Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Speech rhythm sensitivity and reading skills: Investigating behavioral mediators and EEG resting-state functional connectivity
Poster Session D, Saturday, September 13, 5:00 - 6:30 pm, Field House
1 Stephanie Powell1, 1 Garrett McNeil, 1 Alexis Shumate, 2 Srishti Nayak, 2 Reyna Gordon, 1 Cyrille Magne; 1Middle Tennessee State University, 2Vanderbilt University Medical Center
Introduction: Sensitivity to speech rhythm cues, such as lexical stress patterns, is essential for effective oral communication. Additionally, growing evidence underscores its significant role in reading, showing predictive value beyond phonemic awareness in developing readers and revealing associated perceptual difficulties in individuals with dyslexia. Recent research also suggests that this connection to reading proficiency persists in older readers. However, the neural mechanisms linking speech rhythm processing to these complex reading skills remain less understood. This study explores the direct and indirect effects of speech rhythm perception on reading comprehension, considering the roles of word reading fluency and phonemic decoding skills. Moreover, we utilized resting-state EEG to examine associations between functional brain connectivity patterns and these reading abilities. Methods: Seventy-three native English speakers (ages 18-50) underwent assessments of word reading, reading comprehension, and lexical stress perception. An individual dyslexia risk index was derived from standardized performances in word reading and pseudoword decoding (higher scores = lower risk). Resting-state EEG (5 minutes, eyes closed) was also collected. We estimated functional connectivity between pairs of electrodes using the debiased weighted Phase Lag Index (dwPLI) in the delta (0.5-4 Hz), theta (4-8 Hz), alpha (8-12 Hz), and beta (12-30 Hz) frequency bands. Minimum Spanning Trees (MSTs) were constructed from the dwPLI values to analyze network topology. Results: Lexical stress perception significantly correlated with reading comprehension (r = 0.24*) and dyslexia risk (r = 0.56***). A GLM mediation analysis revealed that the relationship between lexical stress perception and reading comprehension was mediated by dyslexia risk (indirect effect: B = 0.75, SE = 0.19, 95% CI [0.44, 1.13]). Functional connectivity analyses showed that reading comprehension was associated with brain networks characterized by greater global integration and organization, particularly in the theta frequency band. Higher comprehension scores were linked to lower values of theta kappa (r = -0.25*) and theta mean dWPLI (r = -0.26*), suggesting reduced network randomness and a lesser dependence on central hub nodes. Conversely, comprehension was positively associated with theta mean betweenness centrality (r = 0.24*) and network eccentricity (r = 0.24*), consistent with more efficient information flow and broader network integration. Both dyslexia risk and lexical stress perception showed positive correlations with tree hierarchy in the alpha band (r = 0.27* and r = 0.24*, respectively), indicating a more segregated and organized network linked to better stress sensitivity and lower dyslexia risk. GLM mediation analysis also suggested that the relationship between alpha tree hierarchy and dyslexia risk was indirect and mediated by lexical stress perception (B = 25.42, SE = 11.5, 95% CI [4.16, 44.38]). Conclusion: Lexical stress sensitivity indirectly impacts reading comprehension skills by influencing the risk of dyslexia. Adequate reading comprehension is linked to the efficient integration of theta network activity, while sensitivity to stress and a reduced risk for dyslexia are associated with a well-organized alpha network topology. These findings highlight the differential contributions of distinct frequency-specific networks to the reading process.
Topic Areas: Prosody, Reading