# The Effects of Maturation and Dyslexia Risk on Neural Speech‐Sound Encoding and Discrimination at Preschool Stage

## TL;DR

This study examines how the brain processes speech sounds in preschool children, comparing typical development and those at risk for dyslexia.

## Contribution

The study provides a detailed characterization of neural speech-sound processing maturation and dyslexia risk effects during the preschool stage.

## Key findings

- Typical 4–5-year-olds show P1–N2 and MMN–LDN components in speech processing.
- Children at risk for dyslexia exhibit reduced MMN amplitudes, indicating impaired speech discrimination.
- Age-related changes in ERP and MMR components are similar in both typical and at-risk children.

## Abstract

Event‐related potentials (ERPs), and particularly change‐elicited mismatch responses (MMRs), are valuable tools for assessing early speech processing and promising markers of dyslexia risk. Yet, their maturation during the preschool stage remains poorly characterized. We determined the typical elicitation of ERPs and MMRs to speech sounds at preschool age (4–5 years), their maturation from early childhood (28 months) to preschool age, and the impact of dyslexia risk on them. To this end, we recorded obligatory ERPs to a repeating pseudoword and MMRs to five deviances at 4–5 years and compared them with a previously reported follow‐up at 28 months, in subgroups with versus without dyslexia risk in a large sample (n ~ 150). In the full sample, including both control and at‐risk children, the 4‐ to 5‐year obligatory ERPs showed a P1–N2 pattern, while the MMRs included a mismatch negativity (MMN) followed by a late discriminative negativity (LDN). From 28 months to 4–5 years, P1 amplitude increased and latency decreased, whereas the N2 amplitude increased. MMN and LDN amplitudes increased and LDN latency decreased with age, whereas a positive MMR reported at 28 months was no longer evident at 4–5 years. Crucially, at‐risk children exhibited reduced MMN amplitudes at 4–5 years across deviants, suggesting deficient speech discrimination. Changes from 28 months to 4–5 years were similar in both subgroups. These findings establish a solid description of typical/atypical neural speech‐sound processing during preschool years, serving as a reference for future studies including interventions or clinical groups.

Event‐related potentials (ERPs) to a repeating pseudoword and mismatch responses (MMRs) to five deviances were recorded at 4–5 years and compared with a previously reported follow‐up at 28 months, in subgroups with versus without familial dyslexia risk (n ~ 150). P1–N2 and MMN–LDN components increased in amplitude and decreased in latency with age, whereas the P‐MMR was no longer evident at 4–5 years. At‐risk children showed reduced MMN amplitudes at 4–5 years, suggesting deficient speech discrimination in dyslexia risk.