Musical rhythm perception relates to typical and atypical language development: a multi-method approach

Introduction: There is growing evidence that individual dif erences in music aptitude explain significant variance in speech-language task performance. In particular, children’s musical rhythm skills are correlated with their early literacy and spoken language (Politimou et al., 2019). Rhythm and grammar are both organized hierarchically with rule-based structure that unfolds over time. Rhythm deficits are emerging in recent studies (Ladanyi et al., in revision) as a potential feature of developmental language disorder (DLD), a prevalent communication impairment characterized by dif iculties in grammar and vocabulary. The goal of this study was to test neural and cognitive mechanisms underlying the relationship between rhythm and language ability in children across a wide range of language development. Methods: We conducted a series of experiments in children ages 5-8, in which expressive grammar, working memory, and rhythm (musical rhythm, speech rhythm, neural rhythms) were assessed. Children had non-verbal IQ in the normal range (>78) and normal hearing; language characterization was conducted under the supervision of a Speech-Language Pathologist. In Experiment 1, we examined the respective contributions of sensitivity to speech rhythm (a prosody-matching task) and auditory working memory (number recall) as mediators between musical rhythm perception and expressive grammar in typically developing (TD) children. In Experiment 2 (EEG) we tested the relationship between neural entrainment to musical rhythms and language ability in children with developmental language disorder (DLD) and TD peers. Results: Experiment 1 used a path model in TD children (N=110) to test whether speech rhythm and/or working memory mediate the association between musical rhythm and expressive grammar. Interestingly, musical rhythm had a direct ef ect on grammar (direct ef ect; β=.24), but neither speech rhythm nor working memory mediated this association (indirect ef ect; β=-.01). In Experiment 2, individual dif erences in musical rhythm processing (measured with EEG time-frequency analysis) predicted grammar (evoked beta: p=.025; evoked gamma: p =.024) in a pooled group of N=66 TD and N=16 with DLD. Moreover, within DLD, evoked gamma EEG predicted complex syntax items (p=.022), while in TD, evoked beta predicted complex syntax (p=.028), suggesting rhythm processing is linked to hierarchical language use. Discussion:These studies converged with earlier work (in smaller samples) showing musical rhythm perception skills as a predictor of individual dif erences in expressive grammar (e.g., Gordon et al., 2015) and more broadly, with a growing number of studies linking language development to musical abilities. Experiment 1 showed that most of the variance is unique to musical rhythm and not explained by sensitivity to prosodic cues or auditory working memory. Experiment 2 pointed to neural entrainment to rhythm as a potential mechanism contributing to grammatical skill in school-aged children, with EEG evoked beta and gamma as predictors of complex syntax. These findings suggest an emerging role of neural oscillatory networks linked to expressive language skills within typical and atypical development.