A Comprehensive Study of Neural Entrainment in Developmental Language Disorder and Reading Disability
Abstract
Prior research has suggested that reading disability (RD, dyslexia) and developmental language disorder (DLD) stem from deficits in rhythmic auditory processing, specifically in synchronizing neural oscillations (Cumming et al., 2015; Goswami, 2011). Speech relies on rhythmic patterns for signaling linguistic information at multiple timescales (e.g., phonemes, syllables; Giraud & Poeppel, 2012). The disruption of regular neural entrainment is hypothesized to lead to difficulties in processing fast acoustic changes in speech, negatively affecting phonological processing, and speech segmentation. In this dissertation, I studied neural entrainment to uncover possible areas of impairment related to speech tracking, which could help inform interventions. In Chapter 2, I present a systematic review summarizing the state of research on neural entrainment in children and adults with DLD or RD. The strongest evidence of impairment was observed in RD, demonstrating atypical neural entrainment to prosody, syllables, and phonemes compared to typically developing (TD) individuals. In contrast, only two studies were conducted on children with DLD, demonstrating the need for more research in this area. Chapters 3 and 4 address some of the gaps highlighted in Chapter 2 by investigating neural entrainment in children with a broad range of language and reading abilities, including children with RD and DLD. I used electroencephalography (EEG) data from the Healthy Brain Network, which allowed me to investigate neural entrainment in many children. In Chapter 3, I measured cerebro-acoustic phase coherence and only found neural entrainment for a small band (3.25-5 Hz), demonstrating syllabic entrainment. However, there were no differences in syllabic entrainment based on language or reading abilities nor clinical diagnoses of RD and DLD. Since coherence was not powerful enough to detect neural entrainment outside of a small band, in Chapter 4, I used encoding models which have shown neural entrainment with shorter amounts of neuroimaging data (Mesik & Wojtczak, 2023). The findings demonstrate robust neural entrainment to acoustic and phonetic information as well as clear differences between RD and DLD compared to TD. Our results provide valuable insights into how neural entrainment to speech is modulated based on the presence of language or reading impairments.