Statistical learning and reading acquisition : a comparison between first-grade students with DLD and typically developing students (Hebrew)

The present study aimed to examine the relationship between nonverbal auditory statistical learning and reading skills among first-grade children. The study compared children with typical language development to children with Developmental Language Disorder (DLD). The research was based on the premise that early reading acquisition relies on the learner’s ability to extract statistical regularities from the spoken and written linguistic input to which they are exposed. These statistical learning processes provide a cognitive foundation that may explain individual differences in reading ability, particularly among children with language difficulties.

Statistical learning was assessed using an auditory artificial grammar task that measured children’s ability to detect grammatical regularities through passive exposure to nonverbal auditory sequences (musical tones). The central hypothesis of the study was that children with higher reading proficiency would be better able to detect statistical regularities in the stimuli, even when these were nonverbal, whereas this ability would be significantly lower among children with language difficulties.

Seventy Hebrew-speaking first graders aged 6–7 participated in the study. Twenty-one of them were diagnosed with DLD by a licensed psychologist and speech-language pathologist, and attended a special education classroom designated for children with developmental language disorder. All children completed screening tests for auditory discrimination difficulties and an intelligence test (WISC), and participated in a series of assessments, including reading measures (accuracy and fluency), phonological awareness (phonemic analysis and phoneme deletion), rapid automatized naming (RAN), and auditory statistical learning.

The results showed that although both groups engaged in statistical learning, there was a significant performance gap between them. Children with DLD achieved significantly lower scores compared to children with typical language development. Significant differences were also observed in reading tasks and phonological awareness. Furthermore, significant positive correlations were found between statistical learning and reading measures, RAN, and phonological awareness. Mediation analyses indicated that phonological awareness served as a mediator between the ability to extract statistical regularities and reading accuracy and fluency.

These findings support the hypothesis that there is a significant association between statistical learning ability and early literacy skills, suggesting that implicit processing of auditory regularities may provide a cognitive basis for reading acquisition, even without direct verbal processing. Moreover, the study highlights the central role of phonological awareness as a mediating ability linking auditory statistical learning to phonological reading.

This study contributes by offering an additional explanatory account for reading acquisition difficulties among children with DLD and by emphasizing the potential for early intervention through strengthening auditory processing and directing attention to recurring patterns in sequential stimuli. Its uniqueness lies in integrating nonverbal cognitive measures with linguistic variables and in focusing on children with DLD, a population at high risk for persistent reading difficulties. These insights may inform assessment, screening, and intervention in reading, tailored to the learner’s cognitive-linguistic profile.