early studies showed that children with ADHD exhibit poor cognitive control associated with lower levels of neural activity in the anterior insula, as compared with TDs (Cubillo et al., 2010; Morein-Zamir et al., 2014).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Cubillo et al., 2010 source

      Not checked in the free run.

    • cites Morein-Zamir et al., 2014 source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    These brain regions (right-MeFG, left-MFG, right-STG, right-AntIns, right supramarginal gyrus) are involved in feedback processing, bottom-up attention, and sensory integration.

    “These brain regions are involved in feedback processing (Ferdinand and Opitz, 2014), bottom-up attention and in sensory integration (Prado et al., 2011; Weissman and Prado, 2012).”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Ferdinand and Opitz, 2014

      Not checked in the free run.

    • cites Prado et al., 2011 source

      Not checked in the free run.

    • cites Weissman and Prado, 2012

      Not checked in the free run.

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

    Premisecitation

    The OFC is involved in the processing of reward-related information.

    “The OFC was also found to be involved in the processing of reward-related information (Schoenbaum and Roesch, 2005; Pauli et al., 2012).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Schoenbaum and Roesch, 2005

      Not checked in the free run.

    • cites Pauli et al., 2012 source

      Not checked in the free run.

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

    Interpretationreasoning

    A multi-task-based fMRI diagnosis session may not be practical for very young children or individuals with severe cognitive deficits.

    “Thus, a multi-task-based fMRI diagnosis session may not be practical for very young children or individuals with severe cognitive deficits.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: Task-based fMRI diagnosis requires the scanned participants to accurately perform a cognitive task while keeping still throughout a relatively long scan.
40 of 40

  • The paper relies on this without support.

    The clinician diagnosis is the ground truth (100% accurate) for ADHD.

    “This accuracy level is not statistically different from perfect accuracy reflecting the clinician's diagnosis (100%), p = 0.12 (one-tailed Fisher exact test; testing the hypothesis “LR classification is the same as the clinician diagnosis”).”

  • The paper relies on this without support.

    The observed group difference in IQ does not account for the fMRI-based ADHD classification.

    “Mean full-scale IQ scores were within normal range in both groups, however the average IQ in the ADHD group was lower than the average IQ in the TD group (Supplemental Table 1), which is not uncommon in ADHD studies (e.g., Hart et al., 2014a).”

  • The paper relies on this without support.

    The four VSWM tasks' reward and feedback manipulations effectively created distinct motivational contexts.

    “The manipulation of feedback and reward provided different motivational contexts, each requiring somewhat distinct executive skills found to be impaired in children with ADHD.”

  • The paper relies on this without support.

    Task-based fMRI diagnosis requires the scanned participants to accurately perform a cognitive task while keeping still throughout a relatively long scan.

    “In order to be of practical clinical use, task-based fMRI diagnosis requires the scanned participants to accurately perform a cognitive task while keeping still throughout a relatively long scan.”

  • The paper relies on this without support.

    Mild ADHD and moderate ADHD are much more common and represent more of a diagnostic dilemma for clinicians.

    “However, mild ADHD and moderate ADHD are much more common and represents more of a diagnostic dilemma for clinicians.”
5 of 5

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?

  • A multimodal logistic regression classifier based on brain activity in 16 regions of interest enabled classification accuracy of 92.5%, with high predicted ADHD probability values for most clinical cases and low predicted ADHD probabilities for most healthy controls.

    “enabled a classification accuracy of 92.5%, with high predicted ADHD probability values for most clinical cases, and low predicted ADHD probabilities for most healthy controls.”

    From Cognitive Neuroscience of Attention Deficit Hyperactivity Disorder (ADHD) and Its Clinical Translation (2018) · cites this paper as Hammer et al., 2015

  • This accuracy level was higher than those achieved by using fMRI data of any single task or behavioral data.

    “This accuracy level was higher than those achieved by using the fMRI data of any single task or the behavioral data (Hammer et al., 2015).”

    From Cognitive Neuroscience of Attention Deficit Hyperactivity Disorder (ADHD) and Its Clinical Translation (2018) · cites this paper as Hammer et al., 2015

Statistics recomputed

Every reported test result, recomputed from its test statistic and degrees of freedom. A result shows as not adding up only once an independent check has confirmed it.

  • Consistentχ²(4) = 28.52reported p < .0001 · recomputed p < .001

    “χ2(4) = 28.52, p < .0001”

  • Consistentt(14) = 14.97reported p < .0001 · recomputed p < .001

    “t(14) = 14.97, p < .0001”

  • Consistentt(14) = 13.59reported p < .0001 · recomputed p < .001

    “t(14) = 13.59, p < .0001”

  • Consistentχ²(4) = 12.37reported p < .01 · recomputed p = .015

    “χ2(4) = 12.37, p < .01”

    The paper states an exact test

4 of 4

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.

  • Significance claim for every single-task comparison contradicts the stated exception

    Minor

    The results identify the small-reward-with-feedback task as an exception to the claim that single-task classification accuracies were significantly lower. The discussion nevertheless says the four-task accuracy was significantly higher than the accuracy from any single task.

1 of 1

Cites

Flagged first, then by citations

2 of the 83 references couldn’t be matched to a known paper.

Cited by

Sorted by citations

Related

Authors
Hammer, R., Cooke, G. E., Stein, M. A., & Booth, J. R.
Affiliations
  • Northwestern University (US)
  • University of Washington (US)
Corresponding author
Hammer, Rubi
ORCID
Venue
NeuroImage: Clinical
Publisher
Elsevier BV
Published
2015
Volume
9
Pages
244-252
ISSN
2213-1582 (ISSN-L)
DOI
10.1016/j.nicl.2015.08.015
PMC
PMC4576365
OpenAlex
W2144003256
Cited by
25
References
83
Citations by year
2025: 1 · 2024: 3 · 2023: 4 · 2022: 2 · 2021: 1 · 2020: 2 · 2019: 1 · 2018: 6 · 2017: 4 · 2016: 1
FWCI
2.98
Type
journal article
Language
en
Open access
Licence
Locations
  • NeuroImage Clinical (publishedVersion, cc-by-nc-nd)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by-nc-nd)
  • PubMed Central (publishedVersion, cc-by-nc-nd)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
NIH (R21-MH080820, T32-NS047987)
Topics
  • Health Sciences › Medicine › Psychiatry and Mental health › Attention Deficit Hyperactivity Disorder
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Neural and Behavioral Psychology Studies
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Functional Brain Connectivity Studies
Keywords
combined type ADHD, visuospatial working memory, functional MRI, neuroimaging biomarkers, ADHD detection, reward magnitude, neurodevelopmental disorders, boys
MeSH terms
Attention Deficit Disorder with Hyperactivity, Brain, Brain Mapping, Child, Female, Humans, Image Processing, Computer-Assisted, Magnetic Resonance Imaging, Male, Memory, Short-Term, Neuropsychological Tests, Oxygen, Photic Stimulation, Visual Perception, Logistic Models, Head Movements, Principal Component Analysis, Feedback, Psychological

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 2026-10-10.

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

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