Phonological abilities in literacy-impaired children: Brain potentials reveal deficient phoneme discrimination, but intact prosodic processing

Claudia Männel,Gesa Schaadt,Franziska K Illner,Elke Van Der Meer,Angela D Friederici

doi:10.1016/j.dcn.2016.11.007

Abstract

Intact phonological processing is crucial for successful literacy acquisition. While individuals with difficulties in reading and spelling (i.e., developmental dyslexia) are known to experience deficient phoneme discrimination (i.e., segmental phonology), findings concerning their prosodic processing (i.e., suprasegmental phonology) are controversial. Because there are no behavior-independent studies on the underlying neural correlates of prosodic processing in dyslexia, these controversial findings might be explained by different task demands. To provide an objective behavior-independent picture of segmental and suprasegmental phonological processing in impaired literacy acquisition, we investigated event-related brain potentials during passive listening in typically and poor-spelling German school children. For segmental phonology, we analyzed the Mismatch Negativity (MMN) during vowel length discrimination, capturing automatic auditory deviancy detection in repetitive contexts. For suprasegmental phonology, we analyzed the Closure Positive Shift (CPS) that automatically occurs in response to prosodic boundaries. Our results revealed spelling group differences for the MMN, but not for the CPS, indicating deficient segmental, but intact suprasegmental phonological processing in poor spellers. The present findings point towards a differential role of segmental and suprasegmental phonology in literacy disorders and call for interventions that invigorate impaired literacy by utilizing intact prosody in addition to training deficient phonemic awareness.

Highlights

Successful literacy crucially depends on intact phonological processing
time windows (TWs) for statistical analyses were defined with respect to the average Mismatch Negativity (MMN) peak (+/−100 ms) at 150–350 ms and the average Late Discriminative Negativity (LDN) peak (+/−200 ms) at 400–800 ms
The omnibus analysis of variance (ANOVA) with the within-subject factors phoneme, condition, region, and hemisphere and the between-subject factor spelling group revealed for both TWs condition differences depending on the phoneme contrast (TW 150–300 ms: F(1,42) = 20.05, p < 0.001; TW 400–800 ms: F(1,42) = 13.92, p < 0.001), indicating differences in the modulation of MMN and LDN depending on the experimental run

Summary

Introduction

In learning to read and write children need to be aware of the sounds of their native language, created from segmental phonemes as the smallest sound elements, as well as the prosody of suprasegmental words and phrases (Bryant and Goswami, 1987). In the reverse grapheme-to-phoneme mapping, that is reading, phonemic awareness is relevant for beginning readers, who decipher words sound by sound (Coltheart et al, 2001; Ziegler et al, 2000). The ability to successfully detect and manipulate phonemes (i.e., phonemic awareness) is essential in both spelling and reading acquisition and has been reported as predictor of literacy outcome across languages (Caravolas et al, 2013; Moll et al, 2014; Seymour et al, 2003; Ziegler et al, 2010)

Methods

Results

Conclusion