Deficits in neural encoding of speech in preterm infants

Abstract

Preterm children show developmental cognitive and language deficits that can be subtle and sometimes undetectable until later in life. Studies of brain development in children who are born preterm have largely focused on vascular and gross anatomical characteristics rather than pathophysiological processes that may contribute to these developmental deficits. Neural encoding of speech as reflected in EEG recordings is predictive of future language development and could provide insights into those pathophysiological processes. We recorded EEG from 45 preterm (≤ 34 weeks of gestation) and 45 term (≥ 38 weeks) Chinese-learning infants 0-12 months of (corrected) age during natural sleep. Each child listened to three speech stimuli that differed in lexically meaningful pitch (2 native and 1 non-native speech categories). EEG measures associated with synchronization and gross power of the frequency following response (FFR) were examined. ANCOVAs revealed no main effect of stimulus nativeness but main effects of age, consistent with earlier studies. A main effect of prematurity also emerged, with synchronization measures showing stronger group differences than power. By detailing differences in FFR measures related to synchronization and power, this study brings us closer to identifying the pathophysiological pathway to often subtle language problems experienced by preterm children.

Keywords: FFR; Myelination; Preterm infants; Speech; Synaptogenesis.

Fig. 1

The distributions (corrected) age (A) and maternal education (B) in the preterm and term groups. No significant group difference was detected in either measure, as indicated by the dash between the violins.

Fig. 2

Two types of FFR measures: Synchronization and Power. Raw EEG signals (the upper panel) are recorded from the listener’s central nervous system, re-referenced, filtered, and aligned with the onsets of a particular external auditory stimulus (indicated by the inverted triangles with dashed lines) to produce the epoched EEG (left lower panel). EEG trials (epochs) are shown as rows at the epoched EEG panel. The ellipses frame Raw and Epoched EEG waveforms to indicate that they show a subset of a larger data set. Epoched EEG was further transformed into time-domain (green) and frequency-domain (orange) representations that included averaged FFR waveforms and ITPC periodograms, FFR Power Spectrum, Root-mean square (RMS) and Autocorrelation. These representations can be mapped onto Synchronization (purple) and Power (blue) FFR measures. See Discussion for a detailed explanation of the mapping. As stated in Results, our a priori classification converged with the statistical clustering results (see Fig. 4).

Fig. 3

The grand-averaged FFR responses of the preterm (D–F) and term (A–C) infants to the three speech stimuli: /ga2/ (A, D), /ga3/ (B, E), and /ga4/ (C, F).

Fig. 4

Clustering of the first two principal component coefficients of the 11 FFR measures. (A) The PC coefficients for each measure are plotted as empty circles connected with the lines to the centroid of their cluster that is plotted as a filled circle. Abbreviations: SN – FFR SNR, NR – Noise RMS, PL – Power LOW, PM – Power MID, mI – maxITPC, RC – Response Consistency, PS – Pitch Strength, IL – ITPC LOW, IM – ITPC MID, dL – Low/High Power Ratio, dM – Mid/High Power Ratio. AU – arbitrary units. (B) Main effects in ANCOVA analysis, representing an effect of one factor on the response from changing the factor value while averaging out the effects of the other factors (MathWorks, 2012). The factors are listed along the y-axes. The bars indicate the confidence intervals. The significant factors (i.e., those whose confidence intervals do not include zero) are marked with asterisks. (C) Synchronization and (D) Power findings of the two experimental groups controlled for Age, Tone, and Maternal Education. Non-parametric Wilcoxon rank-sum tests demonstrated that Synchronization (Z = − 1.88, P =0.030) but not Power (Z = − 0.97, P =0.17) was impaired by prematurity.