Network Models for Assessing the Co-occurrence Between Stuttering and ADHD.
Fjorda Kazazi, Peter Howell
PMID 42616565WHAT IT FOUND
Attention, fluency, and phonological memory connect differently in adults who stutter versus those with ADHD or no condition.
In stuttering, attention and phonological memory are not linked the same way, but fluency and phonological memory are more tightly connected. The formal test was not significant.
Key findings
01The pattern of connections among attention, speech fluency, and phonological working memory differed visually between the three groups, supporting the hypothesis that these features are organised differently in stuttering versus ADHD versus typical development.
02In neurotypical adults and in adults with ADHD, poorer attention was associated with lower phonological working memory performance (correlations of −0.34 and −0.25 respectively), but in adults who stutter this attention–phonological-memory link was essentially absent (correlation of 0.12), while the link between stuttering severity and phonological memory was stronger (correlation of −0.21).
03Formal network comparison tests found no statistically significant differences in edge weights or node centrality between any of the three groups, and the authors note this may reflect limited power with only three nodes and an unbalanced sample.
STILL TO COME
How it was doneWhat they foundWhat it means for SLPs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
The stuttering group had only 33 participants, so the study was underpowered to detect smaller differences in that group, and the authors themselves flag PWS-specific patterns as exploratory. Each network had only three nodes (attention, fluency, phonological memory), which the authors acknowledge may oversimplify the cognitive systems involved in both conditions. Co-occurring conditions such as dyslexia were not screened for, and the authors note that unscreened dyslexic traits may have influenced the phonological memory and fluency associations, particularly in the ADHD group. The phonological working memory measure (UNWR) produces a single accuracy score and does not break down errors by type (substitution, deletion, insertion), so the networks reflect overall performance rather than distinguishing storage, retrieval, or speech-motor breakdown. All participants were English-speaking adults recruited from university pools, online forums, and self-help groups; the authors do not claim generalisability to children, other languages, or culturally diverse cohorts. The ADHD group mixed clinically diagnosed, self-reported, and trait-based participants, and the authors note that socio-demographic, dialect, and print-exposure variables were not fully characterised.
Declared interests
No funding source or conflicts-of-interest declaration is provided in the text supplied. Participants were compensated for their time and travel expenses were reimbursed. The study was approved by the University College London Research Ethics Committee.
The easy way to misread this
Do not read the visually different network diagrams as proof that stuttering and ADHD have fundamentally different cognitive architectures. The formal network comparison tests did not reach statistical significance for any measure, the stuttering group had only 33 participants, and each network contained just three nodes. The authors themselves describe the PWS-specific patterns and all centrality comparisons as exploratory, and they note that the non-significant result may reflect limited power rather than true equivalence. Treat the group-specific edge patterns as a hypothesis to be tested in larger, more balanced samples, not as a basis for changing how you assess or treat a patient.