R9990/2026-03-31/C001/SRC06/E01¶
Peer-reviewed evidence of large-magnitude working memory deficits in ADHD.
URL: https://pmc.ncbi.nlm.nih.gov/articles/PMC7483636/
Extract¶
Kofler et al. (2020) studied 172 children ages 8-13 (81 with ADHD, 91 controls) using bifactor modeling:
Central executive working memory (domain-general): - Very large magnitude impairments: d = 1.62-2.03 - Affects approximately 75%-81% of children with ADHD - Covaries with both inattentive and hyperactive/impulsive symptom severity
Visuospatial short-term memory: - Smaller unique impairment: d = 0.60 - Present in ~38% of ADHD cases
Phonological short-term memory: - No evidence of unique deficit (d = 0.28, nonsignificant)
Key finding: "Working memory deficits appear to be present in upwards of 3 out of every 4 ADHD cases and covary with their ADHD symptom severity based on both parent and teacher report."
Tasks used: Computerized serial reordering tasks requiring mental manipulation of sequentially presented stimuli — children with ADHD showed impairments across all eight performance variables (d = 0.87-1.65; all p < .001).
Relevance to Hypotheses¶
| Hypothesis | Relationship | Strength |
|---|---|---|
| H1 | Supports | Large-magnitude central executive working memory deficits (d = 1.62-2.03) directly undermine the cognitive demands of STAR (sequential recall, temporal ordering, mental manipulation of past events) |
| H2 | Supports | The specificity of deficits (central executive, not phonological storage) suggests the problem is in manipulating and organizing information, not in storing it — precisely what STAR demands |
| H3 | Contradicts | Effect sizes of d = 1.62-2.03 are very large, making it implausible that STAR's cognitive demands would not differentially impact ADHD individuals |
Context¶
This is a pediatric study (ages 8-13). Adult ADHD working memory deficits are also well-documented but may differ in magnitude. The findings establish the neuropsychological mechanism by which STAR would disadvantage ADHD candidates, even though the study does not examine interviews directly.
Notes¶
The effect sizes are notably large. For context, d = 0.8 is conventionally considered "large" in psychology. Central executive deficits at d = 1.62-2.03 represent a profound impairment in exactly the cognitive domain STAR interviews demand.