Phonological working memory involves temporarily storing and manipulating speech-based information. It underlies critical abilities like following instructions and sound-to-meaning mapping in reading.
Deficits in phonological working memory exhibited by children with ADHD are believed to contribute to their challenges regulating emotion, shifting between tasks, taking perspectives, and solving multi-step problems.
Roberts, D. K., Alderson, R. M., & Bullard, C. C. (2023). Phonological working memory in children with and without ADHD: A systematic evaluation of recall errors. Neuropsychology. https://doi.org/10.1037/neu0000899
Key Points
- Children with ADHD exhibited significantly more total recall errors, omission errors, and transposition errors compared to typically developing children, suggesting impairments in phonological storage/rehearsal processes and manipulation/reordering aspects of working memory.
- The increase in total recall, omission, and transposition errors across increasing set sizes was disproportionately larger for children with ADHD compared to typically developing children.
- Intrusion errors were significantly higher in the ADHD group, but there were no between-group differences in specific subtypes of intrusion errors like phonological similarity errors or perseverative errors. This suggests central executive deficits in ADHD may manifest more prominently as difficulties with accurately manipulating and reordering information rather than deficient interference control.
- Findings elucidate specific working memory sub-processes impaired in ADHD to inform development of improved diagnostic metrics and interventions targeting manipulation/reordering rather than storage/rehearsal.
- Sample lacked diversity and included few girls, limiting generalizability. Cross-sectional design precludes analysis of developmental changes.
Rationale
Past research has reliably demonstrated phonological working memory deficits in children with ADHD, with meta-analyses showing moderate-to-large magnitude between-group effects (Kasper et al., 2012; Martinussen et al., 2005).
Performance accuracy serves as the predominant metric, simply quantifying the percentage of correct trials. This approach reliably indexes general working memory ability but lacks specificity regarding underlying neurocognitive processes contributing to impairment (Roberts et al., 2023).
Examining specific error types – omissions, intrusions, transpositions – can clarify which aspects of working memory are disproportionately impaired in ADHD. For example, intrusion errors implicate poor interference control, whereas transposition errors reflect difficulties manipulating information held in mind (McCormack et al., 2000).
Just one prior study has partially parsed working memory subcomponents in ADHD by examining deficits in reordering, updating, and dual-processing (Fosco et al., 2020). Their scoring procedure still lacked nuance by using an average score that did not capture error types contributing to partially incorrect responses.
Comprehensively analyzing multiple error types within the same phonological working memory task is the logical next step toward elucidating specific mechanisms underlying these deficits in ADHD.
Findings can guide development of better diagnostic indicators and interventions targeting the most impaired processes.
Method
Participants who were children completed a computerized phonological working memory task. The task involved remembering and manipulating a series of aurally presented numbers and letters across blocks with increasing set sizes.
Incorrect verbal recall responses were coded for errors of omission, intrusion, or transposition:
- Omission errors occur when a child fails to recall an item that was presented (e.g., leaving out one of the numbers or letters)
- Intrusion errors refer to incorrectly adding an item that was not originally presented (e.g., inserting an extra incorrect number or letter)
- Transposition errors involve recalling stimuli in the wrong order from how they were presented (e.g., reversing the order of numbers or stating the letter out of sequence)
By categorizing errors into these subtypes, the researchers aimed to provide greater specificity regarding aspects of phonological working memory that are disproportionately impaired in ADHD, such as difficulties manipulating order information versus storage deficits.
Between-group differences in total errors and specific error types were analyzed using generalized estimating equations controlling for age.
Recall errors were classified as omissions, intrusions, or transpositions according to definitions established in working memory literature (McCormack et al., 2000).
Sample
The sample included 54 children diagnosed with ADHD and 65 typically developing children between ages 8-12 years. Most ADHD participants were male (83%) and White (76%).
Many had comorbid diagnoses like oppositional defiant disorder and specific learning disorders. The TD group was matched on age, sex ratio, and ethnicity.
Statistical measures
Generalized estimating equation models were used to analyze count data of recall errors, controlling for age. Negative binomial regressions specified autoregressive correlation structure to account for repeated measures over set sizes.
Between- and within-group effects were tested for significance after Bonferroni correction with adjusted alpha of .003. Cohen’s d measured effect sizes for group differences.
Results
- The ADHD group exhibited significantly more total recall errors, with a moderate-large overall effect size (d = 0.72). Total errors increased with set size (3 < 4 < 5 < 6 stimuli) more steeply in the ADHD group (significant interaction effect). Post-hoc analyses showed more total recall errors in ADHD group for all set sizes except 5.
- Significant between-group differences indicated more omission and transposition errors in ADHD group. Omission errors (total, letters, numbers) and transposition errors (total, numerical ordering, replications) also interacted with set size, increasing disproportionately more across set sizes for the ADHD group (significant interactions).
- The ADHD group showed more intrusion errors overall and specifically number intrusions but did not differ in letter intrusion subtypes like phonological similarities or perseverations. Intrusion errors increased with set size similarly across groups (non-significant interactions).
- There were no group differences at Set Size 5 specifically for total recall, omission, and number omission errors, while differences reemerged at the largest Set Size 6.
Insight
This thorough error analysis provides compelling evidence that working memory deficits in ADHD stem from inadequate central executive processes—specifically difficulties manipulating and reordering multiple pieces of information.
As task demands increased, reflected by larger set sizes, the ADHD group showed greater increases in omission and transposition errors. This clearly demonstrates an impaired ability to juggle information in mind while continuously updating responses. Accurately reordering stimuli and avoiding omitted details relies heavily on executive control functions (Baddeley, 2007).
Intriguingly, manipulating several numbers overtaxes central executive resources in ADHD, yet capacity remains less encumbered when simply rehearsing a single letter.
In contrast, the overall intact rehearsal and interference control evidenced by specific intrusion subtypes implies verbal storage and inhibition are not markedly impaired. Rather than an inhibition deficit per se, intrusions could arise from executive resources depleted by inefficient manipulation.
In essence, reordering taxes working memory most heavily in ADHD. Interventions may, therefore, achieve greater functional improvements by targeting manipulation versus basic storage/rehearsal.
Introducing set sizes demanding substantial reordering could also help expose executive deficits for diagnostic and progress monitoring purposes beyond conventional span tasks.
Strengths
- Comprehensively investigated multiple recall error types (omissions, intrusions, transpositions) within the same working memory task, elucidating specific mechanisms underlying phonological working memory deficits
- Included clinical ADHD group plus matched sample of typically developing children
- Used advanced statistical methods appropriate for count outcome variables and repeated measures design
- Findings converge with and expand recent work parsing working memory subcomponents through other paradigms (Fosco et al., 2020)
- Sheds light on predominantly impaired central executive processes of manipulating and reordering information in mind versus intact short-term storage capacity
- Informs intervention targets and progress monitoring metrics emphasizing executive control over storage/rehearsal
Limitations
- Primarily male, White sample limits generalizability
- Modest sample size prevented examining effects of ADHD subtypes or common comorbid diagnoses
- Cross-sectional data cannot address developmental changes
- Many dependent variables inflated Type I error risk despite statistical corrections
- Difficult to code certain intrusion subtypes like perseverative number errors
- Some unexpected null findings at Set Size 5 require replication
- Children discontinued medication for testing but residual effects possible
- Parent ADHD ratings differed across scales (CBCL vs. Conners)
- Working memory task not optimized to parse specific error types
Implications
These findings carry important clinical implications by sharpening the understanding of how working memory falters in youth with ADHD. Children do not appear to struggle retaining information itself generally.
Rather, deficits emerge distinctly when organizing thoughts and appropriately filtering irrelevant details. This executive dysfunction, especially with manipulating information, can profoundly impact a child’s ability to regulate emotions, transition flexibly, take perspectives, and solve multi-step problems.
When such mental flexibility lags, children often cannot achieve their potential academically nor develop strong social skills.
Intervention should emphasize executive skills like reordering and updating knowledge instead of rote rehearsal, which may neglect the most critical weakness. Stressful real-world situations likely overwhelm rehearsal capability secondarily by exhausting executive resources initially.
Prospective working memory training tools can also introduce set sizes with substantial reordering/updating demands, unlike traditional span tasks, to better expose and target manipulation difficulties.
Furthermore, given mounting evidence tying ADHD pathology to the central executive specifically, scientific efforts should continue confirming whether executive function deficits represent a core impaired process versus simply downstream effects.
If future work establishes central executive dysfunction as a cardinal, rather than ancillary ADHD feature, the diagnostic significance would profoundly shift clinical conceptualization of this highly prevalent, often debilitating childhood disorder.
References
Primary reference
Roberts, D. K., Alderson, R. M., & Bullard, C. C. (2023). Phonological working memory in children with and without ADHD: A systematic evaluation of recall errors. Neuropsychology. https://doi.org/10.1037/neu0000899
Other references
Baddeley, A. (2007). Working memory, thought, and action (Vol. 45). OUP Oxford.
Fosco, W. D., Kofler, M. J., Groves, N. B., Chan, E. S., & Raiker, J. S. (2020). Which ‘working’ components of working memory aren’t working in youth with ADHD? Journal of Abnormal Child Psychology, 48(5), 647–660. https://doi.org/10.1007/s10802-020-00621-y
Kasper, L. J., Alderson, R. M., & Hudec, K. L. (2012). Moderators of working memory deficits in children with attention-deficit/hyperactivity disorder (ADHD): A meta-analytic review. Clinical Psychology Review, 32(7), 605–617. https://doi.org/10.1016/j.cpr.2012.07.001
Martinussen, R., Hayden, J., Hogg-Johnson, S., & Tannock, R. (2005). A meta-analysis of working memory impairments in children with attention-deficit/hyperactivity disorder. Journal of the American Academy of Child & Adolescent Psychiatry, 44(4), 377–384. https://doi.org/10.1097/01.chi.0000153228.72591.73
McCormack, T., Brown, G. D., Vousden, J. I., & Henson, R. N. (2000). Children’s serial recall errors: Implications for theories of short-term memory development. Journal of Experimental Child Psychology, 76(3), 222–252. https://doi.org/10.1006/jecp.1999.2550
Keep Learning
Here are some thought-provoking discussion questions for a class about this research:
- This study found children with ADHD show distinct difficulties manipulating and reordering multiple pieces of information in working memory, more so than simply storing information. Do you think strengthening these executive skills should become a frontline psychosocial treatment target for ADHD? Why or why not? What kinds of brain training games or strategies could help improve manipulation abilities?
- How might deficient phonological working memory processes impact children with ADHD across settings like school and home? Can you think of any examples where forgetting instructions or struggling to organize ideas might cause problems for kids with ADHD?
- If future research continues finding central executive impairments are a core feature of ADHD, how might you alter the way you clinically conceptualize or assess the disorder compared to emphasizing attention and hyperactivity behaviors?
- Why might between-group differences disappear at the intermediate Set Size 5 across several error types? What alternative explanations can account for this finding besides the authors’ speculation that demands exceeded capacity regardless of diagnosis?
- Since children discontinued ADHD medication for testing, what type of residual treatment effects might remain even 24 hours later? Could these potentially mask group differences in some areas like interference control?
- How could researchers optimize a working memory paradigm to elicit certain errors like perseverations while minimizing other confounding errors? What key task design changes might you implement?
