Auditory Working Memory Mediates the Relationship Between Musical Sophistication and Speech-In-Noise Perception.
Hasan Colak, Ester Benzaquén, Xiaoxuan Guo and 3 others
Among 203 adults, musical sophistication was weakly linked to speech-in-noise performance, and the association was fully explained by auditory working memory.
This is a cross-sectional correlation — it does not show that music training or AWM exercises will improve a patient's speech-in-noise ability.
Key findings
1Musical sophistication was significantly but weakly associated with speech-in-noise performance (ρ = 0.26, corrected p < 0.001). Auditory working memory fully mediated this relationship: the indirect effect was 0.37 (p < 0.001), while the direct effect of musicality on SPIN became non-significant once AWM was in the model (β = 0.11, p = 0.38).
2Verbal working memory did not mediate the musicality–SPIN link (indirect effect 0.009, p = 0.74), whereas non-verbal intelligence partially mediated it (indirect effect 0.16, p < 0.001; the direct effect of musicality remained significant at β = 0.32, p = 0.007).
3In two structural equation models that controlled for non-verbal intelligence, hearing thresholds, and age, the indirect path from musicality through AWM to SPIN remained significant (β = 0.15, p = 0.014 in the first model; β = 0.16, p = 0.008 in the second), while the direct path was non-significant in both.
to read the rest of this summary — three full summaries a month, no card.
Already have an account?
What it does not show
The study is cross-sectional: all measures were taken at one time point, so the mediation models show association, not causation. The authors explicitly state that longitudinal or intervention studies would be needed to test causal direction. Musical sophistication was measured with a self-report questionnaire (Gold-MSI), not with behavioural tests of musical perception, so it may not reflect actual musical ability. The AWM task was not directly verified in this dataset to be free of basic auditory discrimination contributions, so the 'memory' component may be partly confounded with perceptual acuity. The overall effect size is small (ρ = 0.26), meaning musical sophistication explains only a modest fraction of the variance in SPIN performance. The sample, while large, is a single-site convenience sample of native English speakers with no cognitive or neurological disorder, which limits generalisability to clinical populations with hearing loss, neurological conditions, or non-English-speaking backgrounds.
Declared interests
Funded by the UK Medical Research Council (grant 815 MR/T032553/1). The authors declared no conflicts of interest.
Summarised by AI from the full paper, without a clinician reviewing it. Check it against the source before it changes what you do. Read it on PubMed →