Working memory and short-term memory deficits in ADHD: A bifactor modeling approach

May combine two claims.

Reliability and validity evidence for the working memory tasks includes internal consistency (α=.82-.97), 1- to 3-week test-retest reliability (.76-.90), and expected relations with criterion working memory complex span (r=.69) and updating tasks (r=.61).

“Reliability and validity evidence includes internal consistency (α=.82-.97; Kofler, Sarver et al., 2018), 1- to 3-week test-retest reliability (.76-.90; Sarver et al., 2015), and expected magnitude relations with criterion working memory complex span (r=.69) and updating tasks (r=.61; Wells et al., 2018).”

• cites Kofler, Sarver et al., 2018Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Sarver et al., 2015 sourceNot checked — no full text of the source
• cites Wells et al., 2018 sourceSupported

May combine two claims.

Support for the bifactor model includes replicated evidence that phonological and visuospatial storage systems are functionally and anatomically distinct and that a single domain-general central executive exists rather than separate central executives for each modality.

“Support for this model includes replicated evidence that the phonological and visuospatial storage systems are functionally and anatomically distinct, as well as replicated evidence supporting a single, domain-general central executive rather than separate central executive components for processing phonological and visuospatial information (Baddeley, 2007; Alloway et al., 2006; Smith et al., 1996”

• cites Baddeley, 2007 sourceNot checked — no full text of the source
• cites Alloway et al., 2006 sourcePartly supported

  Split into 2 atoms — 1 confirmed, 1 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

  Experimental — accuracy not yet measured
• cites Smith et al., 1996 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S1 (asserted). May combine two claims.

Previous evidence suggested that short-term memory may be relatively less impaired than working memory in pediatric ADHD and that such deficits, if present, do not appear to covary with ADHD symptom severity.

“Previous evidence suggested that short-term memory may be relatively less impaired than working memory in pediatric ADHD and that such deficits, if present, do not appear to covary with ADHD symptom severity (for review, see Chacko et al., 2014).”

• cites Chacko et al., 2014 sourceSupported

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

It remains possible that short-term memory deficits would be detected or larger on tasks requiring simple storage without active mental manipulation, although such tasks would likely underestimate working memory.

“Thus, although results were generally consistent with findings from alternative assessment methods (e.g., simple span, immediate free recall; Dovis et al., 2013; Gibson et al., 2018), it remains possible that short-term memory deficits would be detected/larger on tasks that require simple storage without active mental manipulation of the internally-stored information, although at the same time”

• cites Dovis et al., 2013 sourceNot checked — no full text of the source
• cites Gibson et al., 2018Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Wells et al., 2018 sourceSupported

May combine two claims.

The study did not evaluate maintenance of information in short-term memory over extended durations or assess the episodic buffer.

“Similarly, the current study did not evaluate the ability to maintain information in short-term memory over extended durations (Bolden et al., 2012) or assess the episodic buffer.”

• cites Bolden et al., 2012 sourceSupported

May combine two claims.

ADHD was associated with a unique but significantly smaller visuospatial short-term memory impairment, with d=0.60 and 38% impaired.

“There was also evidence for a unique, albeit significantly smaller, impairment in visuospatial short-term memory (d=0.60; 38% impaired); however, visuospatial short-term memory abilities did not covary with ADHD symptom severity.”

May combine two claims.

In the current sample, ADHD status was linked to very large central executive working memory impairments, with effect sizes from 1.63 to 2.03 and 75% to 81% of cases impaired.

“ADHD status was associated with very large magnitude impairments in central executive working memory that are present in most pediatric cases (d=1.63–2.03; 75%−81% impaired), and these deficits covaried with ADHD inattentive and hyperactive/impulsive symptom severity based on both parent and teacher report.”

• Also stated in: § Conclusion · boosted

  “These findings provide strong evidence that ADHD is associated with marked central executive working memory deficits that covary with their behavioral symptom presentation across settings.”

May combine two claims.

For attention problems, visuospatial short-term memory had significantly weaker relations than both working memory and phonological short-term memory (all p<.04), while working memory and phonological short-term memory did not differ significantly for parent- (r=-.28 vs. -.25; p=.77) or teacher-reported attention problems (r=-.31 vs. -.12; p=.07).

“In terms of associations with parent- and teacher-reported attention problems, visuospatial short-term memory showed significantly weaker relations relative to both working memory and phonological short-term memory (all p<.04); the difference between working memory and phonological short-term memory did not reach significance for parent- (r= −.28 vs. −.25; p=.77) or teacher-reported attention”

May combine two claims.

The Tier 2 bifactor-(s-1) model had excellent fit, all indicators loaded significantly on their hypothesized factors, and fit improved significantly over the Tier 1 single-factor model (all p<.001).

“Results indicated excellent model fit, all indicators loaded significantly onto their hypothesized factor(s), and model fit was significantly improved relative to the Tier 1 single-factor model (all p<.001; Table 2).”

May combine two claims.

Reliability was high for the general factor (ω=.88) and for both specific factors (ωs=.76-.84).

“Reliability was high for the general factor (ω=.88) and both specific factors (ωs=.76–84).”

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

The ADHD group had slightly lower IQ estimates than the non-ADHD group (102.7 vs. 107.7; p=.02).

“There was no significant evidence to indicate between-group differences in socioeconomic status (p=.44), whereas the ADHD group was slightly younger (10.1 vs. 10.5; p=.05) and had slightly lower IQ estimates (102.7 vs. 107.7, p=.02).”

May combine two claims.

Neither phonological short-term memory (r=-.25 to .26, all p>.10) nor visuospatial short-term memory (r=-.01 to .12, all p>.30) was associated with parent- or teacher-reported ADHD symptom severity.

“In contrast, neither phonological short-term memory (r= −.25 to .26, all p>.10) nor visuospatial short-term memory (r= −.01 to .12, all p>.30) were associated with ADHD symptom severity based on parent or teacher report.”

May combine two claims.

In Tier 1, all eight indicators loaded significantly on the domain-general working memory factor, with standardized loadings from .40 to .81 and all p<.001.

“In Tier 1, all 8 indicators loaded significantly onto the domain-general working memory factor (β=.40-.81, all p<.001) and the model showed adequate fit (Table 2).”

May combine two claims.

The proportion of uncontaminated correlations was below .70 (PUC=.68), supporting multidimensionality of the data.

“The proportion of uncontaminated correlations was <.70, supporting the multidimensionality of the data (PUC=.68; Rodriquez et al., 2016; Watkins, 2017).”

• cites Rodriquez et al., 2016Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Watkins, 2017Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

The ADHD group was impaired on all eight phonological and visuospatial working memory performance variables, with effect sizes from d=0.87 to 1.65 and all p<.001.

“The ADHD group demonstrated impairments on all 8 phonological and visuospatial working memory performance variables (d = 0.87–1.65; all p<.001).”

May combine two claims.

ADHD status was significantly more strongly associated with working memory than with both phonological and visuospatial short-term memory (p<.0001), and the latter two did not differ (p=.15).

“Results of the tests of dependent correlations revealed that ADHD status showed a significantly higher association with working memory than it did with both phonological and visuospatial short-term memory (p<.0001), which did not differ (p=.15; WM>PH=VS).”

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

The ADHD group was slightly younger than the non-ADHD group (10.1 vs. 10.5 years; p=.05).

“There was no significant evidence to indicate between-group differences in socioeconomic status (p=.44), whereas the ADHD group was slightly younger (10.1 vs. 10.5; p=.05) and had slightly lower IQ estimates (102.7 vs. 107.7, p=.02).”

May combine two claims.

For hyperactive/impulsive symptoms, working memory had significantly stronger relations than both phonological and visuospatial short-term memory (all p<.004), which did not differ (both p>.18).

“In terms of associations with parent- and teacher-reported hyperactive/impulsive symptoms, working memory showed significantly stronger relations relative to both phonological and visuospatial short-term memory (all p<.004), which did not differ (both p>.18; WM>PH=VS).”

May combine two claims.

The model explained 53% of total variance, leaving 47% unique variance, which highlights task impurity and the need for multiple indicators.

“Total variance explained by the model was 53%, whereas unique (unexplained) variance was 47%, highlighting task impurity and the importance of using multiple indicators for valid assessment of neurocognitive processes (i.e., 47% of the variance in test scores was not attributable to the constructs of interest).”

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

There was no significant evidence of a between-group difference in socioeconomic status (p=.44).

“There was no significant evidence to indicate between-group differences in socioeconomic status (p=.44), whereas the ADHD group was slightly younger (10.1 vs. 10.5; p=.05) and had slightly lower IQ estimates (102.7 vs. 107.7, p=.02).”

May combine two claims.

The general working memory factor explained 75% of common variance (ECV=.75), while the specific factors explained 25% (phonological ECV=.12, visuospatial ECV=.13), indicating modest but substantive reliable variance for short-term memory factors.

“The general working memory factor explained 75% of the common variance (ECV=.75) vs. 25% for the specific factors (phonological ECV = .12, visuospatial ECV = .13), indicating that the short-term memory factors explained modest but substantive portions of the reliable variance in task performance (Canivez, 2015).”

• cites Canivez, 2015Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

Rapport et al. (2008) found ADHD was associated with very large working memory impairments (d=2.76), large visuospatial short-term memory impairments (d=0.89), and medium phonological short-term memory impairments (d=0.55).

“Using these methods, Rapport et al. (2008) found that ADHD was associated with very large impairments in working memory (d=2.76), large impairments in visuospatial short-term memory (d=0.89), and medium impairments in phonological short-term memory (d=0.55).”

• cites Rapport et al., 2008Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

Shared variance across phonological and visuospatial working memory tasks can be attributed to the domain-general central executive, while unique variance in each task can be attributed to the corresponding short-term memory system.

“Thus, shared variance across phonological and visuospatial working memory tasks can be attributed to the domain-general central executive (referred to as ‘working memory’ in the current study), whereas unique variance in each task can be attributed to phonological and visuospatial short-term memory, respectively.”

May combine two claims.

Gibson et al. (2018) found children with ADHD had large working memory impairments (d=0.73-1.12) but small or no differences in phonological (d=0.38) or visuospatial short-term memory (d=0.29, ns).

“Consistent with regression-based methods, they found that children with ADHD displayed large magnitude impairments in working memory (d=0.73–1.12) but showed small magnitude impairments or did not differ from controls in terms of phonological (d=0.38) or visuospatial short-term memory (d=0.29, ns).”

• cites Gibson et al., 2018Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

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

The Baddeley (2007) model has considerable empirical support and has been widely used in ADHD research.

“the Baddeley (2007) model has considerable empirical support and has been used extensively in the ADHD literature”

• cites Baddeley, 2007 sourceNot checked — no full text of the source
• cites Kofler et al., 2019Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Rapport et al., 2013 sourceNot checked — no full text of the source

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

Early meta-analytic estimates found medium-sized working memory impairments in ADHD, with d=0.55-0.63.

“early meta-analytic estimates indicated medium magnitude impairments (d=0.55–0.63; Willcutt et al., 2005)”

• cites Willcutt et al., 2005 sourceNot checked — no full text of the source

May combine two claims.

The immediate free recall method instructs participants to use a specific strategy, which may affect performance and produces relatively high rates of excluded data, such as 21% of the ADHD group in Gibson et al. (2018).

“However, a primary limitation of this method is that it involves explicitly instructing participants to adopt a specific strategy for completing the task, which may impact children’s performance and produces relatively high rates of excluded participant data (e.g., 21% of the ADHD group in Gibson et al., 2018).”

• cites Gibson et al., 2018Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

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

Those early estimates imply that only about 35%-40% of children with ADHD have working memory impairments.

“which suggest that only 35%−40% of children with ADHD are likely to have working memory impairments based on converting effect sizes into proportion non-overlap (Zakzanis, 2001)”

• cites Willcutt et al., 2005 sourceNot checked — no full text of the source
• cites Zakzanis, 2001 sourceNot checked — no full text of the source

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

Meta-analytic estimates suggest ADHD may have smaller phonological than visuospatial short-term memory deficits, d=0.47 vs. 0.85.

“and/or smaller phonological than visuospatial short-term memory deficits (d=0.47 vs. 0.85 in Martinussen et al., 2005)”

• cites Martinussen et al., 2005 sourceNot checked — no full text of the source

May combine two claims.

Residual phonological and visuospatial short-term memory scores correlate negatively with each other, which appears to contradict their structural and functional independence.

“In other words, the residual phonological and visuospatial short-term memory scores will correlate negatively with each other – a finding that appears contradictory to theory and evidence for their structural and functional independence (Baddeley, 2007; Gibson et al., 2017).”

• cites Baddeley, 2007 sourceNot checked — no full text of the source
• cites Gibson et al., 2017Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

Because ADHD also involves impaired working memory, short-term memory estimates from simple span tasks may be inflated, since even simple span tasks require some controlled-attention working memory processes.

“However, because ADHD is also associated with impaired working memory processes, these short-term memory estimates may be inflated because even simple span tasks (and potentially all cognitive tasks) require at least some working memory processes associated with controlled attention (e.g., Engle et al., 1999).”

• cites Engle et al., 1999Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

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

The abbreviated evaluation subgroup had slightly lower BASC inattention T-scores and SES than the full evaluation subgroup.

“The abbreviated subgroup had, on average, slightly lower BASC inattention T-scores (M=48.0 vs. 55.3, p = .001) and SES (M=45.5 vs. 53.3, p = .01).”

May combine two claims.

Age, gender, and socioeconomic status were controlled in all analyses, and the pattern and interpretation of results did not change when these covariates were removed.

“Age, gender, and SES were controlled in all analyses; the pattern and interpretation of results is unchanged if these covariates are removed.4”

May combine two claims.

The visuospatial short-term memory finding is highly consistent with the meta-analytic estimate of d=0.63 for visuospatial working memory from tasks challenging visuospatial storage/rehearsal, and it extends prior findings by controlling for domain-general working memory.

“This finding is highly consistent with the meta-analytic estimate of d=0.63 for visuospatial working memory obtained primarily from tasks that challenge visuospatial storage/rehearsal (Willcutt et al., 2005), and extends prior findings via improved control for domain-general working memory abilities.”

• cites Willcutt et al., 2005 sourceNot checked — no full text of the source

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

The ADHD group had a higher proportion of ODD cases than the Non-ADHD group.

“The clinically-evaluated Non-ADHD cases did not differ significantly from the ADHD group in the proportion of children diagnosed with anxiety (p=.45), depression (p=.59) or ASD (p=.97); the ADHD group had a higher proportion of ODD cases as expected (p=.03).”

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

In the current sample, internal consistency was .81 for the phonological task and .87 for the visuospatial task.

“Internal consistency in the current sample was .81 (phonological) and .87 (visuospatial).”

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

All ADHD subtypes or presentations were eligible because ADHD subtypes are unstable.

“All ADHD subtypes/presentations were eligible given the instability of ADHD subtypes (Valo & Tannock, 2010).”

• cites Valo & Tannock, 2010 sourceNot checked — no full text of the source

May combine two claims.

Set size 3 from each task was chosen as reference facets because Baddeley's conceptualization holds that central executive demands remain relatively constant as set size increases, while increasing set size mainly manipulates short-term memory demands.

“We chose set size 3 from each task given Baddeley’s (2007) conceptualization that central executive demands remain relatively constant despite increasing set size; increasing set size is viewed as primarily a manipulation of short-term memory demands (see Kofler et al., 2010 for empirical support for this conceptualization in an ADHD sample).”

• cites Baddeley, 2007 sourceNot checked — no full text of the source
• cites Kofler et al., 2010Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

The current effect sizes and impairment estimates are significantly higher than omnibus meta-analytic estimates of d=0.55-0.74 and impairment estimates of about 30%.

“At the same time, these effect sizes and impairment estimates are significantly higher than omnibus meta-analytic estimates of d=0.55–0.74 (Kasper et al., 2012; Willcutt et al., 2005) and impairment estimates of approximately 30% (Coghill, Seth, & Matthews, 2014; Wahlstedt, Thorell, & Bohlin, 2009).”

• cites Kasper et al., 2012 sourceNot checked — no full text of the source
• cites Willcutt et al., 2005 sourceNot checked — no full text of the source
• cites Coghill, Seth, & Matthews, 2014 sourceNot checked — no full text of the source

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

The K-SADS allows differential diagnosis according to symptom onset, course, duration, quantity, severity, and impairment based on DSM-5 criteria.

“The K-SADS (2013 Update) allows differential diagnosis according to symptom onset, course, duration, quantity, severity, and impairment in children and adolescents based on DSM-5 criteria (APA, 2013)”

• cites Kaufman et al., 1997 sourceNot checked — no full text of the source
• cites APA, 2013Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

Applying the bifactor model allowed the study to address prior limitations and to fractionate the working memory system into latent, process-pure estimates of phonological and visuospatial short-term memory.

“By applying the bifactor model (Eid et al., 2018a), we were able to address these limitations and fractionate the working memory system to obtain latent, process-pure estimates of both phonological and visuospatial short-term memory.”

• cites Eid et al., 2018aNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• AssumptionRests on an assumption: The tests used in the study are well-validated and the bifactor modeling maximally isolated reliable variance for each short-term/working memory component.

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

Fractionating test performance into reliable variance linked to all three primary components of the Baddeley model gives an ideal test of whether ADHD is tied to specific short-term memory deficits after accounting for working memory impairments.

“By fractionating test performance into reliable variance associated with all three primary components of the Baddeley (2007) model, this approach provides the ideal test of the extent to which ADHD is associated with specific short-term memory deficits when accounting for their well-documented impairments in working memory, as recommended (Gibson et al., 2017, 2018).”

• cites Gibson et al., 2017, 2018Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

The bifactor-(s-1) model permits maximal separation of the constructs of interest and directly tests how far working memory test scores reflect reliable variance from domain-general central executive processing and from domain-specific phonological and visuospatial storage/rehearsal processes.

“Thus, the bifactor-(s-1) model allows maximal discrimination between our constructs of interest, therefore providing a direct test of the extent to which working memory test scores reflect reliable variance attributable to both domain-general central executive processing (working memory) and domain-specific phonological and visuospatial storage/rehearsal (short-term memory) processes (Eid et al.,”

• cites Eid et al., 2018bNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• AssumptionThe authors state this as an assumption: An individual’s score on an item reflects at least two distinct sources of reliable variance, one attributable to the general factor and one to the specific factor.
• AssumptionThe authors state this as an assumption: In the bifactor model, the general factor is uncorrelated with the specific factors, and the specific factors are uncorrelated with each other.

May combine two claims.

Clinically evaluated Non-ADHD and ADHD groups did not differ significantly in the proportion diagnosed with anxiety, depression, or ASD.

“The clinically-evaluated Non-ADHD cases did not differ significantly from the ADHD group in the proportion of children diagnosed with anxiety (p=.45), depression (p=.59) or ASD (p=.97); the ADHD group had a higher proportion of ODD cases as expected (p=.03).”

May combine two claims.

The computerized phonological and visuospatial working memory tasks predict hyperactivity, attention, impulsivity, ADHD diagnostic group membership, and ADHD-related functional impairments.

“These computerized phonological and visuospatial working memory tasks predict hyperactivity (Rapport et al., 2009), attention (Kofler et al., 2010), impulsivity (Raiker et al., 2012), ADHD diagnostic group membership (Tarle et al., 2017), and ADHD-related functional impairments (Friedman et al., 2017; Kofler et al., 2011, 2016).”

• cites Rapport et al., 2009Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Kofler et al., 2010Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Raiker et al., 2012 sourceNot checked — no full text of the source

May combine two claims.

ADHD is likely not associated with impairments in set shifting, response speeds, or episodic buffer processes, based in part on a subset of the current sample.

“Similarly, we have found that ADHD is likely not associated with impairments in set shifting (Irwin et al., 2019), response speeds (Kofler et al., 2016), or episodic buffer processes (Kofler, Spiegel et al., 2018) based in part on a subset of the current sample.”

• cites Irwin et al., 2019Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Kofler et al., 2016Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Kofler, Spiegel et al., 2018Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

Exploratory analyses showed that the pattern and interpretation of results did not change when different reference facet combinations were selected, suggesting robustness to this methodological choice.

“Exploratory analyses indicated that the pattern and interpretation of results was unchanged when different combinations of reference facets were selected, suggesting robustness of the findings to this methodological decision.”

May combine two claims.

An r of .71 corresponds to Cohen's d=2.03 and 81% population non-overlap, suggesting that about 81% of children with ADHD have working memory scores below the non-ADHD range.

“An r of .71 corresponds to a Cohen’s d of 2.03 and an 81% population non-overlap estimate, suggesting that approximately 81% of children with ADHD have impaired working memory abilities as defined by scores that fall below the Non-ADHD range.”

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

The proportion of non-overlap statistic estimates the share of ADHD cases outside the non-ADHD range and thus estimates the percentage of ADHD children impaired on each short-term/working memory component.

“The ‘proportion of non-overlap’ statistic estimates the proportion of ADHD cases that fall outside of the Non-ADHD range and thus provides an estimate of the percentage of ADHD children with impairments in each short-term/working memory component (Zakzanis, 2001); these estimates tend to align closely with values obtained from within-subject/heterogeneity methods for defining neurocognitive”

• cites Zakzanis, 2001 sourceNot checked — no full text of the source

Classified as probably not a substantial claim.

To properly fit the bifactor model and interpret the general factor, one or more items must load on the general factor but not on any specific factor.

“As required to properly fit the bifactor model and interpret the general factor, one or more items must load onto the general factor but not onto any specific factor (Eid et al., 2018b).”

• cites Eid et al., 2018bNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

About 30% of the ADHD sample was prescribed stimulant medication, somewhat lower than epidemiological estimates, which may have dampened association magnitudes.

“Approximately 30% of our ADHD sample was prescribed stimulant medication, which was somewhat lower than epidemiological estimates and may have dampened association magnitudes when juxtaposing short-term/working memory performance obtained off medication with parent/teacher perceptions that may be influenced by medication.”

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

Prior work suggests that the episodic buffer is likely intact in pediatric ADHD.

“However, prior work suggests that the episodic buffer is likely intact in pediatric ADHD (Alderson et al., 2015; Kofler, Spiegel et al., 2018).”

• cites Alderson et al., 2015 sourceNot checked — no full text of the source
• cites Kofler, Spiegel et al., 2018Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

Effect sizes for short-term/working memory links with ADHD symptoms may be further blunted by measuring cognitive abilities and overt behaviors in different settings, and the cross-sectional associations disallow causal attributions.

“Effect sizes for short-term/working memory links with ADHD symptoms may be further blunted by measuring cognitive abilities and overt behaviors in different settings, and these cross-sectional associations disallow causal attributions.”